JP2003142892A - Method and apparatus for mounting - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately mount by accurately aligning an upper surface side recognition mark of a first material to be connected at a recognition mark of a lower second material to be connected. SOLUTION: The method for mounting comprises the steps of aligning the first material to be connected with a recognition mark for aligning, at the upper surface side to the second material to be connected with a recognition mark for aligning with the first material to be connected and arranged downward; and then mounting the first material on the second material. The method further comprises the steps of reading and storing the relative positional relationship between the recognition mark of the upper surface side of the first material, and the recognition mark adhered to the lower surface side or the profile silhouette of the first material before mounting; reading the lower surface side recognition mark of the first material or the profile silhouette of the first material and the mark of the second material by recognition means of 2 fields inserted between the first material, and the second material at the time of aligning for mounting; and aligning the recognition mark of the second material with the upper surface side recognition mark of the first material as a reference based on the stored relative positional relationship. The apparatus for mounting is provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mounting method and a mounting apparatus for bonding objects to be bonded to each other, and more particularly, to a lower surface of a first object to be bonded having a recognition mark for alignment on its upper surface side. The present invention relates to a mounting method and a mounting device that are mounted after being aligned with a second object to be bonded that has a recognition mark for alignment with the first object to be bonded.

[0002]

2. Description of the Related Art When joining objects to be joined together, for example, when mounting a chip on a substrate, it is necessary to align the chip and the substrate in a predetermined relationship before joining them together. This alignment is usually performed by reading a recognition mark for alignment on the chip and a recognition mark for alignment on the substrate side by a recognition means such as a camera to determine the relative positional relationship between both marks. Is carried out within a predetermined accuracy.

Before joining, for example, when the recognition mark giving surface of the chip arranged above and the recognition mark giving surface of the substrate arranged below face each other, a recognition means, for example, two visual fields, is provided between them. It is possible to insert the recognition means and read the upper and lower recognition marks to align the relative positions of the chip and the substrate.

However, when a recognition mark for alignment is provided on the upper surface of the chip, or when an alignment reference exists inside the chip (for example, an active layer that emits light is provided inside the chip as an alignment reference). In the case of an optical element), the recognition mark on the chip and the substrate cannot be read by the above-described two-view recognition means disposed between them.

[0005]

The object of the present invention is to pay attention to such a problem, and to recognize a recognition mark or alignment reference for alignment on the upper surface or inside (hereinafter collectively referred to simply as "upper surface side recognition mark"). The upper surface side recognition mark of the first object to be bonded is
To provide a mounting method and a mounting device capable of accurately aligning with the recognition mark of the article to be joined, and mounting the first article of such a form with high precision. is there.

[0006]

In order to solve the above-mentioned problems, in the mounting method according to the present invention, a first object to be joined having a recognition mark for alignment on the upper surface side is arranged below. A method of mounting after aligning with a second object to be joined, which is provided with a recognition mark for alignment with the first object to be joined, which is mounted on the upper surface side of the first object to be joined before mounting. The relative positional relationship between the recognition mark and the recognition mark on the lower surface side or the outer shape silhouette of the first object to be joined is read and stored, and at the time of alignment for mounting, the first object to be joined and the second object to be joined are aligned. By the two-view recognition means inserted between the object and the object to be joined, the recognition mark on the lower surface side of the first object to be joined or the outline silhouette of the first object to be joined and the recognition mark on the second object to be joined are recognized. The first contacted area is read based on the stored relative positional relationship. The reference to the upper surface recognition marks of the object 2
The method is characterized by performing alignment with the recognition mark of the article to be joined.

In the above description, the "upper surface side" of the first object to be bonded with the positioning recognition mark means, as described above, the case where the positioning recognition mark is applied to the upper surface itself. The case in which the alignment reference corresponding to the recognition mark for alignment is provided inside the article 1 to be bonded is included. For example, the alignment reference includes an active layer that emits light, which is present inside an optical element or the like.

The first object to be joined is, for example, a chip, and the second object to be joined is, for example, a substrate. However, in the present invention, the chip means, for example, an IC chip,
All forms on the side to be bonded to the substrate are shown regardless of the type and size such as semiconductor chips, optical devices, surface mount components, and wafers. Further, in the present invention, the term "substrate" refers to, for example, resin substrates, glass substrates, film substrates, chips, wafers, etc., and all forms on the side to be joined to chips regardless of type or size.

In this mounting method, the relative positional relationship is determined by the recognition means arranged above and below the first object to be bonded.
It can be read in advance before implementation. At this time, it is preferable to calibrate the relative positional relationship between the upper and lower recognition means in advance by using a calibration object to which reference marks that can be read by the recognition means arranged above and below are attached.

The relative positional relationship can also be read by transmitting infrared light to the first object to be joined. Since the transmitted infrared light is used, only one recognition means is required at this time. That is, it is possible to read the upper surface side recognition mark and the lower surface side recognition mark of the first object to be bonded or the outer shape silhouette of the first object to be bonded by a single recognition means.

When reading the relative positional relationship between the upper surface side recognition mark of the first object to be joined and the outer shape silhouette, for example, the upper surface side recognition mark of the first object to be joined is read using coaxial light. The oblique silhouette can be used to read the outline silhouette of the first object to be joined. In this case, the single recognition means can switch the coaxial light and the oblique light to read the upper surface side recognition mark and the outer shape silhouette of the first object to be bonded.

In the mounting apparatus according to the present invention for carrying out such a mounting method, a first object to be bonded having a recognition mark for alignment on the upper surface side is arranged below the first object to be bonded. A device for mounting after aligning with a second object to be joined, which is provided with a recognition mark for aligning with the object, wherein the recognition mark on the upper surface side of the first object and the lower surface before mounting. Means for reading and storing the relative positional relationship between the recognition mark provided on the side or the outer shape silhouette of the first object to be joined, and the first object to be joined and the second object to be joined at the time of alignment for mounting. By the two visual field recognition means inserted between the object and the object, the lower surface side recognition mark of the first object to be joined or the outer shape silhouette of the first object to be joined and the recognition mark of the second object to be joined are read, A first contacted portion based on the stored relative positional relationship. The reference to the upper surface recognition marks of the object 2
Means for aligning with the recognition mark of the article to be joined,
It is characterized by having.

In the mounting method and mounting apparatus according to the present invention as described above, the recognition mark on the upper surface side of the first object and the recognition mark on the lower surface side or the outer shape silhouette of the first object to be bonded. The relative positional relationship with the object is read and stored in advance, that is, before actual mounting, and at the time of alignment for mounting, the lower surface side recognition mark of the first object to be bonded or the first object to be bonded. The outline silhouette and the recognition mark of the second object to be joined are read, and alignment is performed based on the stored relative positional relationship on the basis of the upper surface side recognition mark of the first object to be joined. The upper surface side recognition mark of the object to be bonded is aligned with the recognition mark of the second object to be bonded. Therefore, it becomes possible to perform alignment with the upper surface side recognition mark reference of the first object to be bonded, which could not be achieved by the conventional method, and it is possible to mount with high accuracy.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT A preferred embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows a mounting apparatus according to one embodiment of the present invention. In this embodiment, the mounting apparatus 1 has a recognition mark 4 for alignment on its upper surface.
A device for mounting a chip 2 as a first object to be bonded, which is attached to a substrate 3 as a second object to be bonded, which is arranged below and has a recognition mark 5 for alignment with the chip 2. It is configured. The chip 2 is held on the tool 6 by suction from its upper surface side. The suction is performed by, for example, air suction through a suction hole 7 provided in the tool 6 and opened on the lower surface of the tool 6. However, this chip holding means is not limited to adsorption holding means, electrostatic holding means by static electricity, magnetic holding means by magnets or magnetic fields, mechanical means for holding the chip by a plurality of movable claws, and a single movable claw for holding the chip. Any means such as a pressing means may be used.

The substrate 3 is held on the stage 8 from its lower surface side. The means for holding the substrate 3 by the stage 8 may be any means such as a suction holding means, an electrostatic holding means, a magnetic holding means, a plurality of movable claws or a mechanical means with a single movable claw. In this embodiment, the stage 8 can be controlled in position and attitude in the horizontal direction (X, Y directions) and the rotation direction (θ direction), and accordingly, the relative position of the substrate 3 with respect to the chip 2 is adjusted. To be done. Tool 6 holding the tip 2
In this embodiment, the movement can be controlled in the vertical direction (Z direction), and accordingly, the chip 2 is pressed against the substrate 3. In addition to pressurization, heating can be performed by a heater (not shown) built in the tool 6.

FIG. 1 shows a state in which the chip 2 and the substrate 3 are aligned for mounting.
In this embodiment, the recognition means 9 having two visual fields is inserted. The two-field-of-view recognizing means 9 reads the recognition mark 10 on the lower surface of the chip 2 located above it or the outline silhouette of the chip 2 and recognizes the recognition mark 5 on the substrate 3 located below.
To read. Then, based on the relative positional relationship between the upper surface side recognition mark 4 of the chip 2 and the lower surface side recognition mark 10 or the outer shape silhouette that has been read and stored in advance (that is, before this alignment), the upper surface side recognition mark 4 is used as a reference. Then, the chip 2 and the substrate 3 are aligned, and the aligned chip 2 is mounted on the substrate 3.

The above-mentioned relative positional relationship stored in advance is
It is read by the following various methods.

In the method shown in FIG. 2, the chip 2 to be mounted is previously transferred from the chip tray 11 onto, for example, a transparent stage 12 (for example, a transparent glass stage), and the upper mark recognizing means 13 is provided. And lower mark recognition means 1
The upper surface side recognition mark 4 and the lower surface side recognition mark 10 of the chip 2 are read by 4, and the relative positional relationship between both marks is read and stored. At this time, the chip 2 may be held by a transparent tool (not shown) from the upper surface side.

When reading the relative positional relationship between both marks, it is necessary, as a premise, to accurately grasp the relative positional accuracy of both mark recognition means 13 and 14. Therefore, it is preferable that the relative positional relationship between the upper and lower mark recognition means 13 and 14 is calibrated in advance. The calibration between the both mark recognition means 13 and 14 is performed by the calibration mark 15 provided on the transparent stage 12, as shown in FIG. 2, for example.
Is read by both recognizing means 13 and 14, the amount of positional deviation between the two is input and corrected later, or the position is adjusted so that the positional deviation disappears in accordance with the detected positional deviation. Can be done by

In the above example, the relative positional relationship between the recognition mark 4 on the upper surface side of the chip 2 and the recognition mark 10 on the lower surface side is read, but as shown in FIG. 3, the recognition mark 4 on the upper surface side of the chip 2 is read. And the outer shape silhouette 16 of the chip 2 (that is, the outer shape silhouette as a shade on the immediate back side of the chip 2), or although not shown,
It is also possible to read the relative positional relationship with the corner portion or the edge portion in the outer shape of the chip 2. In particular, when reading the relative positional relationship with the outer shape silhouette 16 of the chip 2, as shown in FIG. 4, when the outer shape silhouette 16 is read by projection, even if the edge surface 2a of the chip 2 is inclined. , The projected outer shape (outer shape silhouette) from the top and bottom is the same. Therefore, even if the reading direction of the outer shape silhouette at the time of grasping the relative positional relationship in advance and the reading direction of the outer shape silhouette at the time of alignment for later mounting are reversed, the same outer shape silhouette can be read. .

The outer silhouette 16 of the chip 2 can be read by any of the methods shown in FIGS. In the method shown in FIG. 5, the outer shape silhouette 16 projected by the light from the light source 17 can be read by the recognition means 14. In the method shown in FIG. 6, for example, the outer shape silhouette 16 of the chip 2 held by the transparent tool 18 can be read by using the reflecting surface 18 a of the tool 18 instead of the projection.

Further, as shown in FIG. 7, for example, "Gel pack"("Gelpack" is a trade name, and "Gel pack" chip and wafer carrier manufactured by GEL-PAK LLC (USA)) are used. Dust-free case) ("Gel-Pak" Chip / wafer Carrier & Shipping Syste
m)), the upper surface side recognition mark 4 of the chip 2 is read by the coaxial light 21 from the recognition means 20, and the recognition means 20
It is also possible to read the outline silhouette (or the position of the edge) using the oblique light 22 from. In this case, the same recognition unit 20 can switch the coaxial light 21 and the oblique light 22 to read the upper surface side recognition mark 4 and the outline silhouette.

Further, as shown in FIG. 8, the relative positional relationship can be read by transmitting infrared light to the chip. For example, an infrared light source 24 is disposed above the chip 2 held on the stage 23, and the infrared light 25 emitted from the light source 24 causes the upper surface side recognition mark 4 and the lower surface side recognition mark 10 of the chip 2 to move. Both recognition means 26
Can be read by. Further, in this case, even when there is an alignment reference inside the chip 2, for example, even in the case of the above-described optical element, the relative positional relationship between the internal alignment reference and the lower surface side recognition mark 10 is determined by the transmitted infrared light. Can be read. Especially when this infrared transmitted light is used, only one recognition means 26 is required.
In addition to infrared light, for example, X-rays, electromagnetic waves, sound waves, etc. may be used as long as they can pass through the chip and recognize the recognition mark attached to the chip.

Further, in this case, the recognition mark 4 or the internal alignment reference is read with infrared light, and the lower recognition mark or outline silhouette is read by projection of normal light (visible light) or a light source from below. You can also do it. Further, if the infrared light cut filter 27 is provided so as to be able to move in and out, it becomes possible to switch the irradiation light from the same light source 24 between infrared light and normal light, and the external shape silhouette as described above can be obtained by projection with normal light. Can be read.

As described above, in the present invention, the upper surface side recognition mark 4 or the internal alignment reference of the chip 2 and the lower surface side recognition mark 10 or the outer shape silhouette 1 are previously prepared.
The relative positional relationship with 6 is read, stored, and provided for alignment for mounting. Therefore, a series of operations of this method is shown in FIG. 9 for the relative positional relationship reading method shown in FIG. That is, the chip 2 transferred from the chip tray 11 to the transparent stage 12 is measured as shown in FIG. 2, and then, moved from the transparent stage 12 to the tool 6 and held by the tool 6. As shown in FIG. 1, the chip 2 held by the tool 6 is read by the two-field-of-view recognition means 9 while the lower surface side recognition mark 10 of the chip 2 located above or the outline silhouette of the chip 2 is read and The recognition mark 5 on the substrate 3 located at is read. Then, based on the relative positional relationship between the stored upper surface side mark 4 or the internal alignment reference of the chip 2 and the lower surface side mark 10 or the outer shape silhouette, the upper surface side mark 4 or the internal alignment reference is used as a reference and the substrate 3 Alignment with the recognition mark 5 is performed. Therefore, even if the upper surface side recognition mark 4 or the internal alignment reference is not directly read at the time of this alignment, extremely high precision alignment becomes possible and high precision mounting after alignment becomes possible.

In the example shown in FIG. 9, the transparent stage 1
Although the starting point of 2 is set between the axes of the mark recognizing means 13 and 14, the starting point is not limited to this and the starting point may be on the chip tray 11 side. For example, as shown in FIG.
When the transparent stage 12 on which the chip 2 is mounted moves from the starting point on the chip tray 11 side to below the head 6, when the transparent stage 12 passes between the mark recognizing means 13 and 14, the lower surface side recognition mark 10 of the chip 2 is instantly obtained. Alternatively, the outline silhouette of the chip 2 can be recognized.

Further, as shown in FIG. 11, even at the time of alignment for mounting, the outer shape silhouette 16 of the upper chip 2 is recognized by, for example, the two-field-of-view recognizing means 28.
For example, the oblique light 29 can be used for reading.

[0028]

As described above, according to the mounting method and the mounting apparatus of the present invention, the relative positional relationship between the upper surface side recognition mark and the lower surface side recognition mark or the outer shape silhouette of the first object to be bonded in advance. The upper surface side recognition mark is used as a reference to perform alignment with the recognition mark of the second object to be bonded based on the stored relative positional relationship. Even if the side recognition mark is not directly read, both objects can be joined with high accuracy.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a mounting apparatus according to an embodiment of the present invention.

FIG. 2 is a schematic configuration diagram showing an example of means for obtaining a relative positional relationship in advance.

FIG. 3 is a schematic configuration diagram showing a method for reading an outer shape silhouette of a first object to be joined.

FIG. 4 is a schematic configuration diagram showing a state in which the edge surface of the first object to be bonded is inclined by the method of FIG.

FIG. 5 is a schematic configuration diagram showing an example of a method of reading an outline silhouette by projection.

FIG. 6 is a schematic configuration diagram showing an example of a method of reading an outline silhouette by reflection.

FIG. 7 is a schematic configuration diagram showing an example of a method using coaxial light and oblique light.

FIG. 8 is a schematic configuration diagram showing an example of a method using infrared light.

FIG. 9 is a schematic perspective view showing the entire example of the method according to the present invention.

FIG. 10 is a schematic side view showing another overall example of the method according to the present invention.

FIG. 11 is a schematic configuration diagram showing a manner of reading an outer shape silhouette of a first object to be bonded during alignment for mounting.

[Explanation of symbols]

1 Mounting device 2 Chip as the first object to be bonded 2a Edge surface 3 Substrate as second object to be bonded 4 Top side recognition mark 5 Board identification mark Tool to hold 6 tips 7 suction holes Stage for 8 substrates 9 2 Field of view recognition means 10 Bottom side recognition mark 11 chip tray 12 transparent stage 13 Top mark recognition means 14 Lower mark recognition means 15 Calibration mark 16 outline silhouette 17 Light source 18 transparent tools 18a reflective surface 19 chip tray 20 Recognition means 21 Coaxial light 22 Oblique 23 stages 24 Infrared light source 25 infrared light 26 Recognition means 27 Infrared light cut filter 28 2 Vision recognition means 29 Oblique

Claims (8)

[Claims] 1. A first object to be bonded having a recognition mark for alignment on the upper surface side, and a second object which is arranged below and has a recognition mark for alignment with the first object to be bonded. A method of mounting after aligning with a bonded object, the recognition mark on the upper surface side of the first object to be bonded and the recognition mark provided on the lower surface or the outer shape of the first object to be bonded before mounting. The relative positional relationship with the silhouette is read and stored, and at the time of alignment for mounting, the first visual field recognition means inserted between the first object to be bonded and the second object to be bonded causes the first visual field to be recognized. The lower surface side recognition mark of the object to be welded or the outer shape silhouette of the first object to be welded and the recognition mark of the second object to be welded are read, and the first position is read based on the stored relative positional relationship.
2. A mounting method characterized by performing the alignment with the recognition mark of the second object to be bonded on the basis of the recognition mark on the upper surface side of the object to be bonded. 2. The mounting method according to claim 1, wherein the relative positional relationship is read by a recognition unit arranged above and below the first object to be bonded. 3. The mounting method according to claim 2, wherein the relative positional relationship between the upper and lower recognition means is calibrated in advance by using a calibration object to which reference marks that are readable by the upper and lower recognition means are attached. . 4. The relative positional relationship is read by transmitting infrared light to the first object to be joined.
How to implement. 5. The mounting method according to claim 4, wherein the upper surface recognition mark and the lower surface recognition mark of the first object to be bonded or the outer shape silhouette of the first object to be bonded are read by a single recognition means. 6. When reading the relative positional relationship, the upper surface side recognition mark of the first object to be joined is read using coaxial light, and the outline silhouette of the first object to be joined is read using oblique light. 1 mounting method. 7. The mounting method according to claim 6, wherein the single recognition means switches the coaxial light and the oblique light to read the upper surface side recognition mark and the outer shape silhouette of the first object to be bonded. 8. A first object to be bonded having a recognition mark for alignment on the upper surface side, and a second object to be bonded, which is arranged below and has a recognition mark for alignment with the first object to be bonded. A device which is mounted after being aligned with respect to a bonded object, and before mounting, a recognition mark on the upper surface side of the first object to be bonded and a recognition mark attached to the lower surface side or the outer shape of the first object to be bonded. Means for reading and storing the relative positional relationship with the silhouette,
At the time of alignment for mounting, by the recognition means of the two visual fields inserted between the first and second objects to be joined,
The lower surface side recognition mark of the first object to be joined or the outer shape silhouette of the first object to be joined and the recognition mark of the second object to be joined are read, and based on the stored relative positional relationship, the first And a means for performing alignment with the recognition mark of the second object to be bonded with the recognition mark on the upper surface side of the object to be bonded as a reference.