Radiant energy – With charged particle beam deflection or focussing – Magnetic lens
Reexamination Certificate
2000-05-18
2002-12-03
Nguyen, Kiet T. (Department: 2881)
Radiant energy
With charged particle beam deflection or focussing
Magnetic lens
C250S492220
Reexamination Certificate
active
06489620
ABSTRACT:
FIELD OF THE INVENTION
This invention pertains to microlithography (projection-exposure) of a pattern, defined by a reticle, onto a suitable substrate. Microlithography is a key technology used in the manufacture of semiconductor integrated circuits and displays. More specifically, the invention pertains to microlithography performed using a charged particle beam (e.g., electron beam) as an energy beam. Yet more specifically, the invention pertains to apparatus and methods for reducing astigmatism (e.g., deflection astigmatism, hybrid deflection astigmatic distortion, and the like) in charged-particle-beam (CPB) microlithography apparatus.
BACKGROUND OF THE INVENTION
As the sizes of circuit elements in integrated circuits have continued to be further miniaturized, the limitations of optical microlithography (i.e., microlithography performed using light such as ultraviolet light) have become apparent. This has led to much research directed to the development of practical microlithography apparatus and methods that employ an energy beam other than light. Considerable research effort has been directed to microlithography apparatus and methods that employ a charged particle beam such as an electron beam or ion beam.
Various approaches to charged-particle-beam (CPB) microlithography have been investigated. Three approaches include (1) spot-beam exposure, (2) variable-shaped beam exposure, and (3) block exposure. Each of these approaches can provide superior resolution to optical microlithography. However, each provides much lower throughput (number of wafers that can be processed per unit time) than optical microlithography. Specifically, approaches (1) and (2) have limited throughput because they perform exposure by tracing a pattern element-by-element using a beam having an extremely small spot diameter or a square profile. Block exposure (approach (3)) was developed to improve throughput over that of approaches (1) and (2) by utilizing a reticle on which the pattern elements have standard shapes that are exposed in batches. However, batch exposure currently does not provide a sufficiently high throughput because the number of pattern elements that can be defined on the reticle is limited, and because batch exposure typically must be performed in conjunction with the variable-shaped beam approach.
In order to improve throughput, so-called “divided-reticle” CPB microlithography has been proposed. In divided-reticle CPB microlithography, the pattern as defined on the reticle is divided or segmented into multiple exposure units usually termed “subfields” that are individually exposed by respective shots in an ordered manner. Such a reticle is termed a “divided” or “segmented” reticle. During exposure, the images of the individual subfields are positioned contiguously on the substrate (wafer) to form the image of the entire die.
Certain aspects of this divided exposure are shown in FIG.
4
. The wafer is typically exposed with multiple “dies” or “chips.” Each die is comprised of multiple rows of subfields that are arranged in “stripes.” Each die is exposed stripe-by-stripe and (within each stripe) subfield-by-subfield. The reticle defining the pattern to be exposed onto each die is similarly divided.
For projection exposure, the subfields of the reticle are sequentially irradiated (“illuminated”) in an ordered manner by a charged particle beam (e.g., electron beam or ion beam). Upstream of the reticle, the beam (termed an “illumination beam”) passes through an “illumination-optical system.” Downstream of the reticle, the beam (termed a “patterned beam,” formed by passage of the illumination beam through the irradiated subfield) passes through a “projection-optical system.” Each of the illumination-optical system and the projection-optical system comprises multiple lenses.
Certain aspects of exposure are shown in
FIG. 5
, which shows a stripe as defined on the reticle and a corresponding stripe as formed on the wafer. The reticle is mounted on a reticle stage (not shown), and the wafer stage is mounted on a wafer stage (not shown). During exposure of a stripe, the reticle stage and wafer stage are moved synchronously at respective constant velocities along the center line of the stripe. (The respective velocities of stage movement are established and controlled according to the demagnification ratio of the projection-optical system.) The subfields in the stripe on the reticle are illuminated sequentially, row by row, by the illumination beam. To illuminate each row, the illumination beam is deflected in a direction roughly perpendicular to the direction of travel of the reticle stage to sequentially illuminate the individual subfields in the row. After all the subfields in a row are exposed, exposure proceeds to the next row. Hence, exposure is performed in a raster manner. To maximize throughput, exposure of each row is performed by deflecting the beam in opposite directions, as shown in FIG.
5
. When exposure of a stripe is completed, the reticle stage and wafer stage are stopped and then shifted horizontally to the next stripe.
Throughput is remarkably improved with the divided-reticle technique, as shown in
FIGS. 4 and 5
, because each subfield is exposed in a single respective “shot,” and all the elements of a pattern to be transferred are defined on the reticle.
On a divided reticle, each subfield is separate from adjacent subfields. Extending between the subfields are “struts” that provide substantial mechanical rigidity and strength to the reticle. The struts also allow individual subfields to be accurately selected for illumination by the illumination beam.
More complex or larger patterns require more subfields and hence more stripes. Increasing the number of stripes requires a corresponding increase in the number of times that the reticle stage and the wafer stage must be moved back and forth to expose each die. This results in a corresponding increase in the number of times that the stages must be accelerated and decelerated. During accelerations and decelerations of the stages, exposure cannot be performed and the time is wasted. To avoid losses in throughput, the width of each stripe is increased (i.e., each row of subfields has more subfields) to reduce the number of stripes. Such increases in stripe width require corresponding increases in the width of the deflection field of the illumination beam.
The lenses in the illumination-and projection-optical systems, similar to their counterparts in conventional optical systems, can exhibit any of various aberrations corresponding to respective aberrations exhibited by optical lenses. With an increase in stripe width, the distance over which the beam must be deflected laterally is correspondingly increased. i.e., the beam must pass through more off-axis portions of the lenses in the illumination- and projection-optical systems, which results in greater deflection aberrations.
A “hybrid deflection-astigmatic distortion” refers to an astigmatic distortion (aberration) having a magnitude proportional both to lateral deflection distance and the beam size (the lateral dimensions of the subfield). This and related aberrations result in a blurred (defocused) and distorted image as exposed onto the wafer. To correct aberrations, one or more deflectors are conventionally provided in the illumination- and/or projection-optical systems. Image defocusing and/or distortion can be ameliorated somewhat by adjusting the excitation current applied to such a deflector so as to alter the beam trajectory in a manner that reduces aberrations. However, these remedies alone are no longer sufficient to achieve the desired resolution in CPB microlithography.
A conventional approach to providing better correction of deflection-astigmatism aberrations and hybrid deflection-astigmatism distortions involves the use of an astigmatism-correction device. In such a device, a magnetic field is produced that is proportional to cos[2&thgr;]in a magnetic-field distribution in a cylindrical coordinate system (z,r,&thgr;), wherein the optical axis is
Klarquist & Sparkman, LLP
Nguyen Kiet T.
Nikon Corporation
LandOfFree
Astigmatism-correction device and charged-particle-beam... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Astigmatism-correction device and charged-particle-beam..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Astigmatism-correction device and charged-particle-beam... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2980256