We have studied the effect of reducing the implantation energy towards low keV values on the areal density of He and H atoms stored within populations of blister cavities formed by co-implantation of the same fluence of He then H ions into Si(001) wafers and annealing. Using a variety of experimental techniques, we have measured blister heights and depth from the surface, diameter, areal density of the cracks from which they originate as functions of implantation energy and fluence. We show that there is a direct correlation between the diameters of the cracks and the heights of the associated blisters. This correlation only depends on the implantation energy, i.e., only on the depth at which the cracks are located. Using finite element method modeling, we infer the pressure inside the blister cavities from the elastic deformations they generate, i.e., from the height of the blisters. From this, we demonstrate that the gas pressure within a blister only depends on the diameter of the associated crack and not on its depth position and derive an analytical expression relating these parameters. Relating the pressure inside a blister to the respective concentrations of gas molecules it contains, we deduce the areal densities of He and H atoms contained within the populations of blisters. After low-energy implantations (8 keV He þ , 3 keV H þ), all the implanted He and H atoms contribute to the formation of the blisters. There is no measurable exo-diffusion of any of the implanted gases, in contrast to what was assumed at the state of the art to explain the failure of the Smart-Cut technology when using very low energy ion implantation for the fabrication of ultra-thin layers. Alternative explanations must be investigated. V

Hydrogen and helium co-implantation is nowadays used to efficiently transfer thin Si layers and fabricate silicon on insulator wafers for the microelectronic industry. The synergy between the two implants which is reflected through the dramatic reduction of the total fluence needed to fracture silicon has been reported to be strongly influenced by the implantation order. Contradictory conclusions on the mechanisms involved in the formation and thermal evolution of defects and complexes have been drawn. In this work, we have experimentally studied in detail the characteristics of Si samples co-implanted with He and H, comparing the defects which are formed following each implantation and after annealing. We show that the second implant always ballistically destroys the stable defects and complexes formed after the first implant and that the redistribution of these point defects among new complexes drives the final difference observed in the samples after annealing. When H is implanted first, He precipitates in the form of nano-bubbles and agglomerates within Hrelated platelets and nano-cracks. When He is implanted first, the whole He fluence is ultimately used to pressurize H-related platelets which quickly evolve into micro-cracks and surface blisters. We provide detailed scenarios describing the atomic mechanisms involved during and after coimplantation and annealing which well-explain our results and the reasons for the apparent contradictions reported at the state of the art. V

No abstract

The so-called Direct Wafer Bonding (DWB) technique opens new possibilities for the electronic industry but still suffers from the poor knowledge we have of the microstructure of these interfaces and hence of their electrical activity. In this work, we have extensively used Transmission Electron Microscopy techniques in plan-view and cross-section to identify the structure of the interfaces found between two bonded silicon wafers. The general structure of these interfaces is that of a perfect grain boundary and evidently depends on the misorientation between the two bonded wafers. A twist component in the range 0>θ>13˚ creates a square network of pure screw dislocation whereas an unavoidable tilt component (<0.5˚) is compensated by a periodic array of 60˚ dislocation lines perpendicular to the tilt direction. Therefore, the regularity of these networks can be disrupted by the presence of steps (of up to several nanometers) in the interface plane. Silicon oxide precipitates are seen heterogeneously distributed on the interface with preferential nucleation sites on the dislocations.

No abstract

Particular defect structures appear during directional solidification of the Al(fcc)-Al 2 Cu(q) lamellar eutectic composite, the so-called lamellar faults. Each fault corresponds to terminations of Al and q lamellae. These terminations consist of a continuous succession of three types of ribbon-like interfaces, among which two are low angle boundaries: one is located in the fcc phase and the other in the q intermetallic (C16 standard type, a = 0.6063 nm, c = 0.4872 nm). High resolution transmission electron microscopy (HRTEM) of a q/q low angle tilt boundary with an electron beam parallel to h113iq shows that the core structures of the dislocations are undissociated despite the large magnitude of the possible Burgers vectors. The dislocations are of mixed-type character and oriented along the h113iq direction the closest to the growth axis direction of the eutectic. An unusual defect present in the tilt boundary consists of a dense group of three dislocations separated by nanometric distances. This defect acts like a single dislocation at far field distance. The mechanical equilibrium of such a defect has been studied using isotropic and anisotropic elasticities. It is strongly suggested that two of these dislocations form a dipole, which would indicate that the boundary has not a completely relaxed structure.

To weight the importance of a nanocavity buffer in a SiGe deposition substrate, some P type (001) FZ Si wafers are implanted (A samples) or not (B samples) at room temperature with 5×10 16 He + cm -2 at 10keV. They are annealed at 700°C for one hour to form a nanocavity layer close to the Si surface. Then, the wafers are carefully chemically cleaned in a clean room to remove both organic and metallic impurities from the surface. They are coated either by 210 nm (A) or 430 nm (B) Si 1-x Ge x (x=0.20±0.02) alloy grown at 575°C for 0.42 hour by low pressure chemical vapor deposition (LP-CVD) with a growth rate of 8 to 17 nm.mn -1 . Both kinds of samples are studied by cross section transmission electron microscopy, X-rays diffraction, Rutherford backscattering, atomic force microscopy and etch pit counts. The association of these techniques demonstrates that the thin SiGe layer which is deposited on sample A is fully relaxed and that the threading dislocation density (estimated to hardly reach 4×10 3 cm -2 ) is at least one order of magnitude lower than what is obtained so far using ion implantation assistance in SiGe layer growth on silicon. The roughness of the SiGe surface is low enough to stand a further Si epitaxy. Nevertheless, the mechanism involved responsible for the threading dislocation annihilation and/or confinement is still unclear.

Based on the Lambert W-function, an exact analytical solution for the critical thickness of a lattice-mismatched heteroepitaxial layer is presented. The new expression in exact and algebraic closed form eliminates the need for complex iterative computation. Its high accuracy is proved by comparison of the calculated critical thickness versus fractional atomic content of an alloy epilayer with the respective numerical solution.