Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Harold F. Winters.
Journal of Applied Physics | 1979
J. W. Coburn; Harold F. Winters
The extent to which gas‐surface chemical reactions can be enhanced by energetic radiation (primarily ions and electrons) incident on the surface is described. Emphasis is placed on chemical systems which lead to volatile reaction products. In particular, the reactions of Si, SiO2, and Si3N4 with XeF2, F2, and Cl2 are examined experimentally. Possible mechanisms for the radiation‐induced enhancement are discussed and some technological implications of this process in plasma etching technology and lithography are considered.
Journal of Vacuum Science and Technology | 1979
J. W. Coburn; Harold F. Winters
The purpose of the present paper is to review the salient features of our understanding of phenomena which occur in plasma etching situations. The etching process is discussed in terms of three basic steps: adsorption, product formation, and product desorption. Experiments performed in well‐defined (nonplasma) environments are discussed with the goal of clarifying the relative importance of these three steps in the etching process. An attempt is made to relate the resulting concepts to several phenomena generally observed in plasma situations (e.g. etching anisotropy, selective etching, the loading effect, and the role of additive gases). Moreover, the glow discharge, in addition to generating active species which initiate the chemical reactions, also causes the etched surface to be subjected to energetic particle (ions, electrons) bombardment. The role of this radiation in the etching process is emphasized. Speculative comments relating to plasma etching parameters and apparatus are also given.
Applied Physics Letters | 1979
Harold F. Winters; J. W. Coburn
It is shown that silicon is isotropically etched by exposure to XeF2(gas) at T=300 K. Si etch rates as large as 7000 A/min were observed for P (XeF2) <1.4×10−2 Torr and the etch rate varies linearly with P (XeF2). There was no observable etching of SiO2, Si3N4, or SiC, demonstrating an extremely large selectivity between silicon and its compounds. Therefore, thin masks constructed from silicon compounds can be used for pattern delineation. The implication of these experimental results for understanding mechanisms associated with plasma etching (including RIE) will be discussed.
Surface Science Reports | 1992
Harold F. Winters; J. W. Coburn
Abstract Basic studies of the surface science aspects of plasma-assisted etching were initiated over 10 years ago in laboratories throughout the world. Several approaches to this experimentally challenging problem have been taken: (1) simulate the reactive gas glow discharge environment with directed beams of energetic positive ions and thermal energy reactive molecules/radicals in a UHV environment; (2) simulate the reactive gas glow discharge by using a beam of reactive ions (reactive ion beam etching); and (3) carry out careful ex-situ analyses of surfaces etched in reactive gas glow discharges without air exposure. This report will be limited to a review of the work reported using the first of these approaches only and summarizes the status of this virtually unexplored field of surface chemistry. The report includes a discussion of the experimental aspects of this field and then focuses on the gas-solid system which has been studied most thoroughly; i.e., silicon-fluorine. A considerable quantity of new unpublished data is presented and a framework is proposed to explain the many observations associated with the spontaneous reaction of fluorine with silicon. The role of energetic ions in this reaction is then discussed in detail. Other materials combinations which are discussed are silicon-chlorine, silicon-bromine, silicon-hydrogen and SiO2-fluorine. The report concludes with a tabulation of the many other gas-solid systems which have been studied and also a discussion of the extent to which these basic studies relate to actual reactive gas processing environments.
Journal of Applied Physics | 1974
Harold F. Winters; Peter Sigmund
Flash filament techniques and mass spectrometry have been used to measure sputtering yields for nitrogen chemisorbed on tungsten and bombarded with noble‐gas ions in the energy range up to 500 eV. The experimental results show that primarily nitrogen atoms rather than molecules are sputtered. Despite a high binding energy (∼6.7 eV/atom), we find high sputtering yields and low threshold energies. The results are found to be in reasonable agreement with simple theoretical estimates. It is sugguested that, in the investigated energy range, adsorbed nitrogen atoms are sputtered primarily as a consequence of direct knock‐on collisions with impinging and/or reflected noble‐gas ions. Estimates are also given for the yield of nitrogen atoms knocked off by sputtered tungsten atoms. This latter process is expected to dominate at much higher energies.
Journal of Applied Physics | 1967
Harold F. Winters; Eric Kay
The concentration of argon in sputtered nickel films has been obtained as a function of the film‐growth temperature, the discharge pressure, and of the energy (bias voltage) with which the argon ions bombard the growing film. The concentrations vary from about 10−1 argon atoms/Ni atom to 10−4 argon atoms/Ni atom, depending upon the conditions during film growth. The incorporation of both argon and nitrogen into nickel films is interpreted on the basis of results previously obtained from sorption studies in a more‐idealized system on a pre‐existing nickel surface.
Journal of Chemical Physics | 1966
Harold F. Winters
The adsorption of energetic N2+ ions on nickel and molybdenum surfaces has been investigated. Comparison of the sticking probabilities of N2+ and Ar+ as a function of ion energy indicates that the binding mechanism for these two gases is different. The experimental evidence suggests that the mechanism causing adsorption of N2+ may be dissociation upon collision with the surface and the subsequent adsorption of the resulting atomic nitrogen.The total absolute dissociation cross section has also been measured in nitrogen for electron energies from 0 to 300 eV. The cross section has a maximum value of 2. 10−16 cm2 at an electron energy of about 90 eV.
Journal of Applied Physics | 1977
J. W. Coburn; Harold F. Winters; T. J. Chuang
The surface chemistry and the etching behavior of silicon and oxidized silicon bombarded with a CF3+ ion beam (50–4000 eV) have been studied using Auger electron spectroscopy, and a quartz‐crystal microbalance. The conclusions of this study are as follows: (a) the etch rate of Si caused by CF3+ ion bombardment can be accounted for by physical sputtering; (b) the deposition and removal of carbon at the etched surface may be one of the most important phenomena affecting the operation of plasma‐etching systems; and (c) there is reason to believe that ion bombardment of the etched surface enhances the reaction rate of the neutral etching species which are most probably fluorine atoms and CF3 radicals.
Applied Physics Letters | 1989
J. W. Coburn; Harold F. Winters
Very simple vacuum conductance arguments indicate that in the reactive ion etching of high aspect ratio features, the conductance is adequate to allow etch products to flow out of the feature without building up a pressure which would allow gas phase collisions to become important. On the other hand, the conductance can be expected to limit the flow of the reactive species to the bottom of the feature where the etching is taking place, thus creating the possibility of an etch rate dependence on the aspect ratio of the etched feature.
Journal of Applied Physics | 1978
Harold F. Winters
It is shown that the sticking probability (i.e., the probability for dissociative chemisorption plus the probability for nondissociative chemisorption where E≳20 kcal/mole) for CF4, CF3H, and CF3Cl on polycrystalline films of SiO2, Si3N4, and silicon is less than 10−7. The sticking probability for CF2Cl2 and CCl4 is detectable on some of these surfaces but still less than 10−6. Electron collisions with nondissociatively adsorbed molecules (E≳20 kcal/mole) are shown to result in the rapid buildup of adsorbed carbon and chlorine which will then remain on the surface indefinitely. It is further shown that CFx radicals have sticking probabilities between 0.08 and 0.75 on clean silicon. Based on these results, it is concluded that the dissociative chemisorption of stable gases is probably not important in plasma etching while radiation‐induced‐dissociative chemisorption and the chemisorption of CFx radicals are expected to play an important role.