JPS6320380B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a positioning device for a disk-shaped body, and relates to a positioning device for a disk-shaped body, which can easily position orientation flats, which are notches in semiconductor substrates, at the same angular position in a process of transferring a plurality of semiconductor substrates. The semiconductor substrates are aligned and the center of the semiconductor substrate can be positioned at a predetermined position.

When transferring multiple semiconductor substrates (hereinafter referred to as wafers) between manufacturing equipment in the semiconductor manufacturing process, it is necessary to align the notches around the wafers at the same angle and position them at predetermined positions. There is. As shown in FIG. 1, the wafer 1 usually has a substantially circular external appearance, and is provided with a notch for position detection called an orientation flat 2 on its outer periphery.

Therefore, in order to position the wafer 1, the following means have conventionally been adopted. That is, as shown in FIG. 2, rollers 3 and 4 are rotating counterclockwise, and roller 5 is rotating clockwise. Therefore, when the outer circumference of the wafer 1 is in contact with the three rollers 3, 4, and 5 at the same time, the wafer 1 rotates clockwise. As a result, due to the orientation flat 2, the roller 4 separates from the wafer 1 (FIG. 2 shows this state), and the rollers 3, 5
The rotation becomes balanced. Therefore, the rotation of the wafer 1 is stopped and its position can be determined.

However, in this case, if the peripheral edge of the wafer 1 is chipped or the size of the orientation flat 2 is small, a malfunction may occur and reliable positioning cannot be performed. Further, the relative positions of the three rollers 3, 4, and 5 must be adjusted depending on the size of the wafer 1, and the work involved becomes complicated.

On the other hand, the third method that eliminates these drawbacks is
A positioning device as shown in the figure has been proposed (Japanese Patent Application Laid-Open No. 55-65445). When positioning the wafer 1 using the positioning apparatus shown in the figure, first, the output waveform of the wafer for reference is stored in the storage section 11 as digital information. Next, after placing the wafer 1 on the vacuum chuck 7, the centering jig 8 is operated to center the wafer 1. At this time, if the center of the wafer 1 and the rotation axis of the pulse motor 6 do not match, the output waveform of the edge sensor 9 when the wafer 1 rotates once passes through the differentiator circuit 14 and is differentiated. The rotation angle position of the orientation flat 2 is calculated by the second calculation unit 17 from the digital information obtained by A/D exchanged with this differential waveform, and the position of the orientation flat 2 is calculated based on the calculation result. roughly match the position of the edge sensor 9. Next, after releasing the holding state of the wafer 1, the centering jig 8 is operated again to center the wafer 1. Then, the wafer 1 is rotated once, and the output waveform of the edge sensor 9 is converted into an A/D
The converted digital information is compared with the previously stored reference digital information, and the rotation angle of the pulse motor 3 that minimizes the difference between the two pieces of information is calculated, and the orientation is adjusted according to this calculation result. Align the section flat 2 with the position of the edge sensor 9,
The positioning of wafer 1 is completed.

However, in such a positioning device, when centering the wafer 1,
If the center of the pulse motor 6 and the rotation axis of the pulse motor 6 are fixed with a deviation, it is troublesome that a centering operation must be performed again to align the position after a series of signal processing. Furthermore, in the semiconductor manufacturing process, in order to avoid contamination, it is necessary to reduce the number of contact points. Not only is there contact, but also centering is performed by pressing from both sides of the wafer 1, so there is a possibility that the wafer 1 may be damaged, and in this respect it is not preferable as an apparatus for handling semiconductors.

SUMMARY OF THE INVENTION In view of the above-mentioned prior art, an object of the present invention is to provide a positioning device for a disc-shaped body such as a wafer that can easily and quickly position a disc-shaped body such as a wafer at a predetermined position without using a centering jig. . To achieve this object, the present invention rotates at least once about a straight line passing through the center or an arbitrary point near the center of a plate-shaped body having an orientation flat, such as a wafer, and perpendicular to the disk-shaped body as a rotation axis; Based on the positional information of the circumferential edge of the disc-shaped body with respect to the rotation angle of the disc-shaped body obtained by this, the deviation of the center of the disc-shaped body from the center of the rotation axis is calculated as the distance and angle with respect to the center of this rotation axis, and this deviation is calculated. moving the disc-shaped body within the plane to which it belongs, and calculating the deviation of the rotational angular position of the orientation flat from the predetermined rotational angular position, and moving the disc-shaped body within the plane to remove this deviation; The basis of the technical idea is that the disk-shaped body is rotated.

Embodiments of the present invention will be described in detail below based on the drawings. As shown in FIG. 4, a wafer 1, which is a disc-shaped body having a notch called an orientation flat 2 on its outer peripheral surface, is held by a vacuum chuck 18, and this vacuum chuck 18 is attached to the wafer 1. It is rotated by a pulse motor 19 which is a rotating means. At this time, the pulse motor 19
1 and parallel to the wafer 1 due to the rotation of the screw shaft 21a as the pulse motor 20 is driven.
The pulse motor 20 moves in the X-axis direction, which is the horizontal direction of the XY plane, and the Y-direction moving table 2
3, and is moved in the Y-axis direction, which is a direction perpendicular to the XY plane, by rotation of a screw shaft 23a as the pulse motor 22 is driven. The wafer 1 is thus able to move freely on the plane to which it belongs. At this time, the pulse motor 19 is driven by the pulse motor drive circuit 38, the pulse motor 20 is driven by the pulse motor drive circuit 39, and the pulse motor 22 is driven by the pulse motor drive circuit 40. Edge sensors 24 and 25 detect positional changes of the periphery of the wafer 1 as it rotates, and provide position signals S ₂₄ , which are analog signals corresponding thereto.
_S25 , and are arranged facing each other so as to be located on a straight line in a plane perpendicular to the Z axis. These edge sensors 24 and 25 are not particularly limited as long as they have the functions described above.
For example, it is conceivable to have a light-emitting section and a light-receiving section, position the peripheral edge of the wafer 1 between the light-emitting section and the light-receiving section, and detect the position of the peripheral edge based on the weight of light received by the light-receiving section. Furthermore, the edge sensor 24,
25 position signals S ₂₄ and S ₂₅ are sent to A/D converters 26 and 2.
7, the signals are converted into digital signals and written and stored in storage units 28 and 29, respectively, along with rotation angle information representing the rotation angle of the pulse motor 19 corresponding thereto. FIG. 5a is a waveform diagram showing the position signal _S24 which is the output signal of the edge sensor 24, and FIG. 5b is a waveform diagram showing the position signal _S25 which is the output signal of the edge sensor 25. The rotation angle positions of the flat 2 are shown respectively. At the same time, the position signals S ₂₄ and S ₂₅ are output to differentiating circuits 30 and 3, respectively.
1. Passes through the band filter circuits 32 and 33, and further passes through the rectangular wave forming circuits 34 and 35, and is converted into digital information representing the rough rotation angle position of the orientation flat 2, and is stored in the storage units 28 and 29 as the rotation angle. It is written and stored along with the information. Figure 6a
7 is a waveform diagram in which the protruding portion corresponding to the orientation flat 2 is differentiated by the differentiating circuit 30, and FIG. 7a is a waveform diagram showing the corresponding output waveform of the rectangular wave forming circuit 34. Similarly, FIG. 6b is a waveform diagram showing the waveforms of the differentiating circuit 31, and FIG. 7b is a waveform diagram showing the waveforms of the rectangular wave forming circuit 35. First calculation unit 36
calculates the deviation of the center of the wafer 1 from the center of the rotation axis of the pulse motor 19 as a distance and an angle with respect to the center of the rotation axis based on the stored contents of the storage units 28 and 29. Based on this calculation result, the pulse motor drive circuit 3
9 and 40 drive and control the pulse motors 20 and 22. The second arithmetic unit 37 is the first arithmetic unit 36
The accurate rotation angle position of the orientation flat 2 is calculated from the deviation calculated in step 1 and the position signal _S25 of the edge sensor 25 stored in the storage section 29. Based on this result, the pulse motor drive circuit 38 rotates and controls the pulse motor 19 to position the orientation flat 2 at a predetermined rotation angle position.

The operation of this embodiment will be explained in detail together with the positioning mode. First, place wafer 1 in vacuum chuck 1.
Fix it at 8. At this time, there may be a deviation between the center of the wafer 1 and the rotation axis of the pulse motor 19 to the extent that the periphery of the wafer 1 does not protrude from the edge sensors 24 and 25 even if the wafer 1 rotates once. Next, the pulse motors 20 and 22 are operated to move the rotating shaft of the pulse motor 19 to the midpoint of the two edge sensors 24 and 25, and then stopped.
This will be the initial position. Next, the wafer 1 is rotated by the pulse motor 19, and the edge sensor 2 at this time is rotated.
The position signals S ₂₄ and S ₂₅ of Nos. 4 and 25 are digitized at every constant rotation angle Δθ and stored in the storage units 28 and 29. On the other hand, the position signals S ₂₄ and S ₂₅ pass through differentiating circuits 30 and 31 and band filter circuits 32 and 33, respectively, and become differentiated waveforms as shown in FIG. Rectangular waves D ₂₄ and D ₂₅ are formed that roughly indicate the position of the orientation flat 2 as shown in Figures a and b, and the storage unit 2
8, 29. Next, the first calculation unit 36
The position signals S ₂₄ , which are the output signals of the edge sensors 24 and 25 stored in the storage units 28 and 29,
From S ₂₅ and the square wave signals D ₂₄ and D ₂₅ , wafer 1
The deviation between the center of the pulse motor 19 and the rotation axis of the pulse motor 19 is calculated.

To explain the mode of this calculation in more detail, as shown in FIG.
The center of the wafer 1 when the reference point P on the wafer 1 is rotated by θ after aligning the axis of rotation with the origin O is expressed as O'. The reference point P can be set arbitrarily, but it is assumed that it does not change during one revolution of the wafer 1. The distance from the origin O to the periphery of the wafer 1 at this time y ₂₄
and y ₂₅ are expressed by the following equation for the portions other than the orientation flat 2.

y ₂₄ = e cos(θ−)+√ ² − ^{2 2} (−
)
...(1) y ₂₅ = −e cos (θ−) + √ ² − ^{2 2} (−
)
...(2) Here, θ is assumed to be 0° when point P is on the positive y-axis, and then rotated clockwise. R is the radius of wafer 1, e is the origin O and the center of wafer 1
The distance from O' is the angle POO' (the counterclockwise side with respect to the straight line OP is positive.) At this time, equation (2) - equation (1) = -2e cos (θ-). That is, the position signals of the edge sensors 24 and 25
The difference between S ₂₄ and S ₂₅ is the orientation flat 2
In other parts, it becomes a precise cosine wave with an amplitude of 2e. This waveform is shown in FIG. On the other hand, equation (1) x equation (2) = R ² -e ² . That is, edge sensors 24, 25
The product of the position signals S ₂₄ and S ₂₅ has a constant value in areas other than the orientation flat 2, regardless of the rotation angle of the wafer 1. This shape is shown in FIG.

As described above, the storage units 28 and 29 store position signals S ₂₄ and S ₂₅ as shown in FIG. 5 and rectangular wave outputs D ₂₄ and D as shown in FIG. 7 for each constant rotation angle Δθ. ₂₅ is memorized. Therefore, a position signal of a portion other than the orientation flat is selected from the position signals _S24 and _S25 . This selection is made by S ₂₄ ×D ₂₄ ×D ₂₅ and S ₂₅ ×D ₂₄ ×D ₂₅ . In other words, when the square wave outputs D ₂₄ and D ₂₅ are "1" at the same time, the output signal is the position signal.
S ₂₄ and S ₂₅ become the output as they are, and the square wave output D ₂₄
Further, when D ₂₅ is "0", the output is zero.

Next, from the position signals of the portions other than the orientation flat obtained as described above, the least squares method is applied to the following equation to obtain the unknowns e and.

S ₂₅ −S ₂₄ =−2e cos(θ _i −) (3) In equation (3), θ _i is 0° and each rotation angle obtained by multiplying the angle Δθ by an integer.

To find the unknowns e and by applying the least squares method, do as follows.

Position signal S ₂₄ obtained at each constant rotation angle Δθ,
If the difference in S ₂₅ is S _i , equation (3) becomes S _i =−2e cos(θ _i −) (3-1).

If the residual error for each rotation angle is expressed by r _i , then r _i =S _i +2e cos(θ _i −)=S _i +2e cosθ _i cos+2e
sinθ _i sin...(3-2)

The amount of deviation (Δx, Δy) between the origin O and the center O' at a certain rotation angle θ _i is Δx=e cos, Δy=e sin
Therefore, equation (3-2) becomes r _i =S _i +2Δx cosθ _i +2Δy sinθ _i (3-3). The least squares method is applied to this equation (3-3). In other words, Δx, Δy are set so that Σr ² _i is minimized.
Establish.

∂Σr ² / _i / ∂(Δx) = ΣS _i cosθ _i +2ΔxΣcos ²
θ _i +2ΔyΣsinθ _i cosθ _i = 0...(3-4) ∂Σr ² / _i / ∂(Δy) = ΣS _i sinθ _i +2ΔxΣsin
θ _i cosθ _i +2ΔyΣsin ² θ _i =0……(3-5) Δx＝ΣS _i sinθ _i Σsinθ _i cosθ _i −ΣS _i cosθ _i Σs
in ² θ _i /2 {Σcos ² θ _i Σsin ² θ _i −(Σsinθ _i cosθ _i )
² }……(3-6) Δy＝ΣS _i sinθ _i Σsinθ _i cosθ _i −ΣS _i cosθ _i Σc
os ² θ _i /2 {Σcos ² θ _i Σsin ² θ _i −(Σsinθ _i cosθ _i )
² }...(3-7) According to equations (3-6) and (3-7), e is: e=√() ² +() ² ...(3-8) = tan ^-1 (Δy /Δx) ...(3-9)

Based on the deviation thus determined, the pulse motors 20 and 22 are driven via the pulse motor drive circuits 39 and 40 to align the center of the wafer 1 with a predetermined position. On the other hand, the second calculation unit 37 calculates the rotation angle position of the orientation flat 2 by comparing and calculating the position signal S ₂₅ which is the output signal of the edge sensor 25 stored in the storage unit 29 and equation (2). Calculate accurately. More specifically, in addition to the information representing the distance e and the angle, the first arithmetic circuit 36 outputs a value obtained by multiplying the stored contents selected from the storage sections 28 and 29, and the previously determined distance e. Information representing the radius R of the wafer 1 is output from the wafer 1. Based on this information, the distance y ₂₄ at an arbitrary angle θ is calculated, and when this value is compared with the position signal S ₂₅ stored in the storage unit 29, the two values are different. The rotation angle position of the orientation flat 2 can be accurately specified as a range.

This will be explained using a mathematical formula. position signal
If the product of S ₂₄ and S ₂₅ is expressed as m, then m=R ² −e ² ...(4). From formula (3-8), R is R=√+() ² +() ² ...(5). Distance y ₂₄ at a certain rotation angle θ _i according to equation (1)
is found.

Equation (1) is an equation that holds true for the circular portion of the wafer 1, and deviations from equation (1) occur in the orientation flat (straight portion) 2. In other words, this deviation E _i E _i = y ₂₄ - S ₂₄ ≠ 0 (orientation flat part) E _i = y ₂₄ - _{S 24} = 0 (other than orientation flat part), so the orientation flat The rotation angle position of the tube can be determined accurately.

Then, based on the calculation result, the pulse motor 19 is driven via the pulse motor drive circuit 38 to align the orientation flat 2 with a predetermined rotation angle position, thereby completing the predetermined positioning of the wafer 1.

In the above embodiment, the position of the orientation flat 2 was determined by comparing the position signal S25, which is the output signal of the edge sensor ₂₅ , with the theoretical equation (2) in the second calculation section 36. , the reference wafer is 1
The waveform is rotated and corrected using a value calculated by the same method in the first calculation unit 36, and stored in a new storage unit as a reference value, and this and the position signal _S25 of the edge sensor 25 are compared and calculated. The rotational angular position of the orientation flat 2 can also be calculated by . Furthermore, it goes without saying that the object to be positioned is not limited to the wafer 1, but similar effects can be obtained as long as it is a disc-shaped body of the same type.

As specifically explained above with the examples,
According to the present invention, the deviation between the center of a disc-shaped body having an orientation flat, which is a notch provided on the outer periphery, and the center of the rotation axis that rotates the disc-shaped body is calculated, and based on that value, the disc-shaped body is Since the shaped body is positioned, it is possible to position the disc shaped body without requiring a centering jig. At this time, the only contact with the disk-shaped body is the vacuum chuck, which is very suitable for positioning a semiconductor substrate where contamination is averse. Furthermore, since it is possible to position the disc-shaped body even if it is vertical, there is an advantage that the disc-shaped body can be aligned in the same direction while it is being transferred, which reduces the transfer time. Very helpful.

[Brief explanation of the drawing]

Figure 1 is an external view of the semiconductor substrate (wafer), Figure 2
3 is a block diagram showing another positioning device according to the prior art, FIG. 4 is a block diagram showing an embodiment of the present invention, and FIG. 5a and 5b are waveform diagrams showing the output waveforms of the edge sensors 24 and 25, respectively; FIGS. 6a and b are waveform diagrams showing their differential waveforms; and FIGS. 7a and b are rectangular waves based on them. 8 is an explanatory diagram for explaining the calculation principle in the embodiment, and FIG. 9 is a waveform diagram showing the edge sensor 2.
FIG. 10 is a waveform diagram showing the waveform when the difference between the outputs of edge sensors 24 and 25 is calculated, and FIG. In the drawing, 1 is a semiconductor substrate (wafer), 2 is an orientation flat, 18 is a vacuum chuck,
19, 20, 22 are pulse motors, 24, 25 are edge sensors, 28, 29 are storage units, 36 is a first
The arithmetic unit 37 is a second arithmetic unit.

Claims (1)

[Scope of Claims] 1. A device for positioning a disc-shaped body having an orientation flat which is a notch in a part of its outer periphery, the disc-shaped body being held at an arbitrary position, and the disc-shaped body being held at an arbitrary position. a rotating means capable of making at least one rotation around a straight line passing through the center of the body or an arbitrary point near the center and perpendicular to the disk-shaped body as a rotation axis and stopping at a predetermined rotation angle position; and a rotation means perpendicular to the rotation axis of the rotation means. The change in the position of the periphery of the disc-shaped body as the disc-shaped body rotates is arranged opposite to each other on a straight line orthogonal to the rotation axis in a plane,
a pair of edge sensors that detect positional changes in a coordinate system within the plane, with the intersection of the rotation axis and the straight line as the origin and the straight line as the coordinate axis, and output a position signal corresponding thereto; a storage unit that stores position signals output by each edge sensor as the body rotates, along with corresponding rotation angle information; A formula is preset that shows the difference between the distance and the distance to the other edge.
By substituting the difference between the position signals of the pair of edge sensors at a portion other than the orientation flat in the memory contents of the storage section into a preset equation and applying the method of least squares, the disc-shaped body can be calculated. A first step that calculates the deviation of the center from the center of the rotation axis as a distance and angle with respect to the origin of the coordinate system.
a calculation unit that moves the disc-shaped body by moving the rotation unit so as to correct the deviation calculated by the first calculation unit; A disc-shaped body obtained from the distance and angle from the origin of the coordinate system, the product of the position signals of the pair of edge sensors at the portion other than the orientation flat among the contents stored in the storage unit, and the distance from the center of the rotation axis. The distance from the rotation axis position to the periphery at an arbitrary angle of a virtual disk-shaped body without orientation flat is calculated based on the radius of and, and this calculated distance and the position signal stored in the storage section a second calculation for calculating the rotational angular position of the orientation flat by comparing the two values and assuming that the orientation flat position is the orientation flat position when the two values are different, and causing the rotation means to occupy a predetermined rotational angular position. 1. A positioning device for a disc-shaped body, comprising: