Osami Okada

Reaction of atomic fluorine with silicon

The etch rate of Si with F atoms was measured by the use of F2 microwave plasma over a range of discharge pressures between 2.7×10−2 and 17 Pa. Fluorine atom concentration in the plasma was determined over the same pressure range by means of both gas‐phase titration and actinometry using Ar gas. A Si surface etched at 1.0×10−1, 5.3×10−1, 1.3, and 5.3 Pa was analyzed with XPS without exposing the surface to room air. A linear relation was obtained between the Si etch rate and the F atom concentration at discharge pressures between 2.7×10−2 and 2.7 Pa. The reaction probability of F atoms with Si to yield SiF4 was determined from the linear relation to be 0.1 for a Si surface at about 300 K. When the discharge pressure was higher than 1.3 Pa, the surface became rather strongly oxidized by O atoms resulting from residual gases. This surface oxidation results in a slight saturation of the Si etch rate at about 10 Pa.

An experimental system for surface reaction studies in microwave plasma etching

An experimental system for studying surface reactions in the process of microwave plasma etching has been developed. In the system, a surface etched in the microwave plasma can be analyzed with x‐ray photoemission spectroscopy (XPS) without exposure of the surface to room air. In addition, we have developed a procedure for calculating a thickness of a surface layer stoichiometrically different from the substrate material and densities of atoms in the layer. Chemical changes in etched Si and SiO2 surfaces caused by exposing these surfaces to room air are investigated with XPS to show the utility of the system. When the surfaces etched in SF6 microwave plasma are exposed to room air, the chemical states of the surfaces change rapidly. This is mainly due to surface oxidation and adsorption of hydrocarbon compounds to the surfaces. The rapid changes are more clearly shown from increases in surface layer thickness and the number of O and C atoms in the layer. It is clarified that exposure of etched surface to ...

Role of sulfur atoms in microwave plasma etching of silicon

The Si etch rate in an (F2+O2) microwave plasma has been measured as a function of O2 mixing ratio at a fixed total pressure of 5.3×10−2 Pa. The etch rate significantly decreases with the mixing ratio. This etch rate decrease is due primarily to surface oxidation. When sulfur is added to the (F2+O2) plasma, the Si surface is much less oxidized and the etch rate increases by about a factor of 4. Such sulfur‐containing species as S atoms react with O atoms or ions in the plasma and form O atom‐containing species, such as SO2, SO+, SOF+, and SOF+2, thereby reducing the O atom and O+ ion concentrations in the plasma. As a result, the Si surface is scarcely oxidized, so that the etching reaction can easily proceed. Sulfur atoms inhibit surface oxidation and promote Si etching. Sulfur atoms contained in SF6, which is usually used in microwave plasma etching of Si, are expected to have the same role.

Diagnostics of microwave plasma by laser induced fluorescence

CF radicals in a C3F8 microwave plasma were detected by laser induced fluorescence (LIF). CF2 radicals in C3F8 and CF4 plasma were also observed by using the same technique. At discharge pressures of 0.13 and 4.0 Pa, the relative densities of both radicals were measured as a function of microwave input power over a range of 40–240 W. The dependencies of these radical densities on discharge pressure were also measured over a pressure range of 6.7×10−2 to 13 Pa. The reaction mechanism in the plasma is discussed in terms of experimental results.CF radicals in a C3F8 microwave plasma were detected by laser induced fluorescence (LIF). CF2 radicals in C3F8 and CF4 plasma were also observed by using the same technique. At discharge pressures of 0.13 and 4.0 Pa, the relative densities of both radicals were measured as a function of microwave input power over a range of 40–240 W. The dependencies of these radical densities on discharge pressure were also measured over a pressure range of 6.7×10−2 to 13 Pa. The reaction mechanism in the plasma is discussed in terms of experimental results.

Si Etching with a Hot SF6 Beam and the Etching Mechanism

Silicon surface etching using a hot SF6 (SF6*) molecular beam is being investigated in the interest of studying the influence of vibrational energy on surface reactions and developing a damage-free etching technique. The SF6* beam is produced by the free jet expansion of SF6 gas heated in a quartz furnace. It is seen that SF6* vibrational energy enhances the Si etch rate. It is also shown that the Arrhenius model modified to take the vibrational energy effect into account can explain the experimental results. The model parameters are then determined to be 6500 cm-1 for activation energy (Ea), 0.17 for the efficiency (α) of the vibrational energy used to clear the activation energy barrier, and 5.0 for the frequency factor (A).

Si Etching with a hot SF6 beam

Si surface etching using a hot SF6 (SF*6) molecular beam has been studied in the interests of studying an effect of vibrational energy on surface reaction and developing a damage-free etching technique. The SF*6 beam is produced by free expansion of SF6 gas heated in a quartz furnace. It is seen that SF*6 vibrational energy enhances the Si etch rate. This conclusion is supported by an experimental evidence that Si etch rate under a 915°C substrate temperature increases 3.7 times if the furnace temperature is raised from 100°C to 730°C.

Equilibrium and stability of tokamak with arbitrary cross section

An analytical study is made of the equilibrium and stability of tokamaks with an arbitrary cross section. The ideal hydromagnetic model is used. With inverse aspect‐ratio expansion, the analytical expression of the equilibrium magnetic surface which is correct to the second order of the inverse aspect ratio is obtained as a function of the mean radius of the magnetic surface. The specific volume of the magnetic flux is obtained explicitly, and the stability against an interchange mode is discussed.

Separation of Uranium Isotope by Plasma Centrifuge

A method of centrifugal separation of isotopes by electromagnetic means is presented. The principle proposed utilizes electromagnetic acceleration by the interaction between an electric current in a slightly ionized gas and an external magnetic field. The analysis shows that an azimuthal flow of 2.6 km/sec can be realized with a magnetic field of 200 gauss and an electric current of 1.5kA. The resulting centrifugal force is large enough to permit realization of a more compact concentration cascade than the conventional mechanical centrifuge.

Toroidal equilibria with z asymmetric cross section

Toroidal equilibria with cross‐sectional shapes asymmetric to the plane z =0 are studied analytically. A general solution of the toroidal equilibrium is obtained with an inverse aspect‐ratio expansion and the Fourier decomposition as to the poloidal angle of the ideal magnetohydrodynamic equation. In the example of a tear drop rotated around its magnetic axis, the normal tear drop shows the deepest magnetic well, while the depth is only about 70% at 180° rotated ’’Pord‐raet’’ position. A proper combination of z‐asymmetric deformations, on the other hand, can deepen the well.

Gradient B Drift Instability in Toroidal Systems

A low frequency electrostatic instability in toroidal geometry is analysed with the two fluid equations. This instability is driven by the charge separation of plasma due to the toroidicity. The growth rate is largest when the parallel wave number is zero. This instability indicates an oscillation of poloidal mode numbers m =4, 3, 2 on the rational surfaces q =4, 3, 2 in tokamaks.