Nickel metallisation of Si by dynamic ion-beam mixing

Abstract

Thin Ni films were prepared at room temperature by Ni metal vapour deposition and simultaneous irradiation by Ar ions with an energy of 2–20 keV. The reaction of Ni with Si during dynamic ion-beam mixing was studied. The fluences of the ion beam were 4.7 × 10¹⁷ and 8.9 × 10¹⁷ cm^?2, and arrival rate ratios Ni/Ar were 9.7 and 5.1. Concentration profiles of Ni, Si, C, and O were analysed with Auger electron spectroscopy; the surface morphology and the crystalline structure were investigated with a cross-sectional scanning electron microscope and X-ray diffractometry. The theoretical profiles were calculated with the dynamic Monte Carlo simulation T-DYN for comparison with the experimentally obtained profiles. It was possible to observe the ballistic mixing effects and also thermally activated formation of nickel silicide.     

Low beam current density Auger spectroscopy and surface analysis

Auger and secondary electron spectroscopy become a more and more routine technique in surface characterization. Even with primary electron beam current density as low as 10^?2 or 10^?3 A cm^?2 beam damage were reported in both Auger and LEED experiments. So we developed and compared counting method, brightness modulation and Harris' modulation techniques in terms of signal to noise ratio. The two first methods offer the advantage of a primary beam current density decreasing about 10⁴ times. So various mechanisms of beam damage were identified as thermal, chemical and electrical. The advantage of the method is shown with hydrocarbons adsorption layer; the beam cracking of the organic chain produces a chemical shift of the C_KLL maximum Auger line about 5 eV. This progressive shift is observed with current densities of 10^?5 A cm^?2 order of magnitude. The reproducibility of this low current density Auger spectroscopy allowed the study of the background and the true secondary electron yield modifications when adsorbed layers are built up.     

Third harmonic generation and four-wave parametric mixing in sodium vapour

Third harmonic generation of frequency-tunable, mode-locked dye lasers was observed in sodium vapour with an efficiency of ≈10^?6. Phase-matching techniques failed to provide any significant enhancement of the third harmonic output. Several four-wave parametric processes were also detected in the vapour. The most intense of these occurred with a 0.5% efficiency and exhibited a saturation of the output.     

Umwandlung von Beryllium in Be2C und BeO durch Elektronenbestrahlung in Anwesenheit von Kohlenwasserstoff-Polymerisaten

The secondary electron yield from beryllium surfaces in presence of residual hydrocarbon vapour at a vacuum of 10^?4 tor decreases with time fromδ=0.45 to 0.25 when the incident primary electrons hit the target with an energy of 28 keV and an incident angle of 60°. Since the yield from the hydrocarbon polymer contamination layer is 0.3, it is concluded that the beryllium surface reacts with the contamination layer. This is confirmed by electron diffraction and electron microscopic observation of thin beryllium layers on different supporting foils, i.e. C-, Al₂O₃, polymeride from gas discharges, and self supporting Be. When a beryllium foil about 500 A. U. in thickness is first irradiated with low intensity (total beam current in the Siemens Elmiskop II<5 μA), a polymeride layer is formed. After increasing the beam current to 30 μA, the micrographs as well as the Debye-Scherrer diagrams show the formation of Be₂C and BeO. It is supposed that the carbon originates from the polymeride layer, and that the oxygen is supplied by the residual gas.     

Übertragung von Polarisation bei sensibilisierenden Stößen angeregter Natriumatome

The transfer of magnetic polarization from state 3p ² p _1/2 to state 3p ² p _3/2 sensitizing collisions of sodium atoms with noble gas atoms has been observed. Sodium atoms were excited by absorption ofD ₁σ⁺-quanta. When buffer gas was added to the sodium vapour, unequal amounts ofD ₂σ⁺ andD ₂σ^? quanta could be detected in the fluorescent light. The difference of the amounts turned out to be proportional to the polarization transferred. The measurement of this difference as a function of the buffer gas pressure provided a method of determining the cross-sections of the polarization transfer. Values of the cross-sections are 22, 24, 26, 20, 23 Ų for the buffer gases He, Ne, Ar, Kr, Xe, respectively. A preliminary investigation shows that the sign of the polarization transferred is reversed when the vapour is exposed to a strong magnetic field acting along the incident light beam.     

Study of traces of elements on a universal laser photoionization spectrometer

A new complex laser analytical spectrometer operating synchronously in the atomic beam regime and in the flame regime is developed. Analytical potentialities of the spectrometer are demonstrated by the example of determining sodium and calcium content in standard solutions and aluminum alloys. Processes of thermal atomization of Ca and Na in vacuum and the effect of the basis of a sample on the element determined are studied. Efficient excitation via Rydberg states is realized for these elements. The contents of Ca and Na in aluminum alloys are determined at levels of 10^?2 and 10^?3%, respectively.     

Changes in the silicon microhardness induced by a low-intensity electron beam

It is shown that nonequilibrium point defects are of primary importance in the changes in the silicon microhardness induced by a low-intensity (I ～ 10⁵ cm^?2 s^?1) electron beam. It is found that the necessary condition for softening under low-intensity electron irradiation is the presence of an oxide layer on the surface. The thickness of the surface layer in which anomalous changes in the microhardness are observed is determined by the layer-by-layer etching technique.     

Higher order averages of primary recoil distributions

A molecular beam detector is described which ionizes by electron bombardment. The ions are then separated in a quadrupole mass filter and detected by a multiplier. The special advantages of the mass filter allow a high overall transmission so that every 1000th molecule of the beam is measured as an ion. The ion current due to the residual gas is reduced by a factor of 10^?6 for all masses >45. These results were achieved without separately pumping or baking out the ionisation chamber. Since magnets are not used the detector is comparatively light and small in size. The smallest detectable beam was found to have a current density of 1,8·10^?6 molecules/sec mm² corresponding to 4·10³ molecules/cm³, at a vacuum pressure of 8·10^?7 torr in the apparatus, and using a time constant of 1,25 sec.     

Gas mixture effects on ion‐beam flux characteristics from dense plasma focus device: Investigation by the Vlasov–Maxwell equations and experiments

The Vlasov–Maxwell equations were numerically solved to calculate the ion‐beam flux from the plasma of argon and the plasma of mixtures of argon and neon. Some experiments were performed to measure the ion beam from the Amirkabir plasma focus (APF) device. The calculations have shown that the argon ion‐beam flux peaked up to 1.928 × 10³⁰ ions m^?2 s^?1 at the optimum pressure of 1.866 mbar while the neon‐argon mixture's ion‐beam flux reached a maximum of 4.301 × 10³⁰ ions m^?2 s^?1 for 15% neon admixture at the optimum pressure of 1.866 mbar. The calculated kinetic energy of the ion beam has shown a maximum value of 708.7 J for the mixture of 85% argon‐15% neon at the mentioned optimum pressure.     

Surface lifetime degradation in Si and Ge under low energy electron irradiation

The bombardment of n-type Ge and Si by 10–30 keV electron beams is shown to reduce the surface lifetime τ_P of minority carriers but not to affect mobility μ_P or diffusion constant D_P. For experiments, the electron beam in a scanning electron microscope is replaced a chopped light source to perform modified Haynes-Shockley experiment, and the measurements is used to calculate μ, D and τ. The reduction in τ is interpreted in terms of an increase in surface recombination velocity s. At a dose of approximately 10¹⁷ electrons cm^?2 for Ge and 10¹⁵ electrons cm^?2 for Si, a saturation value for s appears to be reached after which further bombardment has no effect. The values of τ obtained from both light and SEM Haynes-Shockley experiments performed in vacuum agree within approximately 5% both prior to and after bombardment. The initial values of lifetime can be restored by etching the samples.     

Upper limit to the cross-section for the process νμ + e− → νμ + e−

An upper limit of 3 × 10^?42E_ν cm² per electron, at 90% confidence level, has been obtained for the cross section of the process ν_μ + e^? → ν_μ + e^?, interpreted in the usual V-A framework, for an electron recoil energy larger than 0.3 GeV. This estimate is based on the observation of one candidate event in exposures of Gargamelle to the CERN PS neutrino beam.     

Intense pulses of polarized electrons produced by Fano effect

A pulsed polarized electron source using the Fano effect on cesium is described in detail. A frequency doubled dye laser producing 2 mJ/pulse at 305 nm is used as the source of circularly polarized light. The light beam interacts with an array of 20 atomic beams. The atomic beam oven operates in a closed cycle thereby increasing running time by a factor of 30. Intensities of 3×10⁹ e ^?/0.5 μs with a polarization of 90% have been routinely obtained.     

Conduction transients in single-crystal octadecane

Charge transients were measured in high-quality single-crystal octadecane after electron beam irradiation. The signals produced were of significantly greater duration than the ionization pulse but were not interpretable by conventional uniform field time-of-flight analysis. By assuming that the specimens contained a strong inbuilt polarization field, it was possible to predict the properties of the observed charge transients and obtain values for the positive and negative drift mobilities. The results indicated a positive species with a mobility of 0.7 × 10^?4 m² V^?1 s^?1 and a negative carrier value of 1.6 × 10^?4 m² V^?1 s^?1. The standard deviations observed in the data were large, being 0.3 × 10^?4 m² V^?1 s^?1 and 0.8 × 10^?4 m² V^?1 s^?1, respectively. This scatter was significantly in excess of the estimated experimental error.     

Some aspects of laser-metal vapour interaction

The vapour beam interaction plays an important role in laser machining process, since the vapour generated on the workpiece surface absorbs some fraction of the incident beam and heats the workpiece surface. Consequently, a study of this interaction mechanism is essential. For this purpose, a computer program was developed to investigate the interaction interaction, it provides a foundation to do further studies. It was found that the vapour locally high pressure gradients. Although the analysis presents a simplified picture of the interaction, it provides a foundation to do further studies. It was found that the vapour temperature reaches 5000 K after 10⁻⁶ s for a laser pulse of 10¹¹ W/m² power intensity. The leading edge of the vapour velocity had a velocity of the order of 4000–7000 m/s.     

Electron beam interactions with CO on W {100} studied by Auger electron spectroscopy

The interaction of 2500 eV electrons with carbon monoxide chemisorbed on tungsten {100} was investigated by rapid-scan Auger electron spectroscopy. When no α state was present the O and C signals from the β state of CO were invariant during electron bombardment, giving an upper limit estimate for the electron stimulated desorption cross section, Q_β of 2 × 10^?21 cm². With the crystal at room temperature and saturated with CO, however, electron-beam induced accumulation of carbon was observed and characterised, the rate of the process being independent of CO pressure at pressures above 2 × 10^?8 Torr. At 450 K the rate was found to be pressure dependent up to at least 6 × 10^?7 Torr. A model is proposed for the accumulation process, which is based on electron beam dissociation of α₂-CO to form adsorbed carbon and gaseous O and the creation of new sites for further α₂-CO adsorption; it is in quantitative agreement with the results and yields a cross section for ESD of α₂-CO (Q_α2 = 1.55×10^?18cm²) in close agreement with direct measurements.     

The interference of an electron beam with the surface reaction between oxygen and germanium

Electron beam assisted adsorption and desorption of oxygen was studied by Auger electron spectroscopy (AES). Beam assisted adsorption was observed on clean as well as on oxidized surfaces. After an oxygen exposure of 1000 × 10^?7 Torr min and continuous irradiation with beam voltage of 1.5 kV and beam current density 2 microA mm^?2, the oxygen 510 eV signal amplitude from the point of beam impact was 2.5 times greater than the signal from the non-irradiated region. The Ge 89 eV signal showed a corresponding decrease. Enhanced adsorption occurred at beam energies as low as 16.5 eV. After irradiation, the oxidized surface was not carbon contaminated. Following an oxygen exposure of 30 min at 0.1 Torr and 550°C and subsequent additional beam assisted exposure of 1000 × 10^?7 Torr min, the maximum oxide thickness was about 18 Å. Beam assisted desorption did not occur from thin oxygen layers (0–510 eV signal strength less than 5 units, calculated oxide thickness about 6 Å), but occurred from thick oxides and stopped after the signal amplitude had decreased to 5 units. Based on these results, a model for the structure of the oxygen layer covering the Ge(111) surface is proposed. Mechanisms for adsorption and desorption are discussed. The implications of beam assisted adsorption and desorption on electron beam operated surface measurements (LEED, AES, ELS, APS etc.) are stressed.     

Field electron emission from Ge-Si nanostructures with quantum dots

Self-assembled arrays of Ge-Si clusters with sizes of ～ 10 nm and a density of ～ 10¹⁰ cm^?2 have been grown by molecular beam epitaxy. Stable steady-state field electron emission from such clusters has been observed and studied. The emission is characterized by resonance current peaks, which are explained by the quantization of the electron energy in nanoclusters. The estimation of the ground level energy from their emission measurements coincides with estimates obtained by other methods.     

Lithium vapour excitation at 2S→3D two-photon resonance

We report study of processes which occur in lithium vapour under two-photon excitation of the Li(3D) state at 639.1 nm. A time-resolved technique has been used to measure the fluorescence from the Li(3D), Li(2P) and Li(3P) states. We have determined radiation rates for lithium atom densities in the range 10¹³–10¹⁴ cm^?3 and laser powers (10⁵–10⁶ Wcm^?2). The ground-state lithium atom density was determined by knowing temperature and vapour pressure in a modified heat-pipe oven. The contribution to radiation rates from different processes and prospect for cross-section determination of homonuclear reverse energy-pooling are discussed.     

Changes in the iron atomic state in the near-surface layers of zirconium alloys irradiated by Ar<Superscript>+</Superscript> ions

The effect of irradiation by an argon ion beam with a broad energy spectrum and an average energy of 10 keV in a dose range of (1?10) × 10¹⁸ ion cm^?2 on the state of iron atoms in the near-surface layers of zirconium alloys Э635, NSF, and Zry-2 was studied by means of conversion electron Mössbauer spectroscopy (CEMS).     

High-purity F atom beam source

Fluorine atoms are generated by 2.45 GHz microwave dissociation of 99.99% F₂ in a synthetic sapphire single-crystal discharge tube. Typical F₂ gas flow and beam intensity are: 0.1 mbar ls^?1 and 9.5×10¹⁷ atoms sterad^?1 s^?1. The measured degree of dissociation into atoms isD>98%. The high total F atom flux together with the purity of the beam seem to make the source a very promising device for both reaction kinetic-and molecular beam-experiments