Mechanisms of ion-beam-enhanced diffusion in amorphous silicon

We have investigated ion-beam-enhanced diffusion of Au in undoped and B doped amorphous Si. The diffusion coefficients depend linearly on ion flux and exibit an Arrhenius-like temperature dependence with an activation energy of 0.37 eV in the temperature range 200–350° C. Moreover the diffusivity is enhanced by a factor of 5 by B-doping at a concentration of 1×10²⁰ atoms/cm³. A similar enhancement is observed in thermal diffusion of Au which has an activation energy of 1.5 eV. On the basis of these results a model for the ion-beam-enhanced diffusion of Au is proposed where the high density of defects present in amorphous Si act as traps for the fast moving interstitial Au atoms. The effectiveness of this trapping process can be changed by the high concentration of mobile defects generated by the beam and also by a change in the charge state of the traps induced by the presence of B.     

The characteristics of platinum diffusion in n-type GaN

The electrical and optical characteristics of platinum (Pt) diffusion in n-type gallium nitride (GaN) film are investigated. The diffusion extent was characterized by the SIMS technique. The temperature-dependent diffusion coefficients of Pt in n-GaN are 4.158 × 10⁻¹⁴, 1.572 × 10⁻¹³ and 3.216 × 10⁻¹³ cm²/s at a temperature of 650, 750 and 850 °C, respectively. The Pt diffusion constant and activation energy in GaN are 6.627 × 10⁻⁹ cm²/s and 0.914 eV, respectively. These results indicate that the major diffusion mechanism of Pt in GaN is possibly an interstitial diffusion. In addition, it is also observed that the Pt atom may be a donor because the carrier concentration in Pt-diffused GaN is higher than that in un-diffused GaN. The optical property is studied by temperature-dependent photoluminescence (PL) measurement. The thermal quenching of the PL spectra for Pt-diffused GaN samples is also examined.     

Application of the modified electrostatic model to the impurity diffusion in nickel

The modified electrostatic model (Neumann and Tölle 1995) is applied to the impurity diffusion in nickel.Z ₀ = 0.4 is used for the effective charge of the nickel ion.The comparison of calculated and experimental diffusion parameters reveals that the sign of J Q, the difference between impurity diffusion and self-diffusion energy, and the sign of the difference between impurity diffusion and self-diffusion coefficient is correctly predicted in all cases. On the other hand the comparison exhibits some systematic deviations for 5p impurities, which cannot be explained in terms of the current impurity diffusion models.     

Annealing behaviour of implanted nitrogen in bulk and evaporated Ni

The diffusion coefficients of nitrogen in N-implanted polycrystalline Ni have been deduced. Both bulk and Ni-evaporated samples implanted with nitrogen were annealed at 150–500° C. The nitrogen profiles were probed using the nuclear resonance broadening technique. The value of 0.99±0.12 eV for the activation energy and (3.0 _–2.8 ⁺⁴⁰ )×10^–6 cm²/s for the frequency factor were obtained for implanted N in bulk Ni. The solubilities for both the bulk and evaporated Ni samples are given. In evaporated Ni nitrogen migration is enhanced due to the defects arising during evaporation.     

Radiation-enhanced diffusion in ion-implanted glass

The Rutherford backscattering technique was used to characterize the purposedly added cesium impurities in soda-lime glass. The impurities were introduced into the glass matrix by an ion-exchange diffusion process at room temperature. The diffusion coefficient of cesium was determined from the measured depth profiles. The diffusion of the cesium impurities stimulated by 280 keV Kr⁺-ion beam irradiation was also studied. The depth distributions obtained are discussed using the model of radiation enhanced diffusion. Results are compared with theoretical values based on transport of ions in matter calculations and other experimental work.On sabbatical leave at Applied Science University, Physics Department, P. O. Box 926296, Amman, Jordan     

Field-enhanced neutralization of electrically active boron in hydrogen implanted Schottky diodes

We have observed that hydrogen implantation in p-type boron-doped silicon material induces a neutralization of boron in a 10 m deep region after the Schottky diodes have been heated at a 90° C temperature under reverse-biasing. The profile of neutralized acceptors can be reversibly shaped by successively applying different reverse biases.     

Microstructure and thermal stability of Fe,Ti, and Ag implanted yttria-stabilized zirconia

Yttria-stabilized zirconia (YSZ) was implanted with 15 keV Fe or Ti ions up to a dose of 8×10¹⁶ at cm^–2. The resulting dopant concentrations exceeded the concentrations corresponding to the equilibrium solid solubility of Fe₂O₃ or TiO₂ in YSZ. During oxidation in air at 400° C, the Fe and Ti concentration in the outermost surface layer increased even further until a surface layer was formed of mainly Fe₂O₃ and TiO₂, as shown by XPS and ISS measurements. From the time dependence of the Fe and Ti depth profiles during anneal treatments, diffusion coefficients were calculated. From those values it was estimated that the maximum temperature at which the Fe- and Ti-implanted layers can be operated without changes in the dopant concentration profiles was 700 and 800° C, respectively. The high-dose implanted layer was completely amorphous even after annealing up to 1100° C, as shown by scanning transmission electron microscopy. Preliminary measurements on 50 keV Ag implanted YSZ indicate that in this case the amorphous layer recrystallizes into fine grained cubic YSZ at a temperature of about 1000° C. The average grain diameter was estimated at 20 nm, whereas the original grain size of YSZ before implantation was 400 nm. This result implies that the grain size in the surface of a ceramic material can be decreased by ion beam amorphisation and subsequent recrystallisation at elevated temperatures.     

High-temperature diffusion of phosphorus and boron in silicon via vacancies or via self-interstitials?

The paper re-examines the effect of oxidation on the diffusion of phosphorus and boron in silicon as well as recent results on redistribution phenomena of these dopants under irradiation and on the emitter-push effect. It is shown that at high temperatures phosphorus and boron diffuse via a defect mechanism involving silicon self-interstitials. These results support the view-point that self-interstitials are the dominating point defects in silicon under thermal equilibrium conditions. Possible generation mechanisms for the self-interstitial supersaturation causing the emitter-push effect are suggested.     

Lead diffusion in α-Zr

Pb diffusion in α-Zr matrix between 823 and 1123 K was measured using heavy ion Rutherford backscattering spectrometry (HIRBS) technique. A deviation from the Arrhenius law was observed, with two different regions. At low temperatures the activation energy Q is close to the expected value for a substitutional diffuser but the pre-exponential factor D₀ is higher than expected. Close to the phase transition temperature the opposite occurs, with a low Q value. This behavior is similar to the one observed for Hf and self-diffusion in α-Zr. Received: 29 September 1998 / Accepted: 29 January 1999 / Published online: 28 April 1999     

Analysis of temperature and enhanced diffusion effects in sputtering of CrSi2

Previously reported experimental results on sputtering and enhanced diffusion processes in CrSi₂ during 100 keVXe⁺ bombardment at different temperatures have been quantitatively analyzed.The framework for the analysis is a simple theoretical model in which the Si atoms are considered mobile in a matrix of Cr atoms whose density is assumed constant and diffusivity is considered zero everywhere. Erosion velocity of the matrix (due to the sputtering of Cr atoms), sputtering and enhanced diffusion processes of Si atoms are taken into account in the mathematical model.In our analysis we show that the time evolution of the total number of sputtered atoms in binary solids cannot yield an unambigous conclusion as to the existence of preferential sputtering.Further, it is found that in the case of CrSi₂ the preferential sputtering of Si atoms depends on the suicide temperature.     

Oxygen diffusion in C60 films

We have performed detailed resistivity measurements as a function of temperature in the range from 12 to 300 K on oxygen loaded C₆₀ films. We observe that two ordering phase transitions (i.e.,T ₀=260 K andT _g =90 K) are present in (T), which, in addition, strongly depends on the oxygen content. We find a decrease of both ordering temperatures with increasing oxygen concentrations. The mechanisms of oxygen diffusion are greatly enhanced in the ordered phase on heating. Finally, the transition to a glassy state atT _g is detected as a point of reversibility of the resistivity curve as a function of temperature.     

On the migration of defects and impurities in microelectromechanical systems subject to a large number of light pulses

The migration of defects in light-irradiated microelectromechanical systems (MEMS) is treated theoretically. The effects of temperature gradients on atomic demixing are considered. It is found that, when the migration energy of defects and impurities is less than about 0.5 eV, their migration has to be introduced in the modelling of ageing of MEMS. Received: 17 July 2000 / Accepted: 31 March 2000 / Published online: 27 June 2001     

Hydrogen implantation and diffusion in silicon and silicon dioxide

Depth profiles of hydrogen implanted into crystalline silicon in random direction at different fluences have been measured by the¹⁵N technique and by SIMS. Whereas hydrogen implanted at a fluence of 10¹⁵ ions/cm² shows some limited mobility, no such mobility is observed for higher implantation fluences. In these cases, ballistic computer codes describe the depth distributions well, within the ranges of both experimental and theoretical accuracy. Annealing up to 510 K does not change the hydrogen distributions.Furthermore, high-fluence hydrogen implantation into silicon dioxide has been examined. There is some indication for radiation-enhanced diffusion during the implantation process. Upon subsequent thermal annealing, the hydrogen is found to diffuse, probably via a trapping/detrapping mechanism associated with an OH/H₂ transformation of the hydrogen bonding.     

Au diffusion in α-Ti

The solubility and diffusion of Au in -Ti have been studied in a 823–1023 K temperature range using the Rutherford backscattering technique. For this purpose we have implanted Au into -Ti samples. Our results show that the solubility of Au varies between 0.2 and 0.35 at.%. In addition, we found that the diffusion coefficients follow a normal Arrhenius behavior with Q=260 kJ/mol and D _o=1.9×10^–5 m²/s¹. These values are typical for a substitutional diffusion mechanism.     

Advanced diffusion studies with isotopically controlled materials

The use of enriched stable isotopes combined with modern epitaxial deposition and depth profiling techniques enables the preparation of material heterostructures, highly appropriate for self- and foreign-atom diffusion experiments. Over the past decade we have performed diffusion studies with isotopically enriched elemental and compound semiconductors. In the present paper, we highlight our recent results and demonstrate that the use of isotopically enriched materials ushered in a new era in the study of diffusion in solids, which yields greater insight into the properties of native defects and their roles in diffusion. Our approach of studying atomic diffusion is not limited to semiconductors and can be applied also to other material systems. Current areas of our research concern the diffusion in the silicon-germanium alloys and glassy materials such as silicon dioxide and ion conducting silicate glasses.     

Diffusion of hydrogen in heterogeneous systems

The effective long-range long-time tracer diffusivity D_eff for interstitial diffusion of hydrogen through heterogeneous systems was studied theoretically for model systems consisting of isolated grains of material G embedded in a matrix of material M. Different solubilities of hydrogen in these two materials as well as different diffusivities are allowed for. Additionally, modified diffusion barriers at the phase boundaries were included in the diffusion model. The effect of different sizes, arrangements, and forms of the grains was also considered. D_eff was determined by Monte Carlo (MC) simulations on simple lattice models of the systems described above. An equilibrium distribution of hydrogen atoms among the two constituent materials was assumed. Our main interest was focused on whether and how D_eff may be related to mesoscopic or macroscopic quantities characterizing the heterogeneous system and its constituent materials, such as the volume fractions of the two materials, the fraction of lattice sites in the immediate vicinity of the phase boundary, the hydrogen concentrations c_G and c_M in the grains and in the matrix and the respective hydrogen diffusivities D_G(c_G) and D_M(c_M). In order to obtain good estimates for these relations in terms of analytic formulas, we attempted to model a heterogeneous system by a network of diffusion elements connected in series and in parallel, in analogy to an electric network. The properties of the basic connections, in parallel and in series, were studied on layered structures, for which analytic expressions for D_eff could be derived. The network formulas for different grain–matrix systems were tested by comparing with results of MC simulations. In general, the network formulas describe the corresponding MC results for D_eff fairly well. It was found that differences in the hydrogen solubilities in the two phases as well as modified energy barriers at the phase boundaries may have dramatic effects on D_eff. Received: 19 September 2000 / Accepted: 6 November 2000 / Published online: 9 February 2001     

Ion mixing and thermochemical properties of markers in Cu

Markers of Nb, Ru, Ag, In, Sb, Hf, Pt, Au, and Bi in Cu were mixed by irradiation with 750 keV Kr at 77 K and analyzed in situ by backscattering of 1.9 MeV He⁺. Cu with Pt and Au markers were also irradiated and analyzed at 7 K. The results were identical to those obtained at 77 K results. The measured mixing efficienciesDt/øF _D , for the various markers correlate with their respective impurity tracer diffusivities and impurity-vacancy binding energies in Cu. The correlation suggests that diffusion by a vacancy mechanism during a thermal spike as an important process in ion mixing of marker atoms in Cu.     

Ion-beam mixing in pure and in immiscible copper bilayer systems

Ion-beam mixing was measured in immiscible Cu bilayer systems after Kr irradiation at 6 K and at 295 K. It was observed that for the systems which form miscible liquids but which have limited solid solubility, Cu-Nb and Cu-Bi, mixing occurs at 6 but not at 295 K. For a system which is not miscible in either the solid or liquid state, mixing does not occur at either 6 or 295 K. Mixing was also measured in pure Cu isotope bilayer specimens,⁶³Cu–⁶⁵Cu, to provide a standard for the other measurements. The results are interpreted on the basis of an atomistic model of ion beam mixing. The model assumes that point defects are created in the initial phases of the cascade evolution, and that these defects migrate during the later thermal spike phase.Work supported by the U.S. Department of Energy     

Modeling of polysilicon diffusion sources during rapid optical annealing

A new model for polysilicon diffusion sources is presented. It considers the following effects: 1) dopant diffusion in grains, in grain boundaries and in the single-crystal silicon substrate, 2) dynamic dopant segregation between grain and grain boundary phases and between the phases of polysilicon and the single-crystal silicon substrate, 3) dynamic dopant activation or clustering in grains and in the single-crystal silicon substrate, 4) dynamic grain growth depending on local grain size and local electron density. These mechanisms with completely different time scales are modeled simultaneously. For the first time this allows the analysis of furnace and rapid optical annealing processes with arbitrary grain growth kinetics even during epitaxial realignment. The advanced model for segregation allows for the effect that dopants in grain boundaries and active dopants in grains as well as in the single-crystal silicon substrate find only a limited number of sites which can be occupied. These limitations are necessary to explain the dopant distributions in polysilicon and in the single-crystal silicon substrate. For the first time the coupling between the concentration of active dopants in grains, between the concentration of dopants in grain boundaries and between the local grain size is shown during doping enhanced grain growth.     

Determination of the copper diffusion coefficient in silicon from transient ion-drift

We use the transient ion drift in a depletion region of a Schottky barrier to determine ion diffusivities at moderate temperatures. The pulsed reverse bias leads to temperature dependent capacitance transients similar to deep level carrier emission transients. A simple theoretical model together with classical transient signal analysis provide the means to extract the ion diffusion constant. When applied to copper in silicon, diffusion data are obtained in a not yet investigated temperature range (280–400 K) which agree well with both low and high temperature diffusion data.