JVB Nameplate
  • Volume/Page
  • Keyword
  • DOI
  • Citation
  • Advanced
   
 
 
 
Search Issue | RSS Feeds RSS
Previous Issue

Nov 1998

Volume 16, Issue 6, pp. 2915-3953


Electrical transport and far-infrared transmission in a quantum wire array

J. Lefebvre, J. Beerens, Y. Feng, Z. Wasilewski, J. Beauvais, and E. Lavallée

J. Vac. Sci. Technol. B 16, 2915 (1998); http://dx.doi.org/10.1116/1.590368 (13 pages) | Cited 1 time

Full Text: | Download PDF

Show Abstract
A wide set of data obtained on a two-dimensional electron gas submitted to a tunable lateral modulation, induced using a split-gate technique, is presented. Owing to a unique design of the sample, it has been possible to combine in a single experimental run, far-infrared transmission measurements and electrical transport measurements in both directions parallel and perpendicular to the lateral modulation. The discussion of the results emphasizes the correspondence between various features observed in both types of measurements. Based on these features, three regimes of modulation are clearly identified, namely the weak, intermediate and strong modulation regimes. Far-infrared transmission data show that each of these regimes is characterized by plasmon modes with a distinctive behavior. These behaviors are analyzed further with the use of transport data, which allow to determine the electron concentration in the structure for every condition of gate voltage. In the weak modulation regime, a quantitative analysis shows that the collective mode energy is consistent with that of a classical 2D plasmon at q=2π/a (where a is the period of the split gate), using the average electron concentration under the gate as the relevant parameter. In the intermediate regime, the collective modes are confined plasmons. The observation of “confined Bernstein modes” indicates that the bare confinement potential is nonparabolic in this regime. In the strong modulation regime, the observation of a far-infrared resonance energy which does not depend on the modulation amplitude, while the effective 2D electron concentration (within each wire) varies with gate voltage, shows that the collective mode is a Kohn mode. © 1998 American Vacuum Society.
Show PACS
73.61.Ey III-V semiconductors
78.66.Fd III-V semiconductors
73.21.-b Electron states and collective excitations in multilayers, quantum wells, mesoscopic, and nanoscale systems
71.45.Gm Exchange, correlation, dielectric and magnetic response functions, plasmons
73.20.Mf Collective excitations (including excitons, polarons, plasmons and other charge-density excitations)

Luminescence from erbium implanted silicon–germanium quantum wells

M. Q. Huda, J. H. Evans-Freeman, A. R. Peaker, D. C. Houghton, and A. Nejim

J. Vac. Sci. Technol. B 16, 2928 (1998); http://dx.doi.org/10.1116/1.590320 (6 pages) | Cited 5 times

Full Text: | Download PDF

Show Abstract
We have investigated the luminescence emitted at 1.54 μm from erbium-implanted strained ultrahigh vacuum chemical vapor deposition-grown (UHVCVD-grown) Si1−xGex quantum wells. Germanium fractions of up to 13% were used, and all well widths were below the critical thickness for pseudomorphic growth. A preliminary study was carried out on Si1−xGex quantum wells implanted with amorphizing doses of silicon at 77 K in order to study the regrowth across the interfaces, and subsequent structural and optical recovery. After amorphization and regrowth by a two stage anneal process, transmission electron microscopy (TEM) clearly showed the presence of the quantum wells, with sharp contrast. X-ray diffraction (XRD) studies showed that good regrowth has been achieved, with line widths very similar to the original material. However, the photoluminescence (PL) was found to be dependent upon the duration of the first anneal. Increasing the anneal time resulted in PL spectra being dominated by broad signals between 0.9 and 0.97 eV associated with structural defects. High concentrations of erbium were incorporated into the strained Si1−xGex quantum wells by implantation and solid phase epitaxial regrowth. TEM and XRD studies showed that the quantum wells retained their structure, with negligible segregation or diffusion of the germanium during the recrystallization. Erbium-related emission centered at 1.54 μm was observed in the implanted Si1−xGex layers after regrowth, and generally found to be of similar intensity as that in bulk silicon implanted with more than an order of magnitude higher dose of erbium. © 1998 American Vacuum Society.
Show PACS
78.55.Hx Other solid inorganic materials
78.66.Li Other semiconductors
61.72.up Other materials
61.72.Cc Kinetics of defect formation and annealing
81.15.Np Solid phase epitaxy; growth from solid phases

Optical absorption of Ag nanoclusters in Ag+-implanted c-SiO2

Xiao-Dong Feng, Min-Bo Tian, and Zheng-Xin Liu

J. Vac. Sci. Technol. B 16, 2934 (1998); http://dx.doi.org/10.1116/1.590321 (4 pages) | Cited 2 times

Full Text: | Download PDF

Show Abstract
c-SiO2 samples were implanted with Ag+ at an energy of 200 keV to doses in the range of 2.3–9×1016 ions/cm2 at room temperature. At a dose of 6.7×1016 ions/cm2, the transmission electron microscopy image shows that the implanted layer consists of two major sizes of nanoclusters: the large clusters, found in the deeper layer, are about 20 nm in diameter; the smaller clusters, found near the surface, are about 5 nm in diameter. At the relatively low dose of 2.3×1016 ions/cm2, there is only one optical absorption band caused by surface plasmon resonance. At a higher dose, a splitting of the absorption band and the redshift are attributed to a dipole interaction between nanoclusters for a high density of nanoclusters. © 1998 American Vacuum Society.
Show PACS
78.40.Ha Other nonmetallic inorganics
61.46.-w Structure of nanoscale materials
73.20.Mf Collective excitations (including excitons, polarons, plasmons and other charge-density excitations)
61.72.up Other materials

Silicon nanopillars formed with gold colloidal particle masking

P. A. Lewis, H. Ahmed, and T. Sato

J. Vac. Sci. Technol. B 16, 2938 (1998); http://dx.doi.org/10.1116/1.590322 (4 pages) | Cited 22 times

Full Text: | Download PDF

Show Abstract
Silicon nanopillars were fabricated by a novel natural lithography technique utilizing a gold colloidal particle monolayer as an etch mask. Using SiCl4 based reactive ion etching (RIE), silicon nanopillars with high density and uniformity in height and shape were obtained with 15 and 10 nm diam gold colloidal particles. The uniform pillars obtained from the 15 nm colloidal gold etch mask were subsequently sharpened to less than 5 nm diam tips by oxidation. 5 nm diam colloids were used to obtain nonuniform 5 nm diam pillars directly by RIE, but with 2 nm colloids the limit of pillar formation was reached. The pillars were also fabricated in selected areas by lithographic patterning of the substrate. © 1998 American Vacuum Society.
Show PACS
81.07.-b Nanoscale materials and structures: fabrication and characterization
81.16.-c Methods of micro- and nanofabrication and processing
85.35.-p Nanoelectronic devices
85.40.Hp Lithography, masks and pattern transfer
81.65.Mq Oxidation
81.65.Cf Surface cleaning, etching, patterning

Detection and control of ferroelectric domains by an electrostatic force microscope

J. W. Hong, D. S. Kahng, J. C. Shin, H. J. Kim, and Z. G. Khim

J. Vac. Sci. Technol. B 16, 2942 (1998); http://dx.doi.org/10.1116/1.590323 (5 pages) | Cited 21 times

Full Text: | Download PDF

Show Abstract
An electrostatic force microscopy (EFM) method has been used for the detection and control of the microdomain in ferroelectric single crystal [triglycine sulfate (TGS)] and thin film piezoelectric transducer (PZT). In this method, EFM is operated in a dynamic contact mode that allows a simultaneous measurement of the topographic and domain contrast images. Through the analysis of the force between the tip and ferroelectric surface, the surface charge density of TGS single crystal is obtained. Polarization charge density of TGS obtained in this method is 2.7 μC/cm2 at room temperature. A complex pattern was written on a PZT film by the polarization reversal. The line shape or the intensity of the reoriented domain does not show any noticeable dependence on the writing speed. The threshold bias for writing on a PZT film studied in this work was 4 V. © 1998 American Vacuum Society.
Show PACS
77.80.Dj Domain structure; hysteresis
77.84.Ek Niobates and tantalates
77.84.Cg PZT ceramics and other titanates
77.84.Jd Polymers; organic compounds
07.79.Lh Atomic force microscopes
68.37.Ef Scanning tunneling microscopy (including chemistry induced with STM)
68.37.Ps Atomic force microscopy (AFM)
68.37.Rt Magnetic force microscopy (MFM)
68.37.Uv Near-field scanning microscopy and spectroscopy
77.22.Ej Polarization and depolarization
85.50.-n Dielectric, ferroelectric, and piezoelectric devices
77.55.-g Dielectric thin films

Atomistic study of nickel silicide structures on Si(100) by tunneling microscopy

Izumi Ono, Masamichi Yoshimura, and Kazuyuki Ueda

J. Vac. Sci. Technol. B 16, 2947 (1998); http://dx.doi.org/10.1116/1.590324 (5 pages) | Cited 2 times

Full Text: | Download PDF

Show Abstract
Surface structures of Ni-deposited Si(100) have been investigated using scanning tunneling microscopy (STM). After heating at 670 °C, NiSi2 islands consisting of patch features grow on the (2×n) substrate. We have proposed a new structure model for NiSi2 island by taking account of the correlation of heights and lateral positions between top-layer atoms in the NiSi2 and Si dimers in the substrate. In the model, the top-layer atoms, with a nearest neighbor distance of 0.38 nm, are located at bridge sites on the Ni layer of NiSi2. Patch features consist of two equivalent anti-phase sites shifted by half the unit length of 1×1 along both [011] and [01̄1] directions. In addition, it is found in the empty-state STM images that the center atoms make pairs two by two. © 1998 American Vacuum Society.
Show PACS
68.37.Ef Scanning tunneling microscopy (including chemistry induced with STM)
68.37.Ps Atomic force microscopy (AFM)
68.37.Rt Magnetic force microscopy (MFM)
68.37.Uv Near-field scanning microscopy and spectroscopy
68.35.B- Structure of clean surfaces (and surface reconstruction)
73.40.Ns Metal-nonmetal contacts

Growth of silicon oxide on hydrogenated silicon during lithography with an atomic force microscope

F. Marchi, V. Bouchiat, H. Dallaporta, V. Safarov, D. Tonneau, and P. Doppelt

J. Vac. Sci. Technol. B 16, 2952 (1998); http://dx.doi.org/10.1116/1.590325 (5 pages) | Cited 20 times

Full Text: | Download PDF

Show Abstract
We present an experimental study of growth of silicon oxide strips drawn on hydrogenated silicon under the voltage biased tip of an atomic force microscope operating in ambient atmosphere. Oxide formation was found to occur at negative tip biases above a voltage threshold around ∣−2∣V, corresponding to the minimum electric field required for hydrogen removal from the substrate surface. We show the influence of tip-sample distance and of the chemical composition of the atmosphere on the growth. An ozone enriched atmosphere leads to a growth kinetics enhancement. © 1998 American Vacuum Society.
Show PACS
81.65.Mq Oxidation
85.40.Hp Lithography, masks and pattern transfer
81.05.Cy Elemental semiconductors
07.79.Lh Atomic force microscopes

Plasma polymer films for 532 nm laser micromachining

M. S. Silverstein, I. Visoly, O. Kesler, M. Janai, and Y. Cassuto

J. Vac. Sci. Technol. B 16, 2957 (1998); http://dx.doi.org/10.1116/1.590326 (11 pages) | Cited 3 times

Full Text: | Download PDF

Show Abstract
Laser micromachining with a frequency doubled Nd:YAG laser (532 nm) can replace more complex microlithographic processes for rapid turnaround in the development of prototype application-specific integrated circuits. Plasma polymerization is a rapid, dry, environmentally friendly process that yields crosslinked pinhole-free films. Plasma polymerized films of ethylene and an additional gas [PP(gas/E)] were investigated for their micromachining potential. The deposition rates, molecular structures, physical properties and optical properties of the polymers were characterized. PP(Ar/E), with relatively little oxygen and no nitrogen, with superior substrate adhesion and with no debris generated on laser micromachining was chosen as the optimal laser micromachining film. The PP(Ar/E) coefficient of optical absorption at 532 nm (α532), related to unsaturated group concentration, increased with the ratio of plasma power to ethylene mass flow rate [W/Fm(E)]. α532 reached an asymptote of 2.9 μm−1 at high W/Fm(E) and could be enhanced slightly using postpolymerization ultraviolet exposure. The optimum conditions were using Ar/E=1/1 and 75 W to produce a 0.6 μm thick film for micromachining at 2 J/cm2 focused 0.25 μm beneath the surface. The laser pulse in a 1.2 μm thick film was not fully developed at 2 J/cm2 and exhibited rounded corners at 4 J/cm2, indicating that multiple low energy pulses would be preferable. A complicated and densely packed pattern with several different pulse sizes in which neighboring holes from pulses in close proximity do not merge was accurately reproduced in PP(Ar/E) using laser micromachining. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
82.35.-x Polymers: properties; reactions; polymerization
52.77.Bn Etching and cleaning
52.77.Dq Plasma-based ion implantation and deposition
85.40.Sz Deposition technology
81.65.Cf Surface cleaning, etching, patterning

Nonstatistical degradation and development characteristics of poly(methylmethacrylate) based resists during electron beam exposure

A. Uhl, J. Bendig, J. Leistner, U. Jagdhold, L. Bauch, and M. Böttcher

J. Vac. Sci. Technol. B 16, 2968 (1998); http://dx.doi.org/10.1116/1.590327 (6 pages) | Cited 1 time

Full Text: | Download PDF

Show Abstract
We present an investigation to study the degradation of poly(methylmethacrylate) (PMMA) based resists during electron beam exposure and their development characteristics. After exposure to the electron beam we detect for ARP 610 resist (PMMA 74%, PMAA 26%) and for a homopolymeric PMMA resist, a similar bimodal respectively multimodal molecular weight distribution curve characterized by a shift of the maximum from 105 to 103 g/mol and an increase of low molecular weight fractions (≈103 g/mol) with increasing exposure dose. The model of Greeneich can only be applied to lower deposited energy densities (ARP 610<4 eV/nm3, homopolymeric PMMA <8 eV/nm3). The difference found between experimental data and modeling values at higher deposited energy densities for both resists, results from a formation of stable low molecular weight fractions. Comparing the measured dissolution rate with the calculated one, the determined difference proves the nonapplicability of the empirical formula for the dissolution rate given by Greeneich. The low molecular weight fractions influence the dissolution rate significantly more than given by their portion in the molecular weight distribution. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Electrical conductivity measurement for quantitative evaluation of development speed of a photoresist

T. Takeda and M. Saka

J. Vac. Sci. Technol. B 16, 2974 (1998); http://dx.doi.org/10.1116/1.590328 (3 pages) | Cited 1 time

Full Text: | Download PDF

Show Abstract
In a previous article, we reported on the possibility of predicting the development speed of a photoresist by measuring its electrical conductivity in an organic solvent [T. Takeda and M. Saka, J. Vac. Sci. Technol. B (submitted)]. The electrical conductivity of the organic solution of a photoresist base polymer at various frequencies was found to have a positive correlation with the dissolution speed of photoresist base polymers into an alkaline developer. In addition, the influence of the measurement temperature and the concentration of polymer solution on conductivity was investigated. Based on the collected data, this article derives the calibration equation, which was influenced neither by measurement temperature nor by the concentration of the solution. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
84.37.+q Measurements in electric variables (including voltage, current, resistance, capacitance, inductance, impedance, and admittance, etc.)
82.35.-x Polymers: properties; reactions; polymerization

Fabrication of submicron suspended structures by laser and atomic force microscopy lithography on aluminum combined with reactive ion etching

A. Boisen, K. Birkelund, O. Hansen, and F. Grey

J. Vac. Sci. Technol. B 16, 2977 (1998); http://dx.doi.org/10.1116/1.590329 (5 pages) | Cited 13 times

Full Text: | Download PDF

Show Abstract
We report a simple fabrication method for suspended submicron silicon and silicon oxide structures. The structures are defined by laser and atomic force microscopy (AFM) writing on a 7-nm-thick aluminum film. For laser writing the minimum obtained linewidth is 500 nm, whereas for AFM it is approximately 100 nm. During AFM writing, aluminum oxide is formed, whereas during laser writing, a compound containing aluminum and silicon is formed. For both processes the nonpatterned aluminum can be dissolved selectively in a wet chemical etch leaving the patterned areas as an etch mask. Alternatively, the aluminum oxide can be etched to form a positive etch mask. Aluminum and aluminum oxide are both excellent etch masks for reactive ion etching (RIE) of silicon and silicon oxide. Hence, by combinations of RIE processes, a variety of structures can be fabricated from the aluminum based masks. To illustrate the flexibility of this technique we demonstrate fabrication of submicron cantilevers and bridges defined in silicon and silicon oxide. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
07.10.Cm Micromechanical devices and systems
81.65.Cf Surface cleaning, etching, patterning
07.79.Lh Atomic force microscopes
81.07.-b Nanoscale materials and structures: fabrication and characterization
81.16.-c Methods of micro- and nanofabrication and processing
85.35.-p Nanoelectronic devices

Submicrometer transmission mask fabricated by low-temperature SF6/O2 reactive ion etching and focused ion beam

H. Y. Sheng, D. Fujita, T. Ohgi, H. Okamoto, and H. Nejoh

J. Vac. Sci. Technol. B 16, 2982 (1998); http://dx.doi.org/10.1116/1.590330 (4 pages) | Cited 2 times

Full Text: | Download PDF

Show Abstract
A novel technique is presented to fabricate a silicon submicrometer transmission mask for nanofabrication. One of the applications of the mask is to fabricate a single electron transistor using an ultrahigh vacuum scanning tunneling microscope. The mask fabrication processes involve KOH wet etching, electron beam lithography, low-temperature SF6/O2 plasma-assisted reactive ion etching and focused ion beam techniques. Using this method, we fabricated masks with pattern sizes of 2.5×2.5 mm2 and opaque parts of submicrometer scale. After the evaporation, by using the mask a submicrometer gap on the metal pattern can be obtained. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
81.07.-b Nanoscale materials and structures: fabrication and characterization
81.16.-c Methods of micro- and nanofabrication and processing
85.35.-p Nanoelectronic devices
81.65.Cf Surface cleaning, etching, patterning
52.77.Bn Etching and cleaning
52.77.Dq Plasma-based ion implantation and deposition

Cleaning of CHF3 plasma-etched SiO2/SiN/Cu via structures using a hydrogen plasma, an oxygen plasma, and hexafluoroacetylacetone vapors

Kazuyoshi Ueno, Vincent M. Donnelly, and Yasuaki Tsuchiya

J. Vac. Sci. Technol. B 16, 2986 (1998); http://dx.doi.org/10.1116/1.590331 (10 pages) | Cited 8 times

Full Text: | Download PDF

Show Abstract
Cleaning processes for CHF3 reactive ion etched Cu vias, consisting of exposure to a hydrogen plasma, an oxygen plasma, and hexafluoroacetylacetone [H(hfac)] vapors have been investigated. After each step in the cleaning process, the dielectric surface and the Cu surface of via structures were analyzed by in situ by angle-resolved x-ray photoelectron spectroscopy. A hydrogen plasma was effective in removing carbon and fluorine deposits on all of the surfaces, and CuO and Cu2O on the Cu surface at the via bottom. It was not effective, however, in removing the Cu deposited on the dielectric surfaces. An oxygen plasma is effective in removing all the carbon and some fluorine deposits. Cu deposits on the dielectric surfaces were not removed, however, and the Cu surface was oxidized. Exposure to H(hfac) vapors reduced some of the Cu deposits on the dielectric, however Cu diffusion into SiO2 possibly occurred during this exposure at the elevated temperature of 200 °C. CuO and Cu2O were removed by the H(hfac) exposure above 150 °C. A three-step cleaning sequence was devised which consists of a brief oxygen-plasma exposure, a dilute-HF solution dip, followed by exposure to H(hfac) vapors. The cleaning sequence is effective in obtaining a clean dielectric surface and an oxide-free Cu surface at the via bottom that allows low contact resistances of 5.8–6.8×10−11 Ω cm2. © 1998 American Vacuum Society.
Show PACS
85.40.Ls Metallization, contacts, interconnects; device isolation
81.65.Cf Surface cleaning, etching, patterning
52.77.Bn Etching and cleaning
52.77.Dq Plasma-based ion implantation and deposition
82.80.Pv Electron spectroscopy (X-ray photoelectron (XPS), Auger electron spectroscopy (AES), etc.)
79.60.Bm Clean metal, semiconductor, and insulator surfaces

Endpoint uniformity sensing and analysis in silicon dioxide plasma etching using in situ mass spectrometry

J. J. Chambers, K. Min, and G. N. Parsons

J. Vac. Sci. Technol. B 16, 2996 (1998); http://dx.doi.org/10.1116/1.590332 (7 pages) | Cited 7 times

Full Text: | Download PDF

Show Abstract
Mass spectroscopy is used to characterize the endpoint uniformity of silicon dioxide etching in an electron cyclotron resonance (ECR) plasma etch process. Etch products are observed using a two stage differentially pumped mass spectrometry system attached to the ECR process chamber. Specifically, using CF4 and D2 etch gases, the partial pressure of CO-containing etch products decays near the endpoint, and the rate of signal decay is directly correlated with the uniformity determined from optical interferometry thickness measurements. To correlate the mass spectrometer signal with the etch rate variation across the wafer, etch uniformity is altered by changing the ECR electromagnet geometry and by modifying the initial oxide uniformity. A COF2 etch product material balance is developed to model the observed concentration versus time data, resulting in a quantitative correlation between change in endpoint slope and uniformity. The ability to utilize a process-state sensor, such as a mass spectrometer, for wafer-state information will result in new approaches for sensing, optimizing, and controlling integrated circuit fabrication processes. © 1998 American Vacuum Society.
Show PACS
81.65.Cf Surface cleaning, etching, patterning
52.77.Bn Etching and cleaning
52.77.Dq Plasma-based ion implantation and deposition
07.75.+h Mass spectrometers

BCl3/Ar reactive ion etching for gate recessing of GaInP/InGaAs/GaAs pseudomorphic high electron mobility transistors

C. W. Kuo, Y. K. Su, H. H. Lin, and C. Y. Chin

J. Vac. Sci. Technol. B 16, 3003 (1998); http://dx.doi.org/10.1116/1.590369 (5 pages)

Full Text: | Download PDF

Show Abstract
BCl3 reactive ion etching for gate recessing of GaInP/InGaAs/GaAs pseudomorphic high electron mobility transistors (PHEMTs) is found improved by the addition of an appropriate amount of Ar to the gas flow. The influence of the BCl3/Ar gas flow ratio on GaAs to GaInP etch selectivity, surface roughness, and surface damage was studied. The results indicate that the conditions for minimum plasma damage, as determined by photoreflectance (PR) spectroscopy, corresponded with the conditions for minimum surface roughness, as determined by atomic force microscopy (AFM). The optimal BCl3/Ar gas flow ratio for minimum surface damage and roughness was found to be 6:4. Two BCl3:Ar flow rate ratios, 6:4 (optimal ratio) and 10:0 (pure BCl3) were used for gate recess etching in the fabrication of GaInP/InGaAs/GaAs PHEMTs. From drain–source current to gate–source voltage (Ids–Vgs) measurements, it was found that the plasma-induced damage for the sample Sc dry etched with 6:4 BCl3/Ar is less than that of the sample Se dry etched with pure BCl3. The dc and small signal rf characteristics of PHEMT Sc were superior to those of the wet-etched PHEMT S0 and PHEMT Se dry etched with pure BCl3. The improvement is attributed to the lower parasitic source resistance associated with the tighter recess geometry of the BCl3 plasma recess device. These results show that photoreflectance spectroscopy is a powerful tool for investigating surface damage and can be used to improve the performance of PHEMTs. © 1998 American Vacuum Society.
Show PACS
85.30.Tv Field effect devices
52.77.Bn Etching and cleaning
52.77.Dq Plasma-based ion implantation and deposition
81.65.Cf Surface cleaning, etching, patterning

Interaction between gas rarefaction and metal ionization in ionized physical vapor deposition

S. M. Rossnagel

J. Vac. Sci. Technol. B 16, 3008 (1998); http://dx.doi.org/10.1116/1.590333 (5 pages) | Cited 21 times

Full Text: | Download PDF

Show Abstract
The process known as ionized physical vapor deposition, or I-PVD, consists of the physical sputtering of metal atoms into a dense, inert gas plasma, ionization of the sputtered metal atoms, and subsequent deposition of the films from these metal ions. Measurements have shown a decrease in electron temperature coupled with an unexpected decrease in plasma density as a function of increasing metal flux. Recent plasma modeling work has suggested gas rarefaction as the underlying factor in these declines. Measurements of neutral gas density in the plasma region reported here confirm this model and are consistent with earlier studies of sputtered atom induced gas heating and rarefaction. © 1998 American Vacuum Society.
Show PACS
81.15.Cd Deposition by sputtering
85.40.Sz Deposition technology

Atomically flat gold film surfaces deposited on Si (111) surfaces at room temperature

H. Okamoto and H. Nejo

J. Vac. Sci. Technol. B 16, 3013 (1998); http://dx.doi.org/10.1116/1.590334 (2 pages) | Cited 2 times

Full Text: | Download PDF

Show Abstract
We have observed atomically flat gold surfaces deposited on Si (111) 7×7 surfaces. These films are deposited by conventional evaporative deposition at a pressure below 2×10−6 Pa. The surface topography was taken by scanning tunneling microscopy. Atomically flat terraces as large as 100 Å separated by steps were observed. This surface should provide a good substrate for scanning tunneling microscopy studies. © 1998 American Vacuum Society.
Show PACS
68.35.B- Structure of clean surfaces (and surface reconstruction)
68.55.-a Thin film structure and morphology
68.37.Ef Scanning tunneling microscopy (including chemistry induced with STM)
68.37.Ps Atomic force microscopy (AFM)
68.37.Rt Magnetic force microscopy (MFM)
68.37.Uv Near-field scanning microscopy and spectroscopy

Correlation between the early stage of copper metal organic chemical vapor deposition and the material properties of thin film

Eui Seong Hwang and Jihwa Lee

J. Vac. Sci. Technol. B 16, 3015 (1998); http://dx.doi.org/10.1116/1.590335 (6 pages) | Cited 9 times

Full Text: | Download PDF

Show Abstract
Cu metal organic chemical vapor deposition has been performed on TiN/Si(100) substrates using Cu(I)(hfac)(vtms) at various substrate (110⩽Ts⩽300 °C) and bubbler temperature (0⩽Tb⩽30 °C) conditions, in which the early stage of film formation characterized by scanning electron microscopy and in situ laser beam (λ=6328 Å) reflectance measurement could be well correlated with the surface roughness and the electrical resistivity of the copper films developed in a later stage. All the reflectance versus deposition time curves showed a qualitatively similar shape consisting of three sequential parts: a rapid initial decrease to a minimum, an increase to a maximum, and a gradual decay to almost zero, in which the minimum and maximum roughly correspond to the onset and completion of island coalescence, respectively. As the Ts is raised and/or the Tb is lowered, larger islands were formed in a smaller density at the onset of island coalescence due to a lower nucleation rate. The surface roughness gradually develops with increasing film thickness after forming a continuous film, but it does not saturate at a thickness of ∼2 μm and shows a correlation with the average island size in the earlier coalescence stage, i.e., the smaller islands evolves to a smoother surface. Above the threshold temperature (Ts∼230 for Tb=30 °C) the copper islands begin to show facets and at even higher Ts large truncated polyhedron-shaped copper islands are formed, which upon coalescing leave many interfacial voids to result in a film with a very rough surface and a high electrical resistivity. The copper films deposited at 150⩽Ts⩽200 °C have a very low value of ρ≈2 μΩ cm, which increases by a factor of ∼2 when Ts is lowered to 110 °C. The latter was accompanied by the formation of smaller islands at the onset of coalescence, and therefore the increase in ρ at Ts⩽150 °C was attributed to the electron scattering at the grain boundaries. © 1998 American Vacuum Society.
Show PACS
81.05.Bx Metals, semimetals, and alloys
81.15.Gh Chemical vapor deposition (including plasma-enhanced CVD, MOCVD, ALD, etc.)
68.55.-a Thin film structure and morphology
85.40.Ls Metallization, contacts, interconnects; device isolation
68.35.B- Structure of clean surfaces (and surface reconstruction)
73.61.At Metal and metallic alloys
78.66.Bz Metals and metallic alloys
78.20.Ci Optical constants (including refractive index, complex dielectric constant, absorption, reflection and transmission coefficients, emissivity)
68.55.Ln Defects and impurities: doping, implantation, distribution, concentration, etc.

Properties of sputtered Cr–O and reactively sputtered Cr–N–O as passivation layers against copper oxidation

Jui-Chang Chuang and Mao-Chieh Chen

J. Vac. Sci. Technol. B 16, 3021 (1998); http://dx.doi.org/10.1116/1.590336 (6 pages)

Full Text: | Download PDF

Show Abstract
Passivation layers of 200 Å sputtered Cr–O as well as reactive sputtered Cr–N–O were studied with respect to the passivation capability against thermal oxidation of Cu in flowing nitrogen and flowing oxygen ambients. In a flowing N2 ambient, both Cr–O and Cr–N–O passivation layers were able to prohibit oxidation of Cu at temperatures up to 700 °C. In an O2 ambient, the passivation capability of Cr–N–O layer was found to be 500 °C, which is 150 °C higher than that of Cr–O layer. The superiority of the passivation capability of the Cr–N–O layer is presumably due to decoration of the surface defects and grain boundaries with nitrogen, which provide fast paths for oxygen and copper diffusion. © 1998 American Vacuum Society.
Show PACS
81.65.Rv Passivation
81.65.Mq Oxidation
81.15.Cd Deposition by sputtering

Role of V-shaped stacking faults in Au/n-type ZnMgSSe:Cl Schottky diodes

Ching-Wu Wang

J. Vac. Sci. Technol. B 16, 3027 (1998); http://dx.doi.org/10.1116/1.590337 (5 pages)

Full Text: | Download PDF

Show Abstract
The microstructural, optical, and electronic characteristics of both Cl-doped ZnMgSSe epilayers and Schottky diodes grown by the molecular beam epitaxy method on n+-type GaAs (100) substrates with different thicknesses of Cl-doped ZnSe buffer layers were investigated via three measurements: transmission electron microscopy, photoluminescence, and deep level transient spectroscopy. The V-shaped stacking faults, originating at or near the ZnSe:Cl (buffer layer)/GaAs interface and extending into the ZnMgSSe:Cl epilayer, were discovered to have the evident inclination to exist in samples with thicker ZnSe:Cl buffer layers and higher growth temperatures. Evidence also revealed that the stacking faults not only produced poor optical quality of ZnMgSSe:Cl thin film, but also created more interface state densities at the Au/n-type ZnMgSSe:Cl Schottky junction. Such results are the main factors to damage the I–V characteristics of Au/n-type ZnMgSSe:Cl Schottky diode, including the less forward conduction current density, the bigger turn-on voltage, and the larger reverse leakage current followed by the smaller reverse breakdown voltage. © 1998 American Vacuum Society.
Show PACS
85.30.Hi Surface barrier, boundary, and point contact devices
85.30.Kk Junction diodes
61.72.Nn Stacking faults and other planar or extended defects
81.15.Hi Molecular, atomic, ion, and chemical beam epitaxy

Interface characterization of Si3N4/Si/GaAs heterostructures after high temperature annealing

Dae-Gyu Park, Zhonghui Wang, Hadis Morkoç, Samuel A. Alterovitz, David J. Smith, and S.-C. Y. Tsen

J. Vac. Sci. Technol. B 16, 3032 (1998); http://dx.doi.org/10.1116/1.590338 (9 pages) | Cited 4 times

Full Text: | Download PDF

Show Abstract
We present data on interface characteristics of Si3N4/Si/GaAs metal–insulator–semiconductor (MIS) structures and correlate electrical properties with spectroscopic ellipsometry, x-ray photoelectron spectroscopy (XPS), and transmission electron microscopy (TEM) observations. The interface of Si3N4/Si/GaAs heterostructures was electrically characterized by a combination of capacitance–voltage and conductance methods. The nature of an insulator/GaAs interface and the microstructure of Si3N4/Si/GaAs interfaces after high temperature annealing were investigated by variable angle spectroscopic ellipsometry and high resolution TEM, respectively. The evolution of chemical species in Si3N4/Si/GaAs heterostructures was examined using in situ angle-resolved XPS. The interface trap density (Dit) of the Si3N4/Si MIS capacitor was in the 2×1010 eV−1 cm−2 range near the Si midgap after rapid thermal annealing at 550 °C in N2. However, this density increased to high 1010 eV−1 cm−2 with annealing at 800 °C. The interface characteristics of Si3N4/Si/GaAs structures with Dit in the 7×1010 eV−1 cm−2 range also degraded after annealing at 750 °C in N2 with Dit increasing to 5×1011 eV−1 cm−2 near the GaAs midgap. The spectroscopic ellipsometry results together with high resolution TEM observations appear to suggest that the degradation is due in part to the interface changing from crystalline to amorphous through chemical reaction. XPS measurements revealed that the as-deposited Si interlayer is nitridated during the initial stages of silicon nitride deposition, thus the thinned Si cannot prevent the outdiffusion of Ga and As species. We circumvented thermally induced interface degradation of Si3N4/Si/GaAs structures by employing a novel ex situ/in situ growth approach. © 1998 American Vacuum Society.
Show PACS
73.40.Qv Metal-insulator-semiconductor structures (including semiconductor-to-insulator)
61.72.Cc Kinetics of defect formation and annealing
73.20.At Surface states, band structure, electron density of states
68.35.Fx Diffusion; interface formation
84.32.Tt Capacitors
85.30.Tv Field effect devices

Reduction of oxygen contamination in InGaP and AlGaInP films grown by solid source molecular beam epitaxy

W. E. Hoke, P. J. Lemonias, and A. Torabi

J. Vac. Sci. Technol. B 16, 3041 (1998); http://dx.doi.org/10.1116/1.590339 (7 pages) | Cited 6 times

Full Text: | Download PDF

Show Abstract
Oxygen contamination has been observed in In0.5Ga0.5P and (Al0.23Ga0.77)0.5In0.5P films grown by solid source molecular beam epitaxy with elemental phosphorus. Using a conventional P4 cracking zone temperature of 950 °C, spike contamination levels as high as 1×1019 cm−3 were observed at growth interrupted interfaces with the resultant deactivation of silicon doping pulses. By reducing the phosphorus cracking temperature to 700 °C, the oxygen level in InGaP was reduced to below the secondary ion mass spectrometry background level of 3×1016 cm−3. No measurable accumulation of oxygen was observed at growth interrupted interfaces and efficient silicon pulse doping was obtained. InGaP films grown at the lower cracking temperature exhibited improved mobilities and enhanced photoluminescence intensities. An oxygen level in (Al0.23Ga0.77)0.5In0.5P of less than 1.5×1017 cm−3 was obtained with good mobilities and luminescence. Efficient silicon pulse doping in AlGaInP was demonstrated. The oxygen contamination is in the phosphorus flux and is likely a volatile phosphorus oxide such as P4O6. © 1998 American Vacuum Society.
Show PACS
81.05.Ea III-V semiconductors
81.15.Hi Molecular, atomic, ion, and chemical beam epitaxy
68.55.Ln Defects and impurities: doping, implantation, distribution, concentration, etc.
61.72.S- Impurities in crystals
82.80.Ms Mass spectrometry (including SIMS, multiphoton ionization and resonance ionization mass spectrometry, MALDI)
73.50.Jt Galvanomagnetic and other magnetotransport effects (including thermomagnetic effects)
73.50.Dn Low-field transport and mobility; piezoresistance
72.80.Ey III-V and II-VI semiconductors
73.61.Ey III-V semiconductors
78.66.Fd III-V semiconductors
61.72.uj III-V and II-VI semiconductors
78.55.Cr III-V semiconductors

Surface chemistry of II–VI semiconductor ZnSe studied by time of flight secondary ion mass spectrometry and x-ray photoelectron spectroscopy

J. Zhao, M. H. Na, E. H. Lee, H. C. Chang, J. A. Gardella, and H. Luo

J. Vac. Sci. Technol. B 16, 3048 (1998); http://dx.doi.org/10.1116/1.590340 (7 pages) | Cited 6 times

Full Text: | Download PDF

Show Abstract
The composition and chemical state of the native oxide of single-crystalline ZnSe and the effect of HCl (18.5%)/H2O wet etching have been studied by time of flight secondary ion mass spectrometry (TOF-SIMS) and x-ray photoelectron spectroscopy (XPS). TOF-SIMS depth profile measurements show that the oxide layer of ZnSe is removed by etching in this solution for one minute, followed by a subsequent one minute rinse in deionized H2O. XPS depth profile measurements of the untreated ZnSe surface show that Se oxide only exists at the topmost surface (within the top 10% of the oxide layer). The change of Zn Auger parameter with depth of the untreated ZnSe specimen indicates that the remaining oxygen is chemically associated to Zn. High resolution XPS measurements of the etched ZnSe show no detectable Se oxide at the surface. Meanwhile, the Zn Auger parameter is similar to that of the unetched ZnSe after its oxide layer being removed by Ar+ sputtering. Both experiments show longer wet etching times result in Zn deficiency and more Cl contamination at the sample surface.© 1998 American Vacuum Society.
Show PACS
81.65.Cf Surface cleaning, etching, patterning
81.05.Dz II-VI semiconductors
79.60.Bm Clean metal, semiconductor, and insulator surfaces
79.20.Rf Atomic, molecular, and ion beam impact and interactions with surfaces

Evaluation of development speed of a photoresist by means of electrical conductivity measurement

T. Takeda and M. Saka

J. Vac. Sci. Technol. B 16, 3055 (1998); http://dx.doi.org/10.1116/1.590341 (4 pages) | Cited 1 time

Full Text: | Download PDF

Show Abstract
A method was examined for evaluating the electrical conductivity of a photoresist to predict its development speed. The conductivity of organic solutions of the photoresist base polymer at various frequencies was found to have a positive correlation with the dissolution speed of the photoresist base polymer into an alkaline developer. In addition, the influence of the measurement temperature and of the concentration of the polymer solution on the conductivity was investigated. It is shown that a prediction of development speed of a photoresist by measuring its conductivity is possible. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Investigation of Ta–RuO2 diffusion barrier for high density memory capacitor applications

Dong-Soo Yoon, Hong Koo Baik, Sung-Man Lee, Chang-Soo Park, and Sang-In Lee

J. Vac. Sci. Technol. B 16, 3059 (1998); http://dx.doi.org/10.1116/1.590342 (6 pages) | Cited 8 times

Full Text: | Download PDF

Show Abstract
The properties of both oxygen indiffusion and oxidation resistance in a Ta+RuO2 layer for high density memory devices were investigated by using Rutherford backscattering spectroscopy, four point probe, x-ray diffraction, x-ray photoelectron spectroscopy, and planar transmission electron microscopy. The Ta+RuO2/Si system sustained up to 800 °C without an increase in resistivity. The Ta+RuO2 diffusion barrier showed a Ta amorphous microstructure and an embedded RuOx nanocrystalline structure in the as-deposited state. The Ta+RuO2 film showed the formation of RuO2 phase by reaction with the indiffused oxygen from atmosphere after annealing in an air ambient. The Ta+RuO2 diffusion barrier showed that Ta is sufficiently bound to oxygen in the as-deposited state, but RuO2 consists of Ru and Ru–O binding state. The Ta–O bonds showed little change compared to the as-deposited state with increasing annealing temperature, whereas Ru–O bonds significantly increased and transformed to conductive oxide, RuO2. Therefore, the Ta layer deposited by RuO2 addition effectively prevented the indiffusion of oxygen up to 800 °C and its oxidation resistance was superior to various barriers reported by others. © 1998 American Vacuum Society.
Show PACS
68.35.Fx Diffusion; interface formation
84.32.Tt Capacitors
85.50.-n Dielectric, ferroelectric, and piezoelectric devices
84.30.Sk Pulse and digital circuits
66.30.Ny Chemical interdiffusion; diffusion barriers
81.65.Mq Oxidation
82.80.Yc Rutherford backscattering (RBS), and other methods of chemical analysis
79.60.Jv Interfaces; heterostructures; nanostructures
68.35.Ct Interface structure and roughness
81.40.Gh Other heat and thermomechanical treatments

Nonerratic behavior of overerased bits in flash EEPROM

F. D. Nkansah, E. Prinz, and M. Hatalis

J. Vac. Sci. Technol. B 16, 3065 (1998); http://dx.doi.org/10.1116/1.590343 (4 pages)

Full Text: | Download PDF

Show Abstract
We report the effects of flash “channel” programming, or severe gate disturb, on the threshold voltage of fast or overerased bits. Experiments have been performed to establish that this class of fast bits are nonerratic and remain fast after 250 °C bake. These fast bits exhibit identical subthreshold characteristics similar to that of a normal bit after ultraviolet erase, thus establishing that the initial charge stored on the floating gate is the same for both normal and fast bits. Polysilicon grain boundary enhanced electric fields which result in impact ionization by tunneling electrons, thus generating trapped positive charges in the grain boundary oxide ridges are believed to play an important role in the generation of fast bits. © 1998 American Vacuum Society.
Show PACS
84.30.Sk Pulse and digital circuits

Vacuum properties of a new panel structure for field emission displays

Y. R. Cho, H. S. Kim, J. D. Mun, J. Y. Oh, H. S. Jeong, and S. Ahn

J. Vac. Sci. Technol. B 16, 3069 (1998); http://dx.doi.org/10.1116/1.590344 (4 pages) | Cited 2 times

Full Text: | Download PDF

Show Abstract
A new field emission display (FED) panel structure with an auxiliary space was developed in order to increase the conductance of the system for evacuation and help evacuate the cavity inside the FED more effectively. Gas molecules in the main space of the panel can move easily into the auxiliary space through the multiple openings in the base plate and can then be pumped out through the exhaust tube at the auxiliary tank. For the very thin panel, in which the space between the face plate and base plate is 58 μm, pressure drop characteristics during evacuation are very sensitive to the number of openings in the base plate. The main purpose of multiple openings in shortening evacuation time is to reduce pressure gradient inside the cavity. Additionally, the auxiliary space makes it easy to place various getters inside the FED panel and to maintain high vacuum for a prolonged lifetime of the FED. © 1998 American Vacuum Society.
Show PACS
85.45.Fd Field emission displays (FEDs)

Effect of CH4 on the electron emission characteristics of active molybdenum field emitter arrays

Babu R. Chalamala, Robert M. Wallace, and Bruce E. Gnade

J. Vac. Sci. Technol. B 16, 3073 (1998); http://dx.doi.org/10.1116/1.590444 (4 pages) | Cited 8 times

Full Text: | Download PDF

Show Abstract
Exposing active Spindt-type molybdenum field emission microcathode arrays to CH4 resulted in increased electron emission. Exposures of 250 and 2500 L result in electron emission enhancement of 106.2±41.0% and 200.1±20.7% and a decrease in tip work function by 10.2±5.6% and 17.6±1.7%, respectively. The measured work function changes indicate the formation of molybdenum carbides on the field emitter tips. © 1998 American Vacuum Society.
Show PACS
85.45.Db Field emitters and arrays, cold electron emitters

Easy and reproducible method for making sharp tips of Pt/Ir

J. Lindahl, T. Takanen, and L. Montelius

J. Vac. Sci. Technol. B 16, 3077 (1998); http://dx.doi.org/10.1116/1.590445 (5 pages) | Cited 9 times

Full Text: | Download PDF

Show Abstract
We have investigated a simple yet powerful method for making sharp scanning tunneling microscopy tips of Pt/Ir. It consists of three electrochemical polish/etch steps, not requiring any special micropolishing. The tips, as seen by high-resolution transmission electron microscopy, are sharper than 20 Å. Furthermore, they are smooth, without mini tips and covered with an oxide layer not much thicker than the native one (as seen from a freshly cut wire) when all three steps are applied. © 1998 American Vacuum Society.
Show PACS
07.79.Cz Scanning tunneling microscopes
68.37.Ef Scanning tunneling microscopy (including chemistry induced with STM)
68.37.Ps Atomic force microscopy (AFM)
68.37.Rt Magnetic force microscopy (MFM)
68.37.Uv Near-field scanning microscopy and spectroscopy
81.65.Ps Polishing, grinding, surface finishing
81.65.Cf Surface cleaning, etching, patterning

Microelectron gun with silicon field emitter

Yasuhiro Endo, Ichiro Honjo, and Shunji Goto

J. Vac. Sci. Technol. B 16, 3082 (1998); http://dx.doi.org/10.1116/1.590495 (4 pages) | Cited 6 times

Full Text: | Download PDF

Show Abstract
We evaluated the characteristics of single Si field emitters (Si-FEs) from the perspective of their use in electron guns. Since the current fluctuation of Si-FEs strongly depends on the product of vacuum pressure P and emission current I, just as it does for conventional cold field emitters (CFEs), the mechanism of fluctuation was assumed to be similar to that of CFE. Such fluctuations are divided into two regions, the boundary was virtually equal to CFEs, at approximately 3×10−12 Pa A. However, unlike as for CFEs, the emission current of Si-FEs drops abruptly after a certain duration. We found that the integral of PI with the duration time (PIT integral) is almost constant (approximately 1×10−8 Pa A s) in various conditions and therefore assume that it correlates with the emission area of the emitter tip. The current leakage between emitter and gate electrode was found to be caused because of their small size. Responsible for this leakage are hydrocarbon contamination layers that were generated by emitted electrons or backscattered electrons on the surface of the insulating layer. Finally, we fabricated a miniaturized electron gun with Si-FEs and succeeded in obtaining SEM images with electrons that were emitted from the Si-FEs. © 1998 American Vacuum Society.
Show PACS
85.45.Db Field emitters and arrays, cold electron emitters
79.70.+q Field emission, ionization, evaporation, and desorption
07.78.+s Electron, positron, and ion microscopes; electron diffractometers
81.65.Cf Surface cleaning, etching, patterning
81.65.Mq Oxidation
68.35.Dv Composition, segregation; defects and impurities

Self-heating effects in a InP/CdS/LaS cold cathode

Aashish Malhotra, Yamini Modukuru, and Marc Cahay

J. Vac. Sci. Technol. B 16, 3086 (1998); http://dx.doi.org/10.1116/1.590446 (11 pages) | Cited 7 times

Full Text: | Download PDF

Show Abstract
We analyze the importance of self-heating effects in a new cold cathode emitter which consists of a thin region of cadmium sulfide (CdS) sandwiched between a heavily doped indium phosphide (InP) substrate and a low work function lanthanum sulfide (LaS) semimetallic thin film. We identify the mechanisms leading to power dissipation in a cathode with an emission window of rectangular geometry, including the effects of inelastic scattering in the CdS and LaS layers. The latter are modeled using a temperature dependent mean-free path approach. The analysis includes the effects of current crowding in the emission window due to the finite resistivity of the LaS layer. We determine the cathode parameters which minimize current crowding and self-heating effects in the cathode. We also calculate the relative contribution of the ballistic and inelastic portions of the electron current emitted into vacuum. © 1998 American Vacuum Society.
Show PACS
85.45.Db Field emitters and arrays, cold electron emitters
79.70.+q Field emission, ionization, evaporation, and desorption
73.40.Gk Tunneling

Experimental demonstration of the validity of accelerated radiation damage testing of x-ray mask materials

R. E. Acosta and R. Rippstein

J. Vac. Sci. Technol. B 16, 3097 (1998); http://dx.doi.org/10.1116/1.590447 (2 pages) | Cited 2 times

Full Text: | Download PDF

Abstract Unavailable
Show PACS
85.40.Hp Lithography, masks and pattern transfer
61.80.Cb X-ray effects
61.82.Fk Semiconductors
06.60.Mr Testing and inspecting procedures

Sputtering rate change and surface roughening during oblique and normal incidence O2+ bombardment of silicon, with and without oxygen flooding

Charles W. Magee, Gary R. Mount, Stephen P. Smith, Brad Herner, and Hans-J. Gossmann

J. Vac. Sci. Technol. B 16, 3099 (1998); http://dx.doi.org/10.1116/1.590370 (6 pages) | Cited 20 times

Full Text: | Download PDF

Show Abstract
A sample of low-temperature epitaxial Si grown with five B delta-doped layers 5.4 nm apart has been profiled using secondary ion mass spectrometry under a variety of O2 bombardment conditions. Energies from 400 eV to 1.5 keV were used with angles of incidence from 0° to 70°. Analyses were performed using oxygen flooding of the sample surface during analysis, as well as without using oxygen flooding. The apparent spacing between the B delta layers was used to determine the magnitude and extent of increased sputtering rate at the beginning of an analysis. Changes in depth resolution due to sputter-induced surface roughening are reflected in variations in the apparent width of the B delta layers. It was found that sputtering with 500 eV O2 at an angle of 50° while flooding with oxygen produced no measurable change in sputtering rate and resulted in no unexpected shift towards the surface of the B delta layers. These analysis conditions also resulted in a depth resolution which was as good as that obtained using 400 eV O2 bombardment at 0° incidence without oxygen flooding. The 0° method of analysis, however, resulted in a 1.1 nm shift of the topmost B delta layer toward the surface, and the 0° method had a sputtering rate only 1/5 that of the 50° method of analysis. © 1998 American Vacuum Society.  
Show PACS
79.20.Rf Atomic, molecular, and ion beam impact and interactions with surfaces
61.72.uf Ge and Si
61.72.S- Impurities in crystals
81.70.Jb Chemical composition analysis, chemical depth and dopant profiling
68.55.Nq Composition and phase identification

Ohmic contacts to p-type GaN using a Ni/Pt/Au metallization scheme

Ja-Soon Jang, Kyung-Hyun Park, Hong-Kyu Jang, Hyo-Gun Kim, and Seong-Ju Park

J. Vac. Sci. Technol. B 16, 3105 (1998); http://dx.doi.org/10.1116/1.590448 (3 pages) | Cited 17 times

Full Text: | Download PDF

Show Abstract
We report a new Ni/Pt/Au trilayer metallization scheme for the formation of ohmic contact to p-GaN. Metal thin films with a thickness of 20 nm for Ni, 30 nm for Pt, and 80 nm for Au were deposited on the p-GaN layer (Na=9.4×1016 cm−3) by electron beam evaporation. The samples, annealed at 500 °C for 30 s in a rapid thermal anneal system, showed a high quality ohmic contact with a low specific contact resistance of 2.1×10−2 Ω cm2. Auger electron spectroscopy analysis of the contact layers suggests that Pt plays an important role in the formation of ohmic contact, indicating that a Ni/Pt/Au trilayer can be used and that it is a promising material system for ohmic contact to p-GaN. © 1998 American Vacuum Society.
Show PACS
85.40.Ls Metallization, contacts, interconnects; device isolation
07.50.-e Electrical and electronic instruments and components
73.40.Cg Contact resistance, contact potential
73.40.Ns Metal-nonmetal contacts

Thermally induced interface degradation in (100) and (111) Si/SiO2 analyzed by electron spin resonance

A. Stesmans and V. V. Afanas’ev

J. Vac. Sci. Technol. B 16, 3108 (1998); http://dx.doi.org/10.1116/1.590449 (4 pages) | Cited 23 times

Full Text: | Download PDF

Show Abstract
A comparative electron spin resonance (ESR) study was performed on thermal (111) Si/SiO2 and (100) Si/SiO2 of the vacuum postoxidation-induced interface degradation in terms of interfacial trivalent Si dangling bond creation (ESR-active Pb, Pb0, and Pb1 defects). In (111) Si/SiO2, the degradation mechanism was isolated as pronounced permanent Pb (∘Si�Si3) creation from ∼640 °C onward in densities Nc monotonically increasing with anneal temperature; at ∼1100 °C, about 1.1×1013 Pbs cm−2 are created in addition to the as-oxidized state value No∼4.9×1012 cm−2. The (100) Si/SiO2 interface is found to be much less vulnerable. Only electrically harmless Pb1’s are additionally created, the density reaching Nc(Pb1)∼4.4×1012 cm−2 at ∼1100 °C. By contrast, the density of the electrically adverse Pb0 trap tends to decrease. Together with the recently established electrical irrelevance of Pb1, the results add to provide a fundamental reason for the preference of the (100) Si face in devicing. © 1998 American Vacuum Society.
Show PACS
73.40.Qv Metal-insulator-semiconductor structures (including semiconductor-to-insulator)
73.20.Hb Impurity and defect levels; energy states of adsorbed species
76.30.Mi Color centers and other defects
61.72.Cc Kinetics of defect formation and annealing

Enabling in situ atomic-scale characterization of epitaxial surfaces and interfaces

J. B. Smathers, D. W. Bullock, Z. Ding, G. J. Salamo, P. M. Thibado, B. Gerace, and W. Wirth

J. Vac. Sci. Technol. B 16, 3112 (1998); http://dx.doi.org/10.1116/1.590496 (3 pages) | Cited 18 times

Full Text: | Download PDF

Show Abstract
A custom designed sample handling system which allows the integration of a commercially available scanning tunneling microscope (STM) facility with a commercially available molecular beam epitaxy (MBE) facility is described. No customization of either the STM imaging stage or the MBE is required to implement this design.© 1998 American Vacuum Society.
Show PACS
68.55.-a Thin film structure and morphology
06.60.Ei Sample preparation (including design of sample holders)
68.37.Ef Scanning tunneling microscopy (including chemistry induced with STM)
68.37.Ps Atomic force microscopy (AFM)
68.37.Rt Magnetic force microscopy (MFM)
68.37.Uv Near-field scanning microscopy and spectroscopy
81.15.Hi Molecular, atomic, ion, and chemical beam epitaxy
68.35.B- Structure of clean surfaces (and surface reconstruction)
68.35.Ct Interface structure and roughness

Atom technology project: Recent activities

Kazunobu Tanaka

J. Vac. Sci. Technol. B 16, 3127 (1998); http://dx.doi.org/10.1116/1.590450 (5 pages)

Full Text: | Download PDF

Show Abstract
The “Atom Technology” project started in fiscal 1992 as one of MITI’s 10 year projects, aims at systematically establishing technology for handling individual atoms and molecules on a solid surface or in a three-dimensional space, as a generic technology for various fields of industry. This project, closely adjacent to science, emphasizes the following three key focuses: atom manipulation, nanoscale self-organization, and critical-state phase control, with two basic approaches of in situ dynamical observation (experimental) as well as ab initio calculation (theoretical). In this article, several topics were picked up from recent activities at the Joint Research Center for Atom Technology (JRCAT) for Phase I of the initial 6 years (1992–1997) and some technical details were described: (1) ultrathin SiO2 on Si(001) surfaces; layer-by-layer oxidation, its kinetics, scanning reflection electron microscopy observation, and scanning tunneling microscopy observation of leakage sites; (2) growth and transport of structure-controlled SinHx+ clusters for deposition using a novel ion trap; and (3) colossal magnetoresistance and related phenomena in perovskite-type manganese oxides. Research plans for phase II (1998–2001) of the project will be also touched upon. © 1998 American Vacuum Society.
Show PACS
68.00.00 Surfaces and interfaces; thin films and nanosystems (structure and nonelectronic properties)
68.37.Ef Scanning tunneling microscopy (including chemistry induced with STM)
68.37.Ps Atomic force microscopy (AFM)
68.37.Rt Magnetic force microscopy (MFM)
68.37.Uv Near-field scanning microscopy and spectroscopy
81.65.Mq Oxidation
36.40.-c Atomic and molecular clusters
68.37.Hk Scanning electron microscopy (SEM) (including EBIC)
68.37.Lp Transmission electron microscopy (TEM)
72.15.Gd Galvanomagnetic and other magnetotransport effects
72.20.My Galvanomagnetic and other magnetotransport effects

Nanofabrication in cellular engineering

C. D. W. Wilkinson, A. S. G. Curtis, and J. Crossan

J. Vac. Sci. Technol. B 16, 3132 (1998); http://dx.doi.org/10.1116/1.590451 (5 pages) | Cited 32 times

Full Text: | Download PDF

Show Abstract
Biological cells are strongly influenced by the topography of the surface on which they live, both in cell culture and in an animal. They are guided along micron sized grooves and change their shape becoming more elongated. On the other hand, cells do not adhere to surfaces that are covered in small nanometrically sized pillars. These effects can be used for cellular engineering purposes to determine the behavior of cells and in particular to make prostheses for medical purposes. The differences in micro and nanofabrication techniques necessary to adapt normal semiconductor technology for these purposes are discussed. Patterning of plastics using mechanical methods (embossing and molding) is shown to have excellent resolution, can be used on biodegradable material and a large enough area of patterned material can be produced at a reasonable cost. An application of this technology to the repair of broken tendons is discussed in some detail. It is shown that a biodegradable membrane patterned by embossing with a fused silica master can be used to effect tendon repair. Not only does the tendon reheal, but the synovial channel that should surround the tendon and provide lubrication to the tendon, is reformed correctly. © 1998 American Vacuum Society.
Show PACS
87.17.-d Cell processes
87.90.+y Other topics in biological and medical physics (restricted to new topics in section 87)
81.07.-b Nanoscale materials and structures: fabrication and characterization
81.16.-c Methods of micro- and nanofabrication and processing
85.35.-p Nanoelectronic devices

Challenges and progress in x-ray lithography

Jerome P. Silverman

J. Vac. Sci. Technol. B 16, 3137 (1998); http://dx.doi.org/10.1116/1.590452 (5 pages) | Cited 30 times

Full Text: | Download PDF

Show Abstract
X-ray lithography (XRL) is a very promising technique with the potential to be available for integrated circuit manufacturing as early as the 130 nm generation. As a result of many years of development, the technology is relatively mature. Synchrotron sources have demonstrated performance and reliability; preproduction aligners are available from multiple vendors; significant improvements are being made in mask fabrication; and high resolution imaging has been demonstrated at 100 nm and below. Established vendors already have experience with almost all of the tools that are needed, although improved performance is required for most in order to satisfy the error budgets at 130 nm and below. Significant development activities are continuing in both the United States and Japan, and numerous complex integrated circuits have been fabricated using XRL for one or more critical levels. Nonetheless, there are challenges still to be met. Among the most important are the development and commercial availability of an improved e-beam mask writer; the ability to fabricate defect-free masks satisfying the image placement and critical dimension control requirements with good yields; the stability of the masks in usage (including the issue of possible radiation damage); the ability to correct for magnification errors; and the ability to satisfy the industry’s desire for a technology extendible to 70 nm ground rules. These issues are primarily manufacturing issues, as opposed to issues related to demonstrating proof-of-concept or feasibility, although demonstrating extendibility is still needed before the industry can commit to using XRL at 70 nm ground rules. Because of the existing XRL facilities and experience, effective work to address these and other issues can be accomplished in a timely manner. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Extreme ultraviolet lithography

C. W. Gwyn, R. Stulen, D. Sweeney, and D. Attwood

J. Vac. Sci. Technol. B 16, 3142 (1998); http://dx.doi.org/10.1116/1.590453 (8 pages) | Cited 95 times

Full Text: | Download PDF

Show Abstract
An extreme ultraviolet (EUV) lithography tool using 13.4 nm radiation is being developed by a consortium of integrated circuit (IC) manufacturers to support 100 nm imaging for integrated circuit production. The 4×, 0.1 NA alpha tool has a >1 μm depth of focus, all reflective optics, a xenon laser plasma source, and robust reflective masks. The technology is expected to support feature scaling down to 30 nm. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Ion projection lithography: Status of the MEDEA project and United States/European cooperation

Gerhard Gross, Rainer Kaesmaier, Hans Löschner, and Gerhard Stengl

J. Vac. Sci. Technol. B 16, 3150 (1998); http://dx.doi.org/10.1116/1.590454 (4 pages) | Cited 5 times

Full Text: | Download PDF

Show Abstract
Structure and targets of the European MEDEA project on ion projection lithography as well as related U.S./European cooperation are explained. By assuming 10 μm virtual source size and 1 eV (full width half maximum) energy spread calculations for a multielectrode electrostatic ion–optical system (1.25 m between ion source and stencil mask, ≈1.8 m between mask and wafer) we realize the possibility of 100 nm resolution (line and space) over an exposure field of 22×22 mm2 even when using the MONTEC model for calculating the stochastic blur and when running 3.3 μA He+ ion beam current through the ion–optical column, thus more than twice exceeding target specifications. Thus, for 100 nm resolution and 50% pattern density the raw throughput is ≈12 cm2/s corresponding to >75 WPH (pattern within 80% of 300 mm wafer area). © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
41.85.-p Beam optics
89.20.Bb Industrial and technological research and development

Critical issues in 157 nm lithography

T. M. Bloomstein, M. Rothschild, R. R. Kunz, D. E. Hardy, R. B. Goodman, and S. T. Palmacci

J. Vac. Sci. Technol. B 16, 3154 (1998); http://dx.doi.org/10.1116/1.590455 (4 pages) | Cited 29 times

Full Text: | Download PDF

Show Abstract
Projection lithography at 157 nm is a candidate technology for the 100–70 nm generations, and possibly beyond. It would provide an evolutionary extension to the current primary photolithographic processes and components: excimer lasers, refractive optics, and transmissive masks. This article presents data on the transmission of optical materials at 157 nm, the performance of optical coatings, the issues that must be faced by photomasks, and the considerations related to engineering resists at this wavelength. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
42.70.-a Optical materials
42.79.Wc Optical coatings

Electron beam lithography process for advanced optical masks

Naoya Hayashi, Masa-aki Kurihara, Toshikazu Segawa, Toshifumi Yokoyama, and Tsukasa Abe

J. Vac. Sci. Technol. B 16, 3158 (1998); http://dx.doi.org/10.1116/1.590456 (6 pages)

Full Text: | Download PDF

Show Abstract
The requirements for advanced optical masks have been tightened according to the accelerated lithography roadmap for semiconductor manufacturing. The accuracy of the optical mask may be affected by the pattern exposure tools, the materials, and the processes. Higher acceleration voltage e-beam pattern exposure was evaluated with a chemically amplified resist. With the higher acceleration voltage e-beam system, proximity effect correction is needed to control the critical dimension (CD). CD linearity is maintained down to 400 nm for different kinds of patterns by adopting the 50 kV e-beam system and proximity effect correction. The results show that the method is an attractive candidate for fabrication of next generation optical masks. Dry etching is effective in achieving small feature sizes, such as optical proximity correction (OPC), with high pattern fidelity. Loading effects were evaluated, and inductively coupled plasma dry etching was found to be stable against those effects. Fabrication of the OPC pattern designed for a 0.15 μm rule device was demonstrated with these methods. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
81.65.Cf Surface cleaning, etching, patterning
52.77.Bn Etching and cleaning
52.77.Dq Plasma-based ion implantation and deposition

EL5: One tool for advanced x-ray and chrome on glass mask making

M. A. Sturans, J. G. Hartley, H. C. Pfeiffer, R. S. Dhaliwal, T. R. Groves, J. W. Pavick, R. J. Quickle, C. S. Clement, G. J. Dick, W. A. Enichen, M. S. Gordon, R. A. Kendall, C. A. Kostek, D. J. Pinckney, C. F. Robinson, et al.

J. Vac. Sci. Technol. B 16, 3164 (1998); http://dx.doi.org/10.1116/1.590457 (4 pages) | Cited 3 times

Full Text: | Download PDF

Show Abstract
The state-of-the-art for mask making continues to be driven by 1× x-ray masks. The IBM EL4+ e-beam mask writer at the Advanced Mask Facility in Burlington, Vermont, was originally designed for 0.35 μm ground rules (GRs) direct write at 50 kV, but delivered at 75 kV operation to achieve 0.25 μm GR performance for 1× mask making. Over the next 2 years, with optimization and improvements in each of the subsystems, its performance was enhanced beyond the 0.18 μm GR requirements. It is clear, however, that for 0.13 and 0.1 μm GR mask manufacturing, a new tool is required. It has also become apparent that because of the very high development and tool build costs, and small number of required x-ray mask makers, the same technology must be applicable for chrome on glass (COG) mask making. Based on the experience with EL4+, IBM is designing an EL5 tool which will provide the 0.13/0.1 μm GR performance for 1×, and easily convert to 4× COG exposure for 9 in. glass as well as 300 mm wafer direct write operation. As with previous IBM EL series e-beam systems, it is anticipated that EL5 performance will be extendable beyond 0.1 μm GR. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
85.40.-e Microelectronics: LSI, VLSI, ULSI; integrated circuit fabrication technology

Distributed, multiple variable shaped electron beam column for high throughput maskless lithography

T. R. Groves and R. A. Kendall

J. Vac. Sci. Technol. B 16, 3168 (1998); http://dx.doi.org/10.1116/1.590458 (6 pages) | Cited 20 times

Full Text: | Download PDF

Show Abstract
The ultimate resolution obtainable with focused electron beams is, for practical purposes in lithography, unlimited. Existing e-beam lithography systems are too slow to be practical for high volume manufacturing of semiconductor devices, however. The usable current in probe forming systems is limited by the stochastic Coulomb interaction in the beam path, which causes loss of resolution at high current. This is due to the need to pass all of the writing current through an aperture. Distributed systems, by contrast, do not suffer from this problem, as the current is spread over a large volume. The purpose of this article is to propose a distributed system, employing multiple, variable shaped beams for direct write (maskless) lithography. We call this system DiVa, to emphasize the key attributes of distributed writing current, and variable beam shaping. It utilizes a planar cathode, patterned with a rectilinear array of square emitters. Focusing is accomplished by a uniform, axial magnetic field, oriented along the optic axis. This transfers a one-to-one image of the emitters onto the writing surface. Deflection plates between adjacent rows of beamlets effect motion in one Cartesian axis, while mechanical translation of the stage effects motion in the orthogonal axis. Theoretical resolution is diffraction limited to 4.6 nm at 50 kV. Experimental results are presented which demonstrate the first order optics using a rudimentary apparatus. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Multiplexed blanker array for parallel electron beam lithography

G. I. Winograd, L. Han, M. A. McCord, R. F. W. Pease, and V. Krishnamurthi

J. Vac. Sci. Technol. B 16, 3174 (1998); http://dx.doi.org/10.1116/1.590345 (3 pages) | Cited 3 times

Full Text: | Download PDF

Show Abstract
Alternative electron beam technologies to conventional single beam systems have the potential to increase beam current by an order of magnitude, into the 1–10 μA range. The technique we are investigating is to insert an aperture array at an object plane in a lithography system, thus creating an array of beamlets which are imaged as a dot matrix. Each aperture is surrounded by an independently controlled electrostatic microblanker which can modulate the beamlet passing through the aperture in conjuction with a blanking aperture placed at a crossover further down the column. An existing 20 kV shaped beam column with a LaB6 source providing 5 μA of illumination current over a 150 μm×150 μm square area is used as a test stand. The column is used to test prototype microblanker chips. Power dissipation calculations show less than a 10 K temperature rise over the prototype chips. Work has begun on a second generation prototype blanker array which will overcome many of the processing difficulties encountered with the first prototype. Specifically, the new all-silicon process will allow fabrication of smaller apertures which require less demagnification and a shorter electron-optical column. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

High throughput electron lithography with the multiple aperture pixel by pixel enhancement of resolution concept

P. Kruit

J. Vac. Sci. Technol. B 16, 3177 (1998); http://dx.doi.org/10.1116/1.590346 (4 pages) | Cited 6 times

Full Text: | Download PDF

Show Abstract
A new lithography concept for the sub-100 nm generations is proposed. The bulk of the machine is a traditional deep ultraviolet demagnification scanner–stepper. The mask is illuminated by 106–108 almost diffraction limited subbeams formed by a microlens array. After demagnification these subbeams are focused on a photon–electron converter plate. Each photon subbeam triggers the emission of a narrow beam of electrons. The electron beams are focused individually on the wafer, which is at a distance of about 1 mm from the converter plate. By scanning both mask and wafer through the many beams, the whole wafer is exposed. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
42.79.Bh Lenses, prisms and mirrors
41.85.-p Beam optics

Experimental evaluation of a miniature electrostatic thin-foil electron optical column for high current and low-voltage operation

Dieter Winkler, Claus-D. Bubeck, Andreas Fleischmann, Guido Knell, Yvonne Lutsch, and Erich Plies

J. Vac. Sci. Technol. B 16, 3181 (1998); http://dx.doi.org/10.1116/1.590347 (4 pages) | Cited 4 times

Full Text: | Download PDF

Show Abstract
A miniaturized electrostatic electron optical column, consisting of thin commercial electron-microscope apertures as electrodes [Burstert, Winkler, and Lischke, Microelectron Eng. 31, 95 (1996)] has been successfully constructed and evaluated. An electron probe with a current of 300 nA and a spot size of 0.6 μm is achieved at 3 keV. The system is operated with beam energies as low as 300 eV, yielding 600 nA in a 2 μm spot size. Long term stability and reproducibility are achieved over operation times of several months. This demonstrates that the column is suitable for testing purposes where a high signal-to-noise ratio is required. © 1998 American Vacuum Society.
Show PACS
07.78.+s Electron, positron, and ion microscopes; electron diffractometers
41.85.Ne Electrostatic lenses, septa
41.75.Fr Electron and positron beams

Micromachined ultrasharp silicon and diamond-coated silicon tip as a stable field-emission electron source and a scanning probe microscopy sensor with atomic sharpness

I. W. Rangelow, F. Shi, P. Hudek, P. Grabiec, B. Volland, E. I. Givargizov, A. N. Stepanova, L. N. Obolenskaya, E. S. Mashkova, and V. A. Molchanov

J. Vac. Sci. Technol. B 16, 3185 (1998); http://dx.doi.org/10.1116/1.590348 (7 pages) | Cited 7 times

Full Text: | Download PDF

Show Abstract
A novel combination of silicon micromachining and deposition on silicon whiskers on (111) oriented silicon substrates shows promise for stable field emission electron emitters and sensors for scanning probe microscopy. As a field emission electron emitters show promise for high-density and high-resolution electron beam applications owing to a small (within 3°) divergence half angle of emitted electrons. Due to the sharpening of the diamond coated Si tips the threshold voltage, which is necessary to create a stable emission current (nanoampers) is decreased from the initial 1300 V to about 150 V. The diamond coating stabilizes the field-emission current as large as about 100 μA from a single tip. Training of the emitter at a current of about 10−7 A for several hours results in an additional improvement of the current stability with an additional slight current increase. Second application of vapor–liquid–solid grown silicon whiskers described here is the recent progress toward high aspect ratio (>100) probes and relates a technique for preparation of cantilevers for atomic force microscopy with atomic resolution on the very end of the tip. The radii are less than 2 nm for silicon tips and the angles at the ends are less than 20°. In the case of diamond, the tip radius is 10 nm. © 1998 American Vacuum Society.
Show PACS
85.45.Db Field emitters and arrays, cold electron emitters
07.79.Lh Atomic force microscopes
07.10.Cm Micromechanical devices and systems
07.07.Df Sensors (chemical, optical, electrical, movement, gas, etc.); remote sensing
79.70.+q Field emission, ionization, evaporation, and desorption
68.37.Ef Scanning tunneling microscopy (including chemistry induced with STM)
68.37.Ps Atomic force microscopy (AFM)
68.37.Rt Magnetic force microscopy (MFM)
68.37.Uv Near-field scanning microscopy and spectroscopy
68.70.+w Whiskers and dendrites (growth, structure, and nonelectronic properties)

Patterned negative electron affinity photocathodes for maskless electron beam lithography

J. E. Schneider, P. Sen, D. S. Pickard, G. I. Winograd, M. A. McCord, R. F. W. Pease, W. E. Spicer, A. W. Baum, K. A. Costello, and G. A. Davis

J. Vac. Sci. Technol. B 16, 3192 (1998); http://dx.doi.org/10.1116/1.590349 (5 pages) | Cited 8 times

Full Text: | Download PDF

Show Abstract
This work focuses on two issues crucial to achieving high throughput with a negative electron affinity semiconductor photocathode source. Monte Carlo simulations indicate that for a 50 kV system, as much as 8 μA of current may be delivered to the wafer to achieve a raw throughput of 20 8 in. wafers per hour with 0.1 μm minimum feature size (assuming a resist sensitivity of 10 μC/cm2). In order to achieve the throughput potential of this approach, suboptical emission areas are required; this suggests the use of cathode patterning. Two patterning alternatives have been investigated experimentally, and both approaches have been used to generate arrays of more than 100 electron beams with source sizes as small as 150 nm. However, each type of patterned cathode presents unique challenges to fabrication and performance in a practical multibeam system. Different source configurations (number of beams, beam current, beam spacing, etc.) create a system-level tradeoff between resolution and throughput. Results from patterned cathode experiments and system modeling are presented. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
85.60.Ha Photomultipliers; phototubes and photocathodes
41.85.Ja Particle beam transport
02.70.Rr General statistical methods
41.85.Ew Particle beam profile, beam intensity

Critical dimension control at stitched subfield boundaries in a high-throughput SCALPEL® system

Stuart T. Stanton, J. Alexander Liddle, Warren K. Waskiewicz, and Anthony E. Novembre

J. Vac. Sci. Technol. B 16, 3197 (1998); http://dx.doi.org/10.1116/1.590350 (5 pages) | Cited 1 time

Full Text: | Download PDF

Show Abstract
Scattering with angular limitation projection electron-beam lithography (SCALPEL®) is a stitching lithography system, using a segmented mask pattern which is assembled on the wafer by dynamic imaging. Critical objects may lie on the original pattern segment boundaries (seams), but their stitched images must adhere to conservatively interpreted +/−10% critical dimension (CD) control requirements. Stitching with butted (touching) spatial image parts requires an aggressive placement-error allocation of approximately +/−CD/20, along with allowed values for dose control, mask, resist, and process in the CD control budget. Further analysis has shown that a seam-blending scheme increases the stitching-error tolerance by as much as five times, and allows redistribution of all values in the CD error budget. The seam-blending approach provides overlapped image portions in a small area common to neighboring segments which receive complementary illumination dose portions. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
41.85.Ja Particle beam transport

Pattern dependent alignment technique for mix-and-match electron-beam lithography with optical lithography

Yasuko Gotoh, Yasunari Sohda, Norio Saitou, Tsutomu Tawa, Takashi Matsuzaka, Naoko Asai, Katsuya Hayano, and Norio Hasegawa

J. Vac. Sci. Technol. B 16, 3202 (1998); http://dx.doi.org/10.1116/1.590351 (4 pages)

Full Text: | Download PDF

Show Abstract
An alignment technique for electron-beam (EB) lithography that corrects the writing pattern to match the pattern-dependent lens distortion of an optical stepper is proposed. In mix-and-match EB and optical lithography, precise measurement of positional distortion caused by the stepper lens is necessary. However, recent research in optical lithography has shown that the lens distortion differs depending on the pattern features. Thus, we have enhanced the measurement method to reflect the pattern-feature dependence of positional distortion in optical lithography. To measure the various types of distortions more effectively, we modified the marks used for distortion and overlay measurement. The measured pattern dependence of the lens distortion was in good agreement with our simulation results and the overlay accuracy in mix-and-match lithography was improved with this method. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Electron-optical optimization for Gaussian, high-current, high-dose columns

Marian Mankos, Lee H. Veneklasen, Rudy Garcia, and Henry Pearce-Percy

J. Vac. Sci. Technol. B 16, 3206 (1998); http://dx.doi.org/10.1116/1.590352 (5 pages)

Full Text: | Download PDF

Show Abstract
This article demonstrates the electron-optical optimization of a high-current, high-dose column operating at 10 keV. The goal is to increase the available dose to the resist, which requires increasing the current density to more than 800 A/cm2. Our calculations use the MEBS Ltd. BOERSCH program. We model the complete column as a set of thin lenses, separated by field-free drift spaces. Monte Carlo simulation propagates discrete bunches of electrons through the column, taking into account the mutual repulsion between pairs of electrons. Gaussian spot size and current density in the column were calculated for three beam currents: 314, 75, and 35 nA. The results show that to achieve a higher current density, it is necessary to change the electron gun and the column. Specifically, a low aberration gun and a significantly shorter column are required. At 10 keV, the column performance at high beam currents (∼300 nA) is almost entirely dominated by electron–electron interactions. Major improvements in gun performance only yield minor improvement in total column performance, and a significant decrease in column length is required to reduce the electron–electron interactions enough to meet the design specification. A new, low aberration gun operating in an accelerating lens mode gun was chosen. A major effort was focused on redesigning the middle column to reduce electron–electron interactions. The combination of the low-aberration electrostatic gun and a shorter middle column provides the key improvement in the current density (dose) performance of the column. Results of experimental measurements on the optimized columns are in good agreement with the presented Monte Carlo simulations. © 1998 American Vacuum Society.
Show PACS
02.70.Rr General statistical methods
85.40.Hp Lithography, masks and pattern transfer
41.85.Ew Particle beam profile, beam intensity

Simulation of Coulomb interactions in electron beam lithography systems—A comparison of theoretical models

W. Stickel

J. Vac. Sci. Technol. B 16, 3211 (1998); http://dx.doi.org/10.1116/1.590353 (4 pages) | Cited 11 times

Full Text: | Download PDF

Show Abstract
The Loeffler effect (trajectory displacement) has been calculated for a magnetic lens doublet electron-optic system considered to represent the basic configuration for next generation lithography. The results of two analytic and two numeric (Monte Carlo) models are compared and presented graphically as well as by analytic approximations, revealing significant quantitative differences within the parameter space considered relevant. Less severe discrepancies in the trends appear to be useful as a guide for system design, but experimental verification is needed to determine the validity of any of the approaches compared. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer
02.70.Rr General statistical methods
41.85.Lc Particle beam focusing and bending magnets, wiggler magnets, and quadrupoles
41.75.Fr Electron and positron beams

Performance investigation of Coulomb interaction-limited high throughput electron beam lithography based on empirical modeling

Liqun Han, Mark A. McCord, Gil I. Winograd, and R. F. W. Pease

J. Vac. Sci. Technol. B 16, 3215 (1998); http://dx.doi.org/10.1116/1.590354 (6 pages) | Cited 6 times

Full Text: | Download PDF

Show Abstract
Multibeam direct writing and projection strategies have been proposed for high throughput electron beam lithography; however, the large beam current required may cause severe beam blur due to the stochastic Coulomb interaction effects between the electrons. From viewpoint of both concept proof and practical system design, evaluation of the fundamental system limits and the optimum performance in a large range of multiple system parameter variable space is critical for such high throughput tools. Based on the response surface design approach, well-fitted empirical models of image beam size as a function of various system parameters have been extracted from Monte Carlo simulations of electron interactions for different proposed multi-beam systems. The blurred imaging beam size is examined in a wide range of system parameters: beam currents from 1 to 100 μA, acceleration voltages from 25 to 100 kV, column lengths from 1 to 100 cm, field sizes from 0.5 to 2 mm, and demagnification from 4 to 16. Two general tendencies for such systems were observed: the nonuniformity of image defocusing due to the inhomogeneous interbeamlet space charge effect at large field sizes, and the high sensitivity of maximum beam current to column length at a fixed image beam size. Finally, the throughput versus acceleration voltage is evaluated under different assumptions of resolution, resist thickness, and column length. It is concluded that to achieve the throughput required by manufacturing beyond 0.1 μm resolution, very aggressive design of a short column along with thin resist will be required. © 1998 American Vacuum Society.  
Show PACS
85.40.Hp Lithography, masks and pattern transfer
02.70.Rr General statistical methods

Rigorous simulation of statistical electron–electron interactions with fast multipole acceleration and a network of workstations

Victor S. H. Wen, Owen T. Carmichael, Hiroshi Yamashita, and Andrew R. Neureuther

J. Vac. Sci. Technol. B 16, 3221 (1998); http://dx.doi.org/10.1116/1.590355 (6 pages)

Full Text: | Download PDF

Show Abstract
Rigorous simulation, which in concept is nearly linear in CPU time with microampere beam currents, has been developed for analyzing statistical electron–electron interactions in lens free regions joined by ideal lenses and apertures. The simulation method is based on combining fast multipole acceleration (FMA) with modifications to a message passing interface for use on a multiprocessor and on a network of workstations. As test cases, the distribution of scattered electrons versus deviation from focus is given as well as results for the interaction between simultaneously imaged spots. The performance improvement per level of spatial division was only about a factor of 2 instead of the expected factor of 8 due to the pencil shape nature of the beam. A FMA method that is adaptable to beam density and geometry is needed. It is projected that it will be possible to simulate a 30 μA beam current on a quad-Pentium Pro machine in half a day and in matter of hour on a network of workstations. © 1998 American Vacuum Society.
Show PACS
41.75.Fr Electron and positron beams
85.40.Hp Lithography, masks and pattern transfer
07.05.Tp Computer modeling and simulation

Mask electron modeling for Coulomb interaction by mask-scattered electrons in electron-beam projection lithography

Hiroshi Yamashita, Victor S. H. Wen, Andrew R. Neureuther, and Eiichi Nomura

J. Vac. Sci. Technol. B 16, 3227 (1998); http://dx.doi.org/10.1116/1.590356 (5 pages) | Cited 3 times

Full Text: | Download PDF

Show Abstract
A rigorous statistical Coulomb interaction simulator that can take into account mask electrons was developed. In electron-beam (EB) projection lithography, many electrons incident to an absorber or a scatterer pass through the EB mask. These mask electrons, however, have not been considered in a simulation of Coulomb interaction. In a novel simulation approach, an assessment has been made of the effects of mask electrons on image quality and flare due to electron–electron interactions. The statistical electron–electron interactions from mask electrons under various conditions are simulated by injecting them into a simulation domain and using the fast multipole accelerated method to rigorously follow all interactions as a function of time. It was found that mask electrons contribute to the trajectory displacement effect and increase the number of deviated electrons in the pattern area by four to five times when the accelerating voltage is 125 kV, the beam current density is 0.1 A/cm2, the mask thickness is 10 μm, and the electron transmittance is 95%. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Optical data storage in LiF using electron beam encoding

E. J. Caine and S. D. Miller

J. Vac. Sci. Technol. B 16, 3232 (1998); http://dx.doi.org/10.1116/1.590357 (5 pages) | Cited 10 times

Full Text: | Download PDF

Show Abstract
Electron beam generated color centers in LiF are presented as the basis for a new optical data storage scheme. Data are “read” with a confocal microscope. The alkali halide medium is capable of gray scale and multilevel or volumetric storage demonstrating a thermally stable coloration. This initial investigation shows that data can be resolved to less than 0.25 μ resolution with at least three orders of magnitude data scaling. An assessment is made for future work. © 1998 American Vacuum Society.
Show PACS
42.70.Ln Holographic recording materials; optical storage media
61.80.Fe Electron and positron radiation effects
61.72.J- Point defects and defect clusters
61.82.Ms Insulators
42.79.Vb Optical storage systems, optical disks

Fabrication of electron beam generated, chirped, phase mask (1070.11–1070.66 nm) for fiber Bragg grating dispersion compensator

R. C. Tiberio, D. W. Carr, M. J. Rooks, S. J. Mihailov, F. Bilodeau, J. Albert, D. Stryckman, D. C. Johnson, K. O. Hill, A. W. McClelland, and B. J. Hughes

J. Vac. Sci. Technol. B 16, 3237 (1998); http://dx.doi.org/10.1116/1.590358 (4 pages) | Cited 9 times

Full Text: | Download PDF

Show Abstract
We report on the fabrication of a chirped, phase mask that was used to create a fiber Bragg grating (FBG) device for the compensation of chromatic dispersion in longhaul optical transmission networks. Electron beam lithography was used to expose the grating onto a resist-coated quartz plate. After etching, this phase mask was used to holographically expose an index grating into the fiber core [K. O. Hill, F. Bilodeau, D. C. Johnson, and J. Albert, Appl. Phys. Lett. 62, 1035 (1993)]. The linear increase in the grating period, “chirp,” is only 0.55 nm over the 10 cm grating. This is too small to be defined by computer aided design and a digital deflection system. Instead, the chirp was incorporated by repeatedly rescaling the analog electronics used for field size calibration. Special attention must be paid to minimize any field stitching and exposure artifacts. This was done by using overlapping fields in a “voting” method. As a result, each grating line is exposed by the accumulation of three overlapping exposures at 1/3 dose. This translates any abrupt stitching error into a small but uniform change in the line-to-space ratio of the grating. The phase mask was used with the double-exposure photoprinting technique [K. O. Hill, F. Bilodeau, B. Malo, T. Kitagawa, S. Thériault, D. C. Johnson, J. Albert, and K. Takiguchi, Opt. Lett. 19, 1314 (1994)]: a KrF excimer laser holographically imprints an apodized chirped Bragg grating in a hydrogen loaded SMF-28 optical fiber. Our experiments have demonstrated a spectral delay of −1311 ps/nm with a linearity of +/−10 ps over the 3 dB bandwidth of the resonant wavelength of the FBG. The reflectance, centered on 1550 nm, shows a side-lobe suppression of −25 dB. Fabrication processes and optical characterization will be discussed. © 1998 American Vacuum Society.
Show PACS
42.81.Dp Propagation, scattering, and losses; solitons
42.79.Dj Gratings
42.82.Cr Fabrication techniques; lithography, pattern transfer

Experimental verification of the TEMPTATION (temperature simulation) software tool

Sergey Babin and Igor Yu. Kuzmin

J. Vac. Sci. Technol. B 16, 3241 (1998); http://dx.doi.org/10.1116/1.590359 (7 pages) | Cited 7 times

Full Text: | Download PDF

Show Abstract
Predicting the temperature rise during electron-beam lithography (EBL) is an important practical problem in the development of new EBL systems and in the optimization of new exposure processes. The TEMPTATION software tool was developed. This software uses an analytic solution of heat propagation in a multilayered substrate. The software attains both high speed and good accuracy. Experimental verification of the software was done. Simulated data were verified versus experiments made with variably shaped EBL systems using three different resists. Change of effective absorbed energy due to heating was simulated and compared to published data. Good agreement of predicted and measured data for both long-range and short-range resist heating was shown. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Lie algebraic aberration theory and calculation method for combined electron beam focusing-deflection systems

Kangyan Hu and T.-T. Tang

J. Vac. Sci. Technol. B 16, 3248 (1998); http://dx.doi.org/10.1116/1.590371 (8 pages)

Full Text: | Download PDF

Show Abstract
Lie algebraic deflection aberration theory is presented and applied to combined magnetic focusing-deflection systems. Employing this new algebraic tool, all third- and fifth-order canonical aberrations (including canonical position and momentum aberrations at arbitrary observation planes) for combined systems are derived. The results are complete with concise forms. An example of an in-lens deflection system is calculated for high-order geometric aberrations through fifth order. This article lays the groundwork for further studies on the computation and optimized design for higher-order aberrations of scanning probe systems. © 1998 American Vacuum Society.
Show PACS
41.85.Gy Chromatic and geometrical aberrations
41.85.Lc Particle beam focusing and bending magnets, wiggler magnets, and quadrupoles
85.40.Hp Lithography, masks and pattern transfer
02.10.Ud Linear algebra
02.10.Xm Multilinear algebra
07.05.-t Computers in experimental physics

Characterizing GHOST proximity effect correction effectiveness by determining the worst-case error

G. Patrick Watson, Steven D. Berger, and J. Alexander Liddle

J. Vac. Sci. Technol. B 16, 3256 (1998); http://dx.doi.org/10.1116/1.590360 (6 pages) | Cited 1 time

Full Text: | Download PDF

Show Abstract
Background proximity effect corrections such as GHOST and the proposed SCALPEL correction technique have been analyzed to determine how well they can perform in situations where the correction dose distribution is not ideal. A technique has been developed that evaluates the largest possible correction dose error by determining the worst-case exposure pattern. This pattern is found to be unlikely to occur in integrated circuit patterns, so that the dose errors found in realistic situations are expected to be much smaller. For Gaussian shaped GHOST correction energy distributions, the worst-case pattern resembles a disk or a bull’s eye. The optimal correction dose distribution has a width parameter of about 90% of the backscatter width. The worst-case background dose error is less than +/−2% for a backscatter coefficient of 0.40. For a correction dose that resembles a disk of constant intensity, the worst-case error is typically larger than the case of a Gaussian shape. The smallest worst-case error occurs with a correction disk radius of 1.2 times the backscatter range for a backscatter coefficient of 0.50. The smallest worst-case error is +/−10% or one half of that, +/−5%, for practical pattern density ranges. The SCALPEL correction dose profile is ring shaped and the worst-case background error can be nearly as large as the backscatter coefficient. The worst-case error concept is used to search for the optimum ring shape; a worst-case error as small as +/−3% is possible with the appropriate correction ring characteristics. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Accuracy and efficiency in electron beam proximity effect correction

S. J. Wind, P. D. Gerber, and H. Rothuizen

J. Vac. Sci. Technol. B 16, 3262 (1998); http://dx.doi.org/10.1116/1.590361 (7 pages) | Cited 9 times

Full Text: | Download PDF

Show Abstract
We present a suite of electron beam proximity effect correction algorithms which incorporate elements of various shape-to-shape techniques as well as field-based techniques. They are applied to a set of prototypical pattern data at simulated electron beam exposure conditions ranging from 25 to 100 keV on bulk substrates. The results are compared in terms of correction accuracy and computational efficiency. In addition, we show how these algorithms, when used in combination with one another, can be applied to hierarchical pattern design data such that the correction can be obtained with a high degree of accuracy without requiring unnesting of the hierarchy. © 1998 American Vacuum Society.
Show PACS
85.40.Hp Lithography, masks and pattern transfer

Adaptive selection of control points for improving accuracy and speed of proximity effect correction

Soo-Young Lee and Jayesh Laddha

J. Vac. Sci. Technol. B 16, 3269 (1998); http://dx.doi.org/10.1116/1.590362 (6 pages) | Cited 1 time

Full Text: | Download PDF