HK1230787B - Method for mounting semiconductors provided with bumps on substrate locations of a substrate - Google Patents

Description

Method for mounting a semiconductor device provided with bumps on a substrate positioning device of a substrate

Technical Field

The present invention relates to a method for mounting a semiconductor chip provided with bumps on a substrate location of a substrate as a flip chip.

Summary of the Invention

The invention is based on the object of developing a method for mounting a semiconductor chip as a flip chip on a substrate which allows, on the one hand, a very high placement accuracy and, on the other hand, the highest possible throughput.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings, which are incorporated into and constitute a part of this specification, illustrate one or more embodiments of the present invention and, together with the detailed description, serve to explain the principles and implementations of the invention. The drawings are schematic and not drawn to scale. In the drawings:

FIG1 schematically shows a side view of an apparatus for mounting a semiconductor chip provided with bumps as a flip chip,

FIG2 shows the camera support in top view, and

FIG3 shows a pixel coordinate system and a machine coordinate system.

DETAILED DESCRIPTION

FIG1 schematically shows a side view of an apparatus for mounting semiconductor chips 2 provided with bumps 1 as flip chips 3, which is configured for carrying out the method according to the invention. The apparatus comprises a wafer stage 4 for providing the semiconductor chips 2, a flip-chip device 5 with a pick-up head 6, a first transport system 7 with a transport head 8, a second transport system 9 with a bonding head 10, a transport system (not shown) for supplying and providing a substrate 11 on a support 12, an apparatus 13 for wetting the semiconductor chips with flux, a first camera 14, and a second camera 15. The apparatus 13 comprises a camera support 16, a plate 17 having a cavity 18, the base of which is transparent, and a downwardly opening flux container 19. The position of the bonding head 10 is described by machine coordinates. The apparatus is controlled by a control device (not shown).

The first transport system 7 is set up to move the transport head 8 in at least two spatial directions. The second transport system 9 is set up to move the bonding head 10 in three spatial directions.

In another device that is also suitable for carrying out the method according to the invention, the wafer stage 4 and the flip-chip device 5 with the pick-up head 6 are not present, but are replaced by a feeding device (also referred to as a feeder) that directly provides the semiconductor chips 2 as flip-chips 3. In this device, the element shown with the reference numeral 4 in FIG1 represents the feeder.

The camera support 16 is arranged in a fixed manner on the device and comprises a base 20 and at least two side walls 21, to which the first camera 14 is fastened. A plate 17 is detachably mounted on the camera support 16. FIG2 shows the camera support 16 in a top view. The camera support 16 comprises a first optical mark 22 and optionally also at least one further optical mark 23. The camera support 16 is formed in a mechanically rigid manner in such a way that the first camera 14 and the optical mark 22 and the optional optical mark 23 are in a rigid geometrical relationship with respect to one another, so that the position and orientation of the pixel coordinate system of the image assigned to the first camera 14 are in a fixed relationship (i.e., assumed to be non-variable in this case) with respect to the position of the optical mark 22 and the optional optical mark 23.

Preferably, the optical mark 22 and, optionally, the optical mark 23 are arranged in a direction extending perpendicularly to the surface of the support 12 for the substrate 11, at a height substantially equal to the substrate positioning height. This offers the advantage that when the second camera 15 records images of the optical mark 22 and, optionally, the substrate positioning, or the substrate markings of the substrate, the bond head 10 is positioned substantially at the same height. This means that the bond head 10 does not need to be raised to a different height in order to bring the object to be imaged into the focal plane of the second camera 15.

The pixel coordinates of the flip chip 3 are determined from the image of the flip chip 3 recorded by the first camera 14 and converted to machine coordinates of the bonding head 10 using first geometric data. The first geometric data includes the position of the first optical mark 22 and a vector A with fixed values (u, v) indicating the direction and distance of the first optical mark 22 relative to a reference point of the pixel coordinate system of the first camera 14. The first geometric data further includes a fixed angle Ψ that describes the twist between the pixel coordinate system of the first camera 14 and the machine coordinate system of the bonding head 10. If more than one optical mark is present, the first geometric data includes the position of each additional optical mark and an associated vector with fixed values indicating the direction and distance of the additional optical mark relative to the reference point of the pixel coordinate system of the first camera 14.

Fig. 3 schematically shows the machine coordinate system MS of the bond head 10, the pixel coordinate system PS of the first camera 14, the first optical mark 22, the vector A and the angle Ψ. The values (u, v) of the vector A are numbers in the machine coordinate system MS.

As will be explained in more detail below, in the method according to the invention, a flip chip 3 is placed in a cavity 18, wherein the bumps 1 of the flip chip 3 are immersed in a flux, an image is recorded with the first camera 14, and after expiration of the wetting period, the flip chip 3 is removed from the cavity 18 and mounted on a substrate 11. During this stage, the cavity 18 is located in a fixed position above the first camera 14, and the field of view of the first camera 14 is oriented towards the base of the cavity 18 so that its image shows the bottom side of the flip chip 3 with the bumps 1.

In a first embodiment, the flux container 19 is arranged in a fixed manner. In this case, the device 13 includes a drive for reciprocating the plate 17. The plate 17 is moved to such an extent that the cavity 18 is filled with flux, the cavity 18 is located below the flux container 19 or on the opposite side of the flux container 19, and then the plate 17 is moved back again so that the cavity 18 is located in the aforementioned position above the first camera 14.

In a second embodiment, the plate 17 is arranged in a fixed manner with the cavity 18 located above the first camera 14. In this case, the device 13 includes a drive for moving the flux container 19 from one side of the cavity 18 to the opposite side of the cavity 18. The flux container 19 slides on the plate 17 and fills the cavity 18 with flux.

The second camera 15 is fastened to the bonding head 10. The optical axis of the camera 15 extends parallel to the clamping axis of the bonding head 10. The second camera 15 is mechanically fastened to the bonding head 10 in such a way that the pixel coordinate system of the image assigned to the second camera 15 is oriented in a fixed geometric relationship to the clamping axis of the bonding head 10. The pixel coordinates of the substrate positioning, which are determined with the aid of at least one image of the substrate positioning or markings on the substrate recorded by the second camera 14, are converted into machine coordinates of the bonding head 10 with the aid of second geometric data.

The second geometrical data comprises a vector B with values (x, y) indicating the direction and distance of a reference point of the pixel coordinate system of the second camera 15 relative to a reference point of the machine coordinate system of the bonding head 10. The second geometrical data further comprises an angle describing a twist of the two coordinate systems.

The first and second geometric data further include a scaling factor that enables values in the pixel coordinate system of the respective camera to be converted into values in the machine coordinate system of the bond head 10. The first and second geometric data are determined in a calibration phase, which is performed before the installation phase. In order to increase the long-term stability of the device and method, the calibration phase can be performed at different points in time.

The described embodiments of the apparatus are capable of carrying out the method according to the invention for mounting semiconductor chips as flip chips on a substrate. The method according to the invention comprises the aforementioned calibration phase and the mounting phase, wherein the first and second geometrical data are determined in the calibration phase, and the following steps are carried out for each semiconductor chip in the mounting phase:

Alternatively: providing the semiconductor chip 2 at a predetermined position using the wafer stage 4;

removing the provided semiconductor chip 2 with a pickup head 6 of a flip chip device 5 and twisting the semiconductor chip 2 by 180° to provide the semiconductor chip 2 as a flip chip 3;

Alternatively: using a feeding device to provide the semiconductor chip 2 as a flip chip;

Receive the flip chip 3 from the pick-up head 6 or the feeder with the transport head 8;

filling a cavity 18 with flux, the cavity 18 being arranged in the plate 17 and being formed by a transparent substrate, wherein the plate 17 is arranged in a fixed manner or is moved after filling the cavity 18 so that the cavity 18 is located above the first camera 14 in both cases;

Placing the flip chip 3 in the cavity 18 with the bump 1 facing the base of the cavity 18;

recording an image of the flip chip 3 with a first camera 14 and determining the actual position of the flip chip 3 relative to the machine coordinate system of the bonding head 10 based on the image and the first geometric data;

The flip chip 3 is removed from the cavity 18 by the bonding head 10;

The actual position of the substrate positioning with respect to the machine coordinate system of the bond head 10 is determined by,

or via:

The bond head 10 is moved to a position above the substrate location where the substrate location is in the field of view of the second camera 15,

recording at least one image by means of the second camera 15, and

calculating an actual position of the substrate location based on the substrate location of the at least one image and the second geometric data;

or via:

The actual position of the substrate positioning is calculated with the aid of the actual positions of at least two substrate marks, wherein the actual position of each of the at least two substrate marks is determined after a new substrate 11 is supplied to the support 12 in the following manner:

The bond head 10 is moved to a position above the substrate 11 where the substrate markings are in the field of view of the second camera 15,

Record the image with the second camera 15, and

determining the actual position of the substrate marking using the image and the second geometric data; and

Calculating the position that the bond head 10 will reach based on the determined actual position of the flip chip 3 and the determined actual position of the substrate positioning; and

The bond head 10 is moved to the calculated position and the flip chip 3 is placed on the substrate.

Assembling the apparatus with the transport head 8 and the bonding head 10 allows for an increase in the throughput of the apparatus, since the transport head 8 receives the flip chip 3 from the pick-up head 6 or the feeder and places the chip in the cavity 18, and the bonding head 10 removes the flip chip 3 from the cavity 18 and places the chip on the substrate 11. The control device is designed to control the movement of the transport head 8 and the bonding head 10 so that the two heads move at least partially simultaneously without colliding with each other. With regard to the highest possible throughput of the apparatus, the control device is particularly programmed to control the sequence of the various steps of the method so that once the bonding head 10 has removed the next flip chip 3 to be mounted from the cavity 18, the transport head 8 places the next subsequent flip chip 3 into the cavity 18 as quickly as possible, based on the duration of the various process steps.

FIG1 shows the apparatus at a point in time at which the pickup head 6 of the flip chip apparatus 5 has taken the semiconductor chip 2 from the wafer stage 4 , the flip chip 3 has been placed in the cavity 18 , and the bonding head 10 transports the flip chip 3 wetted with flux to the substrate 11 .

The image of the flip chip 3 recorded by the first camera 14 can be used, in addition to determining the actual position of the flip chip 3, to check whether all bumps 1 are present and/or properly wetted. Furthermore, the first camera 14 can record an image after other flip chips 3, and the image processing software can evaluate the image and check whether all bumps 1 have been properly wetted. If this is the case, a message can be issued, indicating that the bond head 10 should immediately remove the flip chip 3 from the cavity 18 and place it on the substrate.

If the viewing angle of the second camera 15 is relatively small so that the entire substrate assembly does not fit into an image, it is advantageous to move the bond head 10 to different positions and record an image at each position containing a portion of the substrate assembly. The position and orientation of the substrate assembly is then determined based on these images.

In the first production mode, the position of the substrate positioning of the flip chip to be positioned is determined based on at least one image of the substrate positioning. In the second production mode, once a new substrate is supplied, its position is determined based on the substrate markings, and then the individual target positions of the flip chips are calculated with the help of geometric material data. This application is "wafer-level packaging" (WLB), in which the substrate is a wafer on which plastic is cast. The wafer does not contain any position markings for the individual substrate positioning, but does contain substrate markings attached close to the edge of the wafer.

In order to eliminate the positioning error of the flip chip 3 on the substrate caused by temperature changes, the position of the first optical mark 22 is determined in the calibration phase and updated at one or more predetermined time points in the following way:

moving the bond head 10 to a position where the first optical mark 22 is in the field of view of the second camera 15;

Recording images with a second camera 15;

determining a position of the first optical marker 22 based on the image and the second geometric data; and

The determined position is stored as the new position of the first optical marking 22 .

If the optical markings are covered by the plate 17 when the cavity 18 of the plate 17 is in position above the first camera 14 , the method further comprises moving the plate 17 to a position where the optical markings 22 , 23 are exposed before the positions of the optical markings 22 , 23 are updated.

Thus, the present invention utilizes the discovery that one or more optical marks attached to the camera support 16 to which the first camera 14 is fixed are sufficient to reduce the effect of variations between the pixel coordinate system of the first camera 14, the pixel coordinate system of the second camera 15 and the machine coordinate system of the bonding head 10 on the positioning of the flip chip 3 on the substrate positioning to a level that meets the current requirements.

If one or more further optical markers are present, for example the optical marker 23 , the position of these further optical markers is determined in a similar manner during the calibration phase and updated at the aforementioned point in time.

Advantageously, two pick and place systems are provided, each comprising a flip chip device 5 with a pick head 6, a first transport system 7 with a transport head 8, a second transport system 9 with a bonding head 10, a device 13 for wetting the flip chips with flux, and a first camera 14 and a second camera 15, which systems collect semiconductor chips 2 from the wafer stage 4 in an alternating manner and mount the semiconductor chips 2 as flip chips 3 on a substrate 11 arranged on a support body 12 in an alternating manner.

The method according to the invention offers the following advantages:

- The flip chip is placed in the cavity at the same location as the location from which it is removed, ensuring that the flux distribution in the cavity does not change as the cavity is moved from the first location to the second location and that the flip chip does not shift within the cavity, which could negatively impact wetting of the flip chip bumps or result in a decrease in the production throughput of the device.

The duration of immersion of the bumps of the flip chip in the flux can be adjusted independently of the other process steps. This is important for obtaining optimal wetting of the bumps of the flip chip on the one hand and for obtaining the highest possible throughput on the other hand.

- The throughput of the device is increased by the assembly of the transport head and the bonding head and by the simultaneous parallel operation of the transport head and the bonding head.

While the embodiments and applications of the present invention have been shown and described, it will be apparent to those skilled in the art having the benefit of this disclosure that many more modifications than those described above are possible without departing from the inventive concepts herein. Accordingly, the present invention is not limited thereto but is limited by the spirit of the claims and their equivalents.

Claims (4)

1. A method for mounting a semiconductor chip (2) with bumps (1) on a substrate positioning of a substrate (11), wherein first geometric data and second geometric data are determined in a calibration phase, wherein the first geometric data is used to convert pixel coordinates of a first camera (14) into machine coordinates of a machine coordinate system of a bonding head (10), and the second geometric data is used to convert pixel coordinates of a second camera (15) into machine coordinates of the machine coordinate system of the bonding head (10), and the following steps are performed for each semiconductor chip (2) in the mounting phase: Alternatively: the semiconductor chip (2) is provided at a predetermined location using a wafer stage (4); The provided semiconductor chip (2) is removed by the pick-up head (6) of the flip device (5), and the semiconductor chip (2) is rotated 180° to provide the semiconductor chip (2) as a flip chip (3); Alternatively: the semiconductor chip (2) is supplied by a feeding device as a flip chip (3); The flip chip (3) is received from the pick-up head (6) or the feeding device by the transport head (8); A cavity (18) is filled with flux, the cavity (18) being arranged in a plate (17) and formed of a transparent substrate, wherein the plate (17) is either fixedly arranged or moved after the cavity (18) is filled, such that the cavity (18) is located above the first camera (14) in both cases, the first camera (14) being fixedly arranged. The flip chip (3) is placed in the cavity (18), wherein the substrate on which the bump (1) faces the cavity (18); The first camera (14) records an image of the flip chip (3), and the actual position of the flip chip (3) relative to the machine coordinate system of the connector (10) is determined based on the image and the first geometric data. The flip chip (3) is removed from the cavity (18) using the connector (10); Using the second camera (15) fastened to the coupling head (10), the actual position of the substrate relative to the coupling head (10) in the machine coordinate system is determined by the following method: Or through: The connector (10) is moved to a position above the substrate positioning, where the substrate positioning is within the field of view of the second camera (15). At least one image is recorded using the second camera (15), and The actual position of the substrate is calculated based on the position of the substrate in the at least one image and the second geometric data; Or through: The actual position of the substrate is calculated using the actual positions of at least two substrate markers, wherein after the new substrate (11) is supplied to the support (12), the actual position of each of the at least two substrate markers is determined in the following manner: The bonding head (10) is moved to a position above the substrate, in which the substrate marking is within the field of view of the second camera (15). Images are recorded using the second camera (15), and The actual location of the substrate mark is determined using the image and the second geometric data; and The location to which the bonding head (10) will reach is calculated based on the determined actual position of the flip chip (3) and the determined actual position of the substrate positioning; and The bonding head (10) is moved to the calculated position and the flip chip (3) is placed on the substrate positioning, wherein The transport head (8) and the coupling head (10) move at least partially simultaneously. 2. The method according to claim 1, wherein the first geometric data includes the position of a first optical marker (22) and a first fixed vector, the first fixed vector indicating the direction of the first optical marker (22) from a reference point in the pixel coordinate system of the first camera (14) and the distance from the reference point in the pixel coordinate system of the first camera (14), and wherein the position of the first optical marker (22) is updated at least at a predetermined time point by: Move the connecting head (10) to a position where the first optical mark (22) is within the field of view of the second camera (15); Images are recorded using the second camera (15); The position of the first optical marker (22) is determined based on the image and the second geometric data; and The determined location is stored as the new location of the first optical mark (22). 3. The method according to claim 2, wherein the first geometric data includes the position of at least one additional optical marker (23) and an additional fixed vector indicating the direction of the additional optical marker to the reference point of the pixel coordinate system of the first camera (14) and the distance from the reference point of the pixel coordinate system of the first camera (14), and wherein the position of the additional optical marker (23) is updated by: Move the connecting head (10) to a position where the additional optical mark (23) is within the field of view of the second camera (15); Images are recorded using the second camera (15); The position of the additional optical marker (23) is determined based on the image and the second geometric data; and The determined location is stored as the new location of the additional optical marker (23). 4. The method according to claim 2 or 3, wherein the optical marker is covered by the plate (17) when the cavity (18) of the plate (17) is located above the first camera (14), the method further comprising moving the plate (17) to a position where the optical markers (22, 23) are exposed before the position of the optical markers (22, 23) is updated.