JPS5933969B2 - Electron beam exposure method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for drawing a trapezoid, a rectangle, or a triangle as an accumulation of a plurality of linear or band-shaped drawing areas.

Each time a line segment is drawn, the position to be drawn is shifted one step at a time by a certain amount in the direction perpendicular to the line segment, and the next line segment is drawn.In this way, multiple line segments are arranged closely enough. It is practiced to draw trapezoids, rectangles, triangles, etc. as an accumulation of these. In such an exposure method, for example, when drawing five trapezoids as shown in FIG. 1, the following procedure has conventionally been used.

That is, the coordinate data of one end A of the base (XA
O, YO), data L representing the length of the base, data H representing the height of the trapezoid, data XD representing the length of AK (however, DKIAB), data αL representing the difference in length between AB and DC ( However, data such as AD/MC) and the shift amount ΔY in the Y direction from one line segment to the next line segment are supplied. First, a line segment of length L is drawn starting from point A based on these data. Next, the coordinate data of the next line segment (XAO+XD゜ΔY/H, Y0
+△Y) is calculated in hardware, the length of the next line segment L-αL-AY/H is calculated in hardware, and the coordinates (X
A line segment u of length L-αL-6H/H is drawn using A01-XD·°Y/H, YO+ΔY) as a base point. To draw the next line segment, calculate the coordinates of the next point on the straight line AD and the length of the next line segment, draw a line with the calculated length starting from this point, A trapezoid is drawn by repeating similar drawings (in reality, the intervals between each line are sufficiently dense). By the way, the above-mentioned calculation of the coordinates of the starting point and the calculation of the length of a line segment are both done digitally, so the calculated coordinates and length may vary depending on the accuracy of digitization. It is unavoidable that an error of one unit quantity (denoted as LSB) will intervene.

Therefore, in the figure actually drawn, an error of maximum 1/2·LSB occurs for each point on the side AD. By the way, as is clear from the above description, in the conventional method, the drawing position coordinates of each point on the side BC are calculated by calculating the drawing position coordinates on the side AD, and based on the calculated coordinates, This point is determined as a point shifted in the X direction by a length L-αL·ΔY/H. Therefore, if the error number A that occurs when calculating the drawing position coordinates of each point on side BC is maximum, the error that occurs when calculating each point on side AD and the length L-αL・ΔY /H is added to the error that intervenes when calculating. By the way, when calculating the length L-αL・ΔY/H, it is inevitable that an error of up to 1/2・LSB will occur, so the drawing position coordinates of each point on the side BC will have a maximum of 1/2.・LSB+1/2・LSB=LSB error intervenes. Therefore, the position of each point on side BC may deviate significantly (LSB) from the ideal position,
This was one of the reasons that prevented higher precision drawing. The present invention has been made to solve these drawbacks of the conventional method.
An embodiment of the present invention will be described in detail below with reference to the drawings.

FIG. 2 is for showing an example of an apparatus for carrying out the present invention. In the figure, 1 is an electronic computer, and the computer uses the computer to draw a point Af) of a trapezoid as shown in FIG. Data XAO representing the value is supplied to the register 2.

From the electronic computer 1, a line segment is shunted by ΔY in the Y direction.
Data ΔXA (=ΔYt
anα) is supplied to register 3. The output of register 2 is supplied to register 6 via switch 5, and
The output of register 3 is supplied to adder/subtracter 7. The output of the adder/subtractor 7 is supplied to a register 6 via a switch 5, and the output of the register 6 is supplied to one input terminal of the adder/subtractor 7. When the switch 5 first transfers the contents of the register 2 to the register 6, it is switched from the connection state shown by the dotted line to the connection state shown by the solid line. As will be described later, addition by the adder/subtractor 7 is performed each time one line segment is drawn, since data is read into each register at a fixed timing. Therefore, the contents of register 6 are the initial stored value
Increases by ΔXA from AO. Therefore registers 2, 3,
6, switch 5, and adder 7 are on side A.
This is a first X-coordinate calculation circuit for calculating the X-coordinate of each starting point along D. The output data of the first X-coordinate calculation circuit 8 is transferred to the up-down counter 4. Side BC when drawing one line segment at a time starting from each point along side AD
A first X coordinate calculation circuit 8 to calculate each end point along
A second X-coordinate calculation circuit 9 is provided completely symmetrically. The circuit 9 has a register 10 in which the X coordinate zeta BXO of point B is stored, and data ΔXB representing how much the X coordinate of each end point on side BC must decrease each time it is shifted by a line segment. The switch 12 operates in the same manner as the switch 5, the register 13, and the adder/subtractor 14. Therefore, the contents of the register 13 decrease from the initial stored value XBO by ΔXB every time one line segment is drawn. A Y coordinate calculation circuit 15 is provided similarly to the first and second X coordinate calculation circuits. The circuit 15 is at point A and point Bf)
It consists of a register 16 to which Y coordinate data YO is supplied from the electronic computer 1, a register 17 to which data ΔY representing the amount of movement in the Y direction for each drawing of a line segment is supplied, a switch 18, a register 19, and an adder/subtractor 20. The contents of the register 19 also increase by ΔY from YO every time one line segment is drawn. Reference numeral 21 denotes a drawing end detection circuit, and this circuit 21 is connected to a register 22 to which data H representing the height of the trapezoid is supplied from the electronic calculator 1.
, a switch 23 that functions similarly to the switch described above;
It consists of a register 24, a subtracter 25, and a comparator 26.

The output of the register 17 and the output of the register 24 are supplied to the subtracter 25 . Therefore, the value stored in the register 24 decreases by ΔY every time one line segment is drawn from the initial data H transferred from the register 22. The comparator 26 has the register 2
The comparator 26 compares the stored value of the register 24 with ΔY, and generates a drawing stop pulse when the stored value becomes smaller than ΔY. , is supplied to the control circuit 27. The output signals of the first and second X coordinate calculating circuits 8 and 9 are supplied to a subtracter 28. The subtracter 28 is for generating data corresponding to the length of each line segment by subtracting the X coordinate of the starting point from the X coordinate of the end point of each line segment, and the contents of the subtracter 28 are stored in the down counter 29. be transferred. Reference numeral 30 denotes a clock pulse generator for electron beam scanning, and the clock pulses from the generator 30 are supplied to the up-down counter 4 as well as to the down counter 29.

The output of the up-down counter 4 is sent to the DA converter 31. The signal is supplied to an X-direction deflector 33 via an amplifier 32. Further, the output signal of the Y coordinate calculation circuit 15 is supplied to a Y direction deflector 36 via a DA converter 34 and an amplifier 35. The output signal of the down counter 29 is supplied to a zero detection circuit 37. The zero detection circuit 37 generates a response pulse when the content of the down counter 29 becomes 0, and is a circuit for detecting the completion of scanning for one line. The output signal of the zero detection circuit 37 is supplied to the reset terminal of the flip-flop circuit 38, and the circuit 38 is reset by the response pulse from the zero detection circuit 37, and the output signal of the flip-flop circuit 38 is switched from high level to low level. It will be done. The response pulse from the zero detection circuit 37 is delayed by a predetermined time in a delay circuit 39 and then supplied to the control circuit 27. The control circuit 2
7, when the signal arrives from the delay circuit 39, the counter 4,
A signal is sent to the set terminal of the flip-flop circuit 38 after a delay time necessary for the flip-flop circuit 29 to take in the data, and the flip-flop circuit 38 is put into a reset state and its output signal is switched to a high level. The output signal of the flip-flop circuit 38 is supplied to the up-down counter 4 and the down counter 29, and these counters count clock pulses only during the period when a high level signal is supplied. The output pulse of the delay circuit 39 is supplied to both counters 4 and 29, and these counters 4 and 29 take in the contents of the register 2 and the subtraction circuit 28, respectively, upon arrival of the pulse. In such a configuration, first the electronic computer 1
Depending on the shape of the trapezoid to be drawn, the data XAO, △
XA, XBO, ΔXB, YO, ΔY, and H are supplied to registers 2, 3, 10, 11, 16, and 17, respectively, and stored.

As a result, data XAO is stored in register 6 and up/down counter 4, and data XBO is stored in register 13.
is stored, data YO is stored in the register 19, and data H is stored in the register 24. Therefore, the down counter 29 also contains the contents of the subtracter 28, XBO-XA.
O is placed. Next, the electronic computer 1 supplies the control circuit 27 with a start signal as shown in FIG. 4a, instructing the start of drawing. As a result, a set signal is supplied from the control circuit 27 to the flip-flop circuit 38, and the circuit 38 is set. Therefore, the circuit 38 supplies high level pulses indicated by A in FIG. 4B to the up/down counter 4 and the down counter 29, and these counters start counting the pulses supplied from the clock pulse generator. Therefore, the up-down counter 4 sequentially increases the count value from the initially placed value XAO, and therefore DA
The deflection signal supplied to the X-direction deflector 33 via the converter 31 also increases sequentially from the value corresponding to the coordinate XAO.
On the other hand, the content YO of the register 19 of the Y-coordinate calculation circuit 15 is supplied to the Y-direction deflector 36 after being DA-converted. Therefore, the electron beam is deflected by the deflectors 33 and 36, and drawing is performed along the base AB of the trapezoid starting from point A. On the other hand, the down counter 29 has the initially placed value XAO
- Count down the count value from XBO to point B or G where the scanning point of the electron beam corresponds to the data XBO.
When U is reached, the count value of the counter becomes O at that moment. As a result, the zero detection circuit 37 generates a response pulse as shown in FIG. After this moment, the electron beam is blanked by a blanking mechanism not shown. At the same time that the flip-flop circuit 38 is set and the counters 4 and 29 start counting, the control circuit 27 sends a control signal to read the zeta as shown by A in FIG. 24 (the signal supply line is not shown), and these are connected to adders and subtracters 7 and 14, respectively.
, 20 and the subtracter 25 are read. Therefore, at the time when scanning of the base AB is completed, the values stored in registers 6, 13, 19, and 24 are each X.
AO+△XA, XBO+△XB, YO+ΔY, H-△Y
Therefore, the output zeta of the subtractor 28 is XB
O+ΔXB-(XAO+ΔXA). Therefore, the response pulse indicated by A in FIG.
is slightly delayed as shown by A in FIG.
XAO+ΔXA) is transferred. Once the transfer is complete,
The control circuit 27 supplies a set pulse as shown by the arrow in FIG. 4f to the flip-flop circuit 38, and the flip-flop circuit 38 is set again as shown by the arrow in FIG. 4b. Therefore, the up-down counter 4 has a count value of XAO+ΔX
The up count starts from A, and at this time the output of the Y coordinate calculation circuit 15 is kept at YO+△Y, so the next line segment is a point (XAO+ΔXA) at a position shifted by △Y in the Y direction.
, YO+ΔY) as the starting point. The drawing position of the electron beam for this line segment is (XBO+ΔXB, YO+ΔY)
When this happens, the zero detection circuit 37 generates a detection pulse as shown by the arrow in FIG. 4c, so that scanning is stopped. In exactly the same way, each time a line segment is drawn, a trapezoid is drawn while the drawing position is shifted by ΔY in the Y direction. Each time one line is drawn, the output data of the register 24 decreases by △Y from H. Finally, the comparator 26 supplies a pulse informing the end of drawing to the control circuit 27, and the control circuit 27 starts drawing. Stop supplying the signal for supply. In the electron beam exposure method based on the present invention described above,
The drawing position coordinates of each point on side BC are
Instead of being determined depending on the drawing position coordinate calculation value of each point above and the calculation value of the length from each point, it is determined independently as when calculating the coordinates of each point on side AD. Since it is determined by the calculation of , the position coordinates have 1/
Only an error of 2.LSB intervenes, and therefore, the deviation of the side BC from the ideal position can be kept to a minimum (1/2.LSB) and accurate drawing can be achieved.

In the above-described embodiment, a trapezoid is drawn as an example, but the present invention can be similarly applied to a case where other shapes such as a triangle are drawn.

[Brief explanation of the drawing]

FIG. 1 is a diagram for explaining a conventional method, FIG. 2 is a diagram for showing an example of an apparatus for implementing the present invention, and FIG. 3 is a diagram for illustrating a trapezoid drawn by the present invention. and FIG. 4 are diagrams for illustrating output signals of each circuit element of the apparatus shown in FIG. 2. 1: Electronic computer, 2, 3, 6, 10, 11, 13, 16
, 17, 19, 22, 24: Register, 4: Up-down counter, 5, 12, 18, 23: Switch, 7, 1
4, 20: adder/subtractor, 8: first X coordinate calculation circuit, 92 second X coordinate calculation circuit, 25, 28: subtractor, 15: Y
Coordinate calculation circuit, 21: drawing end detection circuit, 26: comparator, 27 two control circuits, 29: down counter, 30: clock pulse generator, 31, 34: DA converter, 32, 3
5: Amplifier, 33: X-direction deflector, 36: Y-direction deflector, 37: Zero detection circuit, 38: Flip-flop circuit, 3
9: Delay circuit.

Claims (1)

[Claims] 1. In a method of drawing a trapezoid, rectangle, or triangle by shifting the position to be drawn by one step in a direction perpendicular to the line segment each time a line segment is drawn with an electron beam and drawing the next line segment, an arbitrary After drawing one line segment, when drawing the next line segment, add a constant increment ΔX to the X coordinate of one end of the arbitrary line segment to calculate the X coordinate of one end of the next line segment, and Constant increase ΔX to the X coordinate of the other end of the line segment
', calculate the X coordinate of the other end of the next line segment, calculate the difference from the calculated X coordinate of one end of the next line segment, and the X coordinate of the other end, and calculate the next line segment. An electron beam exposure method characterized in that the length of is controlled based on the difference.