Cu thin films with a thickness around 850 Å were prepared on Ti45N55/Ti/Si(100) substrates at room temperature by partially ionized beam deposition (PIBD) with an ion energy of 3 keV at pressure of 8×10−7–1 × 10−6 Torr. The Cu/Ti45N55/Ti/Si samples were annealed at 8× 10−6–1 × 10−5 Torr with annealing temperature of 500 to 700 °C for 30 min.. Thermal stability of the PIB-Cu films was investigated with Rutherford backscattering spectrometry (RBS), Auger electron spectroscopy (AES), Scanning electron microscopy (SEM), and X-ray diffraction (XRD).

The as deposited Cu films had a (111) texture and there was no change of phase in annealed Cu films regardless of annealing temperature. Grain size of the annealed Cu films increased with annealing temperature. SEM studies show no hillock and no voiding on the Cu film surface up to annealing temperature of 700 °C. For PIB-Cu/Ti45N55/Ti/Si samples, all the layers were intact and there was no indication of interdiffusion by conventional depth profiling techniques (RBS, AES) up to 700 °C in vacuum for 30 minutes.

The stability of cobalt silicide formed on narrow polysilicon line degraded rapidly with shrinkage of linewidth as shown by the increase of sheet resistance after annealing. Stability was improved by adding a titanium interlayer under cobalt with or without a Ti or TiN cap layer. The stability was affected by silicide thickness and substrate doping. In best cases an increase of sheet resistance after annealing was still observed. Adding an Ar ion bombardment step improved the thermal stability drastically. This structure did not show any sheet resistance change after a 30 min annealing at 850°C, either with or without interlayer or cap layer.

A plasma immersion ion implantation (Pill) hydrogenation process using an inductively-coupled plasma (ICP) source is implemented for defect passivation in polycrystalline silicon (poly-Si) thin film transistors (TFT's). Device parameter improvement saturates in 4 minutes, which is considerably shorter than for other reported hydrogenation methods. Stress test indicates that the devices hydrogenated by this novel technique have much better long-term reliability. The hydrogenation effects on two types of trap states are analyzed the current-voltage characteristics of the devices. The densities of deep states and tail states are significantly reduced after short time hydrogenation.

Since Fall 1995, a state-of-the-art intermediate voltage electron microscope (IVEM) has been operational in the HVEM-Tandem Facility with in situ ion irradiation capabilities similar to those of the HVEM of the Facility. A 300 kV Hitachi H-9000NAR is interfaced to the two ion accelerators of the Facility, with a demonstrated point-topoint spatial resolution for imaging of 0.25 nm with the ion beamline attached to the microscope. The IVEM incorporates a Faraday cup system for ion dosimetry with measurement aperture 6.5 cm from the TEM specimen, which was described in Symposium A of the 1995 MRS Fall Meeting. The IVEM is now employed for a variety of in situ ion beam studies ranging from low dose ion damage experiments with GaAs, in which damage zones individual displacement cascades are observed, to implantation studies in metals, in which irradiation-induced noble gas precipitate mobility is studied in real time. In this presentation, the new instrumentation and its specifications will be described briefly, several basic concepts relating to in situ experiments in transmission electron microscopes will be summarized and examples of in situ experiments will be presented which exploit the experimental capabilities of this new user facility instrumentation.

We have carried out systematic experiments to explore the range of applicability of the two direct methods which are available for the determination of the nature of small clusters. Experiments in heavy-ion irradiated silver and copper, including direct comparisons with the results of Black-White contrast analysis (B-W), forced us to the conclusion that use of the 21/2D technique for the analysis of faulted loops is effectively impossible because of reciprocal-lattice spike and other effects. We used B-W analysis to investigate the nature of the clusters produced by room-temperature heavy-ion irradiation of copper. In-situ irradiations were performed in the Argonne HVEM-Tandem Facility using 40keV and 80keV Kr+ ions. Nearly all of the analysable first-layer loops were found to be vacancy in nature. Some deeper clusters were also certainly vacancy. There is no unequivocal evidence for interstitial loops under these conditions. The near impossibility of establishing that interstitial loops are not present is shown by the fact that a large fraction of loops - over 50% - could not be analysed either because they did not display clear B-W contrast under any condition or sufficiently consistent B-W contrast under several different operating conditions.

Atomic scale computer simulations are used to investigate the intracascade evolution of the defect distributions of cascades in copper over macroscopic times. Starting with cascades generated using molecular dynamics, the diffusive transport and interactions of the defects are followed for hundreds of seconds in stochastic annealing simulations. The temperature and energy dependencies of annihilation, clustering and free defect production are determined for individual cascades. The simulation results illustrate the strong influence on intracascade evolution of the defect configuration existing in the primary damage state. Other significant factors affecting the evolution of the defect distribution are the large differences in mobility and stability of vacancy and interstitial defects and the rapid one-dimensional diffusion of small, glissile interstitial loops produced directly in cascades. The latter factor introduces a cascade energy dependence of defect evolution that is apparent only beyond the primary damage state, amplifying the need for performing many more cascade simulations at higher energies.

Molecular dynamics (MD) simulations have been used to study the number and arrangement of defects produced by displacement cascades as functions of irradiation temperature, Tirr, in α-iron. The continuum treatment of heat conduction was used to adjust the temperature of the MD boundary atoms throughout the cascade process. This new hybrid model has been applied to cascades of either 2 or 5 keV at 100K, 400K, 600K and 900K. The number of Frenkel pairs decreases by about 20–30% as Tir increases from 100K to 900K, due to the increase in the lifetime of the thermal-spike phase. The same effect also brings about an increase in the proportion of selfinterstitial atoms that form clusters.

Neutrons and high-energy ions incident upon a solid can initiate a displacement collision cascade of lattice atoms resulting in localized regions within the solid containing a high concentration of interstitial and vacancy point defects. These point defects can collapse into various types of dislocation loops and stacking fault tetrahedra (SFT) large enough that their lattice strain fields are visible under diffraction-contrast imaging using a Transmission Electron Microscope (TEM). The basic mechanisms driving the collapse of point defects produced in collision cascades is investigated in situ with TEM for fcc-Cu irradiated with heavy (100 keV Kr) ions at elevated temperature. The isothermal stability of these clusters is also examined in situ.

Areal defect yields were observed to decrease abruptly for temperatures greater than 300°C. This decrease in defect yield is attributed to a proportional decrease in the probability of collapse of point defects into clusters. The evolution of the defect density under isothermal conditions appears to be influenced by three different rate processes active in the decline of the total defect density. These rate constants can be attributed to differences in the stability of various types of defect clusters and to different loss mechanisms. Based upon observed stabilities, estimations for the average binding enthalpies of vacancies to SFT are calculated for copper.

The defect microstructures and tensile properties of high purity copper single crystals have been studied after irradiation at room temperature with 600 MeV protons in the dose range from 10-3 to 10-1 dpa. The defect number density N of clusters shows a linear dependence with dose up to a saturation value of 4×1023 m-3, with a broad transition region between doses of 3×10-3 and 10-1 dpa. The increase in critical resolved shear stress (CRSS) induced by the irradiation is proportional to the square root of the dose, up to a dose of 7×10-3 dpa. These results are compared to those of both fission and fusion neutron irradiations, showing that in room temperature irradiations at low fluences, no recoil spectra effects can be detected.

Stereo transmission electron microscopy has been used to characterise the distribution in depth of disordered zones and associated dislocation loops in the ordered alloys Ni3Al and Cu3Au after heavy ion irradiation, most extensively for Ni3Al irradiated with 50 keV Ta+ ions at a temperature of 573 K. The Cu3Au specimen was irradiated with 50 keV Ni+ ions at an incident angle of 45° at a temperature of 373 K. In Ni3Al the defect yield, i.e. the probability for a disordered zone to contain a loop was found to be strongly dependent on the depth of the zone in the foil, varying from about 0.7 for near-surface zones to about 0.2 in the bulk. The sizes and shapes of disordered zones were only weakly dependent on depth, except for a small population of zones very near the surface which were strongly elongated parallel to the incident ion beam. In Cu3Au the surface had a smaller but still significant effect on the defect yield. The dependence of the tranverse disordered zone diameter d on ion energy E for Ta+ irradiation of NiA was found to follow a relationship d = k1, E1/α with k, = 2.4 ± 0.4 and α = 3.3 ± 0.4. A similar relationship with the same value of α is valid for a wide variety of incident ion/target combinations found in the literature.

Defect clusters in Al during electron and ion irradiation have been investigated using highresolution transmission electron microscopy (HRTEM). An ION/HVEM system which consists of a high-voltage TEM and ion implanters was used for in-situ observation of damage evolution under 1000 keV electrons and 15 keV He+ irradiation at room temperature. HRTEM of Al in [110] orientation showed many planar defects along { 111 } planes during electron irradiation, while a high density of small polyhedron-shaped cavities (He-bubbles) was observed in addition to the planar defects after He+ irradiation. Multi-slice image simulation of various models of dislocation loops indicated the planar defect as an interstitial-type Frank loop.

The role of helium on the microstructural evolution of V-3.8Cr-3.9Ti has been investigated by in-situ transmission electron microscopy observations of as-prepared and He implanted (<10 appm) samples subjected to 200keV He irradiation at room temperature. Quantitative analysis of the defects showed an increase in the defect density and size with irradiation in both types of samples. The unimplanted sample showed a defect density consistent with electron irradiation experiments. In comparison, the He preimplanted sample had slightly larger defects and a substantially greater increase in the number density of defects. This result is consistent with a mechanism of He trapping by the formation of He-vacancy-X (X= C, N, O) complexes.

The V-4Cr-4Ti alloy displays excellent mechanical properties, including a ductile-to-brittle transition temperature (DBTT) below -200°C in the unirradiated condition. Specimens of this alloy were fission neutron-irradiated in the High Flux Beam Reactor (HFBR) at Brookhaven National Laboratory (BNL) to a dose of 0.4 dpa at temperatures from 100 to 275°C. Mechanical testing showed significant irradiation hardening in the alloy which increased with increasing irradiation temperature. Charpy impact testing also showed a dramatic increase in the DBTT on the order of 200 to 350°c. The mechanical property changes are correlated with preliminary results from the transmission electron microscopy (TEM) analysis of the defect microstructure resulting from the low-dose neutron irradiations. The TEM analysis of the irradiated material showed a nearly constant defect density of ∼1.6 × 1023 m−3, with an average defect diameter of slightly greater than 3 nm.

V-4 wt.% Cr-4 wt.% Ti alloy is a promising candidate material for first-wall and structural applications in magnetic fusion reactors. In the past, fast neutron sources were used to evaluate postirradiation properties of fusion candidate materials. The recent shutdowns of the Fast Flux Test Facility (FFTF) and the Experimental Breeder Reactor (EBR-II) left U.S. researchers without local experimental facilities for such tests. Under such circumstances, ease of experimental control, availability, and relatively low cost make heavy-ion irradiation an attractive alternative, provided its limitations are appreciated. We selected 3-MeV V+ and 4.5-MeV Ni++ ions to investigate the effects of irradiation on the microstructure of V-4Cr-4Ti alloy in the temperature range of 200 – 420°C. The main interest is in the evaluation of this alloy's dimensional stability and susceptibility to irradiation embrittlement. In this paper, we report results of ion irradiation experiments and compare them with available data on fast-neutron irradiation. From transmission electron microscopy (TEM) analysis of ion-irradiated specimens, we found that the dominant feature of the postirradiation microstructure was a high density of dislocation loops and pointdefect clusters. Density and defect size depended on irradiation dose and temperature. Precipitates and voids/bubbles were not observed, even in specimens that were simultaneously injected with He and exposed to heavy-ion irradiation. Increased transport of point defects to internal interfaces was observed, as manifested by defect denuded zones along grain boundaries, Defect denuded zones along grain boundaries could lead to segregation of impurities and solutes and formation of precipitates on grain boundaries.

In situ transmission electron microscopy has been used to follow room temperature irradiation of thinned bulk and 62 nm thick gold films with Xe ions in the energy range 50 to 400 keV. Energy spikes resulting from single ion impacts give rise to surface craters and holes which exist until annihilation by subsequent ion impacts. Video recording provided details with a time-resolution of 33 milliseconds. Craters were produced on the irradiated surface at all ion energies and on the opposite surface when the ions had sufficient energy to traverse the specimen. Crater sizes were as large as 12 nm for the higher energy irradiations. On average, approximately 6% of impinging ions result in craters.

A single 200 keV Xe ion may produce a hole in thin gold foils. Hole formation involves the movement by plastic flow of massive amounts of material; on the order of tens of thousand Au atoms per ion impact. Individual ion impacts also result in a filling of both holes and craters as well as a thickening of the gold foil. Change in morphology during irradiation is attributed to a localized, thermal-spike induced melting, coupled with plastic flow under the influence of surface forces.

The changes in surface topography of thin metallic films during ion bombardment were investigated with STM. Pd and Au thin films, having defect-free, atomically flat areas over at least 300×300Å2 were bombarded with Ar+, Kr+, Xe+ ions. Ion energies and target temperatures were varied. The roughening after high fluence irradiation shows scaling with a roughness exponent 0.84. For 300 K target temperature we observe, both for Au and Pd, 3 D erosion, for 700 K, on the contrary, layer-by-layer erosion. The differences in erosion morphologies between Au and Pd surfaces are supposed to stem from differences in surface mobility of adatoms.

High-energy cascades have been simulated in gold using molecular dynamics with a modified embedded atom method potential. The results show that both vacancy and interstitial clusters form with high probability as a result of intracascade processes. The formation of clusters has been interpreted in terms of the high pressures generated in the core of the cascade during the early stages. We provide evidence that correlation between interstitial and vacancy clustering exists.

(Ni0.67Cr0.33)1-xFex alloys with x = 0, 5 and 10 at.%, have been irradiated at temperatures between 300°C and 500°C with the electrons of a 2.5MeV Van de Graaff accelerator and of a 1MeV High Voltage Electron Microscope (HVEM). All of them have been successively ordered at short and at long range under irradiation. The transition temperature is 525°C for Ni0.61Cr0.34Fe0.05 and 445°C for Ni0.59Cr0.31Fe0.10. An Inconel sample irradiated up to 1 d.p.a. in the 1MeV microscope did not show any superlattice peak although its composition is very close to the 10% iron model alloy. The kinetics of ordering has been followed by resistivity measurements and fitted by the Dienes kinetic model.

Atom transport under irradiation is determined by the concentration of freely migrating defects, which depends on the dynamical equilibrium between production and annihilation rates. In order to determine effective values of both of these quantities for the case of ion irradiation, spatially resolved self-diffusion measurements were performed on single crystals of nickel which contained several thin tracer layers at different depths. By fitting the solution of a diffusion equation to the depth dependent measurements effective fractional production rates of freely migrating defects and effective sink strengths have been obtained.

A computer program MODEX has been developed to predict the evolution of the irradiation defects in materials during ion implantation and subsequent annealing. The evolution of the defect clusters is simulated through a number of trapping and dissociation events. Two examples of the program application are presented: (1) 1 keV helium ion implantation in molybdenum up to doses of 1013 cm-2 with the subsequent ramp annealing up to 1400 K; and (2) 2.5 keV helium ion implantation in silicon with 1016 cm-2 with ramp annealing up to 1600 K. The simulation results are compared to the experimental results obtained by gas desorption techniques. The threshold dose needed for micro-cavity formation in silicon is discussed.