JP4681704B2 - Method for dividing semiconductor wafer - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method for dividing a semiconductor wafer in which cutting grooves are formed in advance on a street on the surface of a semiconductor wafer, and the back surface is ground to expose the cutting grooves, thereby dividing the wafer into individual chips.
[0002]
[Prior art]
Semiconductor chips such as ICs are widely used in various electronic devices, especially for products that are strongly demanded to be smaller, thinner and lighter, such as mobile phones, notebook personal computers, smart cards, etc. Therefore, there is a demand for further thinning of semiconductor chips used for these. As the semiconductor chip becomes thinner, not only the device such as a mobile phone is made thinner, but also a plurality of semiconductor chips can be stacked. It is also possible to improve.
[0003]
Normally, a semiconductor chip is formed by cutting and dicing a street of a semiconductor wafer having a predetermined thickness by grinding the back surface. However, when this method is used even when the semiconductor wafer is thinned, Chipping or the like is likely to occur in each semiconductor chip. Therefore, when this method is used, there is a limit to thinning the semiconductor wafer.
[0004]
Therefore, instead of such a method, before grinding the back surface of the semiconductor wafer, a cutting groove that does not penetrate the front and back surfaces of the semiconductor wafer is formed on the surface relatively shallowly, and a protective member is attached to the surface. A method of dividing the semiconductor wafer into individual semiconductor chips by setting the surface of the semiconductor wafer downward and grinding the back surface of the semiconductor wafer to expose the cutting grooves has been developed and put into practical use.
[0005]
[Problems to be solved by the invention]
However, in this method, grinding is performed until the cutting groove appears, and grinding is performed until the chip has a desired thickness after the cutting groove is exposed. May come off and enter a cutting groove, i.e., between adjacent chips, and in this case, the invading abrasive grains may cause chipping in the outer periphery of the chip, and thus damage the chip. There is a problem that sometimes.
[0006]
In addition, after the chips are ground to a predetermined thickness, they are transported to the next process while being stuck to the protective member, so that adjacent chips rub against each other at the time of transport, resulting in chipping. There is also a problem.
[0007]
Furthermore, after grinding the chip to a predetermined thickness, the grinding surface may be chemically etched to remove distortion generated on the grinding surface. In this case, etching is performed from the cutting groove. There is a problem that the agent penetrates and erodes the side surface or the surface (circuit surface) of the chip, thereby deteriorating the quality of the chip.
[0008]
In this way, before grinding the back surface of the semiconductor wafer, cutting grooves that do not penetrate the front and back surfaces of the semiconductor wafer are formed relatively shallow on the surface, and a protective member is attached to the surface so that the surface faces downward. When the chip is set in a grinding machine and the back surface of the semiconductor wafer is ground to divide the individual semiconductor chips by exposing the cutting grooves, the chips may be chipped due to abrasive grains entering the cutting grooves or rubbing between the chips. There is a problem in improving the quality of a semiconductor chip by preventing the occurrence of damage and damage, and further preventing the side surface and surface of the chip from being eroded by an etching agent.
[0009]
[Means for Solving the Problems]
As a specific means for solving the above-described problems, the present invention provides a semiconductor wafer that divides a semiconductor wafer that is partitioned by a street and has a plurality of circuit areas into chips for each circuit area along the street. A cutting method for cutting a semiconductor wafer street to form a cutting groove in the street, and an interference preventing member layer on a sticking surface on the surface of the semiconductor wafer on which the cutting groove is formed. A protective member adhering step for adhering the protective member on which the protective member is formed , and a surface side on which the protective member is adhered is placed facing the chuck table of the grinding apparatus, and the semiconductor wafer is held on the chuck table A semiconductor wafer held on the chuck table is attached to a grinding wheel constituting a grinding wheel mounted on a grinding device on the back surface of the semiconductor wafer held on the chuck table. Grinding the wafer to a predetermined thickness, thereby expose the cutting groove formed in該切Kezumizo forming step and a grinding step of dividing into individual chips, the interference prevention member layer, temperature changes and chemical It is composed of a material that becomes flexible by action, and when the protective member is adhered to the surface of the semiconductor wafer, it is bonded to the surface of the semiconductor wafer in a flexible state by heating, chemical reaction, or by using a solvent. When the pressure is applied, the interference prevention member constituting the interference prevention member layer enters the cutting groove, and the heating is stopped, the chemical atmosphere is removed, or the interference is prevented after a certain time elapses. Provided is a semiconductor wafer dividing method in which a prevention member is solidified, and the protective member attaching step and the cutting groove embedding step are simultaneously performed.
[0010]
In this semiconductor wafer dividing method, after the grinding step, a chemical etching process is performed on the back surface of the semiconductor wafer while the protective member is stuck on the surface of the semiconductor wafer divided by the grinding step. An additional requirement is to perform the etching process.
[0011]
According to the method for dividing a semiconductor wafer configured as described above, the back surface of the semiconductor wafer is cut until the cutting groove is exposed in order to fill the cutting groove with the interference preventing member before the back surface is ground. Even if the semiconductor chip is ground to a desired thickness, the abrasive grains constituting the grinding wheel do not enter the cutting groove, and the chip of the semiconductor chip due to the penetration of the abrasive grain into the cutting groove is eliminated. .
[0012]
Further, even when the semiconductor chip is transported in a state where it is adhered to the protective member, since the interference preventing member is filled between the adjacent semiconductor chips, the chipping caused by the semiconductor chips rubbing does not occur. .
[0013]
Further, when the grinding surface is chemically etched after the grinding process, the etching agent does not enter the cutting groove by performing the chemical etching process in a state where the interference preventing member is filled in the cutting groove. Therefore, the side surface of the semiconductor chip is not eroded, and the etching agent does not wrap around the surface through the cutting groove, so that the surface is not eroded. Only the grinding surface is eroded and the distortion can be removed.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described with reference to the drawings. In the present invention, first, a cutting groove that does not penetrate to the back surface is formed on the street of the surface of the semiconductor wafer by using the cutting apparatus 10 shown in FIG.
[0015]
When forming the cutting grooves in the semiconductor wafer W, a plurality of the semiconductor wafers W are accommodated in the cassette 11 of the cutting apparatus 10 shown in FIG. 1 and unloaded by the loading / unloading means 12 and then the chuck table 14 by the first conveying means 13. And is held with the surface facing up.
[0016]
The chuck table 14 holding the semiconductor wafer W is positioned immediately below the alignment means 15 by movement in the X-axis direction to detect a street where a groove is to be formed, and is further moved in the same direction while rotating at high speed. When the cutting means 17 having the above is lowered and the cutting blade 16 cuts into a predetermined depth, a cutting groove having a predetermined depth is formed in the detected street. Further, as the cutting means 17 moves in the Y-axis direction, the chuck table 14 reciprocates in the X-axis direction, whereby cutting grooves having a predetermined depth are sequentially formed in the street.
[0017]
Further, when the chuck table 14 is rotated 90 degrees to form the cutting grooves on the street in the same manner as described above, the cutting grooves 18 are formed vertically and horizontally on the surface of the semiconductor wafer W as shown in FIG. Cutting groove forming process).
[0018]
When the cutting groove 18 is thus formed on the surface of the semiconductor wafer W, the cutting groove 18 is then filled with an interference preventing member (interference preventing member filling step). The interference preventing member filling step is performed, for example, by sticking the protective member 20 to the surface of the semiconductor wafer W as shown in FIG.
[0019]
The protection member 20 is a member composed of a protection layer 21 and an interference prevention member layer 22. The interference prevention member layer 22 has adhesiveness, and the interference prevention member layer 22 is adhered to the surface of the semiconductor wafer W. As a result, the protective member 20 and the semiconductor wafer W are integrated as shown in FIG. 4 (protective member attaching step).
[0020]
The pressure-sensitive adhesive constituting the interference preventing member layer 22 is made of a material that becomes flexible due to temperature change or chemical action, and the surface of the semiconductor wafer W in a state that becomes flexible by heating, chemical reaction, or by using a solvent. Then, as shown in FIG. 5, the interference preventing member 22a enters the cutting groove 18 and is filled.
[0021]
Then, the filled interference preventing member 22a is solidified by stopping heating, removing the chemical atmosphere, or passing a certain time, and is filled in the cutting groove 18 without a gap. Thus, when the protective member 20 including the protective layer 21 and the interference prevention member layer 22 is attached to the semiconductor wafer W, the interference prevention member filling step and the protection member attachment step are performed simultaneously. Therefore, productivity can be improved.
[0022]
The interference preventing member filling step can also be performed using a spin coater 30 shown in FIG. The spin coater 30 is generally composed of a rotation driving unit 31, a holding unit 32 connected to the rotation driving unit 31, and a dropping unit 33. The holding unit 32 holds a semiconductor wafer W having a cutting groove formed on the surface thereof with the surface facing up. Is done.
[0023]
Then, the holding unit 32 is driven to rotate at a high speed by being driven by the rotation driving unit 31, and the curable liquid resin 34 is spread over the entire surface of the semiconductor wafer W by dropping the curable liquid resin 34 from the dropping unit 33. As shown, the cutting groove 18 is filled. Further, the curable liquid resin 34 filled in the cutting groove 18 is cured with time. Thus, the curable liquid resin 34 becomes an interference preventing member.
[0024]
After the cutting groove 18 is filled with the curable liquid resin 34 as an interference preventing member as described above, as shown in FIG. 8, unlike the protective member 20 shown in FIG. The member 40 is stuck on the surface of the semiconductor wafer W (protective member sticking step). Thus, when filling the cutting groove 18 with the curable liquid resin 34 as an interference preventing member using the spin coater 30, the interference preventing member filling step and the protective member attaching step are performed in separate steps.
[0025]
Next, as shown in FIG. 5, the back surface of the semiconductor wafer W in which the cutting groove 18 is filled with the interference preventing member 22a and the protective member 20 is adhered is ground using, for example, a grinding device 50 shown in FIG. .
[0026]
In the grinding device 50 of FIG. 9, a wall body 52 is provided upright from the end of the base 51, and a pair of rails 53 are arranged in the vertical direction on the inner surface of the wall body 52, As the slide plate 54 moves up and down along the rail 53, the grinding means 55 attached to the slide plate 54 moves up and down. Further, a chuck table 57 that holds a semiconductor wafer is provided on a turntable 56 that is rotatably arranged on the base 51.
[0027]
In the grinding means 55, a grinding wheel 60 is mounted on a mounter 59 at the tip of a spindle 58 having a vertical axis, and a grinding wheel 61 is fixed to the lower part of the grinding wheel 60 in an annular shape. The grinding wheel 61 is configured to rotate as the spindle 58 rotates.
[0028]
When grinding the back surface of the semiconductor wafer W by using the grinding device 50, the semiconductor wafer W is held with the front surface facing the chuck table 57 (holding step). Then, the chuck table 57 is positioned immediately below the grinding means 55, the spindle 58 is rotated, and the grinding means 55 is lowered. As the spindle 58 rotates at a high speed, the grinding wheel 60 rotates at a high speed, and the grinding wheel 61 that rotates at high speed comes into contact with the semiconductor wafer to apply a pressing force, whereby the back surface is ground by the grinding wheel 61.
[0029]
Then, when the back surface is ground by a predetermined amount, as shown in FIG. 10, the cutting groove 18 appears and penetrates the front and back surfaces, and the semiconductor wafer W is divided into individual semiconductor chips C by this cutting groove 18 (grinding step). ).
[0030]
Further, the grinding is continued in order to make each semiconductor chip C have a desired thickness. At this time, since the cutting groove 18 is filled with the interference preventing member 22a, even if the abrasive grains constituting the grinding wheel 61 are detached, the cutting groove 18 does not enter the cutting groove 18. Accordingly, no chipping occurs in the outer peripheral portion of the semiconductor chip C.
[0031]
After the semiconductor chip C has reached a desired thickness, it is transported to the next pick-up process while being adhered to the protective member 20, but the cutting groove 18 is filled with the interference preventing member 22a in the transport process. Therefore, adjacent semiconductor chips do not contact each other during the conveyance process.
[0032]
Further, when picking up the semiconductor chips C one by one in the pickup process, the side surfaces of the semiconductor chips C are not bonded to the interference preventing member 22a, and the interference preventing member 22a is integrated with the protective member 20. Therefore, only the semiconductor chip C can be easily peeled off.
[0033]
When distortion occurs on the ground surface of the semiconductor wafer W in the grinding process, the distortion is removed by performing chemical etching on the ground surface while the protective member 20 is stuck on the surface. At this time, since the protective member 20 remains attached to the surface, the cutting groove 18 is filled with the interference preventing member 22a. Therefore, since the etching agent does not enter the cutting groove 18, the side surface of the semiconductor chip C is not eroded, and the etching agent does not wrap around the surface through the cutting groove 18, so that the surface is eroded. It is never done. That is, the etching agent acts only on the grinding surface, and only the grinding surface is eroded to remove distortion, so that the quality of the semiconductor chip C can be improved.
[0034]
【The invention's effect】
As described above, according to the method for dividing a semiconductor wafer according to the present invention, since the interference preventing member is filled in the cutting groove before grinding the back surface, the back surface of the semiconductor wafer is cut until the cutting groove appears, Furthermore, even if each semiconductor chip is ground to a desired thickness, the abrasive grains constituting the grinding wheel do not enter the cutting groove. Therefore, there is no chipping caused by such penetration of abrasive grains, and the quality of the semiconductor chip can be improved.
[0035]
Further, even when the semiconductor chip is transported in a state where it is adhered to the protective member, since the interference preventing member is filled between the adjacent semiconductor chips, the chip caused by the semiconductor chips rubbing does not occur. Also in this respect, the quality of the semiconductor chip can be improved.
[0036]
Further, when the grinding surface is chemically etched after the grinding process, the etching agent does not enter the cutting groove by performing the chemical etching process in a state where the interference preventing member is filled in the cutting groove. Therefore, the side surface of the semiconductor chip is not eroded, and the etching agent does not go around the surface through the cutting groove, so that the surface is not eroded. Therefore, the etching agent acts only on the grinding surface of the semiconductor chip, and only the grinding surface is eroded to remove distortion, and further the quality of the semiconductor chip can be improved, and the semiconductor chip is polished thinly to around 50 μm. Can do.
[Brief description of the drawings]
FIG. 1 is a perspective view showing an example of a cutting apparatus used in a cutting groove forming step constituting a semiconductor wafer dividing method according to the present invention.
FIG. 2 is a perspective view showing a semiconductor wafer having a cutting groove formed on the surface by the cutting groove forming step.
FIG. 3 is a perspective view showing the semiconductor wafer and a protective member attached to the semiconductor wafer.
FIG. 4 is a perspective view showing a state in which a protective member is attached to the semiconductor wafer in a protective member attaching step constituting the semiconductor wafer dividing method according to the present invention.
FIG. 5 is a cross-sectional view showing a state in which an interference preventing member is filled in a cutting groove by the protective member attaching step and the interference preventing member filling step.
FIG. 6 is a schematic perspective view showing an example of a spin coater used in the second example of the interference preventing member filling step constituting the semiconductor wafer dividing method according to the present invention.
FIG. 7 is a cross-sectional view showing a semiconductor wafer in which a cutting groove is filled with an interference preventing member using the spin coater.
FIG. 8 is a cross-sectional view showing a state in which a protective member is attached to a semiconductor wafer in which the cutting groove is filled with an interference preventing member.
FIG. 9 is a perspective view showing an example of a grinding apparatus used in a grinding process constituting the semiconductor wafer dividing method according to the present invention.
FIG. 10 is a cross-sectional view showing a semiconductor wafer divided into individual semiconductor chips by the grinding step.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 ... Cutting apparatus 11 ... Cassette 12 ... Carry-in / out means 13 ... First conveyance means 14 ... Chuck table 15 ... Alignment means 16 ... Cutting blade 17 ... Cutting means 18 ... Cutting groove 20 ... Protection member 21 ... Protection layer 22 ... Interference Prevention member layer 22a ... Interference prevention member 30 ... Spin coater 31 ... Rotation drive part 32 ... Holding part 33 ... Drip part 34 ... Curable liquid resin 40 ... Protective member 50 ... Grinding device 51 ... Base 52 ... Wall body 53 ... Rail 54 ... Slide plate 55 ... Grinding means 56 ... Turn table 57 ... Chuck table 58 ... Spindle 59 ... Mounter 60 ... Grinding wheel 61 ... Grinding wheel

Claims (2)

A method of dividing a semiconductor wafer in which a semiconductor wafer partitioned by a street and formed with a plurality of circuit areas is divided into chips for each circuit area along the street,
A cutting groove forming step of cutting a street of a semiconductor wafer to form a cutting groove in the street;
A protective member adhering step of adhering a protective member having an interference preventing member layer formed on the adhering surface to the surface of the semiconductor wafer in which the cutting grooves are formed;
A holding step in which the surface side to which the protective member is attached is placed facing the chuck table of the grinding device, and the semiconductor wafer is held on the chuck table;
The semiconductor wafer held on the chuck table is ground to a predetermined thickness by applying a grinding wheel constituting a grinding wheel mounted on a grinding device to the back surface of the semiconductor wafer, and formed in the cutting groove forming step. A grinding step of exposing the cutting grooves and dividing them into individual chips,
The interference preventing member layer is made of a material that becomes flexible by temperature change or chemical action, and is flexible by heating or chemical reaction or by using a solvent when the protective member is stuck to the surface of the semiconductor wafer. In this state, the interference prevention member constituting the interference prevention member layer enters the cutting groove by applying pressure to the surface of the semiconductor wafer , and stops heating or removes the chemical atmosphere. Or a method of dividing a semiconductor wafer in which the interference preventing member is solidified after a predetermined time and the protective member attaching step and the cutting groove embedding step are simultaneously performed.   2. The etching step of performing a chemical etching process on the back surface of the semiconductor wafer while the protective member is stuck on the surface of the semiconductor wafer divided by the grinding step after the grinding step. A method for dividing a semiconductor wafer as described.

Images