Proximity printing with soft x rays is a leading contender for very large scale integrated circuit lithography below 0.25 μm. It will require overlay accuracy far below the 100 nm (3σ) typical of today’s systems. I...
Proximity printing with soft x rays is a leading contender for very large scale integrated circuit lithography below 0.25 μm. It will require overlay accuracy far below the 100 nm (3σ) typical of today’s systems. In addition, it should use wafer marks which permit planar resist flow, so their apparent position is not distorted. We report an alignment system which uses a linear zone plate on the mask to focus laser light into a line on the wafer. The wafer mark is the boundary between two adjacent fine pitch gratings. These gratings diffract the light into photodiodes. The gratings either differ slightly in pitch or by 180° in phase. This configuration minimizes the disturbance to the resist as it flows over the alignment mark. In the case of gratings with slightly different pitch the light from each grating is detected by one‐half of a split photodiode. This arrangement has good sensitivity and a very wide capture range. The phase shift gratings have reduced capture range but enhanced sensitivity because of the improved resolution from the phase shift. We observe that the full width at half‐maximum (FWHM) of the alignment signal is 0.6 μm when the FWHM of the zone plate focus is 1.0 μm. By collecting all of the light focused by the zone plate and diffracted by the gratings we obtain strong alignment signals with signal to noise ratios of more than 1000:1. We routinely obtain a reproducibility in a bench setup equivalent to less than 1 nm (3σ). This alignment technique is currently undergoing installation in a commercially available x‐ray exposure tool.
We report on the fabrication and measurement of a novel type of quantum dot device in which the shape and size of the dot can be controlled. The device consists of four uniformly spaced quantum point contacts which ca...
We report on the fabrication and measurement of a novel type of quantum dot device in which the shape and size of the dot can be controlled. The device consists of four uniformly spaced quantum point contacts which can be biased to produce quantum dots ranging in size from 600 nm to below 200 nm. The capacitances of the gates controlling the dot are small. Measurements were made at a temperature of 0.3 K for three different dot sizes. Devices were fabricated in three lithography steps, with x‐ray nanolithography used to define the final gate layer. The device exhibits conductance maxima whose spacing depends on which quantum point contacts are used to define the quantum dot.
This paper applies the Bayesian conditional decision feedback estimator (BCDFE) to rapidly fading frequency selective channels. The BCDFE is a model-based deconvolution algorithm which jointly estimates the transmitte...
详细信息
This paper applies the Bayesian conditional decision feedback estimator (BCDFE) to rapidly fading frequency selective channels. The BCDFE is a model-based deconvolution algorithm which jointly estimates the transmitted data and channel parameters. The BCDFE applies joint symbol-by-symbol MAP demodulation and channel estimation to equalize channels, and consequently provides robust performance in rapidly fading channels. We provide a brief derivation of the BCDFE and characterize the performance on the land mobile radio channel. We assess the BCDFE's principle design characteristics and the resulting performance in both transient and steady-state operation. The effects of delay spread, Doppler spread, and co-channel interference on the BEP performance are also presented. The BCDFE demonstrates many of the desirable characteristics of an equalizer for mobile radio.< >
We explore several issues relevant to the benchmarking and comparison of machine learning algorithms. We illustrate those issues with a case study using the decision tree induction algorithms C4.5 (J. Quinlan, 1993) a...
详细信息
We explore several issues relevant to the benchmarking and comparison of machine learning algorithms. We illustrate those issues with a case study using the decision tree induction algorithms C4.5 (J. Quinlan, 1993) and multiscale classification (MSC) (A.P. Bradley and B.C. Lovell, 1994), multilayer perceptrons (MLP) and multivariable regression (MVR). Then for a "real world" problem we compare estimates of the true error rates for each classifier, first on a single train and test partition, and then using cross validated subsampling techniques. The relevance of the /spl chi//sup 2/ test is then discussed in relation to comparing the classifier accuracies. The paper concludes by evaluating the performance of these four fundamentally different approaches to the solution of this regression problem.< >
Image formation in x‐ray lithography has been studied extensively. A previous theoretical study predicted that 0.1 μm features can be printed at large gaps (≳10 μm) with absorbers attenuating less than 10 dB. This ...
Image formation in x‐ray lithography has been studied extensively. A previous theoretical study predicted that 0.1 μm features can be printed at large gaps (≳10 μm) with absorbers attenuating less than 10 dB. This study seeks to verify rigorous electromagnetic simulations of image formation by directly measuring the aerial image. Exposures of features with linewidths ranging from 0.15 to 0.075 μm were performed on the Helios synchrotron. Pedestal‐style x‐ray masks, consisting of SiNx membranes and a Au absorber, were patterned with e‐beam lithography at 100 and 50 kV. By careful dose control and inspection of the resulting features, one can directly determine the aerial image (the image at the resist surface). This is verified using a string model of the resist development. Aerial image measurements correlate reasonably well with modeling results.
作者:
IVANCIC, WDCHU, PSHYY, DJNational Aeronautics and Space Administration
Lewis Research Center Cleveland Ohio 44135 USA. A design engineer in the Digital Technology Branch of the Space Electronics Division. He has extensive experience in satellite communications systems
modulation and coding on-board switching and routeing and high speed digital design. Since joining NASA in 1982 Mr. Ivancic has been responsible for the development of a variety of matrix switch controllers ground terminal equipment numerous pieces of special test equipment a beacon controller for the SARSAT program (Search and Rescue Satellite) and a time-shared decoder for a processing satellite. In addition Mr. Ivancic has been technical contract monitor on a number of advanced communications technology contracts including the Advanced Modulation Technology Development a 5 GHz fibre-optic link and multi-programmable modem study. He was awarded the B.S.E.E. and M.S.E.E. degrees by Cleveland State University in 1982 and 1986 respectively. Cleveland State University
Cleveland Ohio 44115 USA. Currently an assistant professor of Electrical Engineering Department at Cleveland State University. He obtained the M.S. and Ph.D. degrees from Iowa State University
majoring in Electrical and Computer Engineering. His research interests include high-speed computer networks digital systems and neural networks applications. He is a member of Sigma Xi Phi Kappa Phi as well as the IEEE and ACM. ComSearch
Reston Virginia 22091 USA. Received the B.S. degree in electrical engineering from National Chiao-Tung University
Hsin-Chu Taiwan in 1983 and the M.S. and Ph.D. degrees in electrical engineering from Georgia Institute of Technology Atlanta GA in 1986 and 1990 respectively. From June 1987 to October 1987 he worked for the Department of Neurology Emory University Atlanta as a Programmer. From September 1989 to December 1989 he worked for the Advanced Development Laboratory. AMP Inc. Atlanta as a Test Engineer. From February 1990 to August 1993 he wor
Many concepts for advanced communication satellite networks have recently been proposed. Critical technical issues relating to satellite packet switching for meshed very small aperture terminal networks and broadband ...
详细信息
Many concepts for advanced communication satellite networks have recently been proposed. Critical technical issues relating to satellite packet switching for meshed very small aperture terminal networks and broadband networks are addressed. Hardware considerations, networking and testing issues are discussed.
An updated survey of plasma immersion ion implantation (PIII) applications in semiconductor processing is presented. PIII is a technique in which a negative bias extracts ions directly from a plasma to be implanted in...
An updated survey of plasma immersion ion implantation (PIII) applications in semiconductor processing is presented. PIII is a technique in which a negative bias extracts ions directly from a plasma to be implanted into a substrate immersed in the plasma. This technique offers the advantage of high dose-rate implantation at low energies and large area processing capability over conventional implanters with more compex design. PIII has been applied to semiconductor processing in the formation of shallow junctions, poly doping, impurity gettering, Pd seeding for Cu interconnects and conformal doping of trenches. While PIII has inherent advantages, such as built-in wafer charge neutralization, issues such as contamination, wafer heating, secondary electron generation, and substrate interactions with the plasma remain concerns.
A combined high frequency and quasi-static C-V technique is used to monitor wafer charging of MOS structures during plasma immersion ion implantation (PIII). The test chip used consists of MOS capacitors spanning over...
A combined high frequency and quasi-static C-V technique is used to monitor wafer charging of MOS structures during plasma immersion ion implantation (PIII). The test chip used consists of MOS capacitors spanning over five decades of area (25 mum2 to 4 x 10(6) mum2). Measured interface trap density (Q(it)) is found to be dependent exponentially on plasma exposure time and linearly on the plasma ion density. The PIII pulsed bias voltage has no effect on wafer charging.
Nearly isotropic etching of the 6H-SiC carbon face has been achieved in a remote plasma at 330-degrees-C using a mixture of O2 and NF3 in argon. Using evaporated aluminum as a mask, undercutting has been observed to a...
详细信息
Nearly isotropic etching of the 6H-SiC carbon face has been achieved in a remote plasma at 330-degrees-C using a mixture of O2 and NF3 in argon. Using evaporated aluminum as a mask, undercutting has been observed to a distance equal to the etch depth. The etch rate is a function of the ratio Of O2 to NF3 flow rates and of temperature, peaking strongly to 220 nm/min at 82% oxygen for 330-degrees-C. Smooth surfaces were obtained for gas ratios leading to the maximum etch rate, and also for a NF3-argon mixture, with significant roughening observed for other O2-NF3-argon mixtures. In the absence of a practical wet etch for SiC, this procedure is promising for isotropic etching in SiC device processing.
In direct methods of contrast enhancement, a contrast measure is first defined, which is then modified by a mapping function to generate the pixel value of the enhanced image. Various mapping functions such as the squ...
详细信息
暂无评论