TWI798619B - Die bonding device and method for manufacturing semiconductor device - Google Patents

Abstract

[Problem] Provide a technology that can improve the uniformity of imaging conditions of multiple imaging objects. [Solution] The die bonding apparatus includes: a transport path for transporting the substrate; a plurality of imaging devices fixedly arranged above the transport path in a row along the width direction of the substrate; and a control unit for It is configured to capture multiple images of a row of multiple bonded regions along the width direction on the substrate by a plurality of imaging devices, generate a composite image from the obtained multiple images, and recognize the imaged object in the bonded region based on the composite image. The imaging field of view of each imaging device overlaps on the substrate, and the overlapping imaging field of view is configured to be larger than the bonding area.

Description

Die bonding device and method for manufacturing semiconductor device

The present disclosure relates to a die bonding device, for example, a die bonder that can be applied to a plurality of camera imaging bonding regions.

Part of the manufacturing process of semiconductor devices includes the process of mounting a semiconductor chip (hereinafter simply referred to as a die) on a wiring board or lead frame (hereinafter simply referred to as a substrate) and assembling a package, a part of the process of assembling a package It includes a process (dicing process) of dividing a die from a semiconductor wafer (hereinafter simply referred to as a wafer), and a bonding process of placing the divided die on a substrate. The semiconductor manufacturing equipment used in the bonding process is a die bonding device such as a die bonder.

A die bonder is a device that uses resin paste, solder, and gold plating as bonding materials to bond (place and bond) die on a substrate or bonded die. For example, in a die bonder that bonds a die to the surface of a substrate, it is repeatedly picked up by suctioning a die from a wafer using a suction nozzle called a chuck attached to the front end of the bonding head, and The action (operation) of bonding by placing on a specific position on the substrate, applying a pressing force, and heating the bonding material at the same time.

When using a resin as a bonding material, a resin paste such as Ag epoxy or acrylic is used as an adhesive (hereinafter referred to as a paste adhesive). The paste-like adhesive for bonding the die to the substrate is sealed in the syringe, and the syringe moves up and down relative to the substrate to eject and apply the paste-like adhesive. That is, by applying a specific amount of paste-like adhesive at a specific position with a syringe filled with paste-like adhesive, the die is pressed, baked, and bonded to its paste-like adhesive . Install a recognition camera near the syringe, and use the recognition camera to confirm the position of the paste-like adhesive for positioning. Furthermore, it is confirmed whether the applied paste-like adhesive is in a specific shape and only A specific amount is applied at a specific location.

Furthermore, an identification camera is installed near the bonding head, and the identification camera is used to confirm the position of the substrate on which the die is bonded for positioning, and furthermore, it is confirmed whether the bonded die is bonded at a specific position. [Prior Art Literature] [Patent Document]

[Patent Document 1] Japanese Unexamined Patent Publication No. 2013-197277

[Problem to be Solved by the Invention]

When a single imaging device using a non-telecentric lens is used to image a plurality of imaging objects, the imaging objects located far away from the imaging device directly below are imaged obliquely, and are viewed as side surfaces of a three-dimensional shape. The subject of this disclosure is to provide a technology capable of improving the uniformity of imaging conditions of a plurality of imaging objects. Other problems and novel features are evident from the description in this manual and the attached drawings. [Means to solve the problem]

The summary of the representative content in this disclosure is briefly described as follows. That is, the die bonding apparatus includes: a transport path for transporting the substrate; a plurality of imaging devices fixedly arranged above the transport path in a row along the width direction of the substrate; and a control unit, It is configured to use the above-mentioned multiple imaging device to capture multiple images of a series of multiple bonding regions located on the above-mentioned substrate along the above-mentioned width direction, generate a composite image based on the obtained multiple images, and recognize the above-mentioned For the object to be imaged in the bonding area, the imaging field of view of each imaging device overlaps on the substrate, and the overlapping imaging field of view is configured to be larger than the bonding area. [Effect of Invention]

By using the above-mentioned die bonding device, the uniformity of imaging conditions of a plurality of imaging objects can be improved.

Hereinafter, embodiments, modifications, and examples will be described using drawings. However, in the following description, the same reference numerals are given to the same components, and overlapping descriptions may be omitted. In addition, in order to clarify the description, although the width, thickness, shape, etc. of each part may be schematically shown in the drawing compared with the actual state, this is only an example and is not intended to limit the interpretation of the present invention.

First, using FIGS. 1 to 3 , the technique studied by the present inventors will be described. FIG. 1( a ) is a perspective view showing a normal field of view optical system, and FIG. 1( b ) is a perspective view showing a wide field of view optical system.

In the sequential operations of positioning, bonding (die bonding or paste coating), and inspection of the substrate S, as shown in FIG. , since only an area slightly larger than one joining area AA is captured, imaging and image recognition of only the number of joining areas AA are required. Here, as an example, an example in which the substrate S has six bonding regions AA in one row and five bonding regions in five rows is shown. A low-resolution imaging device is an imaging device that can capture images with sufficient resolution only in an area slightly larger than a joint area, such as an imaging device with a resolution of about 300,000 to about 1.3 million pixels.

Therefore, as shown in FIG. 1(a), it is necessary to repeatedly move the imaging device CML in the width direction (Y-axis direction) of the substrate S to perform imaging and image recognition of other bonding regions in the row, and The action of performing processing based on its image recognition. In addition, after the imaging of one row, the imaging of the next row is performed by moving the board|substrate S in a board|substrate conveyance direction (X-axis direction).

Therefore, a movement mechanism for moving a supporting member (not shown) of the imaging device CML in the width direction of the substrate S is required, and the supporting mechanism of the imaging device CML becomes complicated and large in size and expensive. Furthermore, because it takes time to move in the width direction of the imaging device CML, and the waiting time for imaging until the shaking of the imaging device CML caused by the vibration generated by the movement of the supporting member subsides, it is impossible to bond the die. High speed. Or, in order to prevent vibration, a vibration prevention mechanism is required, and die bonding becomes expensive. Furthermore, if there is a mechanism part that reciprocates above the bonding area and below the imaging device, it is necessary to consider the timing when the mechanism part does not block the imaging field of view. By securing this imaging timing, it is impossible to increase the speed of die bonding.

On the other hand, this is not repeated every time the bonding after the substrate is transported. As shown in FIG. The above-mentioned problems can be solved by improving the efficiency of processing time by imaging and identifying at least one bonding area in the width direction of the substrate. Here, a high-resolution imaging device refers to an imaging device capable of capturing at least all areas of the substrate S divided by one row in the width direction at one time with a sufficient resolution, for example, an imaging device with a resolution of about 5 million pixels or more. One column includes a plurality of joint areas, for example, four joint areas.

However, the wide-field optical system (wide-field camera) has the following problems. This problem will be described using FIG. 2 and FIG. 3 . Fig. 2(a) is a conceptual diagram of a wide-field optical system using a microlens, and Fig. 2(b) is a conceptual diagram of a wide-field optical system using a telecentric lens. FIG. 3( a ) is a diagram showing the image of the three-dimensional shape of the paste, and FIG. 3( b ) is a diagram showing spectral lines on the paste when the substrate is viewed over a wide area with a microlens.

(1) Guarantee of telecentricity Even if the magnification of the microlens is simply reduced, the wide area can be obtained, but the more it deviates from the center of the field of view, the more it is viewed from obliquely above, as shown in FIG. In , the side surface is not visible, but the left side of the imaging object OBL is visible, and the left side of the right imaging object OBR is visible, which is seen as a side surface of a three-dimensional shape. Furthermore, the magnification varies with the height of the imaging object OBL, and when the height changes, it is difficult to perform such alignment. For example, the paste coated on the substrate or the die stacked on the substrate have different heights due to different positions.

These problems can be solved by using telecentric lenses. Since the telecentric lens collects the parallel light, the sides cannot be seen in all the objects to be photographed. However, in this case, as shown in Figure 2(b), generally a lens with a diameter larger than the desired field of view is required. For example, to ensure a field of view of, for example, a 100mm square, a diameter of 141mm or more at the diagonal is required. Accordingly, the focal point The distance becomes longer. Mounting such a large lens on a die bonding device is not ideal from the viewpoint of efficiency.

(2) Irradiation of uniform illumination in the field of view When using a non-telecentric lens to secure a wide area, the one at the end of the field of view is viewed from obliquely above. Since the directions are opposite at the right and left ends of the field of view, or at the upper and lower ends, the irradiation direction of the illumination tends to be uneven. Furthermore, even in the case of irradiating parallel light using telecentric light, it is difficult to obtain a uniformly irradiated image within the imaging field of view. For example, as shown in FIG. 3( a ), the center part of the paste-like adhesive PA system coated on the substrate in a cross shape is higher, and the front part is formed lower. When viewing the substrate coated with paste-like adhesive as shown in Figure 3(a) with a microlens in a wide area, as shown in Figure 3(b), because the center of the imaging field of view is viewed from directly above The part is brighter, and the peripheral part is darker because it is viewed obliquely. Furthermore, because the height of the paste-like adhesive is not uniform, the spectral line on the paste-like adhesive PA moves. Here, the center of field of view of the imaging device is located at the center of the image in FIG. 3( b ). The board|substrate S has 6 tabs TB as a bonding area in a row, and shows 7 rows. On the four rows of tabs TB on the right side of the substrate S, the paste-like adhesive PA is applied.

Next, an embodiment for solving the above-mentioned problems will be described using FIGS. 4 to 5 . Fig. 4(a) is a top view of the multiplexing of the imaging device of the embodiment, and Fig. 4(b) is a side view when viewed from the direction of arrow A in Fig. 4(a). Fig. 5 is a diagram for explaining the lighting device used in the imaging device of the embodiment, Fig. 5(a) is a sectional view of the lighting device, and Fig. 5(b) is a side view.

As an embodiment to solve the above-mentioned problems, multiplex (three-dimensional) imaging devices are performed. For example, as shown in FIG. 4 , above the substrate S, a plurality of imaging devices CM1 to CM4 are arranged in a row in the width direction of the substrate S (Y-axis direction) and fixedly arranged. Here, each of the imaging devices CM1 to CM4 is an imaging device having a resolution equal to or greater than that of the imaging device CML, and cannot capture an entire area divided into one column in the width direction of the substrate S at a time with a sufficient resolution. Furthermore, each of the imaging devices CM1 to CM4 may use a telecentric lens, but it is preferable to use a non-telecentric lens such as a microlens. The imaging devices CM1 - CM4 are at the same height at a certain interval in the horizontal direction, and the optical axes of the imaging devices CM1 - CM4 are parallel to each other and perpendicular to the substrate S. As shown in FIG. In addition, even if the optical axes of the imaging devices CM1 to CM4 are within the allowable out-of-focus range in the imaging field of view, they may be slightly inclined from the vertical line to the substrate S.

In the present embodiment, each of the imaging devices CM1 to CM4 images the imaging objects of the joining regions AA11 to AA14. Furthermore, the imaging fields of view IA1 to IA4 of each of the imaging devices CM1 to CM4 do not overlap. Furthermore, by moving the board|substrate S in a conveyance direction (Y-axis direction), the imaging target object of the remaining row is imaged sequentially. By multiplying the imaging device, imaging can be performed slightly directly above the imaging object, and the uniformity of illumination in the imaging field of view can be improved for inspection. Furthermore, by multiplying the camera device, the same processing efficiency as that of the wide-field optical system can be obtained without moving the camera device.

Here, the substrate S is rectangular and flat as shown in (a) (b), and has many bonding regions AA11~AA14, AA21~AA24, ･･･in vertical and horizontal directions. The substrate S is configured such that the length in the conveyance direction (X-axis direction) is longer than the length in the width direction (Y-axis direction).

It is better for multiple camera devices to have the same lighting system. For example, as shown in FIG. 5( a ), between the imaging devices CM1 to CM4 and the substrate S, an illuminating device LD including a surface-emitting illumination (light source) SL and a half mirror (half mirror) HM is disposed therein. The irradiation light from the surface emission illumination SL is reflected by the half mirror HM on the same optical axis as that of the imaging devices CM1 to CM4, and is irradiated to the imaging objects in the bonding regions AA11 to AA14 of the substrate S. The scattered light irradiated to the imaging object in the bonding area AA11~AA14 with the same optical axis as the imaging device CM1~CM4 is reflected by the imaging object in the bonding area AA11~AA14, and the regular reflection light penetrates the semi-reflection The mirror HM reaches the imaging devices CM1-CM4 to form images of the imaging objects in the junction areas AA11-AA14. That is, the lighting device LD has the function of coaxial epi-illumination (coaxial lighting).

As shown in FIG. 5(b), the length of the Y-axis direction of the surface-emitting illumination SL and the half mirror HM is configured to be slightly wider than the entire imaging field of view of the imaging devices CM1 to CM4 in the width direction of the substrate S, and the surface-emission Illumination SL is divided into light-emitting areas SL1 to SL4 slightly wider than the respective imaging fields of view of imaging devices CM1 to CM4, and each of them can be turned on/off. The light-emitting area of the coaxial lighting device is divided, and each camera device CM1~CM4 can be dimmed. Accordingly, the uniformity of illumination in all areas of the imaging field of view of the imaging devices CM1 to CM4 can be improved. In addition, as described later, when each of the imaging fields of view of the imaging devices CM1 to CM4 overlaps, the light emitting areas SL1 to SL4 also overlap.

Next, repetition of imaging fields of view of a plurality of imaging devices will be described using FIGS. 6 to 8 . Fig. 6 is a top view explaining repetition of imaging fields of view of a plurality of imaging devices. FIG. 7 is a side view when viewed from the direction of arrow A in FIG. 6 . FIG. 8 is a diagram for explaining the amount of repetition of imaging fields of view.

In the said embodiment, the example which the imaging field of view IA1-IA4 of each of imaging devices CM1-CM4 does not overlap was shown. However, when multiplexing the imaging devices, since the pitch of various products (the pitch between the joining regions) and the pitch between the imaging devices do not necessarily have to be the same, as shown in FIGS. better.

6 and 7 show an example in which four imaging devices are installed and six bonding regions are provided in one row of the substrate S, and the distance between the imaging devices is greater than the distance between the bonding regions. Therefore, the imaging fields of view of adjacent imaging devices are overlapped, for example, two or more imaging objects are included in the imaging field of view of one imaging device. Accordingly, each of the imaging devices CM1-CM4 is an imaging device with a resolution higher than that of the imaging device CML. Although each of the imaging devices CM1 to CM4 is an imaging device that cannot image the entire area of the substrate S in a row in the width direction with sufficient resolution at one time, even if it can image the entire area in the width direction of the substrate S at a time with sufficient resolution Area cameras are also available. Here, the substrate S is rectangular and flat as shown in FIG. 6 , and has many bonding regions AA11 to AA16 , AA21 to AA26 , ･･･ vertically and horizontally. The substrate S is configured such that the length in the conveyance direction (X-axis direction) is longer than the length in the width direction (Y-axis direction).

The overlapping region OVL of the imaging field of view may be of a size including the imaging object of the largest size (it may be sufficient if the largest product size is maintained). Accordingly, all the imaging objects capture the entirety of the imaging objects in any imaging field of view. As shown in Figure 8, if the repeated area OVL is the size of the captured object OB2, for example, for a sub-straight substrate, if it is the repetition amount of the maximum tab size, all the tabs are among the tabs. The panorama fits into any one's field of view.

Synthesis of multiple captured images will be described using FIGS. 9 to 15 . FIG. 9 is a diagram illustrating image mosaicing and coordinate matching. FIG. 10 is a schematic diagram illustrating image mosaicing and coordinate transformation using a calibration plate. Fig. 11 is a diagram illustrating distortion. Fig. 12 is a diagram showing transformation determinants of affine transformation and projective transformation. FIG. 13 is a diagram illustrating image mosaicking and coordinate transformation by an object model of a substrate. Fig. 14 is an image of a state in which a paste-like adhesive is coated on the protruding pieces of a multi-lead frame captured by a plurality of imaging devices. Fig. 14(a) is a photographed image without temporal variation, and Fig. 14(b) is an image with A photographic portrait of a situation that has changed over time. Fig. 15 is a diagram for explaining space recorrection, Fig. 15(a) is a diagram showing a state before space recorrection, and Fig. 15(b) is a diagram showing a state after space recorrection.

In order to suppress the amount of overlap, the control unit CNT synthesizes the images captured by the plurality of imaging devices by video mosaic or the like. In general image mosaic, in order to join multiple images together smoothly, the calibration of the images may be lost in the original state.

Therefore, in the embodiment, in synthesizing images captured by a plurality of imaging devices, (A) At the time of shipment or adjustment of the die bonding device, image mosaic and coordinate conversion using the calibration board are performed. (B) During fine adjustment or continuous operation of the die bonding device, the image mosaic by the target model of the substrate and the coordinate transformation by the target model placed on the chute are performed.

The above (A) will be described using FIGS. 9 to 12 . As shown in FIG. 9 , the control unit CNT projects the coordinate mark CMRK as a scale covering all imaging fields of view of the multiplexed imaging device during the adjustment of the bonding device so as to be reflected on the coordinate mark CMRK in the range of the overlapping region OVL. Using the same intersection point IP as a reference, the coordinates of the image are converted by projective transformation or affine transformation, and a single image (composite image) that sequentially joins the images of each camera device is obtained. Here, the coordinate mark CMRK prepares a calibration plate as an adjustment jig, and marks it on the flat plate to be used by the user. The coordinate mark CMRK is, for example, a grid-shaped one. A mark that guarantees the positional relationship of the entire space is required for coordinate conversion. If it is the coordinate mark CMRK covering the entirety of the synthetic field of view shown in Figure 9, the spatial positional relationship of the entirety can be guaranteed during coordinate conversion such as projection transformation, and the image space coordinates can be matched with the actual space coordinates at the distances between intersection points.

As shown in FIG. 10 , the control unit CNT uses, for example, three imaging devices to divide the field of view and image one large calibration plate CP having grid-like textures. The three imaging devices have overlapping areas OV12 and OV23 in which fields of view overlap to some extent due to adjacent imaging devices. The intersection points IP of the lattice pattern in the repeating regions OV12 and OV13 are represented by black dots.

First, the control unit CNT converts the pixel coordinates of the image of the other imaging device with reference to any one of the adjacent imaging devices by affine transformation or projective transformation. Here, the reference image will be described as image IV1, and the converted image will be described as image IV2. The transformation is to calculate the affine transformation or projective transformation by aligning the coordinates (black dot coordinates) of the intersection point IP in the converted images with the coordinates (black dot coordinates) of the intersection point IP in the corresponding reference image. Transform the parameters of the row and column. Here, in general, the transformation matrix of the affine transformation is represented by the equation (1) of FIG. 12 , and the determinant of the projection transformation is represented by the equation (2) of FIG. 12 . The calculation of conversion ranks and columns generally needs to be three-point coordinates. It is not a univocal conversion. Since the conversion can be performed more accurately by performing the conversion on each part of the grid corresponding to the lattice shape, it is better to convert each grid.

In the case that the reference image IV1 has distortions represented by the barrel-shape or spool-shape shown in Fig. 11, the primary image is converted into a right angle to the vertical by first using all the intersection points IP of the correction plate CP that enters the field of view It is preferable to perform distortion correction on the image aligned with the coordinate system (first distortion correction). According to this, the composite image is also combined by simple magnification adjustment in areas other than the overlapping regions OV12 and OV23.

Since the coordinate alignment is carried out with the pixel coordinate system, the converted coordinates of the portrait IV2 are required to be integer values, but the pixels converted by affine transformation or projective transformation do not necessarily have to be suitable for this place, and there are also transformed The coordinates of become intermediate values. At this time, each coordinate system of the converted image undergoes shade value correction typified by nearest neighbor interpolation, bilinear interpolation, bicubic interpolation, etc. from the shade values of the closest converted coordinates.

The above (B) will be described using FIGS. 13 to 15 . Since the conversion in the adjustment operation before the production operation is in the image space within the field of view, the intersection point IP of the calibration plate CP representing the coordinate reference is arranged at a plurality of equal intervals, so it is relatively easy to align the image space and the actual coordinate space. In contrast, during continuous operation or simple adjustment, image synthesis and coordinate alignment are performed without using the calibration plate CP. Since it is only necessary to know the same positions for image synthesis, the positioning target mark TM etc. on the substrate is used. Since the target mark TM for positioning is used as the positioning process of the protruding piece at the time of joining (adhesive or pasting), it has already been registered as a template model, so it is used. As shown in FIG. 13 , when a plurality of protruding pieces enter the overlapping regions OV12 and OV23 of the image, three or more punctuation points are ensured using the positioning target marks TM of adjacent protruding pieces. When only one protruding piece enters the overlapping area, three positioning target marks TM registered in advance on one protruding piece are registered as a template image model.

In this method, although it is possible to synthesize by aligning coordinates between images, it is not possible to align with the actual space. Even if any image is aligned as a reference, if the alignment with the coordinate space is not performed in advance, when the synthesis is performed as it is, as shown in FIG. 11, the image adjacent to the image IV1 of the reference image The image IV2 and the image IV3 adjacent to the image IV2 are affected by the distortion of the image IV1. Displacement due to distortion When the image IV1 is sequentially adjacent to each other, the image IV3 at the farthest position is enlarged the most. In order to suppress the amount of enlargement, mark SMRK is set at the known coordinates of the chute SCT, and the coordinates of the set mark SMRK are measured again from the image, and returned together with the synthesized image. Since the distortion of the image here depends on the lens of the imaging device, the conversion (first distortion correction) at the time of the initial conversion is performed after the image is synthesized. In order to align the overall coordinates, the mark on the chute SCT is used. SMRK is better. That is, the control part CNT grasp|ascertains the coordinate of the mark SMRK provided separately on the chute SCT which is a conveyance path, after performing correction of real space and image space using coordinate mark CMRK. Here, the chute SCT is located outside the both ends of the width direction of the board|substrate S. As shown in FIG.

As mentioned above, the coordinate mark CMRK is aligned before the production operation by the die bonding apparatus, and is made to match as an adjustment operation. However, with the start of work, due to the self-heating of the imaging device and its saturation state, or the deviation of the heat distribution between the imaging devices, as shown in FIG. 14(b), the imaging fields of each imaging device have slightly different The situation of the diachronic displacement of the person. Here, the image within the imaging field of view IA2 of the imaging device CM2 deviates to the right relative to the image within the imaging field of view IA1 of the imaging device CM1. That is, the imaging device CM2 is displaced to the right. In addition, the place enclosed by the square frame shows the vicinity of the lower left corner of each protruding piece. This deviation cannot be ignored in a bonding device requiring a precision of μm. Even during construction, it is necessary to correct this slight deviation. Hereinafter, a method of correcting this slight deviation will be described using FIG. 15 .

As mentioned above, when a deviation occurs between imaging devices during continuous construction, there may be a difference in the pitch between the protrusions of the substrate S having a known pitch. The control unit CNT periodically measures the known pitch between the protrusions to detect the intrinsic displacement due to factors such as heat of the imaging device. Furthermore, even if the control unit CNT periodically measures the known distance between the marks SMRK on the chute SCT, it is possible to detect the inherent displacement due to factors such as heat of the imaging device.

When detecting this displacement, the control unit CNT uses characteristic marks such as the target mark TM for positioning on the substrate S. Then, the matrix of projective transformation or the matrix of affine transformation for synthesizing and transforming the image is calculated again. At this time, the obtained projective transformation matrix or affine transformation matrix can join images, but as shown in FIG. 15( a ), there may be a state where the image space and the actual space cannot be matched. Therefore, the control part CNT uses the mark SMRK on the chute SCT measured first, and performs reconversion based on the coordinate. In this way, a synthetic image as shown in FIG. 15(b) is obtained.

In addition, the magnification changes due to the thickness of the substrate P (the height of the upper surface of the substrate P), and when the height changes, calibration becomes difficult. A method for reducing the influence of the thickness of the substrate will be described using FIGS. 16 and 17 .

Fig. 16 is a diagram explaining the method of changing the height of the calibration plate. Fig. 16(a) shows the state where the substrate is transported and placed on the bonding table and arranged, and Fig. 16(b) shows the cross section of the calibration plate moving up and down picture. Figure 17 is a diagram illustrating the marks provided on the chute, and Figure 17(a) shows that the substrate is transported and The sectional view of the state placed on the joining table, Fig. 17(b) is a top view of the chute where the mark is set, and Fig. 17(b) is a sectional view of the chute where the mark is set in another example.

In order to correspond to the height displacement, the control unit CNT slightly moves the correction plate CP up and down to obtain the projection conversion matrix for each height. The control unit CNT calculates the known substrate thickness or paste height, calculates the predicted height of the calibration pattern position or inspection view position from the grain thickness, etc., and automatically selects which projection transformation row to use to maintain at each height. The control unit CNT recognizes and measures the height of the same point on the substrate in the overlapping area between adjacent imaging devices. The control CNT automatically selects the used projection conversion rank from its measured values.

Specifically, at the time of correction by the first calibration plate CP, as shown in FIG. 16( b ), the bonding table AS on which the calibration plate CP is installed is moved up and down, and the calibration plate is measured at each height of the bonding table AS. Coordinates of the intersection point of CP. The up and down mechanism of the bonding table is configured to be able to move up and down in units of μm.

Furthermore, it is preferable that the height of the mark SMRK provided on the chute SCT be the same as the height of the upper surface of the board|substrate S. As shown in FIG. As shown in Fig. 17(a) and (b), for example, holes (holes) having different depths and diameters are formed in the chute SCT. The holes marked SMRK can be provided individually for each depth as shown in Fig. 17(c). The mark SMRK may be a hole as long as the top surface of the substrate S is the same as the height.

When implemented, it has one or more of the following effects.

(a) Since each object to be imaged can be viewed from almost directly above, it is possible to prevent the height direction of the image produced by the simple low magnification optical system from being inclined at the edge of the field of view. (b) Since it is possible to detect the intrinsic displacement caused by the main factors such as heat of the imaging device, the correction of the imaging device can be implemented during the operation of the die bonder, so the influence of the temporal displacement of the imaging device can be reduced. (c) Since height displacement can be accommodated, it is not affected by changes in thickness depending on the type of substrate, so the influence of different types can be reduced. (d) Stabilization of positioning accuracy and stabilization of inspection can be achieved by at least any one of (a) to (c) above.

(Modification) Hereinafter, typical modification examples of the embodiment will be illustrated. In the description of the following modified examples, the same reference numerals as those in the above-mentioned embodiment can be used for parts having the same configuration and function as those described in the above-mentioned embodiment. In addition, regarding the description of such a portion, it is assumed that the description in the above-mentioned embodiment can be appropriately used within the range of technical compatibility. In addition, a part of the above-described embodiments and all or a part of a plurality of modified examples can be appropriately and compositely applied within a range that is not technically contradictory.

Fig. 18 is a perspective view illustrating multiplexing of an imaging device in a modified example. In the embodiment, although an example in which imaging objects in a row in the width direction of the substrate S are imaged by a plurality of imaging devices is described, the plurality of imaging devices are also arranged in the longitudinal direction of the substrate S, that is, the plurality of imaging devices are arranged as It is also possible to photograph objects in a lattice shape and to photograph plural rows.

For example, as shown in FIG. 18 , four imaging device groups CM10 to CM40 are configured, and each imaging device group has four imaging devices CM1 to CM4 arranged in a row, and 16 imaging devices are configured as lattice. Here, on the substrate S, six imaging target portions in one row are arranged in five rows, and the imaging fields of view of adjacent imaging devices overlap.

As shown in FIG. 18, not only the first row of imaging objects of the substrate S, but also the entire imaging objects of the second row and subsequent rows are imaged in parallel, and the synthesized image can be recognized. Therefore, compared with the implementation type This state can reduce the number of times of imaging while moving the substrate S. [Example]

Embodiments applicable to the above-mentioned implementation forms are described below. Fig. 19 is a schematic top view showing a die bonder in the embodiment. Fig. 20 is a diagram illustrating the operation of the pick-up head and the bonding head when viewed from the direction of arrow A in Fig. 19 .

The die bonder 10 roughly includes a die supply unit 1 for supplying a die D mounted on a substrate S, a pickup unit 2, an intermediate platform unit 3, a preforming unit 9, a bonding unit 4, a transport unit 5, a substrate supply unit 6, The substrate unloading unit 7 and the control unit 8 monitor and control the operation of each unit. The Y-axis direction is the front-back direction of the die bonder 10 , and the X-axis direction is the left-right direction. The die supply unit 1 is arranged on the front side of the die bonder 10 , and the bonding unit 4 is arranged on the deep side. Here, on the substrate S, a plurality of product regions (hereinafter referred to as package regions P) to be the final package are formed. For example, when the substrate S is a lead frame, the package region P has a tab on which the die D is placed.

First, the die supply unit 1 supplies the die D mounted on the package area P of the substrate S. As shown in FIG. The die supply unit 1 has a wafer holding table 12 holding a wafer 11 and a peeling unit 13 shown by a dotted line pushing a die D from the wafer 11 . The die supply part 1 moves in the XY axis direction by a driving means not shown, and moves the picked up die D to the position of the peeling unit 13 .

The pick-up unit 2 has a pick-up head 21 for picking up the die D, a Y drive unit 23 for the pick-up head that moves the pick-up head 21 to the Y-axis direction, and various components (not shown) that move the chuck 22 up and down, rotate, and move in the X-axis direction. A drive unit, and a wafer recognition camera 24 for grasping the pick-up position of the die D picked up from the wafer 11 . The pick-up head 21 has a chuck 22 (also refer to FIG. 14 ) for sucking and holding the pushed-up die D at the tip, picks up the die D from the die supply unit 1 , and places it on the intermediate table 31 . The pick-up head 21 has various drive units (not shown) that move the chuck 22 up and down, rotate, and move in the X direction.

The intermediate stage part 3 has an intermediate stage 31 on which the die D is temporarily placed, and a stage identification camera 32 for identifying the die D on the intermediate stage 31 .

The preforming unit 9 has a syringe 91 , an actuator 93 for moving the syringe 91 in the Y direction and the Z direction, and an adhesive recognition camera 94 as an imaging device for grasping the application position of the syringe 91 and the like. Here, the adhesive agent recognition camera 94 is, for example, multiplexed imaging devices CM1 to CM4 of the embodiment, and the imaging devices CM1 to CM4 are each configured to perform imaging using the lighting device LD. The preforming unit 9 applies a paste-like adhesive such as epoxy resin to the substrate S transported by the transport unit 5 with a syringe 91 . The syringe 91 is filled with a paste-like adhesive, and is configured such that the paste-like adhesive is pushed out from the tip of the nozzle to the substrate S by air pressure to be applied. In the case of the substrate S, for example, a plurality of unit lead frames arranged in a row horizontally to form a series of multiple lead frames, paste-like adhesive is applied to each tab of the unit lead frame.

The bonding part 4 picks up the die D from the intermediate platform 31 and bonds it to the package area P coated with a paste-like adhesive on the transported substrate S. The bonding unit 4 includes a bonding head 41 similar to the pick-up head 21 that sucks and holds the die D at the tip of a chuck 42 (see also FIG. 20 ), a Y-moving portion 43 that moves the bonding head 41 in the Y-axis direction, and an imaging substrate. The position identification mark (not shown) of the packaging area P of S, and the substrate identification camera 44 for identifying the bonding position. Here, the substrate recognition camera 44 is, for example, multiplexed imaging devices CM1 to CM4 of the embodiment, and the imaging devices CM1 to CM4 are configured to perform imaging using the illumination device LD, respectively. With such a configuration, the bonding head 41 corrects the pick-up position and posture based on the imaging data of the platform recognition camera 32 , picks up the die D from the intermediate stage 31 , and bonds the die D to the substrate based on the imaging data of the substrate recognition camera 44 .

The conveyance unit 5 has a substrate conveyance claw 51 for grasping and conveying the substrate S, and a conveyance path 52 for moving the substrate S. As shown in FIG. The substrate S is moved by driving an unillustrated nut provided on the substrate conveyance claw 51 provided on the conveyance passage 52 by an unillustrated ball wheel provided along the conveyance passage 52 . With such a configuration, the substrate S is moved from the substrate supply unit 6 to the bonding position along the conveyance path 52 , and after bonding, is moved to the substrate carry-out portion 7 , and the substrate S is delivered to the substrate carry-out portion 7 .

Next, the structure of the crystal grain supply part 1 is demonstrated using FIG. 21. FIG. FIG. 21 is a schematic cross-sectional view showing main parts of the crystal grain supply unit shown in FIG. 19 .

The die supply unit 1 includes a wafer holding table 12 that moves in the horizontal direction (XY axis direction), and a peeling unit 13 that moves in the vertical direction. The wafer holding table 12 includes an expansion ring 15 for holding the wafer ring 14 and a support ring 17 for horizontally positioning the dicing tape 16 fixed to the wafer ring 14 . In the wafer 11 , the die D diced into a mesh shape is bonded and fixed to the dicing tape 16 . The peeling unit 13 is disposed inside the support ring 17 .

When the die supply unit 1 pushes up the die D, the extension ring 15 holding the wafer ring 14 descends. As a result, the dicing tape 16 held on the wafer ring 14 is elongated, the interval between the dies D is widened, and when the peeling unit 13 pushes up the dicing tape 16 from below the die D, it moves horizontally, and Improve pick-up of Die D.

As shown in FIG. 22 , the control system 80 includes a control unit 8 , a drive unit 86 , a signal unit 87 , and an optical system 88 . The control unit 8 generally has a control and computing device 81 , a memory device 82 , an input/output device 83 , a bus line 84 , and a power supply unit 85 mainly composed of a CPU (Central Processor Unit). The memory device 82 has a main memory device 82a composed of RAM storing processing programs and the like, and an auxiliary memory device 82b composed of HDD storing control data and image data required for control. The input and output device 83 has a screen 83a for displaying device status or information, etc., a touch panel 83b for inputting instructions from the operator, a mouse 83c for operating the screen, and an image capture device for capturing image data from the optical system 88 83d. Moreover, the input/output device 83 has a motor control device 83e for controlling the drive unit 86 of the XY stage (not shown) of the crystal grain supply unit 1 or the ZY drive shaft of the bonding head stage, etc., and signals from various sensors or lighting The I/O signal control device 83f of the signal part 87 of the switch etc. of the device etc. fetches or controls a signal. The optical system 88 includes the wafer recognition camera 24 , the adhesive recognition camera 94 , the platform recognition camera 32 , and the substrate recognition camera 44 shown in FIG. 20 . The control and calculation device 81 captures the required data through the bus line 84, performs calculations, controls the pick-up head 41, etc., or sends information to the screen 83a, etc.

The control unit 8 saves the image data captured by the optical system 88 in the memory device 82 through the image capturing device 83d. According to the software programmed based on the stored image data, the control and calculation device 81 is used to perform positioning of the die D and the substrate S, inspection of the coating pattern of the paste adhesive, and surface inspection of the die D and the substrate S . Based on the positions of the die D and the substrate S calculated by the control and computing device 81, the driving unit 86 is moved by software through the motor control device 83e. Through this process, the positioning of the die D on the wafer 11 is performed, and the driving unit of the die supply unit 1 and the die bonding unit 4 operates to bond the die D on the substrate S. The recognition camera used in the optical system 88 is a gray scale, color camera, etc., and digitizes the light intensity.

Next, the manufacturing method of the semiconductor device using the die bonder concerning an Example is demonstrated using FIG. 23. FIG. FIG. 23 is a flowchart showing a method of manufacturing a semiconductor device using the die bonder shown in FIG. 19 .

(Step S51: Wafer and substrate loading process) The wafer ring 14 holding the dicing tape 16 attached to the die D divided from the wafer 11 is stored in a wafer cassette (not shown), and loaded into the die bonder 10 . The control unit 8 supplies the wafer ring 14 from the wafer cassette filled with the wafer ring 14 to the die supply unit 1 . Furthermore, the board|substrate S is prepared, and it carries in to the die bonder 10. The control part 8 mounts the board|substrate S to the board|substrate conveyance claw 51 by the board|substrate supply part 6.

(Step S52: pick up project) The control unit 8 moves the wafer ring 14 in such a way that the desired die D can be picked up from the wafer ring 14 by the wafer holding table 12, and performs positioning and surface inspection based on the data captured by the wafer recognition camera 24. examine. The control unit 8 peels the die D from the dicing tape by the peeling unit 13 .

The control unit 8 picks up the peeled die D from the wafer 11 by the picker 21 . In this way, the die D peeled off from the dicing tape 16 is sucked and held by the chuck 22 of the pick-up head 21 to be transported to the next process (step BS13 ). And when the die D is returned to the die supply unit 1 from the chuck 22 transported to the next process, the following die D is peeled off from the dicing tape 16 according to the above procedure, and then the die D is peeled off according to the same procedure. D is peeled off one by one from the dicing tape 16 .

(Step S53: Joining process) The control part 8 acquires the image of the surface of the board|substrate S before application|coating by the adhesive agent recognition camera 94, and confirms the surface which paste-form adhesive agent should apply|coat. If there is no problem on the surface to be coated, the control unit 8 applies the paste adhesive from the syringe 91 to the substrate S conveyed by the conveyance unit 5 . When the substrate S is a multi-lead frame, paste-like adhesive is applied to all the tabs. The control unit 8 uses the adhesive recognition camera 94 to reconfirm whether the paste adhesive is applied correctly after application, and checks the applied paste adhesive. If there is no problem in coating, the control unit 8 transports the substrate S to the bonding station BS by the transport unit, and performs positioning based on the image data captured by the substrate recognition camera 44 .

The control unit 8 places the die D picked up from the wafer 11 by the pick-up head 21 on the intermediate platform 31 , picks up the die D from the intermediate platform 31 again by the bonding head 41 , and bonds it to the positioned substrate S. The control unit 8 checks whether or not the die D is bonded at a desired position based on the image data captured by the substrate recognition camera 44 .

(Step S54: Substrate unloading process) The control unit 8 takes out the substrate S of the dies D to be bonded from the substrate conveying claw 51 by the substrate carrying out unit 7 . The substrate S is unloaded from the die bonder 10 .

As mentioned above, although the invention created by the present inventor has been concretely described based on the embodiments, examples, and modifications, the present invention is not limited to the above embodiments, examples, and examples, and various changes are of course possible.

For example, in the embodiment, although the example of coating the paste-like adhesive on the substrate with the preformed part is described, the adhesive for bonding the die to the substrate is used even if the adhesive bonded to the wafer 11 and the dicing A film-like adhesive material called die adhesive film (DAF) between the tapes 16 may be used instead of the paste-like adhesive applied by the syringe 91 . DAF is suitable for a stacked package formed by mounting many chips on a chip on a substrate S.

Furthermore, in the embodiment, although it is described that an intermediate platform portion 3 is provided between the die supply unit 1 and the bonding unit 4, the die D picked up from the die supply unit 1 by the pick-up head 21 is placed in the middle. The platform 31 is an example where the bonding head 41 picks up the die D again from the intermediate stage 31 and bonds it to the substrate S being transported. S is also available.

10: Die bonder (die bonder)

AA, AA11~AA14, AA21~AA24, AA11~AA16, AA21~AA26: junction area

CML, CMH, CM1~CM4: camera device

CNT, 8: Control Department

S: Substrate

SCT: Chute (transport path)

[ FIG. 1( a )] is a perspective view showing a normal field of view optical system, and [ FIG. 1( b )] is a perspective view showing a wide field of view optical system. [Fig. 2(a)] is a conceptual diagram of a wide-field optical system using a microlens, and [Fig. 2(b)] is a conceptual diagram of a wide-field optical system using a telecentric lens. [FIG. 3(a)] is a diagram showing the image of the three-dimensional shape of the paste, and [FIG. 3(b)] is a diagram showing spectral lines on the paste when the substrate is viewed in a wide area with a microlens. [FIG. 4(a)] is a top view of the multiplexing of the imaging device of the embodiment, and [FIG. 4(b)] is a side view when viewed from the arrow A direction in FIG. 4(a). [FIG. 5] It is a figure explaining the illumination device used for the imaging device of embodiment. [FIG. 6] It is a top view explaining the repetition of the imaging field of view of a plurality of imaging devices. [ Fig. 7 ] It is a side view when viewed from the arrow A direction in Fig. 6 . [FIG. 8] It is a figure explaining the repetition amount of an imaging field of view. [FIG. 9] It is a figure explaining image mosaic and coordinate matching. [ Fig. 10 ] is a schematic diagram illustrating image mosaicking and coordinate transformation using a calibration plate. [ Fig. 11 ] is a diagram illustrating distortion. [ Fig. 12 ] is a diagram showing transformation determinants of affine transformation and projective transformation. [ FIG. 13 ] is a diagram illustrating image mosaicking and coordinate transformation of an object model by a substrate. [ Fig. 14 ] is a diagram illustrating the influence of the temporal displacement of the imaging device. [FIG. 15] It is a figure explaining space re-correction. [ Fig. 16 ] A diagram showing a method of changing the height of the correction plate. [FIG. 17] It is a figure explaining the mark provided in the chute. [FIG. 18] It is a perspective view explaining the multiplexing of the imaging device in a modification. [ Fig. 19 ] is a schematic top view showing a die bonder in an embodiment. [FIG. 20] It is a figure explaining the operation|movement of a pick-up head and a bonding head seen from the direction of the arrow A in FIG. 19. [ Fig. 21] Fig. 21 is a schematic cross-sectional view showing main parts of the crystal grain supply unit shown in Fig. 19 . [FIG. 22] It is a block diagram explaining the schematic structure of the control system of the die bonder of FIG. 19. [FIG. [ Fig. 23 ] is a flow chart showing a method of manufacturing a semiconductor device.

AA11~AA16: junction area

CM1~CM4: camera device

CNT: Control Department

S: Substrate

IA1,IA2: camera field of view

Claims (23)

A die bonding device comprising: a transport path for transporting a substrate; a plurality of imaging devices fixedly arranged above the transport path in a row along the width direction of the substrate; and a control unit for It is configured to use the plurality of imaging devices to capture a series of plural bonding regions on the substrate along the width direction to obtain a plurality of images, generate a composite image from the obtained plurality of images, and recognize the bonding region based on the composite image. The object to be imaged is configured such that the imaging field of view of each imaging device overlaps on the substrate, and the overlapping imaging field of view is set larger than the bonding area. The die bonding apparatus according to claim 1, wherein the control unit is configured to generate the composite image based on the coordinate marks located in the overlapped imaging fields of view. The die bonding apparatus according to claim 2, wherein the coordinate mark is a grid-shaped scale covering all the fields of view of the imaging device, and the control unit is configured to project the coordinate mark so as to be reflected on the repeated field of view of the imaging device. The image is projectively converted based on the same intersection point of the above-mentioned coordinate marks, and the images between the imaging devices are joined to generate the above-mentioned composite image. The die bonding apparatus according to claim 3, wherein the conveying path is divided on the outer sides of both ends in the width direction of the substrate. It has a plurality of reference marks, the substrate has a plurality of protrusions arranged at specific intervals, and the control unit is configured to measure the distance between the protrusions or the distance between the reference marks with the imaging device, and detect the displacement between the imaging devices. . The die bonding apparatus according to claim 4, wherein the substrate further has a characteristic mark, and the control unit is configured to recalculate a projection for synthesizing and transforming the images based on the characteristic mark when detecting the displacement between the imaging devices. Convert ranks. The die bonding apparatus according to claim 5, wherein the control unit is configured to recalculate the projection conversion matrix based on the coordinates of the reference marks measured in advance. The die bonding apparatus according to claim 6, wherein the control unit is configured to slightly move the substrate up and down, obtain a projection conversion matrix for each height, and calculate an alignment pattern from the thickness of the substrate, the paste height, or the thickness of the crystal grain. According to the predicted height of the position or the position of the inspection field of view, any one of the above-mentioned projection conversion ranks maintained at each height is selected according to the calculated predicted height. The die bonding apparatus according to claim 7, wherein the control unit is configured to identify the same point on the substrate and measure the height in overlapping imaging fields of view between adjacent imaging devices, according to the measured height. For the above heights, the selection maintains the ranks of the above projection transformations per height. The die bonding apparatus according to any one of claims 1 to 8, further comprising a plurality of lighting devices corresponding to each of the plurality of imaging devices, and the control unit is configured to independently adjust the light of the plurality of lighting devices . The die bonding apparatus according to any one of claims 1 to 8, wherein the control unit is configured to transport the substrate in the longitudinal direction of the substrate and image the bonding regions in the next row with the plurality of imaging devices. The die bonding apparatus according to any one of claims 1 to 8, wherein the object to be imaged is a paste-like adhesive applied to the substrate. The die bonding apparatus according to claim 11, wherein the control unit is configured to perform an appearance inspection of the paste-like adhesive coated on the substrate by the imaging device. The die bonding apparatus according to any one of claims 1 to 8, wherein the object to be imaged is the substrate or a die on a die that has already been bonded. A method of manufacturing a semiconductor device, comprising the following steps: a step of carrying a substrate into a die bonding device, the die bonding device Constructed to include: a transport path for transporting substrates having a plurality of bonding areas; a plurality of imaging devices fixedly arranged above the transport path in a row along the width direction of the substrate; and a plurality of lighting devices, It is installed corresponding to each of the plurality of above-mentioned imaging devices, and the imaging field of view of each imaging device is repeated on the above-mentioned substrate. A process of obtaining multiple images of a plurality of the above-mentioned bonding regions in a row along the above-mentioned width direction, generating a composite image based on the obtained multiple above-mentioned images, and identifying the imaging target of the above-mentioned bonding regions according to the above-mentioned composite images; and the above-mentioned substrate A process of transporting the substrate in the longitudinal direction and imaging the plurality of joining regions in the next row with the plurality of imaging devices. The method of manufacturing a semiconductor device according to claim 14, wherein the synthetic image is generated based on the coordinate marks located in the repeated imaging fields of view. The method of manufacturing a semiconductor device according to claim 15, wherein the coordinate marks are a grid-shaped scale covering all the fields of view of the imaging device, and the coordinate marks are projected so as to reflect the same intersection point of the coordinate marks of the repeated fields of view of the imaging device. The image is projected and converted as a reference, and the images between the imaging devices are joined to generate the above-mentioned composite image. The method of manufacturing a semiconductor device as claimed in claim 16, wherein the conveyance path has a plurality of reference marks on the outer sides of both ends in the width direction of the substrate, the substrate has a plurality of projections arranged at specific intervals, and the interval between the projections or the interval between the reference marks is measured by the imaging device, Detect the displacement between the above-mentioned camera devices. The method of manufacturing a semiconductor device according to claim 17, wherein the above-mentioned substrate further has a feature mark, and when the displacement between the above-mentioned imaging devices is detected, the projective transformation matrix for synthesizing and transforming the image is recalculated based on the above-mentioned feature mark. The method of manufacturing a semiconductor device according to Claim 18, wherein the projection conversion matrix is recalculated based on the coordinates of the previously measured reference marks. The method of manufacturing a semiconductor device according to claim 19, wherein the above-mentioned substrate is slightly moved up and down, the projection conversion matrix is obtained at each height, and the position of the alignment pattern or the position of the inspection field of view is calculated from the thickness of the above-mentioned substrate or the height of the paste or the thickness of the crystal grain. Predicted height, according to the calculated predicted height, select any one of the above-mentioned projection conversion ranks held at each height. The method for manufacturing a semiconductor device as claimed in claim 20, wherein the above Identifying the same point on the substrate, measuring the height, and selecting the above-mentioned projection transformation ranks maintained at each height according to the measured above-mentioned height. The method of manufacturing a semiconductor device according to any one of claims 14 to 21, further comprising a process of applying a paste-like adhesive to the substrate, and the object to be imaged is the paste-like adhesive coated agent. The method of manufacturing a semiconductor device according to any one of Claims 14 to 21, further comprising a step of bonding a die to the substrate or the bonded die, and the object to be imaged is the coated die grain.

Images