No abstract

The expansion of plasmas produced by a mode-locked Nd:glass laser has been studied. With help of the second harmonic radiation of the laser itself the shock waves in the initial stages after breakdown are recorded. With a streak camera the time development of the plasma depending on the gas pressure and the focal length of the focusing lens was studied in He, Ar, Kr, Xe, N2, CO2 and SF6. While the shock waves obey the well known relations for ideal shock waves, the stepwise plasma expansion towards the lens can be described by a simple ionization model.

I 91 /" I 20 */" 3 0 0 3 M 4 M 4 5 0 5 W 0 (B) CTuQ2 Fig. 2. Variation of parameters (a) A and (b) B with temperature. Obtained reaction activation energies are 0.2eV and 1.1 eV for A and B, respectively.molecular beam epitaxy (MBE), similar to the structures used by several researchers to study selectively oxidized VCSELS.~-~,I~-I~ The wafer is dry etched into cylindrical mesas to expose the edge of the AlAs layer and oxidized in N, gas bubbled through water at 83°C. The N, gas flow rate is kept relatively low (1.3 Vmin). Under these conditions, the oxidant concentration inside the furnace is low enough for the oxidation process to be predominantly diffusion limited. We have carried out oxidation at several temperatures to observe the dependence of oxidation rate on temperature. The parameters involved in Eq. (4), A and B, and their temperature dependence relations of the form, T = To exp( -EJkT), (where E, is the equivalent reaction activation energy,',) are extracted as shown in Fig. 2. Incorporating these temperature dependence relations ofA and B in Eq. (4), we have calculated the oxidation reaction rate at different oxidation temperatures for AlAs layers of different thickness as shown in Fig. 3(a). Theoretical calculations predict a decrease in the oxidation rate with the reduction in the thickness of the AlAs layer, which is confirmed experimentally. Oxidation rate for an AlAs layer of a particular thickness is also dependant on the equilibrium oxidant molecule concentration inside the furnace, which depends on the water temperature in the bubbler as well as the N, gas flow rate. We have observed that, the overall oxidation rate can be increased by increasing the water temperature in the bubbler, and also by increasing the N, gas floy rate. We have oxidized a wafer with a 500 A thick pure AlAs layer, in N, gas (flow rate increased to 3.2 l/min) bubbled through water kept at a constant temperature of 96°C. Oxidation rate for this wafer at different oxidation temperatures is shown in Fig. 3(b). We observe that under these experimental conditions, this wafer, although relatively thin, has a higher oxidation rate than the previous results presented in Fig. 3(a). In 10 91 l (b) CTuQ2 Fig. 3. Dependence of oxidation rate on oxidation temperature for (a) AlAs layers of different thicknesses (N, gas flow rate 1.3 Ymin, water temperature 83"C), and (b) a 500 8, thick pure AlAs layer (N, gas flow rate 3.2 l/min, water temperature 96°C). The lines represent the theoretical calculations and the points represent the experimental data.this situation, the density of the oxidant in the atmosphere is much higher. The oxidation process is predominantly reaction limited, and a higher oxidation rate is observed. This situation is also accurately described using the same theory.In summary, the dependence of the wet oxidation rate ofAlAs on different experimental parameters as well as the thickness of the AlAs layer has been studied. A theory to explain the dependence of oxidation rate on different physical parameters for both...

No abstract

The continuum emission of an argon-and nitrogen plasma developed in an electrical discharge has been investigated in the wavelength range from 3000 Å to 6700 Å. To this purpose the time-and radial dependent plasma parameters such as temperature and the total pressure have been determined in the high conducting stage of the spark by measuring several line intensities. The continuum coefficient was calculated from these data according to the Kramers-Unsöld theory. The comparison of the theoretical and the measured values shows deviations which are discussed. In the case of the argon continuum the ξ-factors for T = 14 000 K agree with the calculated values of Schlüter and the experimental ones of Schulz-Gulde. In the case of the nitrogen plasma the ξ-fac­tors have been determined in the temperature interval from 18 000 K to 45 000 K. Since at these temperatures the particles NII-NIV contribute to the total continuum coefficient, the measured ξ-factors can only be correlated to ξII, ξIII, ... in a narrow temperature range. The measured cor­rection factor ξ for λ = 5050 Å has been applied to determine the temperatures and the pressure of a laser produced spark. The plasma parameters agree with those determined by measuring the line-intensity of the NII-line at 5000 A.