JP2001085363A - Method for manufacturing semiconductor device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a method for manufacturing a semiconductor device, capable of preventing a damage of a wafer or the device on the way of manufacturing. SOLUTION: This method for manufacturing a semiconductor device comprises a step of forming a wiring pattern and electrode pads 2 on a wafer 1, a step of assembling by each semiconductor device area 1A of the wafer, a step of forming grooves 1s for partitioning the wafer 1 at the surface 1a side formed with the pads 2 of the wafer 1 by each area 1A, and a step of grinding the surface 1b which is not formed with the pads 2 of the wafer 1 to separate the wafer 1 to the respective devices 10 by the grooves 1s.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for manufacturing a semiconductor device, and more particularly to a method for manufacturing individual semiconductor devices by dividing a wafer on which a wiring pattern and the like are formed and assembled for each semiconductor device area. How to do it.

[0002]

2. Description of the Related Art In recent years, a CSP (chip size package) type semiconductor device has been provided as a semiconductor device meeting the demand for smaller and thinner electronic devices.
Also, in this type of semiconductor device, the manufacturing method has been shifting from the conventional die level to the wafer level. That is, to manufacture this type of semiconductor device, first, a wiring pattern and an electrode pad are formed for each semiconductor device area of the wafer, and then the surface of the wafer where the electrode pads are not formed is ground (back-ground). Then, the wafer is formed thin to the thickness of the semiconductor chip. Thereafter, assembling is performed for each semiconductor device area of the wafer, and then the wafer is separated for each semiconductor device using a dicing saw or the like, whereby individual semiconductor devices are completed.

[0003]

In the above-mentioned conventional manufacturing method, since the thickness of the wafer is reduced by grinding, the assembly is performed for each semiconductor device area of the wafer. For example, when a wafer is transported in order to reduce the strength, there is a possibility that a wafer that has been thinly ground and has reduced strength may be damaged. In addition, in the above-described conventional manufacturing method, when a wafer is separated into individual semiconductor devices, the wafer that has been thinly ground and has reduced strength cannot withstand the impact of cutting by a dicing saw or the like, and the cut semiconductor device is cut. There was an inconvenience that damage such as cracks and chips occurred. SUMMARY OF THE INVENTION An object of the present invention is to provide a method of manufacturing a semiconductor device capable of preventing a wafer or a semiconductor device from being damaged during manufacturing in view of the above-mentioned situation.

[0004]

In order to achieve the above object, a method of manufacturing a semiconductor device according to the present invention comprises the steps of forming a wiring pattern and an electrode pad on a wafer, and assembling each semiconductor device area of the wafer. Performing a step of forming a groove on the surface of the wafer on which the electrode pads are formed on the side where the electrode pads are formed, and grinding the surface of the wafer on which the electrode pads are not formed. Separating the wafer for each semiconductor device by the groove.

[0005]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the drawings. In the method of manufacturing a semiconductor device according to the present invention, first, as shown in FIGS. 1A and 2, each semiconductor device area on the wafer 1 (a semiconductor chip is formed in an individual semiconductor device manufactured through each process described later). Wiring patterns (not shown) on each of the constituent parts 1A
And an electrode pad 2 are formed.

Thereafter, as shown in FIG. 1 (b) and FIG.
For each semiconductor device area 1A on the wafer 1, an assembling operation such as forming a solder bump 3 as an external connection terminal on the above-mentioned electrode pad 2 is performed.

Next, as shown in FIG. 1 (c) and FIG.
A groove 1 s having a predetermined depth is formed by a cutting means such as a dicing saw from the surface (device surface) 1 a on which the electrode pads 2 are formed on the wafer 1, and the wafer 1 is divided into each semiconductor device area 1 A. I do.

Here, the groove 1s is formed to a depth equal to or greater than the thickness of a semiconductor chip (see reference numeral 1c in FIGS. 1 and 5) formed from the wafer 1.

After the grooves 1s are formed in the wafer 1, as shown in FIG.
A support plate (support) H made of glass, plastic, ceramic, metal, or the like is provided on the surface (device surface) 1a on which is formed via a thermosetting or UV (ultraviolet) curable adhesive G. Attach.

The support plate H made of glass or the like described above.
Alternatively, a configuration is also possible in which a protective tape (not shown) as a support is attached to the surface (device surface) 1a of the wafer 1 on which the electrode pads 2 are formed.

As described above, after the support H is attached to the surface 1a of the wafer 1 where the electrode pads 2 are formed, the surface 1 of the wafer 1 where the electrode pads 2 are not formed is attached.
b is set to a predetermined size using a tool T as shown in FIG.
Specifically, grinding / polishing (back grinding) is performed until the thickness of the wafer 1 becomes the thickness of the semiconductor chip 1c.

Here, as described above, since the groove 1s formed in the wafer 1 is formed at a depth equal to or larger than the thickness of the semiconductor chip 1c, the electrode pad 2 is not formed on the wafer 1. By grinding / polishing the surface 1b by a predetermined dimension using the tool T, the wafer 1 is separated into individual semiconductor devices 10 by the grooves 1s as shown in FIG.

Thereafter, the support plate H is removed from each semiconductor device 10 and the adhesive G is removed, whereby each semiconductor device 10 is completed. Thus, a large number of semiconductor devices 10 are manufactured from one wafer 1 as shown in FIG.

In the above-described method for manufacturing a semiconductor device according to the present invention, after the step of assembling each semiconductor device area 1A of the wafer 1, the electrode pads 2
Since the step of grinding / polishing the surface 1b on which the wafer 1 is not formed is performed, when assembling is performed for each semiconductor device area 1A of the wafer 1, the wafer 1 is still reduced to the thickness of the semiconductor chip 1c. The wafer 1 is not thinned.
Since the wafer 1 has sufficient strength, the wafer 1 is not easily damaged even by a load such as transfer.

Further, in the method of manufacturing a semiconductor device according to the present invention, the wafer 1 is separated for each semiconductor device 10 by grinding / polishing the surface 1b of the wafer 1 on which the electrode pads 2 are not formed. Therefore, as compared with the conventional manufacturing method in which the wafer is separated by a cutting process using a dicing saw or the like, the semiconductor devices 10 can be separated without giving a large impact to the wafer 1, and the semiconductor device 10 thus cut out can be separated. Damage such as cracking or chipping of the semiconductor chip 1c is suppressed.

Further, in the method of manufacturing a semiconductor device according to the present invention, when grinding / polishing the surface 1b of the wafer 1 on which the electrode pads 2 are not formed, a support plate H made of glass or the like is attached to the wafer 1. Therefore, it is possible to prevent the wafer 1 from warping or cracking due to a load during grinding / polishing.

FIGS. 6 to 8 show examples in which the present invention is applied to the manufacture of a stacked semiconductor device. To manufacture a stacked semiconductor device, first, similarly to the method of manufacturing the semiconductor device 10 described above, a wiring pattern (not shown) and an electrode pad 2 ′ (see FIG. 7).

Thereafter, as shown in FIG. 1 (a) and FIG.
Surface on which electrode pads 2 'are formed on wafer 1'
(Device surface) From the side of 1a ', a hole (blind hole) 1h for a through hole of a predetermined depth is formed in each semiconductor device area 1A' of the wafer 1 'by a processing method such as laser processing, etching, stamping, drilling or the like. '.

The hole 1h 'is formed at a depth equal to or greater than the thickness of a semiconductor chip (see reference numeral 1c' in FIG. 6) formed from the wafer 1 '. The hole 1
The shape of h 'may be not only a cylindrical shape with a bottom but also a groove shape.

Next, each hole 1h 'formed in the wafer 1'
Is filled with a conductor 1m 'such as a conductive resin, a wire, or a cream metal. Note that the conductor 1 is inserted into each of the holes 1h '.
As a method for filling m ′, wire piercing with a bonder, application of a conductive resin, spin coating, printing, sputtering, or the like can be employed.

Thereafter, as shown in FIGS. 6 (c) and 8,
For each semiconductor device area 1A 'on the wafer 1', an assembling operation such as forming a solder bump 3 'as an external connection terminal on the above-mentioned electrode pad 2' and conductor 1m 'is performed.

Next, as shown in FIG. 6D and FIG.
Surface on which electrode pads 2 'are formed on wafer 1'
(Device surface) From the side of 1a ', a groove 1s' having a predetermined depth is formed by a cutting means such as a dicing saw, and the wafer 1 'is divided into each semiconductor device area 1A'. Here, the groove 1s' is formed at a depth equal to or greater than the thickness of a semiconductor chip (see reference numeral 1c 'in FIG. 6) formed from the wafer 1'.

After the grooves 1s' are formed in the wafer 1 ', as shown in FIG. 6 (e), an adhesive such as a thermosetting adhesive is applied to the surface 1a' of the wafer 1 'where the electrode pads 2' are formed. A support plate (support) H made of glass or the like is attached via G. In addition, instead of the support plate H made of glass or the like,
A thick protective tape (not shown) as a support may be attached to the wafer 1 '.

As described above, after the support H is attached to the wafer 1 ', the surface 1b' on which the electrode pads 2 'are not formed on the wafer 1' is removed by using the tool T as shown in FIG. Then, grinding / polishing (back grinding) is performed until a predetermined dimension, specifically, the thickness of the wafer 1 'is equal to the thickness of the semiconductor chip 1c'.

Since the groove 1s' formed in the wafer 1 'as described above is formed at a depth equal to or greater than the thickness of the semiconductor chip 1c', the surface 1b 'of the wafer 1' is By grinding / polishing only a predetermined size with the tool T,
As shown in FIG. 1E, the wafer 1 'is separated into individual semiconductor devices 10' by the grooves 1s'.

Since the hole 1hs' formed in the wafer 1 'as described above is formed at a depth equal to or greater than the thickness of the semiconductor chip 1c', the surface 1b 'of the wafer 1' is By grinding / polishing only a predetermined dimension by T,
The conductor 1m 'filled in the hole 1h' as shown in FIG.
Is exposed from the surface 1b 'of the wafer 1'.

After separating the wafer 1 'into individual semiconductor devices 10', the support plate H is removed from each semiconductor device 10 ',
By removing the adhesive G, a large number of semiconductor devices 10 ′, that is, the semiconductor chips 1
The stacked semiconductor device 10 'in which the surface and the back surface of c' are electrically connected by the conductor 1m '(through hole) is manufactured.

The stacked semiconductor device 10 'manufactured by the above-described manufacturing method is stacked and mounted on a mounting board B as shown in FIG. 9 to form a stacked package P. Each stacked semiconductor device 10' Solder bump 3 'and conductor 1
The layers can be connected to each other via m '.

Here, the stacked semiconductor device is expected to achieve a next-generation highly integrated package.
As shown in FIG. 10, there is a mounting mode in which the stacked semiconductor devices S are stacked on a mounting board B via a tape t on which a circuit is printed and a board D and mounted.

However, in such a configuration, for example, the thickness of the stacked semiconductor device S is 50 μm, and the thickness of the substrate D is 80 μm.
, The total thickness of the stacked package is 500
Thickness of about μm will result.

On the other hand, in the stacked package P using the stacked semiconductor device 10 'manufactured by the manufacturing method shown in FIGS. 6 to 8, an intermediate material such as a substrate is interposed between the stacked semiconductor devices 10'. Can be connected to each other without intervening, so that mounting with a thickness of about て of that of the stacked package shown in FIG. 10 is possible.

Therefore, in the stacked package P using the stacked semiconductor device 10 ′, since no intermediate material is used, the number of parts is reduced, thereby significantly reducing the manufacturing cost.
In addition to achieving large compactness, the possibility of higher integration of the stacked package is presented.

The stacked semiconductor device 10 shown in FIGS.
Is the same as the method for manufacturing the semiconductor device 10 shown in FIGS.
When assembling for each A ', the wafer 1' is not yet thinned to the thickness of the semiconductor chip 1c ', and the wafer 1' has sufficient strength. The wafer 1 'is not easily damaged by the load.

In the method of manufacturing the stacked semiconductor device 10 shown in FIGS. 6 to 8, similarly to the method of manufacturing the semiconductor device 10 shown in FIGS. 1 to 5, the wafer is separated by cutting using a dicing saw or the like. As compared with the conventional manufacturing method, each semiconductor device 1 does not have a large impact on the wafer 1 ′.
0 'can be separated, and damage such as cracking or chipping of the semiconductor chip 1c' of the semiconductor device 10 'cut out can be suppressed.

The method of manufacturing the stacked semiconductor device 10 shown in FIGS. 6 to 8 includes a step of forming a hole 1h 'for a through-hole in the wafer 1', and filling the hole 1h 'with a conductor 1m'. By grinding / polishing the surface 1b 'of the wafer 1' to a predetermined size with the tool T, the individual semiconductor devices 10 'are separated and through holes are formed in the semiconductor devices 10'. Therefore, the working efficiency related to the manufacture of the semiconductor device 10 'is greatly improved.

[0036]

As described in detail above, the method of manufacturing a semiconductor device according to the present invention comprises the steps of forming a wiring pattern and electrode pads on a wafer, and performing assembly for each semiconductor device area of the wafer. Forming a groove on the surface of the wafer on which the electrode pads are formed, and dividing the wafer into individual semiconductor device areas, and grinding the surface of the wafer on which the electrode pads are not formed by the grooves. Separating the wafer for each semiconductor device. In the above configuration, since the step of grinding the surface of the wafer on which the electrode pads are not formed is performed after the step of performing the assembly for each semiconductor device area of the wafer, the assembly is performed for each semiconductor device area of the wafer. When carrying out, the wafer is not yet thinned to the thickness of the semiconductor chip, and thus the wafer has sufficient strength, so that the wafer can be easily damaged even by a load such as transportation. Absent. Further, in the above configuration, since the wafer is separated for each semiconductor device by grinding the surface of the wafer where the electrode pads are not formed, the wafer is separated by cutting using a dicing saw or the like. As compared with the manufacturing method, the semiconductor device can be separated without giving a large impact to the wafer, and the semiconductor device cut out can be prevented from being damaged, such as cracking or chipping. As described above, according to the method of manufacturing a semiconductor device according to the present invention, it is possible to prevent damage to a wafer or a semiconductor device during manufacturing.

[Brief description of the drawings]

FIGS. 1A to 1E are conceptual diagrams showing working steps in a method for manufacturing a semiconductor device according to the present invention.

FIG. 2 is a plan view of a web showing operation steps in a method of manufacturing a semiconductor device according to the present invention.

FIG. 3 is a plan view of a web showing operation steps in a method of manufacturing a semiconductor device according to the present invention.

FIG. 4 is a plan view of a web showing operation steps in a method of manufacturing a semiconductor device according to the present invention.

FIG. 5 is a plan view showing a semiconductor device manufactured from one web by the method for manufacturing a semiconductor device according to the present invention.

FIGS. 6A to 6E are conceptual diagrams showing working steps in a method for manufacturing a semiconductor device according to the present invention.

FIG. 7 is a plan view of a web showing operation steps in a method of manufacturing a semiconductor device according to the present invention.

FIG. 8 is a plan view of a web showing operation steps in a method of manufacturing a semiconductor device according to the present invention.

FIG. 9 is a conceptual diagram showing a stacked package using a stacked semiconductor device manufactured by the method of manufacturing a semiconductor device according to the present invention.

FIG. 10 is a conceptual diagram showing a conventional stacked package using a stacked semiconductor device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Wafer, 1A ... Semiconductor device area, 1s ... Groove, 2 ... Electrode pad, 10 ... Semiconductor device, 1 '... Wafer, 1A' ... Semiconductor device area, 1h '... Hole, 1m' ... Conductor, 1s' ... Groove, 2 ': electrode pad, 10': semiconductor device, H: support plate (support).

Claims (3)

[Claims] A step of forming a wiring pattern and an electrode pad on a wafer; a step of assembling each semiconductor device area of the wafer; and a step of mounting the wafer on a surface of the wafer on which electrode pads are formed. Forming a groove for partitioning each semiconductor device area; grinding / grinding a surface of the wafer where no electrode pads are formed;
Polishing, and separating the wafer for each semiconductor device by the groove. 2. A step of grinding / polishing a surface of the wafer on which electrode pads are not formed and separating the wafer for each semiconductor device by the groove on the surface of the wafer on which electrode pads are formed. 2. The method for manufacturing a semiconductor device according to claim 1, wherein the method is performed with a support attached. 3. The semiconductor device according to claim 1, wherein the semiconductor device is a stacked semiconductor device having a through hole, the method including a step of forming a hole for the through hole in the wafer and a step of filling the hole with a conductor. 2. The method of manufacturing a semiconductor device according to claim 1, wherein