Radiant energy – With charged particle beam deflection or focussing – Magnetic lens
Reexamination Certificate
1999-09-21
2003-01-14
Anderson, Bruce (Department: 2881)
Radiant energy
With charged particle beam deflection or focussing
Magnetic lens
C250S398000, C250S492230
Reexamination Certificate
active
06507027
ABSTRACT:
FIELD OF THE INVENTION
This invention pertains to microlithography apparatus and methods for use, e.g., in the manufacture of semiconductor integrated circuits, displays, and the like. The subject apparatus and methods utilize a charged particle beam (e.g., electron beam or ion beam) as an energy beam for transferring a pattern, defined by a reticle or mask, onto a sensitive substrate (e.g., semiconductor wafer or the like). More specifically, the invention is directed to such apparatus and methods exhibiting reduced deflection aberrations even while effecting large beam deflections.
BACKGROUND OF THE INVENTION
Conventional exposure formats in charged-particle-beam (CPB) microlithography apparatus can be categorized into the following three types:
(1) Spot-beam exposure
(2) Variable shaped beam exposure
(3) Block exposure
Although these exposure formats exhibit superior resolution compared with conventional batch-transfer systems employing visible light as an energy beam, they exhibit disappointingly low “throughput” (number of substrates or wafers that can be processed per unit time). Throughput is especially low with exposure formats (1) and (2) because the patterns are exposed by being traced with a beam having an extremely small spot radius or having an extremely small square transverse profile.
The block exposure format (3) was developed to improve throughput. In block exposure, uniformly shaped features are defined on a reticle, and portions of the reticle containing such features are exposed one shot at a time in batches. Because the number of features that can be placed on a reticle is limited in this format, a variable-profile exposure system must be used. Consequently, throughput is not improved as much as would otherwise be expected.
In order to further improve throughput, so-called “divided” projection-transfer apparatus have been developed. In such apparatus, the reticle defining a pattern is divided into multiple portions or “exposure units” that are individually projection-exposed onto the substrate. Each exposure unit requires a respective “shot” (exposure) using the energy beam.
Certain aspects of a conventional divided projection-transfer apparatus are depicted in
FIGS. 13 and 14
. Referring first to
FIG. 13
, an entire substrate (wafer) W is shown containing multiple “chips” C. Each chip C contains multiple “stripes” S, and each stripe S contains multiple “subfields” SF (as representative exposure units). The reticle (not shown) defining the pattern for each chip C is similarly divided into multiple stripes and subfields.
Each subfield SF is individually exposed. Exposure is normally performed in a manner as shown in
FIG. 14
, in which the reticle is situated upstream of the substrate. As a charged particle beam (e.g., electron beam) illuminates each subfield on the reticle, the respective portion of the pattern is projected by a projection-optical system (not shown but understood to be located between the reticle and the substrate) onto the substrate, thereby imprinting the pattern portion onto a respective region of the substrate. The subfields in each stripe are arranged in columns. The columns are sequentially exposed, and the subfields in each column are sequentially exposed. To effect serial exposure of the columns, the reticle and substrate (which are mounted on respective stages that are not shown) undergo relative linear motions in respective scan directions at respective constant scan velocities. The respective scan velocities of the reticle stage and wafer stage are established by, inter alia, the demagnification ratio of the projection-optical system.
Before reaching the reticle, the charged particle beam (generated by a suitable source) passes as an “illumination beam” through an “illumination-optical system” located upstream of the reticle. To expose the subfields in each column of a stripe on the reticle, the illumination beam is deflected (by appropriately situated deflectors in the illumination-optical system) in a direction roughly perpendicular to the direction of linear motion of the reticle stage. Thus, the subfields in each stripe are sequentially exposed by the illumination beam in a raster manner. After passing through the reticle, the beam (now termed the “patterned beam”) passes through a “projection-optical system” to the substrate. The illumination-optical system and projection-optical system are collectively termed the “CPB optical system.” As each column of subfields is exposed, the reticle and substrate are moved in opposite directions to position the next column of subfields in the stripe for exposure. To improve throughput, each subsequent column of subfields is exposed by deflecting the charged particle beam in a direction opposite the direction in which the beam was deflected in the previous column, as shown in FIG.
14
.
The subfields on the reticle are separated from one another by struts. The struts strengthen and add rigidity to the reticle, and also facilitate the illumination of only one subfield per shot.
To improve throughput, the illumination beam usually has a relatively high beam current. However, high beam currents tend to introduce significant image blur due to Coulomb effects. A conventional approach for reducing such Coulomb effects is to enlarge the area being exposed per shot and to subject the illumination beam to a relatively high acceleration voltage.
Throughput can also be increased by increasing the maximum beam deflection (i.e., maximum angle with which the beam is laterally deflected) to expose the subfields in each column. By increasing the beam deflection, the length of each column (and thus the width of each stripe) can be increased. I.e., by increasing the width of the stripes, fewer stage movements are required during exposure of the entire reticle pattern. Hence, the cumulative time required to perform stage movements (to expose the subsequent column of subfields or to begin exposing the next stripe of the pattern) during exposure of the reticle pattern is decreased, and throughput is correspondingly increased. Unfortunately, however, a beam experiencing a higher-magnitude deflection must pass through a subfield (located at an end of a column) that is widely separated from the optical axis of the CPB-optical system. Such a high-magnitude deflection generates more deflection aberrations than a lesser-magnitude deflection. The conventional manner of reducing such aberrations is to adjust the excitation current supplied to the deflectors used to deflect the beam and to manipulate the deflection trajectory of the beam so as to minimize deflection aberrations.
Because the magnitude of beam deflection is proportional to the excitation current applied to the deflector, a large excitation current must be impressed on the deflector in order to impart a large-magnitude deflection on the beam. Also, because it is desirable to deflect and scan the beam at high speeds, a driver circuit supplying electrical power to the deflector should be capable of changing the electrical power very rapidly with each subfield. Unfortunately, an electrical circuit capable of performing sufficiently high-speed changes of a high output power is technically difficult and expensive to design. Consequently, there is an urgent need to provide deflectors that can produce as large a deflection as possible using a relatively small electrical current.
The accuracy with which deflectors are manufactured is also crucial for controlling aberrations. Deflectors usually comprise wound coils of an electrical conductor (wire). The conductor itself has a limited thickness, and the accuracy and precision with which most electrical conductors are fabricated are usually not high. As a result, extraneous magnetic fields outside the main deflection field are usually simultaneously generated by the deflector. The distribution of the magnetic field can be expressed using a cylindrical coordinate system (z,r,&phgr;), in which &phgr; is the rotational angle around the optical axis, r is the radial coordinate and z is the axial coordinate. The ref
Kamijo Koichi
Kojima Shin-ichi
Anderson Bruce
Klarquist & Sparkman, LLP
Nikon Corporation
LandOfFree
Apparatus and methods for 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 Apparatus and methods for charged-particle-beam..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and methods for charged-particle-beam... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3033799