Search

The effect of the presence of a passivation layer on a metal rough surface during contact loading is investigated by means of dislocation dynamics simulations. The metal body is modeled as an FCC single crystal with a self-affine rough surface that is either bare, or covered by a thin coating, impenetrable to dislocations. This analysis permits to isolate the effect of surface roughening driven by dislocation motion: when the surface is bare the dislocations can glide out, leaving crystallographic steps at the surface that modify the local roughness; when the surface is passivated, dislocations are stopped by the interface., This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement no. 681813)., Peer Reviewed, Postprint (published version)

Doctorat en Sciences, info:eu-repo/semantics/nonPublished

Doctorat en Sciences, info:eu-repo/semantics/nonPublished

The main content of this thesis is divided into two parts which relate to the two fields of surface science: surface desorption and surface diffusion. In the first part, we have investigated the desorption of diatomic (CO) and polyatomic (NH3, ND3) molecules from Pt(111) surface using femtosecond laser. We find that the femtosecond laser-induced desorption of ammonia is very different from that of carbon monoxide from Pt( 111). The strong laser pulse width dependence of carbon monoxide desorption shows CO desorbs via a hot-electron driven process while NH3(ND3) desorbs by the conventional thermal process even under femtosecond laser irradiation. Little isotope effect between the desorption of NH3 and ND3 also supports our conclusion about ammonia desorption. As we critically examined femtosecond desorption of CO from Pt(111), we find the short-lived antibonding states at lower energy predicted by the “Adiabatic Model” of Jennison et.al. could be excited by the photoelectrons, and the excitation of these states would allow desorption to occur via DIMET mechanism. In the second part, we have studied the surface diffusion of H on W(100) surface. From the low energy electron diffraction (LEED) studies, it is known that the W(100) surface undergoes a reversible phase transition between a c(2x2) structure at low temperature and a (1x1) disordered structure at high temperature with the hydrogen present up to a coverage of 0.3 ML. Our diffusion data show that an anomalous reduction of diffusion occurs near the phase transition temperature(Tc). Combining the diffusion data with the LEED patterns, we conclude that the H diffusion near Tc is affected by phase transition of W(100) substrate which is predicted by previous theoretical studies.

The main content of this thesis is divided into two parts which relate to the two fields of surface science: surface desorption and surface diffusion. In the first part, we have investigated the desorption of diatomic (CO) and polyatomic (NH3, ND3) molecules from Pt(111) surface using femtosecond laser. We find that the femtosecond laser-induced desorption of ammonia is very different from that of carbon monoxide from Pt( 111). The strong laser pulse width dependence of carbon monoxide desorption shows CO desorbs via a hot-electron driven process while NH3(ND3) desorbs by the conventional thermal process even under femtosecond laser irradiation. Little isotope effect between the desorption of NH3 and ND3 also supports our conclusion about ammonia desorption. As we critically examined femtosecond desorption of CO from Pt(111), we find the short-lived antibonding states at lower energy predicted by the “Adiabatic Model” of Jennison et.al. could be excited by the photoelectrons, and the excitation of these states would allow desorption to occur via DIMET mechanism. In the second part, we have studied the surface diffusion of H on W(100) surface. From the low energy electron diffraction (LEED) studies, it is known that the W(100) surface undergoes a reversible phase transition between a c(2x2) structure at low temperature and a (1x1) disordered structure at high temperature with the hydrogen present up to a coverage of 0.3 ML. Our diffusion data show that an anomalous reduction of diffusion occurs near the phase transition temperature(Tc). Combining the diffusion data with the LEED patterns, we conclude that the H diffusion near Tc is affected by phase transition of W(100) substrate which is predicted by previous theoretical studies.

The fcc metals gold, silver, platinum and palladium absorb and form surfilce alloys with the bee transition metal molybdenum, although they do not form bulk alloy with molybdenum. The surface of bulk molybdenum is unstable and readying reconstructs to a c(2x2), we found that it is also possible for the surface to form a c(2x2) surface vacancy. The Density of States ( DOS ) of surface atoms of a p(1x1) Mo(100) show a peak near the Fermi Energy, rather than a dip of those atoms which are beneath the surface. The very stable surface alloys of Au/Mo(100), Ag/Mo(100), Pt/Mo(100) and Pd/Mo(100) are due to the quenching of the surface state near the Fermi level by chemical bonding between the fcc metal and the surface Mo atoms. The high surface DOS near EF of molybdenum can be quenched in two ways, namely the formation of a c(2x2) surface vacancy array, and the formation of a c(2x2) surface alloy. We found high surface DOS near EF of molybdenum that reappears in a cyclic deposition of monolayers of silver or gold as the absorbate, which has been observed in previous experiments. We found that for platinum and palladium, the surface high DOS neax EF persists for coverage over 0.5 monolayers. The surface atoms of clean Mo(100) are displaced in the [11] direction in the c(2x2) reconstruction, in very good agreement with experimental results. It is both interesting and surprising to find that such a reconstruction survives when the Mo surface is covered by a full monolayer of Au. The displacement of the "subsurface" reconstruction is even larger in magnitude than the case of the clean surface.

The fcc metals gold, silver, platinum and palladium absorb and form surfilce alloys with the bee transition metal molybdenum, although they do not form bulk alloy with molybdenum. The surface of bulk molybdenum is unstable and readying reconstructs to a c(2x2), we found that it is also possible for the surface to form a c(2x2) surface vacancy. The Density of States ( DOS ) of surface atoms of a p(1x1) Mo(100) show a peak near the Fermi Energy, rather than a dip of those atoms which are beneath the surface. The very stable surface alloys of Au/Mo(100), Ag/Mo(100), Pt/Mo(100) and Pd/Mo(100) are due to the quenching of the surface state near the Fermi level by chemical bonding between the fcc metal and the surface Mo atoms. The high surface DOS near EF of molybdenum can be quenched in two ways, namely the formation of a c(2x2) surface vacancy array, and the formation of a c(2x2) surface alloy. We found high surface DOS near EF of molybdenum that reappears in a cyclic deposition of monolayers of silver or gold as the absorbate, which has been observed in previous experiments. We found that for platinum and palladium, the surface high DOS neax EF persists for coverage over 0.5 monolayers. The surface atoms of clean Mo(100) are displaced in the [11] direction in the c(2x2) reconstruction, in very good agreement with experimental results. It is both interesting and surprising to find that such a reconstruction survives when the Mo surface is covered by a full monolayer of Au. The displacement of the "subsurface" reconstruction is even larger in magnitude than the case of the clean surface.

The fcc metals gold, silver, platinum and palladium absorb and form surfilce alloys with the bee transition metal molybdenum, although they do not form bulk alloy with molybdenum. The surface of bulk molybdenum is unstable and readying reconstructs to a c(2x2), we found that it is also possible for the surface to form a c(2x2) surface vacancy. The Density of States ( DOS ) of surface atoms of a p(1x1) Mo(100) show a peak near the Fermi Energy, rather than a dip of those atoms which are beneath the surface. The very stable surface alloys of Au/Mo(100), Ag/Mo(100), Pt/Mo(100) and Pd/Mo(100) are due to the quenching of the surface state near the Fermi level by chemical bonding between the fcc metal and the surface Mo atoms. The high surface DOS near EF of molybdenum can be quenched in two ways, namely the formation of a c(2x2) surface vacancy array, and the formation of a c(2x2) surface alloy. We found high surface DOS near EF of molybdenum that reappears in a cyclic deposition of monolayers of silver or gold as the absorbate, which has been observed in previous experiments. We found that for platinum and palladium, the surface high DOS neax EF persists for coverage over 0.5 monolayers. The surface atoms of clean Mo(100) are displaced in the [11] direction in the c(2x2) reconstruction, in very good agreement with experimental results. It is both interesting and surprising to find that such a reconstruction survives when the Mo surface is covered by a full monolayer of Au. The displacement of the "subsurface" reconstruction is even larger in magnitude than the case of the clean surface.

The influence of modal interactions on the response of harmonically excited flexible L-shaped metallic and composite structures has been investigated analytically and experimentally. Each metallic structure possesses a two-to-one internal resonance, while each composite structure possesses a three-to-one internal resonance and either a two-to-one or a one-to-·one internal resonance. For the metallic structures, a weakly nonlinear analysis is used to derive the autonomous system of equations which describe the evolution of the amplitudes and phases of the internally resonant modes. These equations are obtained for primary- and secondary-resonant excitations. The excitation frequency or amplitude is used as a control parameter and the resulting bifurcations (saddle-node, pitchfork, and Hopf bifurcations) are studied. Theoretical analyses for internally resonant systems are used to predict and explain the responses of the composite structures.

The influence of modal interactions on the response of harmonically excited flexible L-shaped metallic and composite structures has been investigated analytically and experimentally. Each metallic structure possesses a two-to-one internal resonance, while each composite structure possesses a three-to-one internal resonance and either a two-to-one or a one-to-·one internal resonance. For the metallic structures, a weakly nonlinear analysis is used to derive the autonomous system of equations which describe the evolution of the amplitudes and phases of the internally resonant modes. These equations are obtained for primary- and secondary-resonant excitations. The excitation frequency or amplitude is used as a control parameter and the resulting bifurcations (saddle-node, pitchfork, and Hopf bifurcations) are studied. Theoretical analyses for internally resonant systems are used to predict and explain the responses of the composite structures.