JP2006261291A - Electron beam lithography device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an inexpensive electron beam lithography device capable of correcting nonlinearity in changing a beam size of a variable formation type electron beam lithography device for rapid and precise lithography. <P>SOLUTION: The electron beam lithography device is equipped with a sensing means for detecting the reflection or transmission current of a variable formed electron beam for calibrating the electron beam by the detection of the sensing means. Data of a plurality of beam sizes are specified in a formation deflection DAC circuit to detect nonlinear components of the detected value by the sensing means and correct a beam irradiation time by a blanking means so that the detected value is linear. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a highly accurate beam control and drawing method of a variable shaping type electron beam drawing apparatus in the field of semiconductor fine processing.

  An electron beam drawing apparatus is an apparatus that draws a fine pattern on a mask substrate, a wafer, or the like with an electron beam with high accuracy. In recent years, in addition to a point beam that finely narrows the electron source with an electron lens, a variable shaping method that controls the beam shape at high speed according to the drawing pattern has been put into practical use.

  In other words, the variable shaping method is a method in which the upper shaping aperture image is moved on the lower shaping aperture with a shaping deflector to change the beam shape and dimensions, and in some cases, these are controlled in multiple stages to highly control the beam shape. It is.

  In recent years, miniaturization has progressed, and it has become necessary to expose a drawing pattern of several tens of nm with high accuracy. Therefore, in particular, 1 nm or less is required as the dimension setting resolution of the electron beam drawing apparatus. In a point beam, it is common to finely control the beam by the aperture exposure amount, but the exposure time increases.

  In the variable shaping method, the beam size is controlled by a DAC (Digital to Analog Converter). Therefore, it is sufficient to increase the DAC resolution, but the circuit settling time for a predetermined output value is increased according to the resolution, and the throughput is deteriorated.

  Japanese Laid-Open Patent Publication No. 10-27749 (Patent Document 1) discloses a method for measuring a beam current by changing dimensions and drawing a fine pattern with good reproducibility from a relation curve of a minute region. Japanese Examined Patent Publication No. 7-107893 (Patent Document 2) also discloses a means for correcting the beam current value by approximating it with a linear function of dimensions.

  However, these techniques still have the problem that the resolution of the DAC limits the set dimensional accuracy. The high bit DAC has problems such as a decrease in measurement reliability and a complicated circuit system resulting in an increase in cost. Particularly in the processing of semiconductor gate layers and magnetic heads, higher dimensional controllability is generally required than position control, which is a major issue.

JP-A-10-27749 Japanese Patent Publication No. 7-107893

  In view of the above problems, the present invention provides an inexpensive electron beam drawing apparatus that corrects non-linearity at the time of beam dimension change of a variable shaping type electron beam drawing apparatus and enables high-speed and high-precision drawing. With the goal.

  The present invention provides an electron beam drawing apparatus that includes a detection unit that detects a reflection or transmission current of a variable shaped electron beam, and calibrates the variable shaped electron beam by detection of the detection unit. Dimension data is set to detect a non-linear component of the detection value of the detection means, and the beam irradiation time is corrected by the blanking means so that the detection value becomes linear.

  Thereby, it is possible to provide an electron beam drawing apparatus that realizes high resolution of dimension setting and enables highly accurate dimension correction.

  In general, there are two methods for increasing the resolution of beam control: a method of increasing the resolution of the shaping deflection DAC circuit itself and a method of adding a correction DAC circuit to the main shaping deflection DAC circuit. In the present invention, the problematic DAC resolution or non-linearity error is measured as a beam current, and is fed back to blanking control for beam irradiation time correction.

  That is, continuous data is set in the dimension setting, that is, the shaping deflection DAC circuit, and the linearity is evaluated by measuring the current of the generated variable shaping electron beam or the amount of reflected current. Since the non-linearity of the shaping deflection DAC circuit is directly measured as an error of the shaping beam current value, high reliability can be measured, and the output measuring system of the shaping deflection DAC circuit is not necessary, and the cost can be reduced.

  The non-linear change of the current obtained here is fed back to the blanking circuit and corrected by beam irradiation. Accordingly, high-precision dimensional control can be performed at high speed using a low-deflection shaped deflection DAC circuit. In particular, when the variable shaped beam width is smaller than the beam blur width, the irradiation time control by the blanking means can provide higher resolution and reproducibility than the DAC setting.

  An object of the present invention is to provide an inexpensive electron beam drawing apparatus that corrects nonlinearity at the time of beam dimension change and enables high-speed and high-precision drawing.

  FIG. 1 shows a variable shaping type electron optical system using a general two-stage shaping diaphragm. The electron beam emitted from the electron source 1 passes through the first molded diaphragm 2 and is imaged on the second molded diaphragm 6 by the first molded lens 3 and the second molded lens 5. The shaping deflector 4 controls the position of the first shaped diaphragm image 7, and the electrons transmitted through the second shaped diaphragm 6 become the variable shaped beam 11. The shaped beam 11 is reduced by the reduction lens 9 and imaged and positioned on the sample surface by the objective lens 12 and the objective deflector 13.

  Here, the moving stage 16 includes a beam ammeter 15 in addition to the drawing sample, and the backscattered electron detector 14 detects a calibration mark on the moving stage. The blanking electrode 8 of the blanking means separates the beam during beam deflection settling and is turned off by the blanking diaphragm 10 to control the exposure time.

  The control computer 22 divides the drawing data into rectangular shot size data and exposure coordinates. The control computer 22 calculates a correction amount from the beam size error and sets it in the shaping deflection DAC circuit 17. The beam irradiation position is set in the objective deflection control circuit 20 in synchronization with the series of beam dimension settings.

  After these circuit operations, the blanking control circuit 19 of the blanking means turns off the blanking electrode 8 and turns on the beam for a desired exposure time. A Faraday cup ammeter 15 and a current detection circuit 21 for verifying beam size errors including errors of the actual shaping deflection DAC circuit 17 are provided. The ammeter 15 can be replaced by a backscattered electron detector 14 that measures the backscattered electron intensity from a uniform reflecting surface.

  As shown below in the configuration of FIG. 1, the shaping deflection DAC circuit 17 is actually driven, and the current change rate is measured and analyzed by the ammeter 15, so that the beam size can be set with high accuracy. The beam dimension is a two-dimensional quantity, but the following will be discussed as a one-dimensional quantity for simplicity.

FIG. 2 shows an example of a beam current measurement result when there is an error in the shaping deflection DAC circuit. That is, FIG. 2 is a current value I _n, measured in the case of increasing the set data W _n with a resolution 1-bit equivalent of the DAC.

  FIG. 2A shows a case where the resolution of the DAC is rough and changes in steps. FIG. 2B shows a case where there is an error of 1/2 of the output LSB (Least Significat Bit) corresponding to 1-bit resolution.

Problem error a is but ^10-4 levels would normally high resolution type of current measurement, that is, beam size to obtain a resolution of 0.1nm at 1um range. Also, if the shaping deflector rotates, a measurement error occurs, but it is a uniform linear component. However, by superposing the moving diaphragm side on the stationary diaphragm as shown in FIG. The influence can be further eliminated.

A specific correction flow example is shown in FIG. Sequentially sets the data W _n along the variable range of the shaping deflection DAC, to measure the beam current I _n. Since the measurement is DAC linearity, an offset can be added to the large current dimension side in order to maintain the accuracy of current measurement.

From the obtained results, the beam current I and the dimension W are
I = AW + B (1)
The coefficients A and B are calculated by the method of least squares by substituting I _n and W _n . Here, A is a linearity coefficient, and B is an offset coefficient.

Consider a case where the dimension increases by a small value ΔWn with respect to the LSB value R of the shaping deflection DAC circuit. If the exposure time T is finished to the dimension W, and the dimension value and the exposure time are proportional to each other in a minute range, in the example of FIG.
R = W _{n + 1} −W _n
As an interpolation calculation that simply performs interpolation,
T _n = T (1 + ΔW _n / R) (2)
It becomes.

In the case of FIG. 2 (b), correction is further performed on individual W _n .
T _n = T (1 + ΔI _n / I) (1 + ΔW _n / R) (3)
Therefore, a DAC nonlinearity correction term may be added. However [Delta] I _n the amount of difference from the I calculated in (1),
ΔI _n = I _n −I
It is.

  For specific correction, an exposure time ratio correction table 18 that is a beam irradiation time ratio table is set in advance and referred to in accordance with the dimensions at the time of exposure. Usually, a high dimension setting resolution is a problem with a minute dimension pattern. As can be seen from the equations (2) and (3), the effect of the correction ratio increases as the current decreases.

  Therefore, it is possible to save a memory amount of the exposure time ratio correction table 18 by setting a certain threshold value and lowering the threshold value.

  Here, by adding an offset to the beam size to be measured and providing a function for varying the measurement current range, high-precision measurement is possible by adjusting the transmitted current or the amount of reflected electrons to the optimum range of the measuring instrument. Also, if the upper shaping diaphragm that moves when the beam dimension changes is made larger than the lower shaping diaphragm and an offset is added so that it overlaps in the direction perpendicular to the dimension change direction, the beam is cut in the vertical direction during current measurement. The effect can be prevented.

  According to the present invention, the dimension setting resolution of actual pattern drawing is effectively improved with good reproducibility, and fine pattern drawing is realized. Current measurement does not require beam scanning or differential processing time, and can measure multiple points for high speed and high accuracy. A high-resolution shaping deflection DAC circuit and a complicated correction DAC circuit that are expensive in cost and a complicated correction DAC circuit are not required and can be realized at low cost.

The figure which concerns on the Example of this invention and shows the outline | summary of an electron beam drawing apparatus. The figure which concerns on the Example of this invention and shows the measurement range of beam current. The figure which concerns on the Example of this invention and shows the example of a shaping | molding drawing arrangement | positioning for avoiding a rotation error. The measurement flowchart figure concerning the Example of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Electron source, 2 ... 1st shaping | molding aperture_diaphragm | restriction, 3 ... 1st shaping | molding lens, 4 ... Molding deflector, 5 ... 2nd shaping | molding lens, 6 ... 2nd shaping | molding aperture, 7 ... 1st shaping aperture image, 8 ... Ranking electrode, 9 ... reduction lens, 10 ... blanking stop, 11 ... shaped beam, 12 ... objective lens, 13 ... objective deflector, 14 ... backscattered electron detector, 15 ... ammeter, 16 ... moving stage, 17 ... shaping Deflection DAC circuit, 18 ... exposure time ratio correction table, 19 ... Blanking control circuit, 20 ... Objective deflection circuit, 21 ... Current measurement circuit, 22 ... Control computer.

Claims (5)

  The upper and lower shaping apertures for shaping the variable shaped electron beam that passes through the electron beam to change the beam shape or beam size, and the direction of the electron beam passing through the upper shaping aperture and going to the lower shaping aperture. A deflector that deflects, a shaping deflection DAC circuit that generates an analog output to be set in the shaping deflector in accordance with the drawing data used for drawing the variable shaped electron beam, and a block that controls the irradiation time of the variable shaped electron beam. Ranking means, objective deflection means for irradiating and positioning the variable shaped electron beam on the surface of the sample, and placing the sample so that the sample is positioned at the irradiation position of the variable shaped electron beam And a detecting means for detecting a reflected or transmitted current of the variable shaped electron beam, and the variable shaping electric current is detected by the detecting means. In an electron beam lithography apparatus that calibrates a beam, a plurality of beam size data are set in the shaping deflection DAC circuit to detect a non-linear component of a detection value of the detection means, and the detection value becomes linear. An electron beam drawing apparatus, wherein the beam irradiation time is corrected by the blanking means.   2. The electron beam drawing apparatus according to claim 1, further comprising: a beam irradiation time ratio table linked to a blanking control circuit of the blanking means for each shaping DAC data of the shaping deflection DAC circuit to correct non-linearity. An electron beam lithography apparatus characterized by   3. The electron beam drawing apparatus according to claim 1, wherein a non-linearity error between each shaping DAC data is corrected by interpolation.   2. The electron beam drawing apparatus according to claim 1, further comprising: a function of changing a measurement current range by adding an offset to the beam size measured by the detecting means.   2. The electron beam drawing apparatus according to claim 1, wherein the upper forming diaphragm that moves at the time of changing the beam size is larger than the lower forming diaphragm and is positioned so as to overlap the direction of dimension change. An electron beam lithography apparatus characterized by adding an offset and setting a measurement current range.

Images