Dynamic Trim Research Articles

Determination of the Ion Sputteringinduced Indepth Distribution by Means of Elastic Peak Electron Spectroscopy

Elastic peak depth profiling was carried out on an Mo/Si multilayer system using a rotating specimen, grazing angle of incidence (86° with respect to the surface normal) and 0.5 keV Ar ion energy. The depth profiling was simulated by dynamic TRIM (T-DYN) code. The T-DYN code provided the in-depth distribution of the elements in any moment of the depth profiling process. Direct Monte Carlo calculation was developed to calculate the intensity of the elastic peak emitted from the distribution determined by the T-DYN code. Good agreement was found between the calculated and measured elastic peak depth profiles, proving that the Monte Carlo routine developed is readily applicable to inhomogeneous layers and the mixing of the Mo/Si layer is well described by the T-DYN code. © 1997 by John Wiley & Sons, Ltd.

Formation of carbon nitride — a novel hard coating

Increasing efforts have been reported on the formation of carbon nitride. Vapor deposition and simultaneous ion bombardment from accelerators or plasmas (IBAD) proved to be a successful technique for the preparation of this material. In our preparation, the properties of the films were controlled by varying the nitrogen ion energy and the flux composition ratio C N . The deposited films with high nitrogen incorporation ( C N = 0.6 ∼ 0.7 ) and low implantation energies (< 1.0 keV) showed high Knoop hardnesses of up to 63 GPa. XPS and FT-IR measurements indicated a high fraction of triple bonded CN. X-ray diffraction showed an amorphous structure. Computer simulations by the dynamic TRIM code are used to study the formation parameters, nitrogen ion energy and C N ratio. This turned on to be useful in understanding the formation process of the carbon nitride films grown on silicon wafers, fused silica and tungsten carbide substrates.

Study of Low-energy Atomic Mixing by Means of Auger Depth Profiling, XTEM and TRIM Simulation on Ge/Si Multilayer System

Low-energy atomic mixing is experimentally studied on a Ge/Si amorphous multilayer system by means of Auger depth profiling and XTEM. The ion etching was performed by using Ar ions of energy 500 eV (XTEM) and 618 eV (Auger) and angle of incidence >84° and the specimen was rotated during sputtering. Applying these conditions it is believed that the interface broadening occurs mainly because of atomic mixing because the surface roughening is negligible. The Auger depth profile was carried out on a specimen containing Ge/Si strips parallel to the surface. The extent of the atomic mixing was determined by comparing the depth profile to profiles provided by dynamic TRIM simulation. For ion milling another geometry was used ; the Ge/Si strips were perpendicular to the surface. A very thin (2 nm) TEM specimen was prepared and the thickness of the completely mixed region could be directly estimated from the TEM image : 1 nm at 500 eV ion energy and 85° angle of incidence. The TEM image also shows that the atomic mixed region is asymmetric.

Analysis of carbon film growth from low energy ion beams using dynamic trajectory simulations and Auger electron spectroscopy

Carbon film growth from low energy C + ions has been investigated by classical trajectory simulations using the dynamic TRIM (TRIDYN) code and by in situ surface analysis of the evolving layers using Auger electron spectroscopy (AES). The calculations were performed for 30–300 eV C + ions impinging on Li, Si, Ni, and Au and experimental data for 150 eV C + ions on Ni are used for comparison. The simulations provide (i) concentration profiles of the carbon film, (ii) backscattering coefficients and sticking probabilities for the impinging C + ions, and (iii) sputtering yields of the composite films during deposition. The results of these simulations are in good agreement with the experimental profiles of the C and Ni concentrations as a function of C + fluence. The film growth mechanism which is in accord with these results is that of “subplantation”, in which the C + ions are deposited in subsurface layers and the outermost layer of substrate atoms is sputtered and diluted by successive impinging C + ions. The agreement between the experimental and calculated results supports the validity of the TRIM code and the binary collision approximation for 150 eV ions.

Ion-assisted selective deposition of aluminium for via-hole interconnections

Recently we have demonstrated the existence of the so-called sputter yield amplification effect, based on the preferential sputtering of a particular species during the ion bombardment of composite solids. This work presents a more elaborate study of the sputter yield amplification effect using the dynamic Trim version T-DYN. General rules are given as to when the amplification effect is most pronounced and when it is suppressed, as well as its dependence on various parameters. The potential applications of the amplification effect are also discussed. In particular, the use of the effect as a planarization technique during ion-assisted deposition of Al is studied both theoretically and experimentally. It is shown that under low-energy argon ion bombardment Al can be selectively deposited onto Al surfaces while at the same time its growth on W covered SiO 2 surfaces is totally prohibited.

Computer simulation of preferential sputtering

Static and dynamic TRIM Monte Carlo calculations are employed to study preferential sputtering. Results are restricted to two-component, amorphous, single phase materials. Total as well as partial sputtering yields, energy and angular distributions of the sputtered species, surface concentration and composition profiles are given. In addition, the fluence dependence of the surface composition and the partial sputtering yields is investigated. Recoil densities are compared with data from analytical calculations, and the isotropy of cascades is studied. Comparisons with experimental data show good agreement for compounds like TaC, but for metallic alloys the computed results do not agree with all experimental findings.