D. Débarre

Planarization of rough silicon surfaces by laser annealing

Abstract Rough silicon (or silicon alloy) surfaces are sometimes obtained after epitaxy or processes for micromachining due to unusual experimental conditions imposed by external constraints. We show that excimer laser annealing has an efficient ability to reduce roughness to nm scale, providing the surface melts under laser irradiation. No large improvement is observed above the melting threshold and only few laser shots (

Pulsed laser etching of silicon: Dopant profile modification and dopant desorption induced by surface melting

Laser‐induced modification of dopant profile and desorption of the dopant are measured for silicon and several dopants (B, As, Sb). The experimental conditions are laser melting and chemical etching by chlorine, and laser annealing. The laser fluence is above the melting threshold, and the Si dosing by chlorine is pulsed under a very low static Cl2 pressure. Successive dopant profiles evolve very differently for these three dopants. The desorption efficiency is shown to depend strongly on the dopant atom, increasing from boron to arsenic to antimony. The implications of etching, doping, and annealing experiments are discussed.

Viscosity of transient melt layer on polymer surface under conditions of KrF laser ablation

Abstract An important parameter controlling the mobility of melt layer in laser ablation of polymers is the viscosity. We report the first experiments to measure it at the extreme conditions of ultraviolet (UV) KrF laser ablation for poly(ethylene terephthalate) (PET). The volume of laser-induced bumps (re-solidified molten material expelled from the irradiated area to its periphery) is measured with AFM microscopy. The comparison of this value with an analytical modelling (based on the solution of Navier–Stokes equations for viscous melt flow induced by the gradient of ablation plume pressure with the use of experimentally measured melt depth) allows to calculate the kinematic viscosity of laser melt layer on the polymer surface. It appears to be about an order of magnitude greater than the kinematic viscosity of water at room temperature.

The role of gas-phase in the laser etching of Cu by CCl4

In the laser etching of surfaces, the gas-surface chemical reaction used may be spontaneous or induced by a laser-gas interaction. We put into evidence a third mechanism allowing to use gas molecules that interact directly neither with the surface nor with the laser but that contains reactive atoms.

Laser modifications of Si(100) : Cl surfaces induced by surface melting : etching and cleaning

Abstract Pulsed laser driven modifications of one Cl monolayer (ML) chemisorbed on Si(100) are studied by time of flight mass spectrometry (TOF) of the desorbed molecules, Auger electron spectroscopy (AES) of the modified surface, and Cl depth profile analysis by secondary ion mass spectrometry (SIMS). The experimental conditions are that of surface melting. Evidence of Cl diffusion to the bulk during surface melting, and of strong segregation of Cl in Si during surface recrystallization, are presented. The following branching ratios for 1 Cl ML initially adsorbed on clean Si are measured independently: (a) (from TOF and depth profiler measurements) 0.58 ± 0.07 ML desorb reactively in the form of SiCl and SiCl2. The SiCl to SiCl2 ratio is 1.3 ± 0.3; (b) (from AES measurements) 0.37 ± 0.13 ML are found after the laser pulse in a thin surface layer of 7 atomic planes; (c) (from SIMS measurements) 0.10 ± 0.10 ML are found in deeper layers, the solubility of Cl in solid Si being lower than 1019 cm−3, the sensitivity of our SIMS for Cl; (d) 0.12 ± 0.07 ML desorb unreactively in the form of Cl and Cl2. By order of increasing importance, unreactive desorption, stoichiometry and diffusion to the bulk during the laser pulse have the practical effect of limiting the maximum etch rate to 0.40 ± 0.03 Si ML per laser pulse under the conditions of surface melting. Although the laser pulse induces diffusion of Cl towards the bulk during melting, the strong segregation of Cl during recrystallization limits Si contamination by Cl at undetectable levels to our SIMS. Cl in Si is a very good case for laser cleaning, as is shown by comparison with C and O.

Growth of a SiC layer on Si(100) from adsorbed propene by laser melting

Carbon is incorporated into Si(100) to form a thin polycrystalline layer of SiC by laser melting the Si surface after adsorption of propene in ultrahigh vacuum. The SiC layer of thickness up to 25 nm is polycrystalline. Crystallites of size ≈100 nm are oriented with respect to the Si substrate and exhibit a diffraction pattern in low energy electron diffraction (LEED). The evolution of the surface is monitored in real time by recording the Si transient reflectivity at 675 nm at each laser pulse, and after exposure to the laser by LEED, IR spectroscopy, and atomic force microscopy. The formation of the SiC layer is accompanied by very strong variations of both the static and transient reflectivities, by the growth and narrowing of the IR peak assigned to β SiC, and by the increase of the C incorporation rate. The SiC overlayer is very stable against photodesorption, while initially small amounts of C on Si are photodesorbed in a few laser pulses. Recording the transient reflectivity during processing allow...

Neutron transmutation as a method to calibrate the infrared absorption of indium in silicon

The integrated intensity of selected absorption lines of indium in silicon can be used to obtain an infrared calibration factor for this impurity. Infrared calibration factors obtained with different electrical methods for measuring the In concentration are given. We have used the comparison of the In absorption in samples before and after partial compensation of In by neutron transmutation to correlate the difference between the integrated intensities of the In lines and the In concentration compensated by P, independently of the electrical measurements. The coefficient obtained gives slightly higher In concentrations than those using the resistivity and capacitance‐voltage measurements as ’’primary standards.’’

Quantitative Determination of B and P in Silicon by IR Spectroscopy

Many of the applications of silicon share the need for a high-purity homogeneous material with a controlled concentration of residual shallow and deep impurities. One of its utilization is the fabrication of coupled charge devices (CCD) linked to extrinsic Si detectors arrays. This has given a new impetus to the study of impurities and defects in this material as well as to the need for accurate compensation techniques since spurious effects arise from the presence of uncompensated and unwanted impurities, even at low concentration. Besides atomic impurities, electrical, optical and photoconductive studies have shown the existence in the band gap of silicon of shallow acceptor levels due to impurity complexes whose nature is not yet wholly elucidated1–3.