JPS6214090B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an electron beam exposure method capable of accurately drawing a drawing pattern including a hatched portion at high speed.

In recent years, electron beams have been widely used to draw patterns for highly integrated semiconductor circuits, etc., and in particular, electron beams that use variable beam dimensions for exposure are attracting attention because of their ability to increase the drawing speed. In addition, the present inventor previously filed Japanese Patent Application No. 51-90182 (Japanese Unexamined Patent Publication No. 53-16578).
proposed high-speed writing by varying the beam width and scanning in a direction perpendicular to the variable width. This allows variable beam width settings for the electron beam.
For example, as shown in Fig. 1, two apertures 1,
2, the aperture image of the electron beam passing through the first aperture 1 is deflected by the deflector 3 and projected onto the second aperture 2, and from the overlap, an irradiation electron beam with variable beam width Bw is obtained. It is done by certain things. However, with this method, there is no problem when the drawing pattern consists of erect figures that are the same as the electron beam shape, but when it includes diagonal lines, it is extremely difficult to draw the shaded areas. It has its drawbacks. Therefore, conventional techniques have been devised, such as changing the direction of the aperture that defines the beam shape, and using an aperture shaped like a baseball home plate to selectively obtain upright and oblique electron beams. However, there are many problems such as the fact that it is extremely difficult to change the direction of the aperture at a high speed, and double exposure areas occur due to electron beams of different shapes, so it has not been put into practical use. Ta. For this reason, it is necessary to perform subdivided drawing using an elongated strip-shaped electron beam, resulting in problems such as a significantly slow drawing speed.

The present invention has been made in consideration of these circumstances, and its purpose is to provide a simple and practical method that allows complex drawing patterns including diagonal line areas to be drawn at high speed with simple control. The object of the present invention is to provide a high-quality electron beam exposure method.

The outline of the present invention is to vary the beam width of an electron beam scanned in a predetermined direction in a direction perpendicular to the scanning direction, and to adjust the irradiation position of the electron beam in the beam width direction at least in relation to the beam width control. By making variable control possible, it is possible to easily draw a diagonal shape, thereby effectively achieving the above-mentioned purpose.

Embodiments of the present invention will be described below with reference to the drawings.

The exposure of the object 5 to be exposed with the electron beam 6 by this apparatus basically involves spreading the entire exposure area 7 of the object 5 into a plurality of frames F and a plurality of subcells constituting the frame F, as shown in FIG. It is divided into SCs, and each subcell SC is used as a basic unit.
That is, the frame F is set by dividing the entire exposure area 7 into each width FW in the electron beam scanning direction (direction H in the figure) using the scanning width of the electron beam 6 as a unit. In addition, the subcell SC uses the length FL direction of the frame F (direction V in the figure) as a pattern shape change point of the drawn pattern, that is, the bending point or pattern edge part as the dividing point, and the maximum width of the electron beam 6.
It is defined by classifying Bwmax as the maximum width SCHmax. In other words, the drawn pattern 8 is divided by its bends and edges, and is divided into individual subcells.
It is divided into SC.

Therefore, the electron beam 6 is the width SCH of the subcell SC.
The beam width Bw is variably set according to the drawing pattern 8, and the subcell SC is subjected to blanking control according to the drawing pattern 8 and scanned. and one subcell SC
When you finish scanning the subcell adjacent to this
The scanning is performed by moving the position to SC by deflection shift. Note that if the drawing pattern 8 does not exist in the subcell SC subjected to electron beam scanning, the subcell SC is not scanned and the process is moved to the next subcell SC. In this way, when exposure of the drawing pattern 8 to all subcells SC within one frame F is completed by electronic deflection scanning of the electron beam 6 and scanning position control,
For example, a sample stage (not shown) on which the object to be exposed 5 is placed
By step movement in the H direction in the figure, the next frame F is aligned with the scanning area of the electron beam 6. Then, pattern exposure of this frame F is performed in the same manner, and this is sequentially repeated to achieve exposure drawing of the drawing pattern 8 over the entire exposure area 7.

By the way, in this method in which electron beam exposure is performed using the subcell SC as a basic unit as described above, the most distinctive feature is that the oblique exposure pattern 8 is formed in the subcell SC, that is, the pattern edge is the same as that of the electron beam 6. If the beam is tilted obliquely to the scanning direction and the width direction, the width Bw of the electron beam 6 is variably set, and the irradiation position is displaced in the width direction of the beam in relation to the variable setting of the beam width Bw. However, diagonal lines are drawn by scanning.

This device equipped with such a function is, for example, a third
It is configured as shown in the figure. Incidentally, FIG. 3 schematically shows the basic components of the electron beam apparatus. The electron beam 6 emitted from the electron gun 11 is subjected to blanking control by a blanking deflection plate 12 and then projected onto the first aperture 13 via a condenser lens (not shown). A first electron lens 14 is located at the position of the first aperture 13.
An aperture image (elongated rectangular shape) of the electron beam formed by the aperture 13 is projected onto a second aperture 15 by an electron lens 14. A second electron lens 1 is provided between the first and second apertures 13 and 15.
6. A paraxial aperture 17 provided in the electron lens 16 and deflection plates 18 and 19 for deflecting the electron beam projected onto the second aperture 15 and displacing its projection position are provided, respectively. . By controlling the projection position of the aperture image by the deflection plates 18 and 19, the beam width Bw of the electron beam is variably adjusted based on the overlap between the image and the second aperture. Note that the first and second apertures 13 and 15 are formed of elongated slits in the width direction (V direction) of the subcell SC, and the aperture image formed by the first aperture 13 is deflected in the V direction and onto the second aperture 15. irradiated. Therefore, the cross-sectional shape of the electron beam 6 passing through the second aperture 15 is such that the edge of the second aperture 15 in the deflection direction of the electron beam 6 is a fixed side, and the edge of the first aperture 13 is a fixed side.
The beam width Bw is controlled using the opposite side regulated by as the moving side. Further, the width of the electron beam can be variably set, for example, with a maximum beam width of 4 μm and a variable step width of 0.125 μm with quantization precision.

Thus the first and second apertures 13,1
The electron beam 6 whose beam width Bw is variably set via the scanning deflection plate 20 is deflected in the H direction by the scanning deflection plate 20 and scanned.
The beam is slightly deflected in the beam width direction (V direction), and the irradiation position is moved in the beam width direction. The irradiation position control of the electron beam 6 by the deflection plates 21 and 22 is performed in conjunction with the beam width control of the electron beam 6 during the diagonal line writing described above. The electron beam 6 subjected to such deflection control is then irradiated onto the object 5 to be exposed via a reduction lens and an objective lens 23 (not shown).

That is, the feature of this device is that, in addition to the conventional electron beam device, it can deflect the beam-shaped electron beam 6 in the beam width direction, for example, by using deflection plates 21 and 22 (deflection coils may also be used). The present invention has a configuration in which the irradiation position can be variably set in the beam width direction in connection with the variable setting of the beam width Bw of the electron beam 6.

Note that the deflection accompanying the beam width variation in the beam width direction may be performed by independently providing the deflection plate or deflection coil described above, but the scanning deflection function of deflecting the electron beam in the width direction may be used as is. Take advantage of
Of course, it is also possible to perform control by combining the deflection signal for the width direction displacement with the deflection scanning signal.

One subcell with a device configured in this way
SC drawing is performed as follows. FIG. 4 schematically shows this drawing, and shows the electron beam 6 as a block body that is thin in the scanning direction (H direction), long and thin in the width direction (V direction), and is bent by deflection. be. That is, the electron beam 6 irradiated onto the first aperture 13 is subjected to blanking control in the direction of arrow B. This blanking control signal BIk is given according to the drawing pattern 8 in the subcell SC, and is turned off when the scanning position of the electron beam 6 is related to the exposure area (drawing pattern 8). When the exposure object 5 is allowed to be irradiated, and there is no drawing pattern 8 at the scanning position, the electron beam 6 is controlled to be ON.
It acts to block the irradiation of Therefore, the electron beam 6 is applied only at the timing when the drawing pattern 8 exists.
irradiation is performed. On the other hand, the electron beam 6 subjected to the blanking control is subjected to deflection control in the beam width direction shown by arrow D between the first and second apertures 13 and 15. By this deflection, the beam width Bw of the electron beam 6 can be variably set as described above. This beam width Bw
The deflection control signal Def responsible for variable setting has a signal level corresponding to the drawing pattern width at the scanning position of the electron beam 6, and is controlled so that the beam width Bw becomes zero at the maximum deflection. In addition, the beam width Bw is the subcell within one subcell SC.
Needless to say, a deflection control signal of a predetermined level is always applied so as not to exceed the width SCH of SC. Further, the positions of the first and second apertures 13 and 15 are shifted in advance in the V direction,
By obtaining the electron beam 6 in a superimposed shape using the deflection control signal Def, it is possible to obtain a proportional correspondence between the signal level and the beam width Bw. It is also useful to limit the value based on the subcell width SCH.

In this manner, one of the edges in the width direction of the electron beam 6 whose beam width Bw has been variably controlled becomes a fixed side as described above, and the irradiation position is kept constant. Then, the width Bw of the electron beam 6 expands and contracts in the V direction based on the irradiation position of this fixed side, and the beam width Bw changes. Therefore, it is possible to vary the beam width Bw by following the drawing pattern 8 shown in FIG. 4, for example, a rectangular parallelepiped pattern or a pattern having an upward-left oblique side, but conversely, a pattern having an upward-left oblique side Although the beam width Bw can be made to correspond to the pattern width for a pattern, a difference will occur in the drawing position. Therefore, in this device, the beam width
The electron beam 6 with variably set Bw is directed to the deflection plate 2.
1 and 22 in the direction of arrow C in the figure, making it possible to displace the electron beam irradiation position in the width direction. Position correction signal responsible for this deflection
Corr performs position correction in response to the drawing pattern 8, for example, in the above example, the pattern 8 having the upper left oblique side, and its signal level corresponds to the pattern width. ing. By this signal Corr, the irradiation position of the fixed side edge of the electron beam 6 is shifted in the width direction, and the other moving side edge is aligned with the segmented position of the subcell SC. In this way, by scanning the electron beam 6 with variable beam width Bw and correction of the irradiation position, the above-mentioned pattern 8 having the upper left oblique side
is drawn. Incidentally, the scanning control signal Scn is applied to the scanning deflection plate 20, and the scanning control signal Scn
The electron beam 6 is deflected in the direction of arrow S,
The subcell SC is scanned.

Incidentally, the signal level of the position correction signal Corr influences the shape of the drawing pattern 8 as shown in FIGS. 5a to 5d. In other words, when the signal level is zero (characteristic a) and no position correction is applied, the irradiation position of the fixed edge of the electron beam 6 does not change, so only the irradiation position of the moving edge changes the beam width.
It will be displaced as Bw changes. Therefore, as shown in FIG. 5a, the drawing pattern 8 including the oblique side always has a wide bottom side and widens toward the end, making it impossible to obtain the desired pattern. Furthermore, when the signal level is less than the narrow level of the beam width Bw (characteristic b), the irradiation position of the fixed side edge of the electron beam 6 is only displaced to the center of the subcell SC, so the moving side edge ends up being the first. As shown in FIG. 5b, the desired pattern cannot be obtained since the dividing position of the subcell SC is not reached.
On the other hand, if the signal level is equal to or higher than the narrow level of the beam width Bw, the irradiation position of the moving edge of the electron beam 6 is located in another subcell, as shown in FIG. 5d.
This results in entering the SC area. Therefore subcell
Not only is it impossible to form a desired pattern within the SC, but there is also a risk that the exposure areas of other subcells SC will be infringed upon, causing trouble in multiple exposure.
Therefore, the signal level of the position correction signal Corr is related to the beam width Bw, and the edge of the moving side of the beam 6 is always positioned on the dividing line of the busel SC to obtain the desired pattern as shown in FIG. 5c. It is necessary. Therefore, for example, the amount of deflection required to vary the beam width Bw of the electron beam 6 can be
Control may be performed such that it is combined with the deflection amount corresponding to the width SCH of SC, and this combined value is given as the signal level of the position correction signal Corr. Depending on the deflection method, it is also possible to use the deflection amount for beam width control as it is as a position correction signal.

In this way, the position of the fixed side edge can be changed by regarding the moving edge as the beam width Bw changes as an auxiliary fixed edge, making it possible to effectively obtain a pattern with the desired shape. It becomes possible.
It should be noted that, although the explanation has been given here by taking the upper left oblique side as an example, it goes without saying that the same applies to the case where the pattern edge is in the downward right corner state.

In this way, according to this device, one subcell
Not only a rectangular drawing pattern 8 included in the SC, but also a trapezoidal drawing pattern 8 including an upper right oblique side or a lower left oblique side can be drawn extremely quickly. Furthermore, scanning of the electron beam 6 can be continued with high precision and at high speed through simple control such as varying the beam width Bw and correcting the irradiation position of the electron beam 6 whose beam width has been varied. Furthermore, since the above control can be performed purely electronically, high-speed exposure is possible due to faster control response. By the way, conventionally, aperture 1
It has been considered to form an oblique beam by changing the direction of the beams 3 and 15 according to the pattern shape, but this involves mechanical operation, so it is impossible to increase the exposure speed. Also, by changing the width and shape of the electron beam 6,
In addition, a method in which the irradiation position is set independently in the V and H directions has the disadvantage that control becomes considerably complicated. In other words, the present exposure method has extremely great practical advantages, such as enabling high-precision and high-speed electron beam drawing without complicating the device configuration and exposure (drawing) control as in the past. In addition, this can be achieved by simply adding a position correction function to a conventionally configured device.
Full of versatility. Further, by varying the scanning speed according to the width of the subcell, the drawing speed can be further increased.

By the way, according to this apparatus which performs pattern drawing as described above, its control can be performed very simply as follows. 6a to 6c show the above-mentioned control form in comparison with the conventional method, and a shows a pattern to be drawn. When drawing such a pattern, this apparatus first divides the pattern into a plurality of subcells SC at the corner positions and edges of the pattern shape. in this case,
It is divided into a first subcell SC extending from line _L1 to _L4 and a second subcell SC extending from line _L5 to _L7 . Therefore, the first and second subcell SC
Each drawing pattern shown in the drawing pattern can be expressed by information on whether or not a pixel pattern exists at the edge of the subcell SC, so for the first subcell SC, lines _L1 and _L4 information, and for the second subcell SC lines L ₅ and L ₇
Once you have this information, you can control it. In FIG. 6, "1" indicates that a pattern exists, and "0" indicates that there is no pattern. Therefore, for example, the line
Blanking control is performed based on whether or not there is a "1" level signal at L ₁ and L ₅ , and the slope is controlled based on the difference in the "1" level signal position and the distance between lines L ₁ and L ₅ . By determining the inclination and accordingly varying the beam width Bw and correcting the position as the electron beam 6 scans, it becomes possible to draw a pattern in the desired shape as described above.

In contrast, in the conventional method, the drawing pattern must be divided into pixels, as shown in FIG. 6c, and information indicating whether or not to draw should be given to each pixel. Therefore, in order to control this, it is necessary to store a huge amount of information in a large-capacity memory, and it is necessary to sequentially read out this information, which has the drawback that the control becomes extremely complicated and the responsiveness deteriorates.

As is clear from such a comparison, it can be seen that the control in this device is very simple. It is also shown that the memory capacity for storing information necessary for control can be set small, and the device configuration can be simplified.

Figure 7 shows the beam width Bw variable control information (deflection control signal) according to the information on the two lines mentioned above.
This figure shows an example of the configuration of a control circuit that obtains Def). If there is a shift in the position of information indicating the presence of a pattern between the two lines defining the subcell SC, first the length of the shift M is determined. This shift length M is generally given as a bit length. This shift length M data is input to the coefficient unit 25 and subjected to calculation with the width SCH (bit length h) of the subcell SC, and the ratio, that is, the slope is determined. After that, this slope data is sent to the cumulative adder (ADD) 2.
6 and are cumulatively added together with scanning, and calculated as beam width data at that scanning position.
The D/A converter 27 converts the beam width data into analog and inputs it to the coefficient unit 28. In this coefficient multiplier 28, a coefficient N/h corresponding to the deflection sensitivity of the electron beam 6 is added to the analog voltage value, and the resultant value is output as a deflection voltage in the width direction of the electron beam. Therefore, the deflection voltage changes with scanning according to the slope of the oblique side of the pattern, and the oblique side is drawn here. Further, if there is no difference in the pattern information between the two lines, an electron beam with the maximum width defined by the width of the subcell will be obtained according to the full scale data of the adder 26.

In this way, with only very simple control calculations,
Control information (deflection control signal Def) that continuously varies the pattern width necessary to draw the hypotenuse pattern
can be obtained, and the position correction signal Corr can also be easily obtained from this information.

As explained above, according to the present apparatus, it is possible to extremely smoothly draw a pattern including a hypotenuse portion, which has conventionally been considered to be difficult to draw at high speed. Furthermore, the above-mentioned object can be effectively achieved by simply electronically controlling the deflection of the electron beam while maintaining a constant scanning state of the electron beam. Furthermore, since there is no need for mechanical rotation of the aperture, it is possible to easily increase the speed of drawing. For example, if the above-mentioned control calculation is performed in about 25 nsec, real-time control is possible, and there are effects that cannot be expected from conventional devices, such as great practical advantages.

Note that the present invention is not limited only to the above-described embodiments. For example, the settings of frames and subcells may be made based on the writing pattern and the deflection speed specifications of the electron beam. Furthermore, a mechanism for varying the electron beam intensity may be provided to perform drawing according to pattern accuracy. Furthermore, this device can be applied not only to exposure devices but also to CRT devices as shown in FIG. In this case, if the beam width is controlled using the cathode 31 having an elongated emission part and the beam shielding plate 32, and the deflection-controlled electron beam is projected onto the target surface 33, a high-speed, high-density CRT device can be realized. realizable. Furthermore, the beam width
It goes without saying that control calculations for obtaining Bw control information, etc. may be performed using a microcomputer or the like, and other calculation processes may also be performed as appropriate according to specifications. In short, the present invention can be implemented with various modifications without departing from the gist thereof, and the system configuration of the electron optical lens barrel is not particularly limited.

[Brief explanation of the drawing]

Figures 1a and b are diagrams for explaining the principle of changing the electron beam width, Figure 2 is a diagram showing the drawing pattern frame and subcell division, and Figure 3 is a schematic diagram of the configuration of one embodiment of this device. Figure 4 is a diagram showing the variable width of the electron beam and control of the irradiation position displacement, Figures 5 a to d are diagrams showing the effect of irradiation position correction, and Figure 6 is a diagram showing the effect of irradiation position correction.
Figures a to c are diagrams showing drawing patterns and examples of their information in comparison with a conventional device, Figure 7 is a diagram showing an example of the configuration of a control calculation circuit, and Figure 8 is a diagram showing another embodiment. DESCRIPTION OF SYMBOLS 1, 2... Aperture, 3... Deflection plate, 6... Electron beam, 7... Whole exposure area, 11... Electron gun, 13... First aperture, 15... Second aperture, 1
8, 19... Deflection plate (beam width control), 20... Scanning deflection plate, 21, 22... Deflection plate (position correction), 23
...objective lens, 25, 28...coefficient unit, 26...cumulative adder, 27...D/A converter, 31...cathode,
32... Shielding plate, F... Frame, SC... Subcell,
Bw…Beam width.

Claims (1)

[Claims] 1. The electron beam exposure area is divided into a plurality of frames corresponding to the scanning width of the electron beam, and the area is divided into a plurality of frames according to the change point of the exposure pattern included in each frame and the maximum dimension of the cross-sectional width of the electron beam. Therefore, the frame is divided into a plurality of subcells to set the basic unit of exposure, and the cross-sectional width of the electron beam is varied according to the change in the width of the exposure pattern, and the two ends of the electron beam in the cross-sectional width direction are The electron beam irradiation position is displaced in the cross-sectional width direction of the electron beam so that one end is a fixed end and the other is a variable end, and the fixed end and the variable end are switched according to the exposure pattern. Line exposure method. 2 Information on the exposure pattern is given as information on the presence or absence of an exposure pattern at the edge that divides the subcells, and the pattern shape is determined from the information on the two edges to control the displacement of the beam cross-sectional width and irradiation position. The electron beam exposure method according to claim 1, wherein the electron beam exposure method is 3. The electron beam exposure method according to claim 1, wherein the scanning of the electron beam proceeds to the scanning of the next subcell without scanning the subcells that do not include an exposure pattern with the electron beam.