Pulsed-laser sputtering of atoms and molecules. Part I: Basic solutions for gas-dynamic effects

When gases are released from a pulsed nozzle or when solids are sputtered with intense laser pulses, effusion-like expansions take place which terminate abruptly. The resulting gas-dynamic processes depend on , the heat capacity ratio, as well as on whether particles backscattered to the effusing surface are subject to recondensation or reflection. Certain aspects of these terminating expansions have already been treated but we consider it appropriate to examine the problem further. In particular the following topics are emphasized. (a) Following previous work, the expansions are shown to consist of a series of regions separated by lines of contact, i.e. abrupt changes of slope. (b) For conditions of recondensation, there are two regions separated by one line of contact, the first region lying in part behind the effusing surface. For conditions of reflection, there are three regions, the first of which begins at the surface. Both types of expansion terminate with a region which is a remanent of the release process. (c) The nearsurface region under conditions of reflection permits an analytical approximation valid for all in which the sound speed is invariant with distance and the flow velocity is linear with distance. (d) The surface itself under conditions of recondensation permits an analytical approximation valid for all for the sound speed. More generally the near-surface region can be resolved by the method of Stanyukovich. (e) The various analytical solutions and approximations are shown to compare favorably with numerical results. (f) Plots of density and flow velocity versus distance are found to be roughly independent of , thence of the nature of the sputtered particles. (g) Tabulated results are presented to enable a more general use of gas-dynamic ideas.For Part II, which deals with recondensation, see [1]     

Laser photochemical etching of silicon

Argon laser induced chemical etching of single crystal silicon with chlorine is studied. Etch rates are determined as a function of gas pressure, crystal orientation, laser power and wavelength. Analysis of gas phase and surface products by Fourier transform IR and X-ray photoelectron spectroscopy are used to probe the reaction mechanism. Contrary to previous reports, no thermally enhanced etch rate is observed for Si (111) and the presence of oxide on the surface is found to inhibit etching even at high laser power.     

Picosecond laser ablation of thin copper films

The ablation process of thin copper films on fused silica by picosecond laser pulses is investigated. The ablation area is characterized using optical and scanning electron microscopy. The single-shot ablation threshold fluence for 40 ps laser pulses at 1053 nm has been determinated toF _thres = 172 mJ/cm². The ablation rate per pulse is measured as a function of intensity in the range of 5 × 10⁹ to 2 × 10¹¹ W/cm² and changes from 80 to 250 nm with increasing intensity. The experimental ablation rate per pulse is compared to heat-flow calculations based on the two-temperature model for ultrafast laser heating. Possible applications of picosecond laser radiation for microstructuring of different materials are discussed.     

Combined SIMS,AES, and XPS investigations of tantalum oxide layers

Thick layers of tantalum oxide prepared by thermal and anodic oxidation have been studied by combined SIMS, AES, and XPS during depth profiling by 3keV Ar⁺ ion sputtering. The chemical composition of these films is revealed by the OKLL and O 1s signals and by the “lattice valence” parameter determined from the TaO _n ^± intensities. Thus the anodic film consists of a contamination layer, an oxygen-rich reactive interface and a thick homogeneous oxide layer followed by an interface to the Ta metal. The thermal oxide shows an oxygen concentration decreasing with depth and a broad oxide-metal interface. In both cases, carbon contamination (carbide) prevents the application of the valence model to the clean Ta substrate. The sputtering yield of the oxides was found to be 0.6 Ta₂O₅/ion.     

Analysis of rapid dry oxidation of silicon

By analyzing the most recent models on rapid initial oxidation and the experimental data at low temperatures we prove unambiguously that neither enhanced nor retarded oxygen diffusion nor any kind of additional oxygen transport flux can account for anomalous initial regime of silicon dry oxidation.The rapid growth is mainly due to the enhanced oxygen solubility and partly to the enhancement of the reaction rate constantk _s. We argue that the reaction rate depends linearly on the oxygen solubility for low solubilities pertinent to dry oxidation but that it saturates at high solubilities characteristic for the wet oxidation.     

Thermochemical laser etching of stainless steel and titanium in liquids

Laser-induced wet chemical etching of stainless steel 304 and Ti in phosphoric acid, sulfuric acid and aqueous KOH has been investigated using cw Ar and cw Nd : YAG lasers. Two different phases of laser-induced etching of Ti in phosphoric acid were found by electrochemical investigations. Laser-enhanced electrochemical dissolution of stainless steel was observed in the passivation and transpas sivation region. In the latter case, laser heating accelerates the metal dissolution. In the passivation region, laser heating results in a breakthrough of the passivation layer near the Flade potential. By multiple scanning microstructures of high quality and an aspect ratio > 10 have been produced. By EDX analysis the laser-etching process was found to be practically free of chemical residues on etched surfaces.     

A study of blast waveforms detected simultaneously by a microphone and a laser probe during laser ablation

We examine blast waves generated in air during irradiation of absorbing samples with Nd: YAG laser pulses of fluences exceeding the ablation threshold. Blast waves were detected simultaneously by a wideband microphone and a laser beam deflection probe. By a comparative analysis of both signals in the time and frequency domain we investigate characteristic features of their nonlinear waveform evolution. To explain the observed phenomena we employ the weak shock solution of the point explosion model.     

Time-resolved experiments on the photoablation of polystyrene and PMMA by ArF-laser radiation

The photoablation of polystyrene (PS) and polymethylmethacrylate (PMMA) was studied in real-time during the uv laser pulse at 193 nm. The transmission and total reflection of thin polymer layers on quartz glass substrates was measured time-resolved. From the results for the strongly absorbing PS it can be concluded that the emission of material starts within the first few nanoseconds of the laser pulse. Photoablation of PMMA, which is a relatively weak absorber at 193 nm, is accompanied by strong modifications of the transmission by the first several ten laser pulses.     

ELS study on initial oxidation of Ni(100) surfaces

Electron energy loss spectra of clean and oxygen-covered Ni(100) surfaces were observed with concomitant measurements of LEED, work function change, and Auger peak height ratio O(KL_{2, 3}L_{2, 3})/Ni(L_{2, 3}VV). The observed electronic transitions are interpreted on the basis of primary election energy dependence, and of comparison with the loss spectrum for a UHV-cleaved NiO(100) surface and optical data of Ni. The observed loss peaks at 9.1, 14, and 19 eV in the clean surface spectrum are ascribed to the bulk plasmon of the 4s electrons, the surface plasmon, and the bulk plasmon of the coupled 3d + 4s electrons, respectively, and the weak but sharp peak at 33 eV is tentatively attributed to the localized many-body effect in the final state. Three oxygen-derived peaks at 6.0, 8.0, and 10.3 eV in the low oxygen exposure region (?4 L) are ascribed to the O 2p(e) → Ni 3d, O 2p(a₁) → Ni 3d, and O 2p → Ni 4s transitions, respectively. In the high oxygen exposure region (?50 L), the spectra become quite similar to that of the UHV-cleaved NiO(100) surface. The oxidation process consistent with LEED, Auger peak height ratio and work function change measurements is discussed.     

Mechanism of and method to avoid discoloration of stainless steel surfaces in laser cleaning

2 O₃. Based on Auger Electron Spectroscopy analyses and optical microscopic observations, γ-Fe₂O₃ and Fe₃O₄ are formed on the stainless steel surface in laser cleaning in the air. Since laser can induce high temperature rise in stainless steel surfaces, the above phenomenon can be explained by a thermochemical reaction between oxygen in the air and the stainless steel. With increasing laser fluences, the temperature rise in the irradiated area of stainless steel surface increases, which enhances oxygen diffusion into the surface and oxidation reaction within the irradiated area. In order to avoid discoloration of stainless steel surfaces, a vacuum system was used to reduce the oxidation reaction between oxygen and stainless steel. Received: 7 June 1996/Accepted: 30 September 1996     

LEED and ELS study of the initial oxidation of Pd(100)

Oxygen-Pd(100) interactions were studied by means of LEED, AES and ELS to characterize the initial oxidation and elucidate the relationship between surface-oxide reconstructions and bulk oxidation. Five oxygen-related LEED patterns were identified and classified into three types. First, the chemisorption structures included the usual p(2 × 2) which was replaced by a c(2 × 2) upon increased exposure to oxygen at room temperature. Second, surface-oxide reconstructions included a p(5 × 5) which formed above 200°C and which was replaced by a (√5 × √5) R27° above 300°. These two structures are attributed to PdO(110) and PdO(001) coincidence lattices, respectively. Above approximately 500°C the oxygen was removed due to thermal desorption and possibly absorption, but a multilayer oxide, distinct from PdO, could form upon oxygen exposure above 800°C. The ELS study indicated that this last species had significantly different electronic properties than that of the other structures. It is quite plausibly a Si-stabilized oxide as has been suggested recently by Niehus and Comsa for the Pt-oxygen system.     

Rapid growth of ultra thin SiO2 films by a large-area electron beam

Rapid growth of ultra thin oxide films (40–180Å) of silicon using a low-energy large-area electron beam has been performed with a pressure ratio of 31 (O₂/He) and a total pressure of 0.5–0.7 Torr. A higher oxidation rate of about 625Ų/s is found for shorter irradiation time of the e-beam in the e-beam dose range 0.75–3 Coulomb/cm² and at lower substrate temperature 540–740°C. AES and XPS demonstrated a rapid electron-stimulated oxidation process of the Si surface. For the grown ultra thin oxide films, C-V characteristics, dielectric strength, uniformity of the film over the entire Si wafer and its thickness as a function of the processing time of the e-beam are also presented.     

Negative thermal feedback in laser-assisted oxidation of thin Zn films

The processes leading to the control of the lateral dimension of laser-assisted oxidation of Zn films on glass are examined. It is shown that it is determined by negative feedback between temperature and optical absorption.     

A model for the oxidation of Mg(0001) based upon LEED,AES, ELS and work function measurements

The Mg(0001) face is subjected to oxygen adsorption from 0 to 10³ L. Three characteristic stages of oxygen adsorption are detected from 0 to 10 L. The AES signal of clean Mg decays exponentially against exposure with slopes α _ai such that α_A2 (0.75 → 3 L) >α_A1 (0 → 0.75 L)>α_A3 (3 → 10 L). For increasing exposures, they correspond to: (1) a clear (1 × 1)-Mg(0001), (2) a diffuse (1 × 1)-Mg(0001) and (3) a (1 × 1) with a weaker (1 × 1)-R30°-MgO(111) LEED patterns, respectively. At the end of the third stage, a supplementary ($\text{7}\text{×}\text{7}\text{2}\text{)?R19°?MgO(111)}$ pattern is observed. In ELS, a very fast intensity decrease of energy loss peaks due to surface and bulk plasmon excitations of the clean metal is recorded during the first stage. The energy loss peak due to the oxidized surface plasmon excitation reaches a maximum intensity at the end of the second stage. Energy loss peaks to be attributed to excitations in bulk MgO appear during the third stage. The work function of the sample decreases and shows a minimum around 6 L, and then slowly increases. Beyond 10 L, a logarithmic relation between oxide thickness and exposure seems to exist. These results are interpreted by the following sequential processes: stage 1: random oxygen chemisorption followed by oxygen incorporation (α_A1); stage 2: assembling into islands and lateral island growth (α_A2); stage 3: oxide formation (α_A3) and stage 4: oxide thickening. Lattice models describing these processes are proposed and discussed. The influence of surface roughness on the results is emphasized.     

AES investigations of Ar+ ion retention in Si during Ar sputtering

Argon retention in silicon has been studied by AES in the energy range between 1 and 15 keV at bombardment fluences up to ∼10¹⁸ ions/cm². AES data of implanted argon in silicon near the surface region, as obtained during sputtering, can be interpreted qualitatively by a simple model of ion collection. Discrepancies between calculated and measured saturation values of collected argon ions indicate that during implantation at high fluences addition surface effects become important and that the simple model of ion collection has to account for this. Quantitative AES correlated with RBS indicates pronounced concentration gradients of argon in silicon near surface regions.     

Surface-science aspects of plasma-assisted etching

Plasma-assisted etching methods have been used in the manufacture of integrated circuits for more than 10 years and yet the surface-science aspects of this technology are poorly understood. The chemistry must be such that the reactive species generated in the plasma react with the surface being etched to form a volatile product. The chemistry is usually dominated by atoms, molecular radicals and low-energy (20–500 eV) positive ions. In microstructure fabrication, the positive ions are accelerated from the plasma towards the etched surface arriving essentially at normal incidence. Thus, the bottom surface of a very small feature being etched is subjected to both energetic ions and reactive neutral species, whereas the sidewalls of the feature are exposed to reactive neutral species only. The role of the energetic ions is primarily to accelerate the reaction between the neutral species and the etched surface (i.e., accelerate the etch rate), thereby reducing the steady-state top-monolayer coverage of the etching species on the etched surface. On the sidewalls, however, the reacting-species coverage is a saturation coverage. The present understanding of some of the surface-science aspects of this complex environment will be summarized, often using the Si-F system as an example, and some phenomena which are not well understood will be described.     

UV-laser ablation of ductile and brittle metal films

Single-shot laser damage of Ni and Cr films on fused silica substrates has been studied as a function of film thickness, utilizing 248 nm/14 ns pulses and detection by probe beam deflection. Threshold fluences for visible damage and vaporization are compared to predictions of the heat diffusion model. The model fits thresholds for visible damage well and identifies their origin, which is melting for Ni films and brittle-to-ductile phase transition for Cr films. When predicting thresholds for vaporization, the diffusion model is of limited success in case of Ni films but fails completely for Cr films, indicating that transient thermal properties of the material should be taken into account. Microscopic inspection shows that Cr films rupture at low fluences before entering the common sequence of melting and vaporizing with increasing fluence. 17 December 1996/Accepted: 17 December 1996     

CW laser-assisted oxidation of thin Cd,In, Sn,and Zn films in air

Thin films of Cd, In, Sn, and Zn are deposited onto glass and irradiated in air by means of a cw-Ar⁺ laser beam. The films are oxidized. The variations of the diameter of the oxidized zones are measured as a function of time and laser beam power, P. The temperature is measured by an interferometric method. It is shown that oxidation proceeds rapidly at some critical temperature, independent of P over some range of P, in the cases of Cd, In, and Zn. These critical temperatures correspond to the melting temperatures of Cd and In. No relation to any specific temperature of the Zn-O phase diagram is found. Feedback effects are also discussed.     

Realization of silicon quantum wires by selective chemical etching and thermal oxidation

Ultra-fine silicon quantum wires with SiO₂ boundaries were successfully fabricated by combining SiGe/Si heteroepitaxy, selective chemical etching and subsequent thermal oxidation. The results are observed by scanning electron microscopy. The present method provides a very controllable way to fabricate ultra-fine silicon quantum wires, which is fully compatible with silicon microelectronic technology. As one of the key processes of controlling the lateral dimensions of silicon quantum wires, the wet oxidation of silicon wires has been investigated, self-limiting wet oxidation phenomenon in silicon wires is observed. The characteristic of the oxidation retardation of silicon wires is discussed.     

Investigation of titanium and titanium hydride by AES and EELS

Effects of hydriding on the Auger spectrum of titanium are investigated and can clearly be distinguished from the changes induced by oxygen contamination of the surface. New transitions occur about 5 eV below all transitions involving metal valence band electrons. Beam induced decomposition of the hydride could be avoided by proper experimental procedure. The composition of the investigated hydride was inferred from the measured bulk plasmon energy shift.