JP2914926B2 - Charged particle beam exposure method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a charged particle beam exposure method in an electron beam exposure apparatus or the like.

[0002]

2. Description of the Related Art In recent years, direct exposure of an LSI by electron beam exposure, that is, exposure has been widely performed without using a mask, but there are many variable rectangular beams, and there are also point beams or fixed rectangular beams. The latter two are generally for research use. As shown in FIGS. 10 and 11, the variable rectangular beam exposure apparatus includes an electron gun 1, two masks 2 and 3 each having a rectangular aperture, and a beam size deflecting device that deflects the electron beam to change the cross-sectional area of the rectangular beam. It comprises an electrode 4, a main deflection lens (coil) 5, and a sub deflection lens (electrode) 6, and irradiates a rectangular beam of an arbitrary size onto a target such as a wafer placed on a sample stage 7. A control unit that controls these generates a CPU 10, a magnetic disk 11 for storing data, a magnetic tape 12, an interface 13, a data memory 14, and a pattern into rectangular patterns to generate size information and positional information of each rectangular pattern. It is composed of a pattern generation circuit (shot decomposition) 15, a pattern correction circuit 16 for performing necessary correction, and a driver (DAC / DAC) for driving the beam size deflection electrode 4, the main deflection coil 5, and the sub deflection electrode 6.
AMP) 17, 18, 19, and a stage controller 21 for controlling a motor 20 for moving the sample stage 7.

That is, in rectangular beam exposure, as shown in FIG. 12, data transferred from the magnetic disk 11 or the magnetic tape 12 to the data memory 14 corresponds to the entire main field. This main field is divided into a plurality of subfields, and the pattern of each subfield is further divided into a plurality of shots (rectangular patterns). At this time, the electron beam movement between each subfield in the main field is performed by the main deflection coil 5, and the electron beam movement between each rectangular pattern in the subfield is performed by the sub deflection electrode 6, and the size of the rectangular pattern is changed. Is adjusted by the beam size deflection electrode 4.

[0004]

SUMMARY OF THE INVENTION However, LSI
In particular, when the size of the LSI is reduced to 0.2 to 0.3 .mu.m, the rectangular beam exposure has the following problems. 1) There is an upper limit on the current density of the rectangular beam that can be used. 2) In drawing an LSI pattern having an enormous number of shots, the number of adjustments of each driver (DAC / AMP) for the main deflection, the sub deflection, and the beam size deflection increases, and therefore, the waiting time for that increases. Take too long.

Therefore, the variable rectangular beam is no longer in the range of 0.2 to
It is not suitable for mass production of 0.3 μm fine patterns. Furthermore, 3) the 0.2-0.3 μm rectangular beam has no difference from the point beam, and there is no merit of the variable rectangular beam.

4) In the case of a 0.1-0.2 μm beam, 2
Since the apertures of the two masks 2 and 3 overlap each other, the dose amount of the small beam becomes relatively unstable. Therefore, an object of the present invention is to provide a stable,
It is an object of the present invention to provide a charged particle beam exposure method which enables uniform exposure at high speed and high accuracy.

[0007]

The means for implementing the method of the present invention to solve the above-mentioned problems is shown in FIG. That is, a charged particle beam generating unit generates a charged particle beam for a charged particle beam exposure apparatus for exposing a charged object to an object on the stage. This charged particle beam passes through a blanking aperture array. In the blanking aperture array, a plurality of apertures each provided with a blanking electrode are arranged in a line in the figure. The blanking electrodes of these apertures are independently turned on and off by blanking electrode driving means. Here, for all the apertures whose blanking electrodes were turned on, the turning on of each blanking electrode was performed so that the product of the intensity of the charged particle beam passing through the aperture and the ON time of the blanking electrode of the aperture became constant. Control the time.

The operation of the above means is represented by FIG. That is, the charged particle beam that has passed through the apertures arranged in a line in the blanking aperture array is divided into a number of small beams (line beams). For example, the line beam is deflected at right angles to the stage movement direction, and with this movement, the blanking electrodes of each aperture are turned on and off independently. At that time, the product of the intensity of the charged particle beam and the on time of the blanking electrodes is turned on. Is driven to be constant. In FIG. 2, for example, a blanking electrode a-
Blanking signals are generated for a ', bb', cc ', and dd' as shown, and the beam is turned on and off.

[0009]

FIG. 3 is a diagram showing an example of an apparatus for implementing the method of the present invention. In FIG. 3, reference numeral 30 denotes an electron gun including a cathode 31, a grid 32, and an anode 33, and a beam emitted from the cathode 31
2. Create a crossover image between the anodes 33. This crossover image is magnified by the crossover magnifying lens 34. In the crossover magnifying lens 34, a quadrupole beam collapsing lens 35, a limiting aperture 36, and an alignment coil 37 are provided. The quadrupole beam crushing lens 35 is, for example, a magnetic lens or an electrostatic lens shown in FIG.
As a result, the crushed crossover image 38 is formed on the blanking aperture array 39. When the electron gun 30 itself generates a crushed crossover beam, the quadrupole beam crushing lens 35 is unnecessary.

The blanking aperture array 39 shown in FIG.
5 × □ 256 apertures 39 _{0 to} 39 ₂₅₅
Are arranged in a line shape in a so-called harmonica shape.
The blanking aperture array 39 is made of, for example, silicon single crystal, an aperture is formed on the single crystal by trench etching, and an electrode is formed on the inner wall. The blanking aperture array 39 is provided inside the converging lens 40. Note that the apertures may be arranged in two rows.

Reference numeral 41 denotes a large blanker which turns on and off the entire beam. When no electric field is applied to the large blanker 41, the on-beam 41 a having passed through a portion of the blanking aperture array where no voltage difference is applied between the electrodes passes through the blanking aperture 43 via the reduction lens 42. On the other hand, the off-beam 41b passing through a portion of the blanking aperture array where a voltage difference is applied between the electrodes is blocked by the blanking aperture 43 after passing through the reduction lens.
Conversely, when an electric field is applied to the large blanker 41, all passing beams are blocked by the blanking aperture irrespective of the state of the individual electrodes of the blanking aperture array 39.

The reduction lens 42 transmits the beam that has passed through the blanking aperture 43 together with the reduction lens 44 and the immersion type lens 50 onto the continuously moving stage 51 on the wafer 53 that has been attracted by the electrostatic chuck 52 at a ratio of 1/500.
Is formed as a reduced image of the aperture of the blanking aperture array 39. In the reduction lens 44, a refocus coil 45 is provided. The refocus coil 45 corrects beam blur due to Coulomb interaction by passing a refocus current determined by the total number of cells in which the aperture of the blanking aperture array 39 is on.

Reference numeral 46 denotes a horizontal scanning deflector, which scans a beam in a direction perpendicular to the movement of the continuously moving stage 51.
Further, the dynamic focus coil 47 and the dynamic stig coil 48 correct beam deflection blur in the horizontal scanning. Reference numeral 49 denotes a stage feedback 8-pole deflector, which performs deflection distortion correction, feedback of continuous movement of the stage 51, and speed correction.

The continuously moving stage 51 is provided in the immersion type lens 50 to increase the resolution. FIGS. 5 and 6 are diagrams (parts 1 and 2) showing the control unit for controlling the main part of FIG. 3 in detail.
The same components as those in FIG. 11 are denoted by the same reference numerals. That is, the pattern generation circuit 15, the pattern correction circuit 16, and the driver (D
AC / AMP) 17, 18, 19, instead of a bitmap generator 61, a blanking generator 62, a sequence controller 63, a blanking controller 64, a deflection controller 65, and drivers (DAC / AMP) 66, 6.
7. A laser interferometer 68 is provided.

The bit map generation circuit 61 reads information (256 bits) for each line beam from the data memory 14 and sends it to the blanking generation circuit 62. As a result, the electrodes of the apertures 39 _{0 to} 39 ₂₅₅ of the blanking aperture array 39 are turned on and off independently. Since the intensity I _i of the beam passing through each aperture differs depending on the position, the ON times τ ₀ , τ
_1, ..., _{τ 255} is circuit 62 _0, 62 _1, ..., it is set to 62 _255. This will be described later.

The blanking control circuit 64 operates at a timing based on a command signal from the sequence controller 63,
Each of the circuits 62 ₀ , 62 ₁ ,
..., 62 ₂₅₅ is started. The deflection control circuit 65 drives the horizontal scanning deflector 46 by the driver 66 based on the horizontal coordinates of the line beam information generated by the bitmap generation circuit 61. In this case, the sample stage (stage) 51 is continuously moved by the stage control unit 21. Therefore, it is necessary to feedback correct the position of the line beam accompanying the continuous movement of the stage 51. Therefore, the deflection control circuit 65 determines that the difference between the position of the stage 51 detected by the laser interferometer 68 and the target position is zero.
The 8-pole deflector 49 is driven so that That is, as shown in FIG. 7, the deflection control circuit 65 includes a register 651 for storing the target position (X, Y) from the sequence controller 63, position information (X ', Y') from the laser interferometer 58, and a register. And a difference calculator 652 for calculating a difference between the deflector and the diverter 61. As a result, the stage error is fed back at high speed. Note that the eight-pole deflector 49 actually adjusts the two-dimensional position of the beam. However, in FIGS. 5, 6 and 7, the eight-pole deflector is shown as one-dimensional for simplicity of description. It is.

Further, a blanking generation circuit 62 shown in FIG.
Will be described with reference to FIGS. That is, the beam are crushed crossover in a line shape in the critical irradiation of crossover image but blanking each aperture (cell) 39 _0-3 of aperture array 39
Beam intensity I _0, I ₁ at 9 _255, ..., I ₂₅₅ is
As shown in FIG. 8A, only a uniformity of 80 to 90% can be obtained. For this reason, as shown in FIG. 8B, the beam-on pulse lengths τ ₀ , τ ₁ ,..., Τ ₂₅₅ of each cell are changed to the beam intensities I ₀ , I ₁ ,. Determined in inverse proportion to ₂₅₅ . That is, τ _i × I _i = constant. As a result, the dose N of each line beam becomes constant as shown in FIG.
Therefore, the circuit 62 _i of the blanking generator 62 is constructed as shown in FIG.

In FIG. 9, 621 is a register for storing the beam on pulse length τ _i , 622 is a down counter, 6
23 is an AND circuit, 624 is a flip-flop, 625
Is a driver. That is, the bitmap generation circuit 6
When the data from 1 is "1" and a start pulse is input from the blanking control circuit 64, the flip-flop 624 is set via the AND circuit 623, and the output pulse rises and corresponds. The blanking electrode _39i is driven. At the same time, the starting pulse from the blanking control circuit 54 is the value τ _{i of the} register 621.
Is set in the down counter 622. The down counter 622 decreases the value while counting the high-speed clock signal CLK.

When the value of the down counter 622 becomes 0,
In other words, when the time corresponding to τ _i elapses, down counter 622 generates a carry signal, which acts as a reset signal for flip-flop 624. As a result, the flip-flop 624 is reset, and the driving of the blanking electrode 39 _i is turned off. Of course, when the data of the bitmap generation circuit 61 is “0”, the flip-flop 624 is not set, and no drive pulse τ _i is generated.

As described above, each circuit 62 _i generates a drive pulse for a time corresponding to a predetermined τ _i according to the data of the bit map generation circuit 61. Note that the register 62
The value τ _i of 1 can be set in advance or can be changed by the CPU 10 as needed. The above configuration, for example, 0.1 [mu] m □ beam 256 side by side present a current density of 200A / cm ^2, was irradiated with shot time of 25ns in the resist to each point of the sensitivity of 5μC / cm ^2, (4
0MH _Z) scanning area of 2mm width continuously moving at 50 mm / s, it was possible to obtain an exposure speed of 1 cm ² / s.

[0021]

As described above, according to the method of the present invention, the ultra-miniaturized L of, for example, about 0.2 μm rule is used.
Even in depiction in SI, stable, high-speed, and high-precision uniform exposure can be performed.

[Brief description of the drawings]

FIG. 1 shows a means for implementing the method of the invention.

FIG. 2 is a diagram showing an operation by means of FIG. 1;

FIG. 3 is a diagram showing an example of an apparatus for implementing the method of the present invention.

FIG. 4 is a diagram showing a quadrupole beam crushing lens.

FIG. 5 is a diagram (part 1) illustrating in detail a control unit that controls a main part of FIG. 3;

FIG. 6 is a diagram (part 2) illustrating in detail a control unit that controls the main part of FIG.

FIG. 7 is a diagram illustrating an example of a deflection control circuit of FIG. 5;

8 is a graph showing a beam intensity (A), a beam-on pulse length (B), and a dose (C) of each aperture (cell) of the blanking aperture array of FIG.

9 is a diagram showing an example of a blanking generation circuit 62 of FIG.

FIG. 10 is a diagram (part 1) showing a conventional variable rectangular electron beam exposure apparatus.

FIG. 11 is a view (part 2) showing a conventional variable rectangular electron beam exposure apparatus.

FIG. 12 is a diagram illustrating rectangular beam exposure.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 30 ... Electron gun 35 ... 4-pole beam crushing lens 39 ... Blanking aperture array 43 ... Blanking aperture 46 ... Horizontal scanning deflector 49 ... 8 pole deflector 62 ... Blanking generation circuit 68 ... Laser interferometer

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-53-117387 (JP, A) JP-A-62-118526 (JP, A) JP-A-61-42128 (JP, A) JP-A Sho 57- 7930 (JP, A) JP-A-1-248617 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) H01L 21/027

Claims (1)

(57) [Claims] A blanking electrode provided in each of a plurality of apertures arranged in a blanking aperture array to be irradiated with a charged particle beam is independently turned on and off to turn on the blanking electrode. Wherein the on-time of each blanking electrode is controlled such that the product of the intensity of the charged particle beam passing through the aperture and the on-time of the blanking electrode of the aperture is constant. Particle beam exposure method.