Radiant energy – Irradiation of objects or material – Irradiation of semiconductor devices
Reexamination Certificate
1999-07-15
2001-06-05
Nguyen, Kiet T. (Department: 2881)
Radiant energy
Irradiation of objects or material
Irradiation of semiconductor devices
Reexamination Certificate
active
06242751
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention generally relates to charged-particle-beam exposure devices, and particularly relates to a charged-particle-beam exposure device which forms a pattern on a wafer by exposing the wafer to charged particles.
2. Description of the Related Art
As the circuit density of semiconductor integrated circuits increases, a finer processing technique is required. Compared to the light exposure method widely used in the manufacturing of LSI chips, the charged-particle exposure method has much superior characteristics in terms of the resolution and the focus depth. With respect to the resolution, a processing limit of the photolithography method is about 0.3 &mgr;m, while processing as fine as 0.1 &mgr;m can be achieved in the charged-particle-beam exposure method.
However, the charged-particle-beam exposure method is inferior compared to the light exposure method in terms of an exposure positioning accuracy, an overlay accuracy, and a field stitching accuracy. Because of this, the charged-particle-beam exposure method is not widely used in the field for manufacturing purposes of LSI chips.
The charged-particle-beam exposure device has a smaller area to be able to be exposed at one time, compared to a light exposure device such as a stepper. (This area is called a deflection field hereinafter.) Thus, in order to expose one LSI chip, stage movement is required to successively shift the deflection field on the LSI chip. In doing so, if the connecting precision across borders of the deflection fields is low, severance of wires and/or short-circuits are generated which greatly degrades the yield of the chips.
In order to improve the yield, the connecting precision at the field borders must be enhanced, which requires a higher precision of the deflection of the charged-particle beam. In the charged-particle-beam exposure device, the charged-particle beam is generally deflected by a magnetic field generated by coils. The coils include two systems for x-direction deflection and for y-direction deflection. Separate currents are applied to these two systems to deflect the beam in the x direction and the y direction independently. Unfortunately, the amount of beam deflection is not in proportion to the amount of current applied to the deflection coils, but is represented as a complex function of the current amount.
In order to deflect the beam with a high precision, the amount of the current applied to the deflector must be corrected. There are two types of corrections. One is a distortion correction for establishing a linear relation between the input and the deflection amount, and the other is a deflection-efficiency correction for correcting coefficients for linear factors. The distortion correction is a time consuming process since it requires data collection at various points within the field. However, the data needs to be collected only one time since a time variation of the distortion is small. On the other hand, correction coefficients can be obtained in a short period of time for the deflection-efficiency correction. However, the deflection-efficiency-correction coefficients must be frequently obtained because the deflection efficiency varies over time due to a change in thermal distribution of the deflector, etc.
In order to calibrate the deflection field, coordinates of the deflector are generally matched with coordinates of the stage, whose measure and orthogonality are guaranteed through the laser-interferometer system. In order to measure the coordinates of the deflector, an actual position of the charged-particle beam must be obtained by directing the beam to mark positions on a wafer and detecting reflected charged particles.
FIG. 1A
is an illustrative drawing for explaining a method of detecting mark positions through the charged-particle-beam scan. As shown in
FIG. 1A
, the charged-particle beam is scanned by the deflector over a mark
306
formed as a groove in a reference chip
305
. Reflection detectors
300
and
301
, symmetrically arranged with respect to the axis of the beam optical system, detect reflected charged particles. Outputs of the detectors are added by the adder
302
. A signal after the addition is successively obtained in synchronism with the scan of the deflector, providing a reflection signal form to be analyzed. When such a process is conducted by using the position-detection mark
306
as shown in
FIG. 1A
, a reflection signal form as shown in
FIG. 1B
is obtained. The reflection signal form obtained in this manner is analyzed by an analyzing device
303
to detect a center position of the mark. A result of the analysis is sent from the analyzing device
303
to a control-purpose computer
304
, which uses the result in processes such as a correction of the beam. In general, a groove (dent) formed in a wafer (silicon) is used as a mark.
The detection of the position mark described above is conducted at various points by shifting the mark on a wafer through stage movement. In this manner, the deflection-efficiency-correction coefficients for correcting the linear factors and a distortion map of the deflector for the distortion correction are obtained.
In the mark-position-detection method described above, the detected mark positions contain errors. This is because a relative position of the mark with respect to the reflection detectors changes when the mark is detected at various points.
When the mark is detected at various points, an angle at which charged particles are reflected by the mark toward a reflection detector varies depending on a relative position of the mark with respect to the reflection detector. When reflected charged particles are detected in a configuration as shown in
FIG. 2A
, signal forms as shown in
FIG. 2B
are obtained. As shown in figures, a reflection signal having a symmetric form without a distortion can be obtained when the mark is positioned at an equal distance from the two reflection detectors. When the mark is positioned at other locations, however, a reflection signal form having an asymmetry is obtained. This is because the angle of the reflection is different for the different reflection detectors.
In addition to the problems of errors regarding the mark-position detection, there is a problem concerning the focusing of the charged-particle beam in the charged-particle-beam exposure device.
FIG. 3
is an illustrative drawing showing a configuration for the focusing of the beam in the related-art charged-particle-beam exposure device. As shown in
FIG. 3
, an optical system
310
, using a type of light not affecting a resist, is provided between a wafer and a charged-particle lens. The optical system
310
includes a light source
311
and a light detector
312
. When the wafer is exposed to the charged-particle beam, the light source
311
illuminates light on the wafer, and the light detector
312
detects light reflected from the wafer to measure the height of an exposed surface. Based on the height of the exposed surface, a focusing distance of the charged-particle lens is changed.
Such a related-art charged-particle-beam exposure device has such problems as:
a) when the focusing distance of the reflection path is changed, the deflection path of the charged-particle beam is affected to cause a displacement of the beam position on the wafer surface; and
b) since structures under the exposed surface have complex patterns in a LSI device, light reflected from these patterns has an adverse effect of causing errors in the detection of the height.
The problem a) will be described below. In the charged-particle-beam exposure device, deflection coordinates X=(X, Y), having an origin at the axis of the beam optical system, are entered into a correction circuit to obtain corrected deflection coordinates X′=(X′, Y′).
X′=Gx*X+Rx*Y+Dx(X, Y) (1)
Y′=Ry*X+Gy*Y+Dy(X, Y) (2)
Here, G=(Gx, Gy) are correction coefficients concerning the gain, R=(Rx,
Abe Tomohiko
Kawakami Ken-ichi
Nasuno Hideki
Ohkawa Tatsuro
Ooaeh Yoshihisa
Fujitsu Limited
Nguyen Kiet T.
Staas & Halsey , LLP
LandOfFree
Charged-particle-beam exposure device and... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Charged-particle-beam exposure device and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Charged-particle-beam exposure device and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2513200