JP2003007661A - Apparatus and method for machining planar surface - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an apparatus and method for machining a planar surface of a wafer, capable of improving throughput without damaging the wafer. SOLUTION: In the apparatus for machining planar surface 10, a grinding stage 22 is set on a body 12, on which a rough grinding stage 18 and a finishing grinding stage 20 are set, and rough grinding, finish grinding, and polishing of a wafer 28 are implemented in the identical apparatus for machining the planar surface 10. A cleaning stage 23 for cleaning a polishing cloth 56 of the polishing stage 22 set in the apparatus for machining the planar surface 10 cleans the polishing cloth 56 in the identical apparatus 10, when it is smeared. A set of an etching apparatus 150 in the apparatus for machining planar surface 10 enables a series of planar machining from rough grinding to etching to be processed using a single apparatus.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a flat surface processing apparatus and a flat surface processing method, and more particularly to a semiconductor wafer manufacturing process.
The present invention relates to a flat surface processing apparatus and a flat surface processing method for grinding a back surface of a semiconductor wafer on which chips are not formed.

[0002]

2. Description of the Related Art A flat surface processing apparatus for grinding a back surface (one surface) of a semiconductor wafer includes a chuck for adsorbing and holding a wafer, a grinding wheel for rough grinding, a grinding wheel for fine grinding, a back surface cleaning device, and the like. According to such a flat surface processing apparatus, after the front surface (the other surface) of the wafer is adsorbed and held by the chuck, the rough grinding grindstone is pressed against the back surface of the wafer and the back surface of the wafer is rotated by rotating the chuck and the grindstone. Is roughly ground. Then, the wafer after the rough grinding is removed from the chuck, then held by the chuck for fine grinding, and finely ground here by the grinding wheel for fine grinding. After the fine grinding, the wafer is transported to the back surface cleaning device and the back surface thereof is cleaned. This completes the back surface grinding process for one wafer by the flat surface processing apparatus.

By the way, the wafer whose back surface grinding has been completed is transferred from the flat surface processing apparatus to the etching apparatus in order to shift to the next step, that is, the etching step. The altered layer is removed.

However, when a wafer is ground into an ultrathin wafer close to a standard product in a planar processing apparatus,
When the wafer is transferred from the flat surface processing apparatus to the etching apparatus, there arises a problem that the wafer is damaged (cracked) due to the work-affected layer.

Therefore, in the conventional flattening apparatus, in order to prevent the above problems, the wafer is ground to a thickness that does not damage the wafer during transportation. Here, explaining the thickness of the wafer, which was cut from the ingot,
For example, a wafer having a thickness of 725 μm is roughly ground to a thickness of 250 μm by rough grinding of a flat surface processing apparatus, and is then ground to a thickness of 200 μm by fine grinding. Then, in the final etching step, the standard thickness of 50 μm is processed.

[0006]

However, in the conventional flat surface processing apparatus, the wafer cannot be processed to a thickness close to the standard thickness in order to prevent the wafer from being damaged during transportation. As a result, the machining allowance in the etching process (the allowance of 150 μm in the above example) becomes large, so that the etching takes a long time and the throughput cannot be improved.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a flat surface processing apparatus and a flat surface processing method capable of improving throughput without damaging a work.

[0008]

In order to achieve the above object, the present invention has a holding means for holding a work and a grinding means for grinding one surface of the work with a grinding stone for grinding the work. And a polishing means for polishing one surface of the work, the polishing means rotating means for rotating the polishing head, and a gap between the polishing head and the work. It is characterized in that it is composed of a positioning feed mechanism for setting.

In order to achieve the above object, the present invention provides a holding means for holding a work, a rough grinding means for roughly grinding the work held by the holding means, and a work roughly ground by the rough grinding means. A fine grinding means for fine grinding while holding the holding means, and a polishing means for polishing the work finely ground by the fine grinding means while being held by the holding means,
And a moving means for moving the holding means to a rough grinding position by the rough grinding means, a fine grinding position by the fine grinding means, and a polishing position by the polishing means.

In order to achieve the above object, the present invention provides a holding means for holding a work, a grinding means for grinding the work held by the holding means, and a polishing for polishing the work ground by the grinding means. And a moving means for moving the holding means to a grinding position by the grinding means and a polishing position by the grinding means, the grinding means uses the grinding process calculated with twice the standard deviation. The amount of work more than the amount required to remove the work-affected layer by 6 times the standard deviation, and the amount required to remove the work-affected layer and the variation in the thickness during grinding, or the work within a large amount of 20 μm It is characterized by polishing.

In order to achieve the above object, the present invention provides a holding means for holding a work, a rough grinding means for roughly grinding the work held by the holding means, and a work roughly ground by the rough grinding means. Fine grinding means for fine grinding, polishing means for polishing the work finely ground by the fine grinding means, rough holding position by the rough grinding means for the holding means, fine grinding position by the fine grinding means, and polishing And a moving means for moving to a polishing position by means of the means, the fine grinding means is calculated by twice the standard deviation, and is equal to or more than the amount necessary for removing the work-affected layer by rough grinding. , Standard deviation of 6
Finely grinds the work within a large amount calculated by multiplying the variation in the thickness during rough grinding and the removal of the work-affected layer, or a large amount of 150 μm, and the polishing means uses a standard deviation of twice. More than the calculated amount necessary for removing the work-affected layer by the fine grinding process, calculated by 6 times the standard deviation, and the amount required for removing the work-affected layer and the variation in the thickness during the fine-grinding process, or 20 μm The feature is that the work is polished within a large amount.

According to the first aspect of the present invention, since the polishing means is mounted on the plane processing apparatus on which the work grinding means is mounted, the work grinding and polishing can be performed in the same apparatus. . In the method of polishing a work by the polishing means, the positioning feed mechanism positions the polishing head with respect to the work, the polishing head is brought into contact with the work, and the polishing head is rotated by the rotating means. As a result, the work is polished. Therefore, in the present invention, since it is not necessary to convey the work from the flat surface processing apparatus to the etching apparatus, the work can be ground by the grinding means to a thickness close to the standard. Therefore, the time required for polishing can be shortened, and the throughput can be improved. Further, by carrying out polishing, it is possible to remove the work-affected layer of the work caused by grinding, so that it is possible to eliminate the post-etching step, and as a result, the overall structure of the work manufacturing processing line. Can be simplified and can be made compact.

Further, at the time of polishing the work, a constant pressure processing for pressing the polishing head to polish the work, or a quantitative processing for polishing the work while feeding the cutting head by cutting may be performed.

According to the second aspect of the invention, the work is ground by the grinding means while being held by the holding means,
Then, after the holding means is moved to the polishing position by the moving means, the work is polished by the polishing means. That is, according to the present invention, since the work is ground and polished while being held by the holding means, the work can be accurately processed without damaging the work. On the other hand, in an apparatus that moves a work from a holding means dedicated to the grinding means to a holding means dedicated to polishing to process the work, the work may be damaged by an external force when the work is transferred. Also,
In the transfer device, the accuracy of the holding surface of the holding means changes each time, and the accuracy affects the processing accuracy of the work, so that it cannot be processed accurately. The present invention can eliminate this point.

According to the third aspect of the present invention, since the cleaning means and / or the dressing means for cleaning the polishing cloth of the polishing means are provided in the flat processing apparatus, when the polishing cloth is soiled or clogged, the polishing is performed. The cloth can be cleaned and dressed in the same device to eliminate its dirt and clogging.

According to a fourth aspect of the present invention, the grinding and / or polishing means processes the work-affected layer on one surface of the work ground by the pre-working, or works the work-affected layer. Processes variations in work thickness.
Thereby, a highly accurate work can be obtained.

According to the fifth aspect of the present invention, a plurality of holding means are provided, and these holding means are sequentially moved from the grinding position to the polishing position by the moving means, so that grinding and polishing can be performed simultaneously. The operation rate is higher than that of grinding and polishing the work with one holding means.

According to the sixth aspect of the invention, since the holding means is moved by the moving means while the spindle is connected to the holding means, the spindle is separated from the holding means each time the holding means is moved, It is possible to save the trouble of connecting the holding means to the spindle installed at the next movement position.

According to the invention described in claim 7, the holding means is detachably connected to the spindle, the holding means is separated from the spindle each time the holding means is moved, and only the holding means is moved by the moving means. Therefore, it is possible to reduce the load applied to the moving means, and it is sufficient to have a spindle corresponding to each processing, and the manufacturing cost of the device can be kept low.

According to the invention described in claim 8, as the holding means, a holding means for sucking and adsorbing and holding the work is applied, or a holding means for freezing and holding the work through an ice film is applied, or the work is held. Since the holding means for holding the workpiece with static electricity is applied, the work can be held firmly.

According to the ninth aspect of the present invention, since the partitioning member for partitioning the polishing position by the polishing means and the adjacent portion of the polishing position is provided, the grinding processing liquid or grinding waste used in the grinding means is placed at the polishing position. It does not enter, and the polishing processing liquid used in the polishing means does not enter the grinding position.
Therefore, it is possible to prevent a processing defect due to the mixing of both the processing liquid and the processing waste. In particular, when the polishing means is a means for performing chemical mechanical polishing, the polishing liquid contains a chemical polishing agent. Therefore, if the polishing liquid is mixed with such a polishing liquid, the concentration of the chemical polishing agent will increase. However, there is a problem in that the processing time becomes long and the processing time becomes long. Therefore, by providing the partition plate, the above-mentioned problem can be solved.

According to the tenth aspect of the present invention, since the surface processing apparatus is provided with the rough grinding means, the fine grinding means, and the polishing means, the work is automatically rough-ground, fine-ground, and polished by these means. You can Further, since the moving means for moving the work holding means to the rough grinding position, the fine grinding position and the polishing position is provided, a single or a plurality of rough grinding means, fine grinding means and polishing means are arranged in combination. Even with the configuration, it is possible to effectively operate the processing means without lowering the operation rate.

According to the eleventh aspect of the present invention, since the work etching means is provided in the surface processing apparatus provided with the rough grinding means, the fine grinding means, and the polishing means, a series of works from rough grinding to etching is performed. Can be processed by one plane processing device. In this case, the work that has been finely ground before polishing may be etched by an etching means, or the work that has been polished may be etched. That is, light etching for cleaning after grinding may be performed before polishing, and after polishing for the purpose of enhancing the gettering effect for removing impurities, heavy metals, point defects, etc., which are harmful to the device characteristics in the chip. Etching may be performed.

The invention of a flat surface processing method according to claim 12 is
The present invention is intended for a flat surface processing apparatus provided with a grinding means and a polishing means, and the polishing means has a standard deviation of 6 times or more, which is calculated by multiplying the standard deviation by more than the amount necessary for removing the work-affected layer by fine grinding. Since the work is polished within the calculated amount required for removing the variation in the thickness during the fine grinding process and the work-affected layer, or within a large amount of 20 μm, the predetermined processing can be performed without reducing the operating rate of the device. Can be given.

The invention of a flat surface processing method according to claim 13 is
Targeting a flat surface processing apparatus provided with a rough grinding means and a fine grinding means as a grinding means, the fine grinding means has an amount equal to or more than the amount necessary for removing the work-affected layer by rough grinding, calculated by twice the standard deviation, The amount required to remove the variation in thickness and the work-affected layer during rough grinding, calculated with 6 times the standard deviation, or 15
The work was finely ground within a large amount of 0 μm, and the polishing means was calculated by 2 times the standard deviation, and was calculated by 6 times the standard deviation, which is more than the amount necessary for removing the work-affected layer by the fine grinding process. Since the work is polished within the amount required for removing the variation in thickness and the process-affected layer during the fine grinding process or within a large amount of 20 μm, it is possible to perform a predetermined process without lowering the operation rate of the device. it can.

According to the invention of the flat surface processing method described in claim 14 and the invention of the flat surface processing device described in claim 15, there is provided a sensor for measuring the thickness of the work, and prior to the processing,
Alternatively, since the thickness of the work is measured during processing and the grinding amount or polishing amount is controlled based on the measured value, it is possible to process a wafer having a desired thickness.

[0027]

BEST MODE FOR CARRYING OUT THE INVENTION Preferred embodiments of a flat surface processing apparatus and a flat surface processing method according to the present invention will be described below in detail with reference to the accompanying drawings. FIG. 1 is a perspective view of a semiconductor wafer flattening apparatus to which the present invention is applied, and FIG. 2 is a plan view.

As shown in FIG. 1, the main body 12 of the planar processing apparatus 10 includes a cassette housing stage 14, an alignment stage 16, a rough grinding stage 18, and a fine grinding stage 2.
0, a polishing stage 22, a polishing cloth cleaning stage 23, a polishing cloth dressing stage 27, and a wafer cleaning stage 24 are provided. In addition, the rough grinding stage 18,
The fine grinding stage 20 and the polishing stage 22 are partitioned by a partition plate 25 shown by a chain double-dashed line in FIG. 2 to prevent the machining fluid used in each stage 18, 20, 22 from scattering to the adjacent stage. .

The partition plate 25 is fixed to the index table 34 as shown in FIGS. 4 and 5, and four chucks (corresponding to holding means) 32, 36, 38, 40 installed on the index table 34 are installed. It is formed in a cross shape so as to be partitioned. Further, the polishing stage 22 is covered with a casing 102 having a top plate 100 so as to be separated from other stages. A brush 104 is attached to the side surface of the casing 102 through which the partition plate 25 passes, as shown in FIG.
04 is brought into contact with the upper surface 25A and the side surface 25B of the partition plate 25 when the chuck 40 is located at the processing position. As a result, when the chuck 40 is positioned at the processing position, the polishing stage 22 is held in a substantially airtight state by the casing 102, the partition plate 25, and the brush 104. Can be prevented from entering the polishing stage 22, and the polishing processing liquid used in the polishing stage 22 can be prevented from scattering from the polishing stage 22. Therefore, it is possible to prevent a processing defect due to the mixing of both processing liquids. The polishing stage 22 of this example performs chemical mechanical polishing, and the polishing liquid contains a chemical polishing agent. Therefore, if the polishing liquid is mixed into such a polishing liquid, the concentration of the chemical polishing agent will increase. However, there is a problem in that the processing time becomes long and the processing time becomes long. Therefore, by providing the partition plate 25, the above-mentioned problem can be solved.

The rough grinding stage 18 is shown in FIGS.
As described above, it is surrounded by the side surface of the main body 12, the top plate 106, and the partition plate 25, and the fine grinding stage 20 is also surrounded by the side surface of the main body 12, the top plate 108, and the partition plate 25. These tops 100, 106, 108
Through holes 101 through which the heads of the respective stages are inserted,
107 and 109 are formed. Reference numeral 110 in FIG.
Is a brush for isolating the rough grinding stage 18 from the outside, and the brush 110 is in contact with the upper surface and the side surface of the partition plate 25.

The cassette storage stage 1 shown in FIGS. 1 and 2.
Two cassettes 26, 26 are detachably set in 4, and a large number of wafers 28 before backside grinding are stored in these cassettes 26, 26. This wafer 28
Are held one by one by the hand 31 of the transfer robot 30 and sequentially transferred to the alignment stage 16 in the next process. The transfer robot 30 may be suspended and supported by a beam (not shown) standing on the main body 12 via an elevating device, or may be installed on the upper surface 12A of the main body 12. When the transportation robot 30 is suspended and supported,
Cassette storage stage 14 and alignment stage 16
Since the space between and can be narrowed, the flat surface processing apparatus 1
0 size reduction can be achieved. The robot 30 is
Since it is a general-purpose 6-axis joint robot and its configuration is well known, its explanation is omitted here.

The alignment stage 16 is a stage for aligning the wafer 28 conveyed from the cassette 26 at a predetermined position. The wafer 28 aligned by the alignment stage 16 is sucked and held by the hand 31 of the transfer robot 30 again, then transferred to the empty chuck 32, and sucked and held by the chucking surface of the chuck 32. .

The chuck 32 is installed on the index table 34 and has the same function.
Nos. 6, 38, 40 are installed at intervals of 90 degrees on the circumference of the index table 34 around the rotary shaft 35 shown by the broken line in FIG. A spindle (not shown) of a motor (corresponding to moving means) 37 shown by a broken line in FIG. 2 is connected to the rotary shaft 35. The chuck 36
Is positioned on the rough grinding stage 18, and the sucked wafer 28 is roughly ground here. Further, the chuck 38 is positioned on the fine grinding stage 20, and the adsorbed wafer 28 is subjected to finish grinding (fine grinding, spark out) here. Further, the chuck 40 is positioned on the polishing stage 22, and the adsorbed wafer 28 is polished here, and the work-affected layer caused by the grinding and the variation in the thickness of the wafer 28 are removed.

The chucks 32, 36, 38 and 40 are
As shown in FIG. 3, a spindle 94 and a rotation motor 9 are provided on the lower surface thereof.
2 are connected to each other and rotated by the driving force of these motors 92. The motor 92 is supported by the index table 34 via a support member 93. Therefore, the planar processing apparatus 10 of the present embodiment is an apparatus in which the chucks 32, 36, 38, 40 are moved by the motor 37 while the motor 92 and the spindle 94 are connected to the chucks 32, 36, 38, 40. Is. This allows
Each time the chucks 32, 36, 38, 40 are moved by the motor 37, the spindle 94 is moved to the chucks 32, 36, 3
It is possible to save the trouble of disconnecting the chucks 32, 36, 38, 40 from the spindles 8, 40 or connecting the chucks 32, 36, 38, 40 to the spindle 94 installed at the next movement position.

Although FIG. 3 shows an example in which the spindle 94 and the motor 92 are directly connected to the chucks 32, 36, 38, 40, the present invention is not limited to this, and as shown in FIGS. The spindle 94 is detachably connected to the chucks 32, 36, 38, 40 via the connecting member 112,
Each time the chucks 32, 36, 38, 40 are moved, the connecting member 112 is separated from the chucks 32, 36, 38, 40, and only the chucks 32, 36, 38, 40 are moved by the motor 37. Good. This allows
The load on the motor 37 can be reduced, and a spindle and a motor can be provided for each of the rough grinding, fine grinding, and polishing processes, and the cost of the device can be reduced.

In FIG. 7, the chucks 32, 36, 3
Nos. 8 and 40 are placed on the stepped portion 116 of the opening 114 formed in the index table 34, and the piston 12 of the cylinder device 118 is provided below the motor 92.
0s are connected. When the piston 120 is extended as shown in FIG. 8, the connecting member 112 passes through the opening 114 and is fitted and connected to the recess 122 formed in the lower portion of the chucks 32, 36, 38, 40. Then, the chucks 32, 36, 38, 40 are moved upward from the index table 34 by the continuous extension operation of the piston 120, and are positioned at the grinding positions by the grindstones 46, 54.

The chucks 32, 36, 3 of this embodiment
8 and 40 are formed by a porous material 124 whose adsorption surface is made of a sintered body such as ceramics. When the connecting member 112 is connected to the recess 122, a fluid coupling (not shown) is simultaneously coupled, and the suction force of a suction pump (not shown) coupled to the fluid coupling acts on the porous material 124 via the air passage 126. , This allows wafer 2
8 is firmly adsorbed and held on the surface of the porous material. Also, when the connecting member 112 is separated, the suction force is maintained as it is by a check valve (not shown).

In the present embodiment, the chucks 32, 36, 38, 40 for sucking and holding the wafer 28 have been described, but the present invention is not limited to this, and instead of this, the frozen chuck device shown in FIG. 128 may be applied.

The freezing chuck device 128 comprises a chuck plate 130, a controller 132, and a cooling water supply device 134. A voltage is applied to the chuck plate 130 by the controller 132, and the Peltier effect produced by this is used to make a wafer. 28 is frozen and held on the chuck plate 130 via an ice film. The chuck plate 130 connects two types of metal (for example, Cu and Bi) plates to form a closed circuit, and causes a current to flow at the junction to freeze-hold the wafer 28 on the thermoelectric element (Cu plate) side. Further, the cooling water supply device 134 is
Cooling water is supplied to one side (Bi plate) of the thermoelectric element to cool the heat generated on the one side. Instead of the freeze chuck device 128, an electrostatic chuck device that holds a wafer with static electricity may be applied.

The chuck 32, which is located at the chuck position of the wafer 28 shown in FIG. 2, has a suction surface before the wafer 28 is transferred, and the chucking surface of the chuck 32 (see FIG. 2).
). The cleaner device 42 is slidably provided on a rail 44, and is moved along the rail 44 when the suction surface is cleaned.
Located on top. The cleaner device 42 has a removing member 43, and the removing member 43 abuts the suction surface of the chuck 32 to remove dust such as sludge attached to the suction surface. When the suction surface of the chuck 32 is a porous material made of a sintered body such as ceramics, the removing member 43 is made of a porous material.

Wafer 2 sucked and held by chuck 32
The thickness of No. 8 is measured by a pair of measuring gauges 136 and 138 shown in FIG. These measuring gauges 13
6, 138 respectively have contacts 140, 142,
The contact 140 is in contact with the upper surface (back surface) of the wafer 28, and the contact 142 is in contact with the upper surface of the chuck 32. These measurement gauges 136 and 138 can detect the thickness of the wafer 28 as a difference between in-process gauge readings with the upper surface of the chuck 32 as a reference point.

The wafer 28 whose thickness has been measured is positioned on the rough grinding stage 18 by the 90 ° rotation of the index table 34 in the direction of arrow A in FIGS. 1 and 2, and is moved by the cup-shaped grindstone 46 of the rough grinding stage 18. The back surface of the wafer 28 is roughly ground. As shown in FIG. 1, the cup-shaped grindstone 46 is connected to an output shaft (not shown) of the motor 48, and is attached to the grindstone feeding device 52 via a support casing 50 of the motor 48. The grindstone feeding device 52 moves the cup-shaped grindstone 46 up and down together with the motor 48, and the cup-shaped grindstone 4 is moved by this downward movement.
6 is pressed against the back surface of the wafer 28. Thereby, the back surface rough grinding of the wafer 26 is performed. The descending movement amount of the cup-shaped grindstone 46, that is, the grinding amount by the cup-shaped grindstone 46, is the reference position of the cup-shaped grindstone 46 registered in advance and the thickness of the wafer 28 detected by the measurement gauges 136 and 138. It is set based on.

The thickness of the wafer 28 whose back surface is roughly ground by the rough grinding stage 18 is measured by the thickness measuring gauge having the same structure shown in FIG. The wafer 28 whose thickness has been measured is positioned on the fine grinding stage 20 by the 90 ° rotation of the index table 34 in the same direction, and finely ground and sparked out by the cup-shaped grindstone 54 of the fine grinding stage 20. The structure of this fine grinding stage 20 is
Since the structure is the same as that of the rough grinding stage 18, its description is omitted here. Although two grinding stages are provided in this embodiment, one grinding stage may be provided.
Further, the thickness measurement with the measurement gauge may be performed in-line.

The thickness of the wafer 28 whose back surface has been finely ground by the fine grinding stage 20 is measured by the thickness measuring gauge having the same structure shown in FIG. The wafer 28 whose thickness has been measured is positioned on the polishing stage 22 by rotating the index table 34 by 90 degrees in the same direction, and the polishing cloth 56 of the polishing stage 22 shown in FIG. 3 and the slurry supplied from the polishing cloth 56. And is removed by polishing to remove the work-affected layer generated on the back surface thereof. The thickness measurement with the measuring gauge may be performed in-line.

Management of the wafer thickness in the planar processing apparatus 10 will be described with reference to FIG. 11. First, the initial wafer thickness is measured before the rough grinding (S100), and the processing in the rough grinding is performed based on the measured thickness. The amount is set, and rough grinding is performed on the rough grinding stage 18 (S110). Next, the thickness of the wafer after rough grinding is measured (S120), a processing amount for fine grinding is set based on the measured thickness, and fine grinding (finish grinding) is performed on the fine grinding stage 20 (S130). Then, the thickness of the wafer after the fine grinding is measured, the polishing time is set based on the measured thickness, the polishing conditions, and the final thickness (S140), and polishing (polishing) is performed on the polishing stage 22 (S150). The above is the plane processing apparatus 10
Is a flow of controlling the thickness of the wafer 28 in FIG.

In such wafer thickness control, the processing amount in the precision polishing stage 20 is calculated as twice the standard deviation, which is more than the amount necessary for removing the work-affected layer by rough grinding, and 6 times the standard deviation. The amount required to remove the variation in the thickness during rough grinding and the work-affected layer calculated by
It is preferable to set it within a large amount of m. It has been found that finely grinding with this amount can remove the work-affected layer generated in the rough grinding without lowering the operating rate.

If the amount required for removing the work-affected layer by rough grinding is set to an amount calculated with a value less than twice the standard deviation, the work-affected layer may not be reliably removed. If the variation in thickness during rough grinding and the amount required to remove the deteriorated layer are set to a value calculated by more than 6 times the standard deviation or a value exceeding 150 μm, which is a large value, Is longer and the operating rate is lower.

Further, the processing amount on the polishing stage 22 is calculated by twice the standard deviation, which is equal to or more than the amount required for removing the work-affected layer by the fine grinding process, and is calculated by 6 times the standard deviation. It is preferable to set the variation within the thickness at the time and the amount necessary for removing the work-affected layer, or within the larger amount of 20 μm. It has been found that polishing with this amount can remove the work-affected layer generated in the fine grinding process without lowering the operation rate.

If the amount required for removing the work-affected layer by fine grinding is set to a value calculated with a value less than twice the standard deviation, there is a drawback that the work-affected layer cannot be reliably removed. If the amount of thickness variation during fine grinding and the amount required to remove the work-affected layer are set to a value calculated by more than 6 times the standard deviation or a value exceeding 20 μm, the processing time Is longer and the operating rate is lower.

As an example, when a wafer having a diameter of 200 mm and an initial thickness of 725 μm is processed to have a thickness of 50 μm, the rough grinding speed at this time is 22 as shown in FIG.
5 (μm / sec) and fine grinding speed 65 (μm
/ Sec) and the polishing (polishing) speed is 6 (μ
m / sec), rough grinding amount of 510 μm, fine grinding amount of 150 μm, and polishing amount of 14.9 μm, the respective processing times (2.27 to 2.48) become substantially equal, so The wafer 28 can be processed to a thickness of 725 μm to 50 μm without reducing the rate. In this case, for example, the amount of polishing processing is such that the variation in thickness during fine grinding is 2.25 μm in standard deviation, which is 13.5 μm when calculated by 6 times the standard deviation, and the depth of the work-affected layer during fine grinding is Has an average value of 0.7 μm and a standard deviation of 0.11 μm. Therefore, when calculated six times, it is 0.66 μm, and the maximum value of the work-affected layer is 1.36 μm. Therefore, the amount required for removing the variation in the thickness and the process-altered layer during the fine grinding process can be set to 14.9 μm.

However, in the processing time obtained by the above calculation, there are cases where the variation in thickness and the work-affected layer cannot be reliably removed.

Therefore, in the present embodiment, the processing amount in the fine grinding is 150 μm and the amount required for removing the variation in the thickness and the work-affected layer at the time of the rough grinding calculated by 6 times the standard deviation described above. When the former is large, the processing amount is set to 150 μm, and when the latter is large, the processing amount is set to the latter amount. As a result, it is possible to reliably remove the variation in thickness and the work-affected layer during fine grinding.

Further, the amount of processing in polishing was compared with the variation in thickness at the time of fine grinding calculated by 6 times the standard deviation and the amount required to remove the work-affected layer. The processing amount is set to 20 μm, and the latter is 20 μm.
When it is larger than m, the processing amount is set to the latter amount. As a result, it is possible to reliably remove the variation in thickness and the work-affected layer during polishing.

FIG. 3 is a structural diagram of the polishing stage 22.

The polishing cloth 56 of the polishing stage 20 shown in FIG.
Is attached to a polishing head 61 connected to an output shaft 60 of a motor (corresponding to rotating means) 58. Further, guide blocks 62, 62 forming a linear guide are provided on the side surface of the motor 58, and the guide blocks 62, 62 move up and down on a guide rail 66 provided on the side surface of the support plate 64. It is freely engaged. Therefore, the polishing cloth 56 is attached to the support plate 64 together with the motor 58 so as to be vertically movable.

The support plate 64 is provided at the tip of the horizontally arranged long arm 68. The base end of the arm 68 is connected to the output shaft 74 of a motor 72 arranged in the casing 70. Therefore,
When the motor 72 is driven, the arm 68 can rotate around the output shaft 74. As a result, the polishing cloth 56 can be moved within the range of the polishing position shown by the solid line in FIG. 1, the polishing cloth cleaning position by the polishing cloth cleaning stage 23, and the dressing position by the polishing cloth dressing stage 27. When the polishing cloth 56 is moved to the polishing cloth cleaning position, the polishing cloth cleaning stage 23 cleans the surface of the polishing cloth 56 and removes polishing dust and the like adhering to the surface. Examples of the polishing cloth 56 include foamed polyurethane, polishing cloth, and the like. The polishing cloth cleaning stage 23 is provided with a removing member such as a brush for removing polishing dust. The removing member is rotationally driven when the polishing cloth 56 is washed, and the polishing cloth 56 is also rotationally driven by the motor 58 (see FIG. 3). The polishing cloth dressing stage 27 is made of the same material as the polishing cloth 56, for example, foamed polyurethane.

Guide blocks 76, 76 forming a linear guide are provided on the side surface of the casing 70, and the guide blocks 76, 76 are attached to a guide rail 80 provided on the side surface of the screw feed device housing 78. It is engaged so that it can move up and down. Further, a nut member 82 is provided on the side surface of the casing 70 so as to project. The nut member 82 has an opening 79 formed in the housing 78.
It is disposed in the housing 78 via the and is screwed to a screw rod 80 of a screw feeding device (corresponding to a positioning feeding mechanism). The output shaft 84 of the motor 82 is attached to the upper end of the screw rod 80.
Are connected. Therefore, when the motor 82 is driven and the screw rod 84 is rotated, the casing 70 is moved up and down by the feeding action of the screw feeding device and the rectilinear action of the guide block 76 and the rail 80. As a result, the polishing cloth 56 is largely moved in the vertical direction, and the distance between the polishing head 61 and the wafer 28 is set to a predetermined distance.

On the upper surface of the motor 58, a piston 88 of an air cylinder device (corresponding to a pressurizing mechanism) 86 is provided.
Are connected via a through hole 69 of the arm 68. Further, a regulator 90 that controls the internal pressure P of the cylinder is connected to the air cylinder device 86. Therefore, if the internal pressure P is controlled by the regulator 90, the pressing force (pressure contact force) of the polishing pad 56 against the wafer 28 can be controlled.

Wafer 2 polished by polishing stage 22
8 is a robot 9 shown in FIG. 2 after the polishing cloth 56 is retracted from the position above the wafer 28 by the rotation of the arm 68.
It is adsorbed and held by the hand 97 of No. 6 and conveyed to the wafer cleaning stage 24. The robot 96 is omitted in FIG. Since the process-affected layer has been removed from the wafer 28 that has been polished, it is not easily damaged,
Therefore, it is not damaged during the transfer by the robot 96 and during the cleaning on the wafer cleaning stage 24.

In the present embodiment, the polishing cloth 56 is used as the polishing body, but the polishing body is not limited to this, and a polishing grindstone or electrophoresis of abrasive grains may be applied. When the polishing grindstone or the electrophoresis of abrasive grains is applied, it is preferable to perform quantitative polishing.

As the wafer cleaning stage 24,
A stage having a rinse cleaning function and a spin drying function is applied. The wafer 28 that has been cleaned and dried in the wafer cleaning stage 24 is handled by the hand 31 of the robot 30.
It is suction-held and stored in a predetermined shelf of the cassette 26. The above is the flow of the wafer processing process in the flat surface processing apparatus 10 of the present embodiment.

As described above, according to the planar processing apparatus 10 of this embodiment, the polishing stage 22 is provided on the main body 12 in which the rough grinding stage 18 and the fine grinding stage 20 are installed.
Since, the rough grinding, the fine grinding, and the polishing of the wafer 28 can be performed in the same planar processing apparatus 10.

As a result, it is not necessary to convey the wafer 28 from the surface processing apparatus 10 to the etching apparatus, so that the rough grinding stage 18 and the fine grinding stage 20 can grind the wafer 28 to a thickness close to the standard. For example, in the conventional flat surface processing apparatus, in order to prevent the damage of the wafer at the time of transportation, grinding is performed while leaving an allowance for etching of 150 μm. For example, grinding may be performed leaving 3 μm.

Therefore, the polishing time is greatly reduced, and the throughput can be improved. Further, by performing the polishing, the work-affected layer of the wafer 28 generated by the grinding can be removed, so that the etching process of the subsequent process can be eliminated, and as a result, the entire structure of the wafer manufacturing process line can be eliminated. It can be simplified. Further, in the flat surface processing apparatus 10, the polishing stage 2
Since the polishing cloth cleaning stage 23 for cleaning the second polishing cloth 56 and the polishing cloth dressing stage 27 for dressing the polishing cloth 56 are provided, when the polishing cloth 56 becomes dirty or clogged, the same polishing cloth is used. It can be cleaned and dressed in the apparatus 10 to eliminate its dirt and clogging.
This facilitates the handling of the polishing cloth 56. Further, a means is provided for detecting that the polishing cloth 56 is clogged (clogging that adversely affects the processing), and when this means detects the clogging, the polishing cloth 56 is polished by the polishing cloth dressing stage 27. If the automatic control is performed so that the dressing is performed, the entire planar processing apparatus 10 can be fully automated. Examples of the detection means include a polishing cloth 5
A means for detecting the rotation torque of the motor 58 of No. 6 can be illustrated. When this rotation torque exceeds the reference value,
The polishing cloth 56 may be washed.

Further, according to the planar processing apparatus 10, the wafers 28 are chucked by the same chuck 32 (36, 38, 40).
While being held at, the index table 34 can be rotated to perform rough grinding, fine grinding, and polishing. This makes it possible to prevent the wafer from being damaged due to the transfer of the wafer and to process the wafer with high accuracy. On the other hand, if the wafer is transferred to another chuck for each stage and processed, the wafer is damaged when the transfer is performed. Further, when the wafer is transferred and processed, the accuracy of the chucking surface of the chuck changes each time, and the accuracy affects the wafer processing accuracy, so that the processing cannot be performed accurately.
Therefore, the planar processing apparatus 10 of the present embodiment can eliminate the conventional drawbacks.

Further, it goes without saying that the work-affected layer of the wafer 28 is polished on the polishing stage 22.
By extending the processing time, it is possible to polish the variation in the thickness of the wafer 28.

In the present embodiment, the wafer is exemplified as the work, but the present invention is not limited to this, and if the work needs to be surface-polished and then surface-polished, the plane-processing apparatus of the present invention is used. Can be applied.

FIG. 13 is a plan view showing a first embodiment of a flat surface processing apparatus 152 equipped with an etching apparatus 150, and the same or similar members as those of the flat surface processing apparatus 10 shown in FIG. 2 are the same. The reference numerals are given and the description thereof is omitted.

The plane processing apparatus 152 of FIG. 13 is an apparatus for etching the wafer 28 that has been polished by the polishing stage 22. When the polished wafer 28 is positioned at the position of the chuck 32 by the clockwise rotation of the index table 34 by 90 °, the robot 97
Held by Then, the wafer 28 is transferred to the cleaning stage 24 by the robot 97, cleaned here, and then transferred to the etching apparatus 150. Since the planar processing device 152 does not perform etching processing while it is held on the chuck, the robot 97
Although there is a concern that the wafer 28 will be damaged during transportation by the wafer, since the work-affected layer has been removed by the polishing stage 22 in the pre-process of the etching process, the wafer 28 is transported during transportation.
Will not be damaged.

The etching apparatus 150 is a spinner type etching apparatus, and its configuration is as shown in FIG.
A motor 156 and a spindle 158 for rotating 54,
It is composed of a nozzle 162 for supplying the etching liquid 160, an etching tank 164, and the like. Etching device 15
At 0, the wafer 28 is attracted to the porous material 155 of the chuck 154, and then the etching solution 160 is supplied from the nozzle 162 to the central portion of the upper surface thereof while being rotated at a predetermined rotation number by the motor 156, and is diffused radially. Etching is performed with the etching liquid 160. An etching liquid tank 168 is connected to the nozzle 162 via a pump 166, and when the pump 166 is driven, the etching liquid 160 stored in the etching liquid tank 168 is supplied from the nozzle 162.

In the etching bath 164, a through hole 172 into which the spindle 158 is inserted is formed at the center of the bottom surface 170, and the bottom surface 170 is formed on the wafer 28.
In order to prevent the etching liquid 160 scattered from the leakage from the through hole 172, the etching liquid 160 is formed in a shape inclined downward toward the outer peripheral portion. A drain pipe 174 is connected to the outer peripheral portion of the bottom surface 170, and the etching liquid is discharged from the drain pipe 174.

In the planar processing apparatus 152 of this example, the wafer cleaning device 176 is mounted on the polishing stage 22 as shown in FIG.
Further, a spin cleaning device 178 is provided near the cleaning stage 24. Wafer cleaning device 17
The wafer 6 is provided on a pair of rails 180, 180 so as to be freely movable, and is mounted on a wafer 28 before or after polishing.
At the time of cleaning, the wafer 28 is moved to a position above the wafer 28 held by the chuck and the upper surface of the wafer 28 is cleaned.
On the other hand, the spin cleaning device 178 cleans the wafer 28 before and after etching. Spin cleaning device 1
The etched wafer 28 transferred to 78 is spin-cleaned here, and then held again by the robot 97 and stored in a predetermined shelf of the cassette 26.

FIG. 15 is a plan view showing a second embodiment of a flat surface processing apparatus 192 equipped with an etching apparatus 190. The flat surface processing apparatus 10 shown in FIG. 2 and the flat surface processing apparatus shown in FIG. The same or similar members as those of 152 are designated by the same reference numerals, and the description thereof will be omitted.

The plane processing apparatus 192 of FIG. 15 is an apparatus for etching the wafer 28 that has been finely ground by the fine grinding stage 20, that is, an apparatus for etching the wafer 28 before polishing. The wafer 28 that has been finely ground is moved in the clockwise direction 9 of the index table 34.
By the rotation of 0 °, it is conveyed to the etching device 190, where it is etched while being held by the chuck 40. The etched wafer 28 is held by the robot 97 at the position of the chuck 32. Then, the wafer 28 is transferred to the cleaning stage 24 by the robot 97, cleaned here, and then transferred to the polishing stage 22. In the flattening apparatus 192, the wafer 28 is not damaged during the transfer by the robot 97 because the processing-affected layer is removed by etching while being held by the chuck.

The etching device 190 is a spinner type etching device, and its configuration is as shown in FIG.
A motor 194 and a spindle 196 for rotating the wafer, a nozzle 200 for supplying the etching liquid 198, an etching tank 202, and the like.

The chuck 40 includes the index table 3
4 is placed on the step 206 of the opening 204 formed in 4,
A piston 210 of the cylinder device 208 is connected to the lower part of the motor 194. When the piston 210 contracts, the piston 210 is in a position retracted from the chuck 40. When the piston 210 is extended as shown in FIG. The member 212 is fitted and connected to a recess (not shown) formed in the lower portion of the chuck 40. Then, the chuck 40 is moved upward from the index table 34 by the continuous extension operation of the piston 210, passes through the through hole 216 formed in the bottom surface 214 of the etching tank 202, and is positioned inside the etching tank 202.

In the etching apparatus 190, after the wafer 28 is adsorbed by the porous material 41 of the chuck 40,
While being rotated at a predetermined rotation speed by the motor 194,
From the nozzle 200 to the center of the upper surface of the etching liquid 19
8 is supplied, and the etching liquid 198 is diffused radially.
Is etched by. An etching solution tank 218 is connected to the nozzle 200 via a pump 217, and when the pump 217 is driven, the etching solution 198 stored in the etching solution tank 218 is discharged from the nozzle 200.
Supplied from

In the etching tank 202, a through hole 216 into which the chuck 40 is inserted is formed in the center of the bottom surface 214, and the etching solution 198 scattered from the wafer 28 does not leak from the through hole 216 in the bottom surface 214. Thus, it is formed in a shape that is inclined downward toward the outer peripheral portion. A drain pipe 220 is connected to the outer periphery of the bottom surface 214.
The structure is such that the etching liquid is discharged from the.

The polishing stage 22 of the flat surface processing apparatus 192 of this embodiment is provided with a chuck 222 for sucking and holding the wafer 28 after etching, which is conveyed by the robot 97, as shown in FIG. The wafer 28 is held by the chuck 224 and rotated by a motor (not shown) that rotates the chuck 224, and the polishing cloth 56 is pressed against the upper surface of the wafer 28 to be polished. The polished wafer 28 is held by the robot 97 and transferred to the spin cleaning device 178, where it is spin-cleaned, and then held by the robot 97 again and stored in a predetermined shelf of the cassette 26.

[0080]

As described above, according to the flat surface processing apparatus of the present invention, the polishing means is mounted on the flat surface processing apparatus in which the work grinding means is mounted, and the work grinding and polishing are performed in the same apparatus. Since it is performed in step 1, the throughput can be improved without damaging the work.

Further, according to the present invention, since the work is ground and polished while being held by the same holding means, the work can be accurately processed without damaging the work.

Further, according to the present invention, since the cleaning means and / or the dressing means for cleaning the polishing cloth of the polishing means are provided in the flat surface processing apparatus, the polishing cloth is cleaned when the polishing cloth is soiled or clogged. It can be cleaned and dressed in the same device to eliminate dirt and clogging.

Further, according to the present invention, the work-altered layer on one surface of the work ground by the grinding means and the variation in the thickness of the work are polished by the polishing means or the etching means. It is possible to obtain a highly accurate work.

Further, according to the flat surface processing method of the present invention, the fine grinding means has a standard deviation calculated by twice the standard deviation, which is 6 times the standard deviation or more, which is more than the amount required for removing the work-affected layer by rough grinding. Since the work is precisely ground within the calculated amount required to remove the variation in thickness during rough grinding and the amount of work-affected layer removal, or a large amount of 150 μm, it does not occur in the rough grinding process without reducing the operating rate. The work-affected layer can be removed. Further, the polishing means was calculated with twice the standard deviation, more than the amount necessary for removing the work-affected layer by the fine grinding, and calculated with 6 times the standard deviation. Since the work is polished within an amount necessary for removing the deteriorated layer or within a large amount of 20 μm, it is possible to remove the processed deteriorated layer generated by the fine grinding without lowering the operation rate.

[Brief description of drawings]

FIG. 1 is an overall perspective view of a semiconductor wafer flattening apparatus according to an embodiment of the present invention.

FIG. 2 is a plan view of the flat surface processing apparatus shown in FIG.

3 is a cross-sectional view showing the structure of a polishing stage of the flat surface processing apparatus shown in FIG.

FIG. 4 is a perspective view showing a partition plate of the plane processing apparatus shown in FIG.

5 is a plan view of the partition plate shown in FIG.

6 is a sectional view taken along line 6-6 of the partition plate shown in FIG.

FIG. 7 is an explanatory view in which a chuck and a spindle are separated by a fluid coupling.

FIG. 8 is an explanatory view in which a chuck and a spindle are connected via a fluid coupling.

FIG. 9 is a configuration diagram of a freeze chuck device.

FIG. 10 is a side view of the wafer thickness measuring cage.

FIG. 11 is a flowchart showing wafer thickness management in a flat surface processing apparatus.

FIG. 12 is a diagram showing a processing speed, a processing amount, and a processing time of rough grinding, fine grinding, and polishing.

FIG. 13 is a plan view showing a first embodiment of a flat surface processing apparatus equipped with an etching apparatus.

14 is a cross-sectional view showing the structure of the etching apparatus shown in FIG.

FIG. 15 is a plan view showing a second embodiment of a flat surface processing apparatus equipped with an etching apparatus.

16 is a sectional view showing the structure of the etching apparatus shown in FIG.

[Explanation of symbols]

10, 152, 192 ... Planar processing device, 12 ... Main body, 1
4 ... Cassette storage stage, 16 ... Alignment stage, 18 ... Rough grinding stage, 20 ... Fine grinding stage, 2
2 ... Polishing stage, 23 ... Polishing cloth cleaning stage, 24 ...
Wafer cleaning stage, 28 ... Wafer, 32, 36,
38, 40 ... Chuck, 150, 190 ... Etching device

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 3C043 BB00 CC04 CC07 CC11                 3C058 AA04 AA07 AA18 BC02 CB01                       CB03 DA17

Claims (15)

[Claims] 1. A flat surface processing apparatus comprising: a holding means for holding a work; and a grinding means for grinding one surface of a work with a grind stone for grinding the work. And a polishing means for polishing one surface of the polishing means, the polishing means comprising a rotating means for rotating a polishing head, and a positioning feed mechanism for setting a gap between the polishing head and the work. Processing equipment. 2. The plane processing apparatus is provided with a moving means for moving the holding means between a grinding position by the grinding means and a grinding position by the grinding means, and the work held by the holding means. The flat surface processing apparatus according to claim 1, wherein the is ground and polished while being held. 3. The plane processing apparatus is provided with a cleaning means for cleaning the polishing cloth attached to the polishing head of the polishing means, or a dressing means for dressing the surface of the polishing cloth. Claim 1 or 2
The plane processing device described. 4. The grinding and / or polishing means processes a work-affected layer on one surface of the work ground by pre-working, or works the work-affected layer and causes variation in the thickness of the work. The flat surface processing apparatus according to claim 1, 2 or 3, wherein the minute portion is processed. 5. The flat surface processing apparatus according to claim 2, wherein a plurality of the holding means are provided, and the plurality of holding means are sequentially moved from the grinding position to the polishing position by the moving means. 6. The plane processing apparatus according to claim 2, wherein a spindle for rotating the holding means is connected to the holding means, and the spindle is moved together with the holding means. 7. The holding means is detachably connected to a spindle for rotating the holding means, and when the holding means is moved, the connecting portion is disconnected and only the holding means is moved. The flat surface processing apparatus according to claim 2 or 5. 8. The holding means is a holding means that sucks and holds a work by suction, or a holding means that freezes and holds the work through an ice film or an electrostatic holding means that holds the work with static electricity. The planar processing apparatus according to claim 1. 9. The flat surface processing apparatus according to claim 1, wherein the flat surface processing apparatus is provided with a partition member for partitioning a polishing position by the polishing means and a portion adjacent to the polishing position. 10. A holding means for holding a work, a rough grinding means for roughly grinding the work held by the holding means, and a work coarsely ground by the rough grinding means while being held by the holding means. Fine grinding means for grinding, polishing means for polishing the work finely ground by the fine grinding means while being held by the holding means, rough grinding position of the holding means by the rough grinding means, fine grinding by the fine grinding means A planar processing apparatus comprising: a position and a moving unit that moves the polishing unit to a polishing position. 11. The flat surface processing apparatus is provided with a work etching means, and the work before polishing or the work after polishing is etched by the etching means. 10. The flat surface processing apparatus according to item 10. 12. A holding means for holding a work, a grinding means for grinding the work held by the holding means, a polishing means for polishing the work ground by the grinding means, and a holding means for grinding the holding means. A plane processing apparatus provided with a moving means for moving the grinding position and the polishing position by the polishing means is used, and the polishing means is an amount necessary for removing the work-affected layer by the grinding processing calculated by twice the standard deviation. As described above, the planar processing method is characterized in that the work is polished within the amount required for removing the variation in the thickness during the grinding process and the work-affected layer calculated by 6 times the standard deviation, or within a large amount of 20 μm. . 13. A holding means for holding a work, a rough grinding means for roughly grinding the work held by the holding means, and a fine grinding means for finely grinding the work roughly ground by the rough grinding means, Polishing means for polishing the work finely ground by the fine grinding means, and moving means for moving the holding means to a rough grinding position by the rough grinding means, a fine grinding position by the fine grinding means, and a polishing position by the polishing means. , And the fine grinding means is calculated at 2 times the standard deviation, and is calculated at 6 times the standard deviation, which is equal to or more than the amount necessary for removing the work-affected layer by rough grinding. Fine grinding of a work within a large amount of variation in thickness and removal of a work-affected layer at the time of rough grinding, or a large amount of 150 μm, and the polishing means calculates fine grinding with twice the standard deviation. Removal of work-affected layer by working It is necessary to polish the work within the amount necessary for removing the variation in the thickness and the process-affected layer during fine grinding, which is calculated by 6 times the standard deviation, which is greater than the amount necessary for the removal, or within 20 μm. The characteristic flat surface processing method. 14. A sensor for measuring the thickness of a work, the thickness of the work is measured prior to or during processing, and the grinding amount or polishing amount is controlled based on the measured value. 4. The flat surface processing apparatus described in 4. 15. A sensor for measuring the thickness of a work, the thickness of the work is measured prior to or during processing, and the grinding amount or polishing amount is controlled based on the measured value. 12. The flat surface processing method according to 12 or 13.