JPH11188589A - Device and method for mirror finishing semiconductor wafer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately and efficiently perform a mirror finishing of a wafer chamfered part. SOLUTION: A semiconductor wafer mirror finishing device comprises a transfer loader 4 which takes a wafer 2 having a chamfered part 2a to be mirror-finished into at least one finishing station ST4 or ST5 and takes a finished wafer 2 out of the finishing stations ST4 and ST5, a polishing drum 12 which is located at the finishing stations ST4 and ST5 and oscillatably moved continuously or intermittently in axial direction by an oscillation means, at least one chuck 9 or 10 which is installed near the polishing drum 12 and holds the wafer 2 to be mirror-finished, and an oscillation means which moves and oscillates the chuck 9 or 10 in two axis directions perpendicular to the axial direction of the polishing drum. Thus the wafer chamfered part can be mirror-finished accurately and efficiently.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and a method for mirror-finishing a semiconductor wafer whose peripheral edge is chamfered.

[0002]

2. Description of the Related Art Conventionally, semiconductor wafers used for integrated circuits and the like have been manufactured by thinly slicing an ingot made of silicon single crystal, but recently, in order to prevent occurrence of pitching at an edge portion, The chamfered edge is mirror-finished to prevent chamfering of the edge and increase the strength of particles.

A method and apparatus for chamfering or mirror-finishing a semiconductor wafer include, for example, Japanese Patent Publication No. 6-37.
No. 025 and Japanese Patent Publication No. 7-61601 are known.

The mirror finishing apparatus disclosed in the former publication includes a chuck table for holding a disk-shaped wafer whose outer peripheral portion is chamfered and rotating it around an axis, and a polishing cloth on the outer peripheral surface of a short cylindrical ring member. The surface of the wafer held on the chuck table while rotating around the axis is disposed with the axis oriented in a direction orthogonal to the axis of the wafer held on the chuck table. A polishing ring for mirror-polishing the chamfered portion, wherein the polishing ring has a diameter sufficiently larger than the chamfer width of the wafer and has a ring width sufficiently smaller than the diameter of the wafer, so that the entire width of the ring can be reduced. It is designed to be able to abut the entire chamfer width of the bevel. Uneven wear and caused by, has the effect of a decrease in working accuracy due to uneven wear can be prevented.

The mirror polishing method disclosed in the latter publication includes a step of holding a disc-shaped wafer having a chamfered outer peripheral portion on a chuck table, a step of rotating the wafer around an axis of the wafer, and a linear movement. A polishing drum supported so that it can be freely separated and contacted with the wafer, by the urging force of a vertically suspended weight, and polishing the chamfered portion of the wafer with a rotating polishing drum; The method has a step of changing the contact position with the wafer by moving the polishing drum in the axial direction, and has an effect that the entire chamfered portion can be surely and uniformly mirror-polished. Have.

[0006]

However, in the mirror finishing apparatus of the former publication, the ring width is formed sufficiently smaller than the diameter of the wafer, so that the wafer is polished at the same contact point of the cloth provided on the polishing ring. As a result, when the cloth is worn locally, the life of the cloth is short, and when the worn cloth is used for mirror finishing,
There were inconveniences such as the surface roughness being deteriorated and a mirror surface with high surface accuracy could not be obtained.

In the mirror polishing method of the latter publication, a polishing drum having a long length in the axial direction is used to move the polishing drum in the axial direction. Since it is only in one direction, there is a problem that a mirror surface with high surface accuracy cannot be obtained.

SUMMARY OF THE INVENTION The present invention has been made to solve such a conventional problem, and an object of the present invention is to provide a method and an apparatus for mirror-polishing a semiconductor wafer, which can perform mirror-polishing of a chamfered portion with high accuracy and efficiency. It is the purpose.

[0009]

In order to achieve the above object, according to the first aspect of the present invention, a wafer to be mirror-finished in a chamfered portion is carried into at least one processing station, and the mirror-finishing is completed. A transfer loader that unloads a wafer from the processing station, and a cylindrical polishing drum that is provided in the processing station and that can be continuously or intermittently oscillated in the axial direction by an oscillating means,
At least one chuck provided near the polishing drum and holding a wafer to be mirror-finished; moving the chuck in two axial directions perpendicular to the axial direction of the polishing drum and continuously or intermittently; And an oscillating means for oscillating.

[0010] With the above configuration, the chamfered portion of the wafer carried into the processing station by the transfer loader is oscillated in the axial direction by the polishing drum provided in the processing station, and the wafer is biaxially orthogonal to the axial direction of the polishing drum. Since the mirror surface processing can be performed while oscillating in the direction, the mirror surface processing of the chamfered portion can be performed automatically and with high efficiency, and the surface roughness of the processed surface is improved, so that the mirror surface processing with high surface accuracy can be performed.

Further, since the mirror-finished wafer can be unloaded from the processing station by the transfer loader, the unprocessed wafer can be quickly unloaded and productivity can be improved.

In order to achieve the above object, according to the second aspect of the present invention, a pair of chucks are provided in the vicinity of the polishing drum so as to face each other or in the same direction.

According to the above configuration, when the chamfered portion on one surface of the wafer held by one chuck is completed, the wafer can be transferred to the other chuck and the chamfered portion on the other surface can be mirrored. In one processing station, the chamfered portions on both sides of the wafer can be mirror-polished, and each chuck is provided so as to face both sides of the polishing drum. After the part is mirror-polished, it can be transferred to the other chuck and the other chamfered part can be mirror-polished, so that both sides of the wafer and the outer periphery can be mirror-polished without the need to transfer the wafer. become.

In order to achieve the above object, the invention according to claim 3 is provided with a notch mirror surface processing means for mirror processing a notch formed in advance on a wafer on the upstream side of the processing station.

According to the above configuration, since the notch portion of the wafer can be mirror-finished, the process of mirror-finishing the notch portion can be automated.

In order to achieve the above-mentioned object, the invention according to claim 4 is provided with a plurality of processing stations, wherein each of the processing stations simultaneously performs the mirror-finishing of the chamfered portion.

According to the above configuration, mirror processing of a wafer can be performed in each processing station in parallel, so that productivity is greatly improved.

According to a fifth aspect of the present invention, there is provided a polishing apparatus, comprising the steps of: loading a wafer having a chamfered peripheral edge on both sides by a transfer loader into a processing station; While the wafer is continuously or intermittently oscillated in two directions orthogonal to the axial direction of the polishing drum, the first chamfered portion on one side is formed on the outer periphery of the polishing drum. In this case, the chamfered portion on the other surface of the wafer is abutted against the outer peripheral surface of the polishing drum, and the chamfered portion on the other surface is mirror-finished. .

According to the above method, the surface roughness of the processed surface is improved, so that a mirror surface with high surface accuracy can be obtained, and the contact position between the polishing drum and the chamfered portion follows the oscillation of the polishing drum and the wafer. In addition, the polishing drum cloth is not locally worn, thereby extending the life of the cloth, which is economical, and at the same time, the frequency of replacing the cloth can be reduced, so that the maintenance and management becomes easy.

In order to achieve the above object, according to a sixth aspect of the present invention, a pair of chucks are provided so as to face each other in the vicinity of a polishing drum provided in a processing station, and a wafer held by one of the chucks is provided. Once the chamfered part on one side is mirror-finished, the wafer is delivered to the other chuck,
The chamfered portion on the other surface is mirror-finished.

According to the above method, since the chamfered portions on both surfaces of the wafer can be mirror-polished at one processing station, there is no need to provide a plurality of processing stations, thereby eliminating the need for a transfer means for transferring the wafer. Thus, equipment costs and installation space can be reduced.

In order to achieve the above object, the invention according to claim 7 is characterized in that the polishing drum is moved in the circumferential direction in accordance with the angle of the chamfered portion, so that the outer periphery of the polishing drum and the chamfered portion are brought into contact with each other. Is changed.

According to the above method, even if the angle of the chamfered portion of the wafer changes, it can be dealt with only by changing the contact position between the polishing drum and the chamfered portion.

In order to achieve the above object, according to the present invention, the wafer is moved in the circumferential direction of the polishing drum, and the outer peripheral surface of the wafer is brought into contact with the outer peripheral surface of the polishing drum. The surface is also mirror-finished.

According to the above-mentioned method, mirror processing of the chamfered portion of the wafer and mirror processing of the outer peripheral surface of the wafer can be performed at the same processing station.

[0026]

Embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a plan view of a mirror finishing device, FIG. 2 is a front view thereof, FIG. 3 is a left side view thereof, FIG. 4 is a right side view thereof, FIG.
FIG. 6 is a sectional view taken along line XX of FIG. 2, and FIG. 6 is a sectional view taken along line YY of FIG.

In these figures, reference numeral 1 denotes a mirror processing apparatus main body, and a loading station ST1 for loading a semiconductor wafer (hereinafter simply referred to as a wafer) 2 on a bed 1a.
And the positioning station ST2, the notch mirror surface processing station ST3, the first processing station ST4, the second processing station ST5, the transfer station ST6, and the unloading station ST7 for unloading the wafer 2 from the upstream side in the wafer transfer direction A to the downstream side. They are arranged sequentially.

The wafer 2 on which the chamfered portion 2a is mirror-finished is
The loading station S in which a plurality of sheets are set in the cassette 3
The wafers 2 loaded into the loading station ST1 and loaded into the loading station ST1 are taken out one by one from the cassette 3 by the transport loader 4, and are transferred to the next positioning station ST.
2 is transported.

The transport loader 4 is provided at the loading station S
It has a movable body 4b movable along a linear guideway 4a laid from T1 to the transfer station ST6, and the movable body 4b is provided with a support arm 4c movable in the front-rear direction and the vertical direction. A suction means 4d for sucking the wafer 2 is provided at the tip of the support arm 4c. The suction means 4d can suck the wafer 2 and sequentially transport the wafer 2 to the stations ST1 to ST6. I have.

The positioning station ST2 includes a center positioning means 5a for positioning the center of the wafer 2 by pressing the outer peripheral surface of the wafer 2 in multiple directions, and a phase shifter such as a notch or an orientation flat formed on the wafer 2 in advance. A positioning device 5 comprising a phase reference detecting means 5b for detecting the reference 2b is provided, and the positioning means 5 performs center and phase positioning.

When the wafer 2 positioned by the positioning device 5 is provided with a notch phase reference 2b formed of a notch, the wafer 2 is transferred to the next notch mirror processing station ST3, and the notch portion is processed by the notch mirror processing means 6. It is designed to be mirror-finished.

The wafer 2 on which the orientation flat is formed as the phase reference 2b is moved to the next first and second processing stations S
Since the outer peripheral edge of the wafer 2 and the orientation flat can be simultaneously mirror-processed at T4 and ST5, the wafer 2 is conveyed to one of the next first and second processing stations ST4 and ST5 through the notch mirror-surface processing station ST3. .

The first processing station ST4 and the second processing station ST5 can simultaneously mirror-process another wafer 2 in order to increase productivity.
5 is provided with a similar mirror surface processing means 7. As shown in FIG. 6, the mirror finishing means 7 has an upper support member 7a and a lower support member 7b vertically separated from each other on the front surface of a column 1b erected on a bed 1a.

The upper and lower support members 7a and 7b are supported by guide rails 1c laid in the vertical direction of the column 1b, and are vertically movable independently by vertical drive means (not shown). The guide rails 7 are arranged horizontally on the lower surface of the lower support member 7a and the upper surface of the lower support
d are laid respectively, and these guide rails 7d
A chuck driving motor 8 is supported by an oscillating means (not shown) so as to freely oscillate in the horizontal direction.

Guide rail 7d of the upper support member 7a
The rotating shaft 8a of the chuck driving motor 8 supported on the lower side is directed downward, and the rotating shaft 8a of the chuck driving motor 8 supported on the guide rail 7d of the lower support member 7b is directed upward. Part, chamfered part 2
Upper chuck 9 for attracting and holding wafer 2 for mirror-finishing a
And the lower chuck 10 are respectively attached.

As shown in FIG. 2, a pair of guide rails 1e are laid along the wafer transfer direction A on the upper surface of the table 1d provided on the bed 1a. A pair of spindle heads 11 which are independently movable in the wafer transfer direction A by a driving means not supported are supported. One end of a main shaft 11a is protruded from the surfaces of the main shaft heads 11 facing each other, and a polishing drum 12 for mirror-finishing the chamfered portion 2a of the wafer 2 is detachably provided at the tip of the main shaft 11a. Installed.

The polishing drum 12 has a wafer 2 on its outer peripheral surface.
A cloth 12a for polishing the chamfered portion 2a is mounted over the entire circumference, and the other end of each of the spindles 11a is connected to a rotary shaft 13a of a drum drive motor 13 mounted on the front side of the spindle bed 11. The main shaft 11a is connected via a power transmission means 14 by a drum drive motor 13 via a power transmission means 14 such as a timing belt.
Are rotated in the same direction.

On the other hand, in the transfer station ST6, the wafers 2 having the chamfered portions 2a mirror-processed in the first processing station ST4 or the second processing station ST5 are stored one by one in the cassettes 3 waiting in the unloading station ST7. The cassette 3 in which a predetermined number of wafers 2 are stored by the transfer means 16 and which is unloaded from the unloading station ST7 to the outside of the main body 1 by manual means or the like. Has become.

Next, with the mirror finishing device configured as described above,
A method for mirror-finishing the chamfered portion 2a of the wafer 2 will be described in detail with reference to FIGS. A chamfering device (not shown) provided in another place of the wafer 2 to be mirror-finished is preliminarily chamfered at the peripheral edge, and a phase reference 2b such as a notch or an orientation flat is also preliminarily processed to obtain a predetermined number of wafers. Is loaded into the loading station ST1 in a state where is set in the cassette 3.

When the cassette 3 in which the wafers 2 are set is carried into the loading station ST1, the support arm 4c of the transfer loader 4 is advanced to the cassette 3 side as shown in FIG. Sucks one of the wafers 2 in the cassette 3 from below and transports it to the positioning station ST2 as shown in FIG.

The wafer 2 transferred to the positioning station ST2 is transferred onto the chuck 5c of the positioning device 5 provided in the positioning station ST2 as shown in FIG. 3), the outer peripheral surface is pressed from multiple directions by the center positioning means 5a.
The center position is determined so that the center of the wafer 2 coincides with the center O of the chuck 5c. When the positioning of the center position is completed, the wafer 2 is attracted to the chuck 5c and the center positioning means 5a is moved to the position shown in FIG. Retreat as shown in (e). Thereafter, the chuck 5c is rotated, and the phase reference 2b attracted to the chuck 5c is detected by the phase reference detecting means 5b as shown in FIG.

When the positioning of the center and the phase of the wafer 2 is completed, as shown in FIG.
Moves forward, and the wafer 2 whose positioning has been completed is sucked by the suction means 4.
d is attracted, the chuck 5c releases the wafer 2, and when the phase reference 2b formed on the wafer 2 is a notch, the process proceeds to the next notch mirror surface processing station ST3, and in the case of the orientation flat, the first processing station ST4 or Second
It is transported to the processing station ST5.

In the case of a wafer 2 having a notch formed on the phase reference 2b, the wafer is transferred from the positioning station ST2 to the next notch mirror surface processing station ST3, and after the notch position is determined, the notch portion is processed by the notch mirror surface processing means 6. Mirror finished. Then, the wafer 2 on which the mirror finishing of the notch is completed is transferred by the transfer loader 4 to the first or second processing station ST4 or ST5.

Next, a mirror processing procedure of the wafer 2 carried into the first processing station ST4 or the second processing station ST5 will be described with reference to FIGS.

The first processing station ST4 and the second processing station ST4
Since the mirror processing procedure in the processing station ST5 is the same, the procedure in the first processing station ST4 will be described.

[0046] (b) in FIG. 8 is mirror finished on the lower surface side of the chamfered portions 2a of the wafer 2 ₁ adsorbed on the chuck 9 is completed, the wafer 2 ₂ adsorbed to the lower chuck 10 is chamfered on the upper and lower surfaces This shows a state where the mirror finishing of the portion 2a has been completed. In this state, the upper chuck 9 and the lower chuck 10 are moved in a direction away from the polishing drum 12, and then (b) in FIG.
The upper chuck 9 is moved upward and the lower chuck 10 is moved downward as shown in FIG.

[0047] Thereafter view vertical chuck 9 as shown in (c) of 8 later that is retracted once to the transfer position, the wafer 2 ₂ completing the mirror-finished as shown in (d) of FIG. 8
When the lower chuck 10 is raised until the height of the lower chuck 10 becomes substantially the same as the horizontal line passing through the center of the polishing drum 12, the transfer loader 4
The supporting arm 4c is advanced between the upper and lower chucks 9 and 10 as shown in FIG.

[0048] Note that the wafer 2 ₃ mirror-finished and then is adsorbed on the upward suction means 4d this time.

[0049] Then the downward suction means 4d provided at the end of the support arm 4c as shown in (f) of FIG. 8, when the processed wafers 2 ₂ held by the lower chuck 10 is attracted, once the support arm 4c is raised as shown in FIG. 8 (g) and then retracted to the standby position as shown in FIG. 8 (h).

[0050] On the one hand the transfer position, the upper chuck 9 is lowered as shown in (i) in FIG. 8, the counter position the wafer 2 ₁ held by the upper chuck 9 to exist a small gap between the lower chuck 10 In this state, after being sucked and held on the lower chuck 10 side as shown in FIG. 8 (N), the upper chuck 9 is raised to the original position.

Thereafter, as shown in FIG. 8 (L), the support arm 4c is advanced to the transfer position, and at the same time, the lower chuck 10 is advanced below the processing position. Then, at the transfer position, the unprocessed wafer 2 sucked by the upward suction means 4d
₃ together is adsorbed as shown in upper chuck 9 (e) in FIG. 8, the lower chuck 10 is raised to a processing position.

[0052] After being raised to subsequently on the chuck 9 also processing position, the upper chuck 9 and the lower chuck 10 is moved simultaneously polishing drum 12 side, a chamfered portion 2a of the raw wafer 2 ₃ lower, already under the upper surface of the wafer 2 ₂ mirror finishing is complete side is in contact with the external surface of the polishing drum 12,
Mirror finishing of the chamfered portion 12a is started.

During the mirror finishing, the spindle head 11 is continuously or intermittently oscillated in the axial direction of the polishing drum 12 at a constant period by an oscillating means (not shown), and at the same time, each chuck motor 8 is operated by the oscillating means (not shown). with the oscillating to continuously or intermittently horizontal direction at a constant period, in the mirror-finished, the wafer 2 ₂ already completed the processing, adsorbed to transfer means 16 from the downward suction means 4d of the support arm 4c Is stored in the cassette 3 waiting at the unloading station ST7.

Each time the polishing of the wafer chamfered surface 2a by the polishing drum 12 is completed, the operations of (a) to (e) in FIG. 8 are repeated.
At T4 and the second processing station ST5, mirror processing of different wafers 2 can be performed at the same time, so that productivity is almost doubled as compared with the case where mirror processing is performed at one processing station, and the waiting time of the transfer loader 4 is reduced. Can be halved.

In the above embodiment, the wafer 2 which has been chamfered to a predetermined angle by the pre-processing is mirror-finished. However, when the angle of the chamfer 2a is changed, or when the outer periphery of the wafer 2 is When the surface is polished, the operation can be performed by the operation shown in FIG. That is, the upper chuck 9 and the lower chuck 10 for holding the wafer 2 are movable in two axial directions, namely, the direction of contact and separation and the vertical direction with respect to the polishing drum 12.

Therefore, when the angle of the chamfered portion 2a of the wafer 2 is small, the wafer 2 held by the upper and lower chucks 9 and 10 is brought into contact with the outer peripheral surfaces on the upper and lower sides of the polishing drum 12 as shown in FIG. And as the angle increases,
By contacting the outer peripheral surface near the horizontal line passing through the center of the polishing drum 12, the wafer 2 having a large angle of the chamfered portion 2a to the wafer 2 having a small angle can be handled. Then, as shown by the imaginary line in FIG. 9, by bringing the wafer 2 into contact with the outer peripheral surface 2c of the polishing drum 12 on the horizontal line, the outer peripheral surface 2c of the wafer 2 can be mirror-finished.

[Brief description of the drawings]

FIG. 1 is a plan view of a mirror finishing apparatus according to an embodiment of the present invention.

FIG. 2 is a front view of the mirror finishing device according to the embodiment of the present invention;

FIG. 3 is a left side view of the mirror finishing apparatus according to the embodiment of the present invention;

FIG. 4 is a right side view of the mirror finishing device according to the embodiment of the present invention;

FIG. 5 is a sectional view taken along line XX of FIG. 2;

FIG. 6 is a sectional view taken along the line YY of FIG. 2;

FIGS. 7A to 7G are explanatory views showing the operation of the mirror finishing apparatus according to the embodiment of the present invention;

FIGS. 8A to 8E are explanatory views showing the operation of the mirror finishing apparatus according to the embodiment of the present invention.

FIG. 9 is an explanatory view showing the operation of the mirror finishing apparatus according to the embodiment of the present invention;

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 2 ... Wafer 2a ... Chamfer surface 4 ... Conveyer loader 6 ... Notch mirror surface processing means 9 ... Upper chuck 10 ... Lower chuck 12 ... Polishing drum ST4, ST5 ... Processing station

Claims (8)

[Claims] 1. A wafer 2 having a chamfered portion 2a mirror-finished.
At least one processing station ST4, ST4
5, a transfer loader 4 for carrying out the mirror-finished wafer 2 from the processing stations ST4 and ST5, and an oscillator continuously or intermittently provided in the processing stations ST4 and ST5 in the axial direction by an oscillating means. A free-standing cylindrical polishing drum 12, at least one chuck 9, 10 provided near the polishing drum 12 and holding the wafer 2 to be mirror-finished, and the chuck 9, 10 being an axis of the polishing drum 12. An oscillating means for moving in two axial directions perpendicular to the direction and for oscillating continuously or intermittently. 2. The mirror finishing apparatus according to claim 1, wherein a pair of chucks 9 and 10 are provided in the vicinity of the polishing drum 12 so as to face each other or in the same direction. 3. The mirror processing apparatus according to claim 1, further comprising a notch mirror processing means for mirror processing a notch formed in advance on the wafer, upstream of the processing stations ST4 and ST5. 4. A plurality of processing stations ST4 and ST5 are provided, and each of the processing stations ST4 and ST5 is provided.
5. The mirror finishing device according to claim 1, wherein the mirror finishing of the chamfered portion 2a is simultaneously performed in step 5. 5. A transfer loader 4 processes wafers 2 whose peripheral edges are chamfered at processing stations ST4 and ST4.
5, the polishing drum 12 provided in the processing stations ST4 and ST5 is oscillated continuously or intermittently in the axial direction of the polishing drum 12, and the wafer 2 is orthogonal to the axial direction of the polishing drum 12. While continuously or intermittently oscillating in two directions, first, the chamfered portion 2a on one surface is brought into contact with the outer peripheral surface of the polishing drum 12, the chamfered portion 2a is mirror-finished, and then the other surface of the wafer 2 is chamfered. A method for mirror-finishing a semiconductor wafer, comprising: bringing a portion (2a) into contact with the outer peripheral surface of the polishing drum (12); 6. A pair of chucks 9, 1 near a polishing drum 12 provided at processing stations ST4, ST5.
When the chamfered portion 2a on one surface of the wafer 2 held by one chuck 9 is mirror-finished, the wafer 2 is transferred to the other chuck 10 and the chamfered portion on the other surface is provided. 6. The mirror finishing method according to claim 5, wherein 2a is mirror-finished. 7. By moving the wafer 2 in the circumferential direction of the polishing drum 12 in accordance with the chamfering angle of the chamfered portion 2a formed in advance, changing the contact position between the outer periphery of the polishing drum 12 and the chamfered portion 2a. 6. The mirror finishing method according to claim 5, wherein the method is adapted to an arbitrary chamfer angle. 8. The outer peripheral surface of the wafer 2 is mirror-finished by moving the wafer 2 in the circumferential direction of the polishing drum 1 and bringing the outer peripheral surface of the wafer 2 into contact with the outer peripheral surface of the polishing drum 1. The mirror finishing method according to claim 5.