Abstract:
A electron beam performance measurement system includes an electron beam generator device for producing an electron beam, a test reticle having a series of openings forming a pattern, a reduction projection imaging device, a reference target having an essentially identical pattern as the test reticle, and a beam current detector. A patterned beam is generated by passing the electron beam through the pattern openings of the test reticle. The patterned beam is reduced and projected by the reduction projection imaging device and the reduced patterned beam is imaged onto a reference target. The reduced patterned beam is then exposed to the reference target, wherein some of the beam may pass through reference target pattern openings. Beam current detector records and measures the amount of beam current that is absorbed on, back-scattered from, or transmitted by the target reference, and determines from the measured beam current the accuracy of the projection system.
Abstract:
An electron beam system for direct writing applications combining the parallel throughput of a projection system and the stitching capability of a probe-forming system employs an electron gun to illuminate an initial aperture uniformly, a first set of controllable deflectors to scan the beam over the reticle parallel to the system axis, impressing the pattern of a subfield of the reticle in each exposure, in which a first variable axis lens focuses an image of the initial aperture on the reticle, a second variable axis lens collimates the patterned beam, a second set of controllable deflectors to bring the beam back to an appropriate position above the wafer, and a third variable axis lens to focus an image of the reticle subfield on the wafer, together with correction elements to apply aberration corrections that may vary with each subfield, thereby providing high throughput from the use of parallel processing of the order of 10.sup.7 pixels per subfield with the low aberration feature of the variable axis lens and the ability to tailor location-dependent corrections that are associated with gaussian systems that stitch the image pixel by pixel.
Abstract:
A dynamic correction arrangement for an electron beam projection/deflection system provides high order correction values for deflection in accordance with a correction equation. Particularly as applied to high accuracy telecentric deflection, the coefficients of terms of the correction equation may be determined by calibration for a small number of test points. Correction values may be stored in a look-up table or computed in real time by using a math co-processor in a processing pipeline. The correction provided corrects landing angle errors through the third order in telecentric projection/deflection systems such as systems utilizing variable axis immersion lenses.
Abstract:
The presence of magnetic fields from lenses and other (e.g. parasitic) sources at locations in a charged particle beam system such as the charged particle emitter, where the particles have little or no kinetic energy, creates disturbances of the charged particle trajectories, generating undesired angular momentum and resulting in excessive aberrations and associated adverse effects on system performance. Solutions are provided by suppression of these magnetic fields, by relocating the charged particle emitter away from the field source and/or counterbalancing the field with a bucking field. In addition, residual irrepressible field asymmetries are compensated by a suitable element such as a stigmator located at the correct location in the beam path, permitted by relocation of the charged particle emitter.
Abstract:
A charged particle beam performance measurement system includes a charged particle beam generator device for producing a charged particle beam, a test reticle having at least two areas of higher transparency than its surroundings that form a pattern, a reduction projection imaging device, a reference target having an essentially identical pattern as the test reticle, and a beam current detector. A patterned beam is generated by passing the charged particle beam through the pattern areas of the test reticle. The patterned beam is reduced and projected by the reduction projection imaging device and the reduced patterned beam is imaged onto a reference target. The reduced patterned beam is then exposed to the reference target, wherein some of the beam may pass through reference target pattern areas. Beam current detector records and measures the amount of beam current that is absorbed on, back-scattered from, or transmitted by the target reference, and determines from the measured beam current the accuracy of the projection system.
Abstract:
An electrostatic deflection plate for charged particle beam systems is formed of a planar semiconductive substrate having a conductive region at the substrate surface. The conductive region is diffused or implanted into the body of the substrate, or one or more conductive layers are deposited upon the substrate surface. The substrate material is preferably silicon and the diffused or implanted region is formed of a nonmagnetic, nonoxidizable metal such as gold or platinum. The deposited conductive region may be formed of a single layer of these or similar metals, one or more conductive underlayers with a nonmagnetic, nonoxidizable overlayer, a single or multilayer structure with a conductive oxide on the outermost layer, or a metallo-organic compound which forms a conductive layer during following heat treatment. The deflection plates are fabricated using conventional semiconductor processes and form durable structures which minimize eddy current effects.
Abstract:
An electron beam system for non contact testing of three dimensional networks of conductors embedded in dielectric material, specifically detection of open and short circuit conditions. Top to bottom and top to top surface wiring is tested electrically without making physical electrical contact. The system comprises two flood beams and a focus probe beam wih one flood beam located at either side of the specimen. Proper choice of acceleration potentials, beam currents and dwell times of the beams allow alteration of the secondary electron emission from the specimen in such a way that electrical properties of the conductor networks can be measured directly. The difference in secondary electron emission resulting from different surface potentials is detected as a strong signal which allows clear discrimination between uninterrupted and interrupted as well as shorted pairs of conductors. This testing system can be applied to the high speed testing of advanced VLSI packaging substrates as well as to the greensheets, sublaminates, and laminates from which they are fabricated.
Abstract:
Proximity (exposure dose) effects and/or local Coulomb (space charge defocussing) effects, both dependent on local pattern density of exposed areas, are simultaneously compensated in a charged particle beam projection device or tool by a projection reticle having an apertured weakly scattering membrane with selective strongly scattering regions between the apertures in the membrane. A reticle so constructed provides at least three independent exposure dosage levels that can be mixed to provide a wide range of exposure levels with high contrast. The more weakly the electrons are scattered (in the extreme, the electrons are not scattered at all through apertures), the greater the number that pass through a given beam contrast aperture, the higher the corresponding dose received at the target plane and the more space charge is contained in the beam bundle. Therefore, to compensate for the proximity effect (i.e. provide dose boost) and the local Coulomb effect (i.e. provide additional space charge) regions with reduced scattering characteristics are employed. Use of a hybrid doped resist allows exposure corresponding to an intermediate exposure dose in a preferred variant implementation exploiting the three dosage levels developed.
Abstract:
A method of optimizing locations of correction elements of a charged particle beam system determines respective corrector element currents to achieve optimum correction as a function of individual corrector location. Substantially complete dynamic correction of FSD and SFD can be obtained consistent with efficiency of operation and minimization of deflection distortion. In particular, FSD and SFD corrections can be sufficiently separated for substantially complete correction of SFD and FSD simultaneously with two stigmators. Both of these types of correction can be provided in complex charged particle beam systems employing curvilinear axis (CVA) particle trajectories and or large area reduction projection optics (LARPO) which cause complex hybrid aberrations in order to achieve high throughput consistent with extremely high resolution supporting one-tenth micron minimum feature size lithography regimes and smaller.
Abstract:
A method is disclosed for improving the electron beam apparatus lithography process wherein the calibration procedure for the apparatus is improved by using the product pattern and stepping sequence used to make the mask on a calibration plate and/or calibration grid and to determine improved apparatus correction errors which errors are used to control the apparatus for making an improved mask. The well-known EMULATION procedure is improved by calculating additional field correction errors based on a two step registration procedure to determine X/Y apparatus stepping errors. The LEARN procedure based on a static calibration grid procedure is improved by employing the duty cycle of the product pattern to calibrate the apparatus to determine deflection beam errors.