Search

Atom probe tomography (APT) is a powerful three-dimensional nanoanalyzing microscopy technique considered key in modern materials science. However, progress in the spatial reconstruction of APT data has been rather limited since the first implementation of the protocol proposed by Bas et al. in 1995. This paper proposes a simple semianalytical approach to reconstruct multilayered structures, i.e., two or more different compounds stacked perpendicular to the analysis direction. Using a field evaporation model, the general dynamic evolution of parameters involved in the reconstruction of this type of structure is estimated. Some experimental reconstructions of different structures through the implementation of this method that dynamically accommodates variations in the tomographic reconstruction parameters are presented. It is shown both experimentally and theoretically that the depth accuracy of reconstructed APT images is improved using this method. The method requires few parameters in order to be easily usable and substantially improves atom probe tomographic reconstructions of multilayered structures.

Chemically resolved atomic resolution imaging can give fundamental information about material properties. However, even today, a technique capable of such achievement is still only an ambition. Here, we take further steps in developing the analytical field ion microscopy (aFIM), which combines the atomic spatial resolution of field ion microscopy (FIM) with the time-of-flight spectrometry of atom probe tomography (APT). To improve the performance of aFIM that are limited in part by a high level of background, we implement bespoke flight path time-of-flight corrections normalized by the ion flight distances traversed in electrostatic simulations modeled explicitly for an atom probe chamber. We demonstrate effective filtering in the field evaporation events upon spatially and temporally correlated multiples, increasing the mass spectrum's signal-to-background. In an analysis of pure tungsten, mass peaks pertaining to individual W isotopes can be distinguished and identified, with the signal-to-background improving by three orders of magnitude over the raw data. We also use these algorithms for the analysis of a CoTaB amorphous film to demonstrate application of aFIM beyond pure metals and binary alloys. These approaches facilitate elemental identification of the FIM-imaged surface atoms, making analytical FIM more precise and reliable.

Lifetime of the nuclear power plant (NPP) is limited by the embrittlement of the reactor pressure vessel steel (RPV) steel. One of the most important parameters determining the embrittlement is the bulk concentration of irradiation sensitive solutes.A set of chemically-tailored (CT) steels with the tuned bulk solute contents were neutron irradiated at 290 °C in high flux BR2 reactor. To study the nanosized solute clusters the state of the art Atop Probe Tomography (APT) technique was utilised.The results of this study show that:- Solute clusters enriched in Mn, Ni, Si and to the less extent in Cu and P were observed in 11 out of 12 conditions. These clusters were formed inside the ferritic matrix, alongside dislocation lines and at low-angle grain boundaries (LAGBs);- Addition of all solutes of interest results in the increase in solute cluster number density. The stronger effect was observed for Ni and Cu that of Mn and P. In addition, a positive synergy between Mn and Ni was evaluated;- Measured by tensile testes irradiation hardening is well explained by the formation of solute clusters. Solute clusters are weak obstacles with an obstacle strength of around 0.14.; La durée de vie d'un réacteur nucléaire à eau pressurisée (REP) peut être limitée par le durcissement et la fragilisation sous irradiation de l'acier de la cuve du réacteur. Un des paramètres importants influençant l'évolution des propriétés mécaniques de l'acier de cuve sous irradiation est la composition chimique de ce dernier, en particulier les teneurs en éléments réputés fragilisant, tels que Mn, Ni, Si, Cu et P.Dans cette thèse, un ensemble d'aciers de compositions chimiques contrôlées, permettant d'isoler les effets du Mn, du Ni, du Cu et du P et les synergies entre ces éléments, a été irradié aux neutrons à une température de 290 °C dans le réacteur expérimental BR2. L'évolution de la microstructure sous irradiation à été étudiée par sonde atomique tomographique. Les microstructures d'irradiation ont été comparées à des résultats d'essais de traction réalisés au SCK.CEN.Les résultats de cette étude montrent que :- Des amas nanométriques de solutés enrichis en Mn, Ni, Si et dans une moindre mesure en Cu et P sont observés après irradiation. Ces amas se situent dans la matrice ferritique, le long des lignes de dislocation et aux joints de grains;- L'augmentation de la teneur initiale en solutés réputés fragilisant entraîne une augmentation de la densité numériques d'amas de solutés. L'effet le plus fort est observé pour le Ni et le Cu. De plus, une synergie positive entre Mn et Ni a été mise en évidence;- La variation de la limite d'élasticité due à l'irradiation s'explique bien par la formation des amas de solutés en considérant ces amas comme des obstacles faibles (force d'obstacle d'environ 0,14).

The Laser Beam Powder Bed Fusion (PBF-LB) process is used for the additive manufacturing (AM) of metal parts and is capable of producing a wide variety of complex parts for aeronautic and aerospatial sectors. Al-Si system is the most common aluminum alloy series used for the 3D manufacturing of parts by means of the PBF-LB method. They are known for their good flowability, high thermal conductivity and good corrosion resistance. However, rapid solidification causes ultrafine microstructure and non-equilibrium phases with supersaturation of Si in the Al matrix beyond the maximum solubility expected for the Al-Si system.In order to take advantage of the high cooling rate and thermal gradient, new aluminum alloys have been developed or modified specifically for the LPBF process. These materials, known as non-weldable, could be formed by additive manufacturing due to the introduction of grain refiners elements such as zirconium and titanium to limit the hot cracking effects.The main objective of this study is to evaluate the impact of the Laser Powder Bed Fusion (LPBF) process and heat treatments on the formation of precipitates and the composition of phases inherited by the rapid solidification process. In this context, clustering of Si and Mg atoms has been analysed for Al-Si alloys as well as the effect of adding Zr and Ti in Al2139 alloy. The LPBF process allows to obtain complex structures with microstructure gradients at the micrometer and nanometer scale, which modifies the precipitation kinetics compared to the conventional process. All these elements contribute to the high mechanical strength of these materials, especially Al2139 AM which can reach a yield strength (Rp0,2) and a tensile strength (Rm) higher than 500 MPa.In this study, a multi-scale microstructural characterization approach was chosen, from scanning electron microscopy to Atom Probe Tomography. From the knowledge of the microstructure inherited from the PBF-LB process, specific heat treatments were designed to improve the hardening response of the alloy and the level of hardening will be put into relation with the evolution of the alloy’s microstructure.; La fusion laser sur lit de poudre (LPBF) est un procédé utilisé en fabrication additive qui permet la mise en forme d’une large variété de pièces complexes notamment pour les secteurs de l’aéronautique et l’aérospatial. Le système Al-Si est la famille d’alliages d’aluminium la plus communément utilisée pour la fabrication de pièces élaborées par LPBF. Ces alliages sont bien connus pour leur bonne coulabilité, leur haute conductivité thermique ainsi que pour leur bonne résistance à la corrosion. Cependant, les fortes vitesses de refroidissement inhérentes au procédé LPBF génèrent des microstructures ultrafines et des phases hors-équilibre avec une sursaturation en Si dans la matrice d’Al bien en-dessus de la limite de solubilité attendue dans les systèmes Al-Si.Afin de tirer profit des fortes vitesses de refroidissement et des gradients thermiques, de nouveaux alliages d’aluminium ont été développés ou modifiés spécialement pour le procédé LPBF. Ces matériaux réputés non soudables ont pu être mis en forme par fabrication additive grâce à l’introduction d’éléments germinateurs tels que le zirconium et le titane afin de limiter les phénomènes de fissuration à chaud.L’objectif de cette étude est d’évaluer l’impact du procédé de fusion laser sur lit de poudre et des traitements thermiques sur la formation de précipités et la composition des phases induites par le refroidissement rapide dans ces alliages d’aluminium. Dans ce contexte, le comportement des atomes de Si et Mg a été analysé pour les alliages Al-Si ainsi que l’effet de l’ajout de Zr et Ti dans l’alliage Al2139. Il est révélé que le procédé LPBF permet d’obtenir des structures complexes avec des gradients de microstructure à l’échelle micrométrique et nanométrique, modifiant ainsi les cinétiques de précipitation par rapport au procédé conventionnel. L’ensemble de ces éléments contribue à la haute résistance mécanique de ces matériaux notamment l’Al2139 AM qui peut atteindre une limite élastique (Rp0,2) et une résistance à la traction (Rm) supérieures à 500 MPa.Pour répondre aux objectifs, une caractérisation microstructurale multi-échelle a été entreprise, de la microscopie électronique à balayage jusqu’à la Sonde Atomique Tomographique. Compte tenu de la microstructure héritée du procédé LPBF, des traitements thermiques spécifiques ont été conçus pour améliorer la réponse au durcissement de l'alliage. Le niveau de durcissement a pu être mis en relation avec l'évolution de la microstructure de l’alliage.

International audience; The present study aims to understand the underlying mechanisms of macroscopic contraction that occurs in γ/γ′ nickel based superalloys submitted to isothermal annealing and resulting in stress relaxation retardation. Samples of the AD730TM alloy were characterized at different scales using dilatometry, X-Ray Diffraction (XRD) with in-situ heating, Electron Channeling Contrast Imaging (ECCI) in the Scanning Electron Microscope (SEM), conventional Transmission Electron Microscopy (TEM) and Atom Probe Tomography (APT). APT analyses revealed chemical composition changes in both the γ matrix and the γ′ hardening precipitates during annealing at 760°C after sub-solvus solution heat treatment. These chemical alterations lead to a decrease of the lattice parameters of both phases, which was captured by XRD analyses. The macroscopic volume contraction is shown to be mainly governed by that decrease in the lattice parameters. In addition, from APT analyses, the chemical alterations seem to be promoted by diffusion of solutes such as Cr, Co and Fe, along dislocations occurring under relaxation testing conditions.

International audience; Tungsten is the primary candidate materials for the high neutron flux, high temperature components of a future demonstration fusion reactor. Despite this, there is a lack of data on W under fusion relevant neutron doses and irradiation temperatures. In this study, single crystal and polycrystalline W samples irradiated at the High Flux Reactor (HFR) at 900 C were characterised using Atom Probe Tomography (APT) and Scanning Transmission Electron Microscopy (STEM). Bulk chemical and isotopic concentration predictions were validated by analysing the mass spectrum from APT experiments. A post irradiation composition of W-1.26 0.15 at.%Re - 0.08 0.02 at.%Os - 0.01 0.01 at.%Ta was measured in the single crystal sample, whereas W-1.09 0.07 at.%Re - 0.08 0.02 at.%Os - 0.01 0.01 at.%Ta was measured for the polycrystalline. APT and STEM showed that a high number density of Re and Os rich precipitates had formed under neutron irradiation. These typically consisted of a core rich in Re and Os, surrounded by a less dense Re rich cloud. Multiple analysis methods were applied to investigate the composition of these clusters. APT showed that the centres of some of the precipitates had a rod shaped core which were rich in both Re and Os. Line profile analysis suggests that in the centre of the precipitates, the threshold composition for phase formation may have been reached, as has been observed in higher transmutation rate experiments. In addition, dislocations, sub grain boundaries and dislocation loops were all shown to be decorated with both Re and Os, in agreement with predictions from DFT simulations.

International audience; The effect of Cr content on intragranular κ-carbide precipitation behaviors in Fe-Mn-Al-Cr-C low-density steels was investigated by X-ray diffraction (XRD), scanning (SEM) and (high-resolution) scanning transmission electron microscopy ((HR)STEM), atom probe tomography (APT), thermodynamic and first-principles calculations. We found that the volume fraction of κ-carbides in near-rapidly solidified samples was decreased with increasing Cr content. The increased Cr content also significantly slowed down the growth rate of κ-carbides during isothermal aging treatment (600 • C/9 h). Thermodynamic calculations showed that the increased Cr content decreased the precipitation temperature of κ-carbides and narrowed the temperature range of κ-carbide formation. The impurity of Cr atoms in κ-carbide structure increased its interfacial energy and elastic strain energy for nucleation, thereby significantly increasing the critical energy for the κ-carbide formation in γ-austenite. Simultaneously, the results of APT and first-principles calculation indicated that the impurity of Cr atoms in the κ-carbide structure preferred to occupy the Al sites, which increased the formation energy of κ-carbides making the κ-carbide formation difficult. In addition, the increased interfacial energy between the κ-carbides and γ-austenite, the improved solubility of Al and C in the γ-austenite and the decreased diffusion coefficient of C in γ-austenite caused by increasing Cr content resulted in a slower growth rate of the κ-carbides.

The development of a cost-effective bipolar transistor requires a collector doped by ion implantation. A low-resistive collector is needed to ease the electron flux towards the contact. The high dose required can only be processed fast enough on “high-current” implant machines which are limited in energy. The choice of the N type dopant has been made regarding this limitation. Phosphorus has been selected as it features a lower mass than arsenic and antimony. Indeed, at a given dose and energy, phosphorus profile is deeper and hence less resistive as compared to arsenic and antimony profiles. However, the high phosphorus diffusion coefficient leads to a dopant pile-up close to the base. This pile-up degrades the electrical properties of the base/collector junction. This side effect is hindered by using carbon co-implantation. Upon the first anneal step, the duration of the phosphorus transient enhanced diffusion regime is reduced by carbon. Upon the second anneal step, carbon establishes a self-interstitial undersaturation. Hence, phosphorus diffusion is greatly reduced. To sum up, low resistance as well as high junction voltages (Early and breakdown) are reached thanks to the combined action of carbon and phosphorus., La conception de transistors bipolaires de moindre coût nécessite entre-autres l’implémentation d’un collecteur réalisé par implantation ionique. Ce dernier doit être faiblement résistif pour faciliter le transit des électrons vers les contacts. Le fort dopage requis nécessite l’utilisation d’implanteurs « fort-courants » qui sont, en revanche, limités en énergie. Cette limitation conditionne le choix du dopant type N : le phosphore est retenu car il présente une masse plus faible que l’arsenic ou l’antimoine. Il contribue donc à l’obtention d’un profil plus profond, et donc moins résistif, à énergie donnée. En revanche, le coefficient de diffusion élevée du phosphore conduit à une dégradation des propriétés électriques de la jonction base-collecteur. La co-implantation du carbone permet de s’affranchir de cet effet. Au cours de la première phase du recuit, le carbone permet de réduire la durée du régime de diffusion accélérée du phosphore. Au cours de la seconde phase, le carbone instaure une sous-saturation en auto-interstitiels qui atténue fortement la diffusion du phosphore. L’action combinée du phosphore et du carbone permet donc d’atteindre à la fois une résistance faible et des tensions de jonction (claquage et Early directe) élevées

A position-energy-sensitive detector has been developed for APT instruments in order to deal with some mass peak overlap issues encountered in APT experiments. Through this new type of detector, quantitative and qualitative improvements could be considered for critical materials introducing mass peak overlaps, such as nitrogen and silicon in TiSiN systems, or titanium and carbon in cemented carbide materials. This new detector is based on a thin carbon foil positioned on the front panel of a conventional MCP-DLD detector. According to several studies, it has been demonstrated that the impact of ions on thin carbon foils has the effect of generating a number of transmitted and reflected secondary electrons that mainly depends on both the kinetic energy and the mass of incident particles. Despite the fact that this phenomenon is well known and has been widely discussed for decades, no studies have been performed to date for using it as a mean to discriminate particles energy. Therefore, this study introduces the first experiments on a potential new generation of APT detectors that would be able to resolve mass peak overlaps through the energy-sensitivity of thin carbon foils., Comment: 36 pages, 13 figures, 2 tables, conference paper

International audience; The effects of boron segregation and austenite grain size on the bainitic isothermal transformation of a high-strength steel were studied independently. Dilatometry, microstructural observations and previous atom probe tomography analyses showed that bainitic transformation rate decreases with increasing boron excess at austenite grain boundaries. We also found that boron segregation causes an unexpected grain size effect: large austenite grains transform faster than small ones. A kinetic model assuming slow nucleation at austenite triple junctions and rapid growth of bainitic nuclei successfully described our experimental data. These results confirm that bainite nucleation is inhibited by segregated boron at grain boundaries. On this basis, austenitic grain size could be optimized to make the most of boron addition in advanced high-strength steels.

International audience; The mechanisms and the sequence of crystallization of a Fe64B24Y4Nb6Al0.4 metallic glass were investigated experimentally. To this end, the microstructure of the metallic glass after interrupted isothermal treatments at a temperature between glass transition and primary crystallization was studied by complementary techniques, namely atom probe tomography, transmission electron microscopy and X-ray diffractometry. During the early stages of crystallization of the glass, some nanometric (Fe,Nb)B crystals nucleate first by rejecting Y at their interface, which hinders their growth. The Fe62B14Y3 dendritic phase starts growing in a second step. Longer annealings lead to the development of chemical gradients in the amorphous matrix surrounding Fe62B14Y3 crystals, which favors the nucleation of additional nanometric (Fe,Nb)2.4B crystals.

International audience; We report on the microscopic behavior of residual hydrogen on nanometric field emitters. By using homogeneous or heterostructured semiconductor specimens analyzed in a laser-assisted atom probe, it is possible to study how the relative abundances of the ionic species H+, H2+, and H3+ depend on the microscopic electric field, estimated through post-ionization statistics. In the case of Ga-containing semiconductors, the relative abundances of H+, H2+, and H3+ follow a common trend, independent of the nonmetallic component of the matrix. The dependence of the total H flux on the electric field exhibits a more complex behavior, which depends also on the spatial direction of the variation of the field (in-depth or on-surface). The analysis of multiple detection events provides further insights into surface chemistry. Noticeably, H+–H3+ ion pairs are both correlated in number and separated by very small distances on the detector space, suggesting a possible reaction 2H2 → H3++H++2e– taking place on the field emitter surface.

International audience; In ferrite/austenite dissimilar metal welds, carbon can diffuse from the ferritic low-alloy steel to the austenitic weld metal during heat treatment and thermal aging due to the large carbon and chromium content differences between the two alloys. This carbon influx leads to the precipitation of carbides in the high-alloy weld metal, near the fusion boundary, which can affect the properties of the joint. This precipitation was investigated using multiple characterization techniques: electron probe microanalysis, small angle X-ray scattering and atom probe tomography. The studied joint consisted of low-alloy steel and stainless steel base metals that were arc welded using a nickel-based filler metal. Given the heterogeneous microstructure of such joints, a microstructure navigation strategy was implemented to ensure all characterizations were carried out in the same regions of interest and the associated data could later be correlated. The precipitates were found to be mostly M23C6. The precipitation was quantified using two different methods that gave comparable results, whose differences highlight the capabilities and limitations of each technique. Precipitate fractions are higher following thermal aging but remain far from equilibrium in all cases, indicating that precipitation kinetics should be accounted for to model the microstructural evolution of such dissimilar metal welds more accurately.

International audience; Diffusion barriers play an important role in numerous phase formation processes. A well known exam- ple in microelectronics is the reactive diffusion growth of silicide thin films which are applied as contact materials. In this work, the effect of a Ti interlayer on the kinetics of the formation of CoSi by reactive diffusion is investigated. Therefore, Co(100nm)/Ti(5nm) deposited on Si(111) is annealed at various tem- peratures. Transmission electron microscopy and atom probe tomography allow to study the evolution of microstructure and local compositions after each annealing. It is observed that the CoSi growth starts at the Ti/Si interface and it is controlled by Co diffusion through the Ti interlayer. The Ti interlayer keeps its microstructure during the growth of CoSi. At higher annealing temperatures, Si diffusion through the Ti interlayer to the Co layer is evidenced. First, it segregates at grain boundaries of the polycrystalline Co layer on top of the Ti interlayer before it reacts to cobalt silicides. Also Ti diffusion to the CoSi/Si in- terface occurs that may affect diffusion processes and phase formation reaction at this interface. Beyond the experimental observations, a model is developed to quantify the diffusivity of Co in the Ti interlayer on the base of the investigated CoSi growth. Furthermore, a model is developed in order to estimate the diffusivity of Si in the Ti interlayer on the base of the formed solid solution and segregation in the Co layer. Both models can be generally applied for similar material configurations to estimate diffusivities in interlayers or to predict phase growth.

Chalcogenide glasses have been used for many years and in various applications. These glasses are particularly known for their infrared transmission property. Today the economic stakes with the evolution of the market of security, surveillance and autonomous cars need the use of thermal cameras, pushing to the reduction of the costs, and for that new materials should be developed, cheaper and easier to make. The objective of this thesis work is to study a material that can answer this problem, the composition 62,5GeS2-12,5Sb2S3-25CsCl has thus been selected. The study of the evolution of the microstructure during physical aging under Tg and crystallization above Tg is carried out from the macro scale to the nano scale in order to give a better understanding of the evolution of the glass and of its microstructure over time at different temperatures. These studies were carried out using different analysis techniques: differential scanning calorimetry (DSC) and with temperature modulation (MT-DSC), fast scanning calorimetry (FSC), X-ray diffraction (XRD), transmission electron microscopy (TEM), atom probe tomography (SAT). The experimental results demonstrate that crystallization under Tg is possible when thermodynamic equilibrium is reached. The study of the crystallization of the glass made it possible to relate the microstructure and the evolution of its optical and mechanical properties, by observing and characterizing the development of two morphologies ofcrystals based on same CsCl structure., Les verres de chalcogénures sont utilisés depuis de nombreuses années et dans diverses applications. Ces verres sont notamment connus pour leur propriété de transmission dans l'infrarouge. Aujourd'hui les enjeux économiques avec l'évolution du marché de la sécurité, de la surveillance et des voitures autonomes nécessitant l'utilisation de caméras thermiques, poussent à la diminution des coûts et pour cela il est nécessaire de se tourner vers de nouveaux matériaux, moins chère et plus simple à réaliser. L'objectif de ce travail de thèse est d'étudier un matériau qui peut répondre à cette demande, la composition 62,5GeS2-12,5Sb2S3-25CsCl a ainsi été sélectionnée. L'étude de l'évolution de la microstructure lors du vieillissement physique sous Tg et la cristallisation au-dessus de Tg est réalisée de l'échelle macro à l'échelle nano afin de mieux comprendre l'évolution du verre et de sa microstructure au cours du temps à différentes températures. Ces études ont été réalisées à l'aide de différentes techniques d’analyses : calorimétrie différentielle à balayage (DSC) et avec modulation de température (MT-DSC), calorimétrie à balayage rapide (FSC), diffraction des rayons X (DRX), microscopie à transmission électronique (MET), sonde atomique tomographique (SAT). Les résultats expérimentaux ont permis de mettre en évidence que la cristallisation sous Tg était possible lorsque l'équilibre thermodynamique était atteint. L'étude de la cristallisation du verre a permis de mettre enrelation la microstructure et l'évolution des propriétés optiques et mécanique, en observant et en caractérisant le développement de deux morphologies de cristaux de même composition.

International audience; During ageing, the ferritic phase of Cast Austenitic Stainless Steels (CASS) undergoes decomposition by spinodal decomposition and G-phase precipitation, which leads to increasing hardness and decreasing impact toughness. Mo-bearing steels are more prone to ageing than Mo-free steels. To study the effects of Ni, Mo, and Mn, duplex model alloys were thermally aged at 350 and 400°C. The microstructure evolution during ageing was characterized by atom probe tomography. Only Ni additions seem to affect the spinodal decomposition kinetics. Ni accelerates the spinodal decomposition kinetics regardless of the presence of Mo or Mn. Additions of Mo and Mn promote the precipitation of G-phase.

International audience; The evolution of radiation damage in steels is a major issue for the safe operation of nuclear power plants. Mn and Ni contribution to the formation and evolution of microstructural features remains a controversial issue. The present study aims at investigating their effects on microstructure by irradiating undersaturated BCC Fe-3.3at.%Ni and Fe-2.8at.%Mn model alloys. Two different ion irradiation conditions have been applied to study the effect of irradiation dose rate in both of these Fe-based alloys at a temperature of 673 K. In all cases, irradiation promotes the formation of dislocation loops. In the FeNi alloy, radiation-induced segregation is observed on loops, leading to precipitation of a FCC phase in the low dose rate irradiation condition only. In the case of FeMn, even if no precipitation was revealed, highly enriched decorated dislocation loops are observed, with a higher Mn segregation at radiation defects than Ni.

Atom Probe Tomography uses field evaporation phenomena to peel off atoms from the surface of a sample. These ions are projected in a divergent electric field and collected by a time and position sensitive detector. Finally, thanks to the use of data processing algorithms, it is possible to map the composition of this sample in three dimensions near the atomic scale, in order to define its physical properties. However, the characterization of these elements depends mainly on the understanding of the field evaporation phenomena, the mass resolving power and the spatial resolution of the instrument. Currently, the use of high voltage pulses to trigger the field evaporation process is not without flaws. Indeed, the mass resolution of a straight flight path atom probe is not optimal compared to an atom probe with an energy compensation system (reflectron). However, it presents the best detection efficiency. This work provides solutions to optimize this performance factor through in-depth studies on the generation of the high voltage pulse and the associated field evaporation phenomena.One part consists in studying the different electronic technologies, components and circuits, allowing the high voltage pulses generation. Then, a geometrical model of the atom probe has been realized in order to visualize the influence of different voltage pulse shapes on the ion flights. The results of the simulations show that the field variation influences the velocity of the ions. This degrades the mass peak shapes and amplifies the chromatic aberrations. These dynamic effects are particularly visible on light ions (M < 15amu).At the end of these investigations, a new configuration of pulse generation is proposed in order to benefit from the advantages of the use of a nanosecond pulse. This has been done experimentally on a test bench in the laboratory. The results show a clear improvement of the mass resolution at 10% and 1%.Moreover, the analysis of the mass spectra has shown that physical information intrinsic to the material is available by measuring the slope of the mass peak tail. Thus, it is possible to extract the energy barrier of the atoms composing the sample from the experimental data. Then, atoms contained in the mass peak tail are identified to image the variation of energy deficits in a reconstructed volume. This new method of Field Evaporation Energy Loss Spectroscopy is applicable to any sample analyzed by a straight flight voltage pulse atom probe., La Sonde Atomique Tomographique utilise le phénomène d’évaporation par effet de champ pour arracher les atomes de la surface de l’échantillon. Les ions sont ensuite projetés dans un champ électrique divergent puis récoltés par un détecteur sensible en temps et en position. Enfin, grâce à l’utilisation d’algorithmes de traitement des données, il est alors possible de représenter la composition de cet échantillon en trois dimension à l’échelle atomique, afin d’en définir les propriétés physiques. Cependant la caractérisation de ces éléments dépend principalement de la compréhension des phénomènes d’évaporation par effet de champ, de la résolution en masse et de la résolution spatiale de l’instrument. Actuellement, l’utilisation d’impulsions haute tension pour déclencher le processus d’évaporation par effet de champ n’est pas sans défaut. En effet, la résolution en masse d’une sonde atome droite n’est pas optimale comparée à celle d’une sonde atomique à compensation d’énergie (réflectron). Cependant, elle présente le meilleur rendement de détection. Ces travaux apportent des solutions afin d’optimiser ce facteur de performance à travers des études approfondies sur la génération d’impulsions haute tension et des phénomènes d’évaporations par effet de champ associés.Une partie de cette thèse consiste à étudier les différentes technologies électroniques, composants et circuits, permettant la génération d’impulsions haute tension. Par la suite, un modèle géométrique de la sonde atomique est réalisé afin de visualiser l’influence de différentes formes d’impulsions sur les trajectoires de vol des ions. Les résultats des simulations montrent que la variation de champ influence la vitesse des ions. Cela dégrade la forme et la résolution des pics de masse et amplifie les aberrations chromatiques. Ces effets dynamiques sont particulièrement visibles pour les ions légers (M < 15amu).A l’issue de ces recherches, une nouvelle configuration de génération d’impulsion est proposée afin bénéficier des avantages de l’utilisation d’une impulsion nanoseconde. Celle-ci est réalisée expérimentalement sur un banc de test au sein du laboratoire. Les résultats montrent une nette amélioration de la résolution en masse à 10% et 1%.Par ailleurs, l’analyse des spectres de masse a démontré que des informations physiques intrinsèques au matériau sont disponibles en mesurant la pente de la traîne de masse. Ainsi, il est possible d’extraire la barrière d’énergie des atomes composant l’échantillon à partir des données expérimentales. Les atomes contenus dans la traîne de masse sont ensuite identifiés pour imager la variation des déficits en énergie dans un volume reconstruit. Cette nouvelle méthode de Spectrocopie des Pertes d’Energie par Evaporation par Effet de Champ est applicable pour tout échantillon analysé en Sonde Atomique droite à impulsion de tension.

International audience; The evolution of radiation damage in steels is a major issue for the safe operation of Nuclear Power Plants. Mn and Ni contribution to the formation and evolution of microstructural features remains a controversial issue. The present study aims at investigating their effects on microstructure by irradiating undersaturated BCC Fe-3.3at.%Ni and Fe-2.8at.%Mn model alloys. Two different ion irradiation conditions have been applied to study the effect of irradiation dose rate in both of these Fe-based alloys at a temperature of 673 K. In all cases, irradiation promotes the formation of dislocation loops. In the FeNi alloy, radiation-induced segregation is observed on loops, leading to precipitation of a FCC phase in the low dose rate irradiation condition only. In the case of FeMn, even if no precipitation occurred, highly enriched decorated dislocation loops are observed, leading to a higher Mn segregation at radiation defects than Ni.

In this study, the formation of a periodic microstructural pattern designed by laser powder bed fusion (LPBF), commonly called selective laser melting (SLM), of an AlSi10Mg alloy is revealed at high resolution using scanning transmission electron microscopy and atom probe tomography. Special attention is paid to the description of non-equilibrium structures and compositional fields resulting from the ultrafast cooling of the LPBF process. Observations reveal the existence of a glass state in eutectic areas, wherein short-range ordering of the diamond-Si structure is observed. The apparent very fast solidification of eutectic regions is found to involve a local strain in adjacent Al cells, which extends up to 100 nm on average. In the supersaturated aluminium solid solution retained by the LPBF process, two populations of clusters are identified, for which the potential role of the selection of hardening phases is discussed. It is proposed that the microstructure of former melt pools outside heat-affected areas is described by the repetition of a periodic microstructural pattern consisting of eutectic regions /strain-hardened Al-crystals /strain-free Al-crystals with a high density of solute-rich clusters.

Boron steels are defined by a high hardenability. This effect is related to the intergranular segregation of boron in austenite. In these steels, two mechanisms of boron segregation generally coexist: equilibrium segregation and non-equilibrium segregation. They have a different driving force and kinetics, and accordingly a different temperature and time dependence. Thus, depending on heat treatment conditions, the boron content at austenite grain boundaries fluctuates. The thesis aims at understanding the boron segregation at austenite grain boundaries according to heat treatment conditions. A heat treatment was studied: austenitization at 1000°C, followed by a quenching at 50°C/s up to 600°C where an isothermal holding is carried out. This study is divided in two parts. First, an experimental part with the interfacial boron quantification by Atom Probe Tomography (APT) analysis. Second, a numerical part with the development of a modeling of the segregation kinetics which integrates both types of boron segregation. Informations about what is happening in the grain and the grain boundary can be extracted from the modeling. Experimental measurements are carried out after a fast quenching (300°C/s) down to room temperature with the ambition of freezing the studied state. This quench leads to an austenite-martensite transformation. In this complex configuration, a correlative microscopy protocol was developed to prepare samples containing a Prior Austenite Grain Boundary (PAGB) for APT analysis. Experimental analysis show that other elements segregate at PAGBs in addition to boron (presence of C, P, Cr, V, Mo and Mn), as well as an absence of boron in martensitic lath boundaries. A study of APT and numerical results shows that quenching at 300°C/s does not allow the thermal state to be frozen, an enrichment of the interface during this quenching takes place. Finally, as expected, a non-monotonic variation of the boron atomic fraction at grain boundary has been observed both experimentally and numerically, for hold times greater than one second. However, time scale between numerical and experimental interfacial enrichment kinetics is divergent. This can comes from both an unsuitable set of numerical parameters and from the differences which exist between the model and the experimental reality (effects of co-segregation or the second quenching at 300°C/s, which are not taken into account by the model). Results are proving to be very encouraging, and open up numerous perspectives., Les aciers trempants au bore sont définis par une importante trempabilité. Cet effet est lié à la ségrégation intergranulaire du bore dans l’austénite. Dans ces aciers, deux mécanismes de ségrégations du bore coexistent généralement : la ségrégation à l’équilibre et la ségrégation hors-équilibre. Ils ont une force motrice et une cinétique différente, et donc une dépendance en température et en temps différente. Ainsi, selon les conditions de traitement thermique, la teneur en bore aux joints de grains austénitiques varie. L’objectif de cette thèse est de comprendre la ségrégation du bore aux joints de grains austénitiques selon les conditions de traitement thermique. Un traitement thermique a été étudié : austénitisation à 1000°C, suivie d’une trempe à 50°C/s jusqu’à 600°C, température à laquelle un traitement isotherme est effectué. L'étude comporte deux volets. L'un expérimental avec la quantification de la teneur interfaciale en bore par analyses en sonde atomique tomographique (APT). L'autre numérique avec le développement d'une modélisation de la cinétique de ségrégation qui intègre les deux types de ségrégations du bore. Une information sur ce qu’il se passe dans le grain et le joint de grains peut être extraite de la modélisation. Les mesures expérimentales sont réalisées après une trempe rapide (300°C/s) jusqu’à température ambiante dans l’ambition de figer l’état à étudier. La trempe conduit à une transformation de l’austénite en martensite. Dans cette configuration complexe, un protocole de microscopie corrélative a été mis au point afin de préparer des échantillons contenant un ancien joint de grains austénitiques (PAGB) pour l'analyse APT. Les analyses expérimentales montrent que d’autres éléments ségrègent aux PAGB en plus du bore (présence de C, P, Cr, V, Mo et Mn), ainsi qu’une absence de bore dans les joints de lattes martensitiques. L’étude des résultats APT et numériques montre que la trempe à 300°C/s, ne permet pas de figer l’état thermique, et qu’un enrichissement de l’interface à lieu lors de cette trempe. Enfin, comme attendu, une variation non monotone de la fraction atomique de bore au joint de grains a été observée à la fois expérimentalement et numériquement, et ce pour des temps de maintiens supérieurs à une seconde. Cependant l’échelle de temps entre la cinétique d’enrichissement de l’interface issue des simulations et celle issue des résultats expérimentaux est très éloignée. Cela peut provenir à la fois d’un jeu de paramètres numériques non adapté et des différences qui existent entre le modèle et la réalité expérimentale (effets de co-ségrégation, ou seconde trempe à 300°C/s, non pris en compte par le modèle). Les résultats obtenus se révèlent être très encourageants, et ouvrent de nombreuses perspectives.

International audience; A multi-microscopy investigation of a GaN tunnel junction (TJ) grown on an InGaN-based light emitting diode (LED) has been performed. The TJ consists of a heavily Ge-doped n-type GaN layer grown by ammonia-based molecular-beam epitaxy on a heavily Mg-doped p-type GaN thin layer, grown by metalorganic vapor phase epitaxy. A correlation of atom probe tomography, electron holography and secondary ion mass spectrometry has been performed in order to investigate the nm-scale distribution of both Mg and Ge at the TJ. Experimental results reveal that Mg segregates at the TJ interface, and diffuses into the Ge-doped layer. As a result, the dopant concentration and distribution differ significantly from the nominal values. Despite this, electron holography reveals a TJ depletion width of ~7 nm, in agreement with band diagram simulations using the experimentally determined dopant distribution.

The Atom Probe Tomography (APT) is an analysis technique allowing the visualization of the atomic structure and the determination of the elemental nature of each atom originating from material samples. In order to get this type of analysis, the APT uses a destructive technique, called the Field Evaporation, allowing the extraction and the detection of atoms constituting the analyzed material. To do so, it is necessary, in a first step, to shape the material sample in the form of a very sharp needle. Then, in a second step, to apply a high electric potential on the tip for ionizing and expelling the atoms from the surface of the tip apex. Surface atoms are then projected towards a position-sensitive detector (PSD), with the aim of reconstructing the internal three-dimensional shape of the analyzed material, layer-by-layer. Concerning the spectrometric aspect of this technique, each detected atom is associated to a time-of-flight (TOF) measurement. Those TOFs are measured between electric (or laser) pulses applied on the tip, and arrival times on the detector. Knowing the resulting electric potential applied on the tip and the projection speed of ions, it is possible to get the mass-to-charge ratio of each collected ion. Regarding the working process of this technique, it may be thought that APT instruments could be the best mean for analyzing materials at the atomic scale. However, since the invention of this technique, more than 30 years ago, APT instruments are still relatively confidential in the panel of high-resolution microscopes used in material sciences. One of the main reasons to this concerns the potential biases brought by the APT detection system. Some of those biases concern the inability of the detector to correctly treat high amounts of ion arrivals in a short period of time. In the case where some elements tend field evaporate in bundles of more than one ion, it is then possible to suffer from selective losses during APT experiments. Secondly, given that the elemental nature of collected ions is determined through time-of-flight mass spectrometry, material analyses may be intrinsically subjected to compositional biases. This is due to interferences between mass peaks originating from different elements, sharing very close mass-to-charge ratios. Based on the foregoing, this thesis is intended to answer to the following issue: How to outperform the limitations of APT detection systems? To do so, theoretical and experimental studies have been conducted for resolving those biases, and finally led to the development of new theoretical models and new type of detectors for APT instruments., La Sonde Atomique Tomographique (SAT) est un outil d’analyse qui permet de déterminer la structure et la composition locale de matériaux à l’échelle atomique. Afin d’obtenir ce type d’analyse, la SAT utilise une technique destructive appelée "évaporation par effet de champ", permettant d’extraire et de collecter un certain volume d’atomes provenant d’un échantillon de matériau. Pour ce faire, il est nécessaire, dans un premier temps, d’usiner le matériau à analyser sous la forme d’une pointe très fine. Puis, dans un second temps, d’appliquer un potentiel électrique très élevé au niveau de cette pointe, afin de d’ioniser et expulser les atomes de surface. Ces atomes de surface sont alors projetés vers un détecteursensible en position, couche après couche, afin d’obtenir une tomographie à l’échelle atomique du volume analysé. Afin d’être chimiquement identifié, chaque atome détecté est associé à une mesure de temps de vol, qui s’effectue entre une impulsion électrique (ou laser) appliqué sur la pointe, et un signal d’arrivé sur le détecteur. Connaissant le potentiel électrique total appliqué à la pointe, ainsi que la vitesse des ions projetés vers le détecteur, il est alors possible de déterminer un rapport masse sur charge (m/n) de chaque ion collecté. Au vu de cette dernière description, la SAT peut apparaitre comme l’un des meilleurs outils d’analyse permettant de caractériser des matériaux à une échelle atomique. Cependant, l’utilisation de cet instrument reste encore restreinte en science des matériaux. L’une des principales raisons à cela concerne les biais apportés par le système de détection de la SAT. En effet, il a été constaté que des pertes d’information pouvaient se produire lorsque le détecteur était soumis à des impacts d’ions multiples sur de très courtes périodes de temps. Ceci peut se traduire par des pertes sélectives dans le cas où certains éléments auraient tendance à s’évaporer par groupe de plusieurs ions sur une même impulsion de départ sur la pointe. D’autre biais ont aussi été constatés au niveau de l’identification chimique des ions détectés. Étant donné que cette identification s’effectue par spectrométrie de masse à temps de vol, on peut s’apercevoir que certains éléments d’un même matériau peuvent se regrouper sous des rapport m/n extrêmement proches. Ce qui a pour effet de créer des interférences entre pics de masses provenant d’éléments différents, et donc de créer de potentiels biais de compositions. Sur la base de ces différentes limites, cette thèse a pour but de répondre à la problématique suivante : Comment peut-on surpasser les limites des différents systèmes de détection de SAT ? Pour y répondre, des études, à la fois théoriqueset expérimentales, ont été effectuées pour traiter l’ensemble de ces derniers biais, jusqu’à l’aboutissement de nouveaux modèles théoriques et de nouveaux prototypes de détecteurs pour la SAT.

Methods are presented determining the Gibbs triple junction excess ( $${\varGamma }^{\text{TJ}}$$ ) of solute segregation in polycrystalline materials from single atom counting in 3D volumes. One method bases on cumulative profile analysis, while two further methods use radial integration of solute atoms. The methods are demonstrated and compared on simulated model volumes which include three grain boundaries joining together at a triple junction with set values for Gibbs grain boundary and triple junction excess. An experimental technique that provides 3D volumes with single atom detection and spatial resolution close to atomic scale is atom probe tomography. An atom probe tomography volume of a $$\hbox {CoSi}_2$$ thin film that contains three grain boundaries and a triple junction has been acquired. Ti segregation is found qualitatively at the grain boundaries and triple junction. The quantification of the Ti excess at the investigated $$\hbox {CoSi}_2$$ triple junction reveals for the three introduced methods positive Gibbs triple junction excess values. It demonstrates that there is an excess of Ti at $$\hbox {CoSi}_2$$ triple junctions and provides opportunities for its quantification.

Correlation between off-axis electron holography and atom probe tomography (APT) provides morphological, chemical and electrical information about Mg doping (p-type) in gallium nitride (GaN) layers that have been grown at different temperatures at a nanometric scale. APT allows access to the three-dimensional distribution of atoms and their chemical nature. In particular, this technique allows visualisation of the Mg-rich clusters observed in p-doped GaN layers grown by metal-organic chemical vapour deposition. As the layer growth temperature increases, the cluster density decreases but their size indicted by the number of atoms increases. Moreover, APT reveals that threading dislocations are decorated with Mg atoms. Off-axis electron holography provides complementary information about the electrical activity of the Mg doping. As only a small fraction of dopant atoms are ionised at room temperature, this fraction is increased by annealing the specimen to 400 °C in situ in a transmission electron microscope (TEM). A strong reduction of the dopant electrical activity is observed for increases in the layer growth temperature. The correlation of APT with TEM-based techniques was shown to be a unique approach in order to investigate how the growth temperature affects both the chemical distribution and electrical activity of Mg dopant atoms.

Radiation induced segregation (RIS) is a process identified to be involved in most of the degradation mechanisms of austenitic stainless steels core components of nuclear reactors. During irradiation within the operating temperature range of present and possible prototypes of future reactors, chemical elements of reactor internal metallic components redistribute heterogeneously and segregate on the pre-existent and newly created defects. Local enrichments or depletions levels depend, inter alia, on the element and nature of the defect affected. While mechanisms involved in RIS have already been identified, their kinetics and mutual weigh are still poorly known. Indeed, current theorical models of RIS suffer from a lack of experimental data. These two aspects motivated this study. First, by the determination of interdiffusion kinetics after thermal ageing and under irradiation on model alloys in the Fe-Ni-Cr system. Irradiation effect on diffusion have been studied at 440°C. At this temperature, conventional methods to study diffusion fail to give access to such slow kinetics under reasonable time. Experimental implementation of the nanolayers method on binary and ternary alloys gave, in this case, encouraging results. However, discrepancies between experiments and predictions revealed the need of optimizations on both sides. In the framework of this study, recommendations have been established to account for the effect of the microstructure and the selected ageing conditions on measured effective kinetics. Experiments on model alloys have the great interest to be directly confronted to simulation. Nevertheless, nuclear steels contain a large variety of minor alloying elements and impurities which have shown to play a key role on RIS behaviour. Thus, in a second part, RIS study on various bulk defects of an optimized nuclear austenitic stainless steel, a 316L(N) has been performed. Consequently, nanoscale coupled and correlated techniques (Atom Probe Tomography and Transmission Eelectron Microscopy) have been implemented. Techniques complementarity offered the possibility to associate enrichments to defect crystallography and revealed singular tendencies of RIS depending on the element and the defect involved. Furthermore, a higher nitrogen and niobium content in the steel nominal composition formed primary Z-phase, its stability under irradiation had also been for the first time studied in austenitic stainless steels., L’implication de la ségrégation induite par l’irradiation (SIR) dans la majorité des mécanismes de dégradation des composants du cœur des réacteurs nucléaires en aciers inoxydables austénitiques est avérée. Aux températures de fonctionnement des réacteurs actuels et possibles prototypes du futur, les éléments chimiques des pièces métalliques des structures internes se redistribuent de manière hétérogène en ségrégeant sur les défauts préexistants et nouvellement créés par l’irradiation. L’intensité des enrichissements ou des appauvrissements locaux dépendent, entre autres, de la nature de l'élément et du défaut concerné. Si les différents mécanismes de ségrégation sont aujourd’hui bien identifiés, leur cinétique et leur degré d’implication respectifs sont encore peu renseignés. En effet, les modèles théoriques actuels de la SIR souffrent d'un manque de données expérimentales. Ces deux aspects sont le moteur de cette étude. Dans un premier temps, les cinétiques d'interdiffusion après vieillissement thermique et sous irradiation ont été déterminées dans des alliages modèles du système Fe-Ni-Cr. L'effet de l'irradiation sur la diffusion a été étudié à 440°C. A cette température les coefficients de diffusion sont faibles et les méthodes classiques d'étude de la diffusion ne permettent pas d'y accéder dans un délai raisonnable. La mise en œuvre expérimentale de la méthode des multicouches à période nanométrique dans des alliages binaires et ternaires a montré, dans ce cas, des résultats encourageants. Toutefois, des divergences entre les résultats expérimentaux et les prévisions théoriques ont révélé la nécessité d'optimiser ces deux approches. Dans le cadre de cette étude, des recommandations ont été formulées pour tenir compte de l’impact de la microstructure et du choix des conditions de vieillissement sur les cinétiques effectives mesurées. Les expériences réalisées sur des alliages modèles offrent l’intérêt de pouvoir être directement confrontées à la simulation. Néanmoins, les aciers de l’industrie nucléaire contiennent une grande variété d'éléments d'alliage minoritaires et d'impuretés qui jouent un rôle clé sur le comportement des éléments impliqués dans la SIR. Ainsi, dans un second temps, une étude de la SIR sur différents défauts intragranulaires d'un acier inoxydable austénitique optimisé, un 316L(N) a été réalisée. Pour cela, des techniques couplées et corrélées à l'échelle nanométrique (Sonde Atomique Tomgraphique et Miscroscopie Electronique en Transmission) ont été mises en place. La complémentarité des techniques a rendu possible l'association des enrichissements locaux observés à la cristallographie des défauts concernés. Des tendances singulières de SIR ont été observées selon l'élément et la nature du défaut impliqués. Une teneur plus élevée en azote et en niobium dans la composition nominale de l'acier a entrainé la précipitation de phase Z primaire. La stabilité de cette phase sous irradiation dans les aciers inoxydables austénitiques a pour la première fois pu être étudiée.

International audience; The behavior of solute atoms in an undersaturated Fe-3.3at.%Ni model alloy upon heavy and light particle irradiations was investigated by Atom Probe Tomography. Whereas the first irradiations, performed using 27 MeV Fe ions at 400°C, highlight significant Ni enrichment on PD clusters, solute-rich features detected after 2.8 MeV protons H+ ions (400°C) and 1 MeV electron (320°C) irradiations are only associated with pre-existent microstructural defects. Nanoindentation tests shows no hardening in light particle irradiated samples, strongly suggesting that no dislocation loop formed. These observations put forward the importance of point defect clusters resulting from displacement cascades in Ni clusters’ formation.

International audience; High resolution imaging by secondary ion mass spectrometry and atom probe tomography have been employed to investigate boron segregation at austenite grain boundaries (γ GBs) after soaking in a high-strength low-carbon steel. The combined use of these two analytical techniques is shown to be powerful for quantifying solute and segregated boron levels. Quenching was performed after soaking aiming to clarify the temperature effect on boron distribution under thermal equilibrium. Boron depletion in the γ GBs vicinity was observed in the as-quenched states from high temperatures, suggesting that the cooling rate was not fast enough to limit boron diffusion during cooling. We found that boron segregation at γ GBs increases with temperature. This is due to the increase of solute boron concentration in the grains, resulting from boride precipitate dissolution. It appears that the segregation magnitude still follows the equilibrium laws as a function of temperature. From our investigations, it was possible to determine the boron equilibrium segregation enthalpy. These results have important practical consequences for controlling the levels of segregated boron in steels.

International audience; The effect of plastic deformation (cold rolling) on the thermal aging of a lean 2202 duplex stainless steel at intermediate temperatures (280 °C-450 °C) was studied using a combination of atom probe tomography and thermoelectric power measurements. In the undeformed condition, spinodal decomposition of ferrite was shown to be the prime aging process with an activation energy of 206 kJ/mol. Plastic deformation enhanced the aging and the austenite partly transformed into martensite. When no precipitation or phase transformation was observed in austenite, both ferrite and martensite decomposed. The spinodal decomposition kinetics of ferrite increased with increasing deformation rate. The FeCr-rich martensite also decomposed at 450 °C, via a nucleation and growth process, owing to a decrease of the Cr content in the matrix. Chromium nitrides, G-Phase, and Cu-rich precipitates were observed in the decomposed martensite.

The change of silicon optical and electrical properties induced by size reduction, due to the quantum confinement of charged carriers, is a well-known effect and allowed to develop new optoelectronic devices. As in bulk silicon, doping should allow to optimize these properties in nanostructured silicon. However, the characteristics of doping of nanostructured silicon still misunderstood and many questions, concerning the location of impurities and their activation state, remain unanswered. Moreover, in these materials, the environment of impurities seems to inuence strongly all of their properties. The purpose of this thesis is to get a better understanding of structural characteristics of doping at the atomic scale in function of the nature of the impurity, the host matrix, and the elaboration technic. In this way, we have investigated two di_erent systems using atom probe tomography. The first concerns a rare earth doping of hafnium silicates. We have evidenced that the clustering of HfO2 nano-grains crystallized in their cubic form induced an efficient energy transfer with praseodymium ions. The second system concern the n and p type doping of silicon nanocrystals embedded in silica. We have demonstrated the important introduction of n type impurities (As, P) in the core of every nanocrystals, independently of the elaboration technic. This introduction of impurities should allow the formation of highly doped silicon nanocrystals. A different behavior has been observed in the case of p type doping, represented by the aggregation of Boron at the interface between the nanocrystals and the silica matrix.; La modification des propriétés optiques et électriques du silicium apportée par la réduction en taille, notamment due aux effets de confinement quantique des porteurs de charges, est aujourd'hui bien connue et a permis le développement de nouveaux systèmes en optoélectronique. Comme dans le cas du silicium massif, le dopage devrait permettre d'optimiser les propriétés du silicium nanostructuré. Cependant, les caractéristiques du dopage dans le silicium nanostructuré sont encore mal comprises et de nombreux questionnements, concernant la localisation des impuretés ainsi que leur état d'activation, restent en suspens. De plus, l'environnement des impuretés semble avoir une influence majeure sur l'ensemble des propriétés. Cette thèse vise à mieux comprendre les caractéristiques structurales du dopage à l'échelle atomique en fonction de la nature de l'impureté, de la matrice hôte et de la technique d'élaboration. Pour cela, nous avons étudié deux types de système en sonde atomique tomographique. Le premier concerne un dopage aux ions de terres rares dans les silicates d'hafnium. Nous avons mis en évidence que la formation de nano-grains de HfO2 cristallisés sous la forme cubique permet un transfert d'énergie efficace vers les ions praséodyme. Le second porte sur les dopages de type n et p de nanocristaux de silicium insérés dans la silice. Nous avons démontré l'introduction des impuretés de type n (As, P) au cœur des nanocristaux, indépendamment de la technique d'élaboration, permettant de réaliser des forts dopages. Un comportement différent a été mis en évidence pour les impuretés de type p, avec l'accumulation de Bore aux interfaces entre les nanocristaux et la matrice.

International audience; Correlative analysis is a powerful way to relate crystallographic and chemical information to the properties of materials. In this work, a procedure is proposed to select and analyze interfaces of polycrystalline thin film materials through correlative transmission Kikuchi diffraction/energy dispersive X-ray spectroscopy (TKD/EDS), transmission electron microscopy (TEM) and atom probe tomography (APT). TKD provides information on the crystallographic orientation. The EDS analysis performed together with TKD in the scanning electron microscope (SEM) makes chemical information available allowing phases of similar crystal structure, but with a different composition to be distinguished. The information of TKD/EDS can be correlated to successive TEM and APT analysis on selected interfaces for structural and chemical analysis at the atomic scale. An interface of an epitaxial orientated grain of a polycrystalline CoSi2 thin film on (111)Si is selected and analyzed. The selected interface has a twin character and shows facets of different orientation and area. Site-specific segregation of Ge to junctions of the facets is evidenced. The correlation between local strain from misfit (defects) at the interface and segregation is discussed.

Stainless steel is an important alloy for the technical development of a modern society, they were discovered in the early 20th century. However, their base alloying system, Fe-Cr, is affected by a low temperature (, L’acier inoxydable est un alliage important pour le développement technique d’une société moderne; cela a été découvert au début du 20ème siècle. Cependant, leur système d'alliage de base, Fe-Cr, est affecté par une lacune de miscibilité à basse température (

International audience; In a previous work, the chemical elements necessary for the achievement of Niobium base superalloys were defined in order to get a structure equivalent to that of Nickel base superalloys, which contain ordered precipitates within a disordered solid-solution. It was especially emphasized that precipitation hardening in the Niobium matrix would be possible with the addition of Ni. The remaining question about the design of such Niobium superalloys concerned the achievement of ordered precipitates in crystalline coherence with the Nb matrix i.e., with a crystalline structure equivalent to the Nb crystal prototype and with a lattice parameter in coherency with that of the Nb matrix. In order to reduce the trial/error experimental work, a reasoning based on various data for the achievement of coherency is presented. Then, starting from the Nb-Hf-Ni ternary alloy thus defined, this paper demonstrates that the precipitation of an ordered Nb phase within a disordered Nb matrix can be achieved with lattice parameter coherency between the ordered precipitates and the disordered matrix. The chemistry and the crystallographic structure of the precipitates were characterized using Transmission Electron Microscopy and Atom Probe Tomography. These results can help to conceive a new family of Nb base superalloys.

International audience; The alloying elements partitioning from martensite to austenite during quenching and partitioning was analysed by coupling in situ High Energy X-Ray Diffraction experiments and 3D atom probe tomography. A rapid and significant carbon partitioning from martensite to austenite without any partitioning of both Mn and Si was highlighted while the interface was immobile. After a relatively short time at 400 °C, a clear Mn partitioning occurs at the vicinity of the α′/γ interface. The analysis conducted indicates that manganese equilibrates its chemical potential during partitioning and raises the issue of the Constrained Carbon Equilibrium model applicability throughout the partitioning process.

International audience; Dewetting or agglomeration is a crucial process in material science since it controls the stability of thin films or can be used for film nanostructuration by formation of islands. The models developed for dewetting usually assume diffusion at the interface and/or at the surface but no direct evidence of such diffusion was demonstrated. Moreover, these models are usually dealing with elemental materials and not with compounds in which several elements can diffuse. The mechanisms behind agglomeration of polycrystalline compounds thin film are still not fully understood. In this work, Si isotope multilayers coupled with atom probe tomography (APT) are used to reveal the agglomeration mechanism of NiSi, a binarycompound. The diffusion of Si, the less mobile species in NiSi, at the NiSi/Si interface is demonstrated through comparison between the three dimension redistribution of the Si isotopes determined by APT and models taking into account grooving and agglomeration. The implication for the understanding and control of agglomeration in poly-crystalline compound thin films are highlighted.

International audience; Intergranular segregation in a Fe-P-C model alloy after thermal ageing at 650°C has been examined using Atom Probe Tomography (APT) and Auger Electron Spectroscopy (AES). The specimens were prepared using site-specific method combining focused ion beam (FIB) and Transmission Kikuchi Diffraction (TKD). Grain boundary's five macroscopic degrees of freedom (DOFs) were determined from the TKD map supplemented by 3D APT reconstruction. The phosphorus intergranular segregation values obtained using APT and AES techniques were in excellent agreement and equal to 1.6 ± 0.7 at/nm2 and 1.4 ± 0.5 at/nm2, respectively. The two techniques are complementary for a fully quantitative description. © 2018 Elsevier B.V.

This PhD thesis focuses on the study of SiC, doped with Fe in order to elaborate a diluted magnetic semiconductor at room temperature for spintronic applications. The iron doping was carried out by ion implantation of multi-energy type (30-160 keV) at different fluences, leading to a 2% constant atomic concentration between 20 to 100 nm, followed by a high temperature annealing in the goal of homogenizing the dopant concentration. The implantation temperature during this process is 550 °C, in order to avoid amorphization. The optimization of the magnetic and electronic properties of SiC-Fe, as well as the understanding of the physical mechanisms at the origin of induced magnetism, require a thorough characterization of the microstructure of the implanted materials. The objectives of this work are, on the one hand, to carry out an atomic scale study of the nanostructure according to the implantation conditions (temperature, fluence) and the post-implantation annealing and the other hand, to characterize the magnetic properties of implanted materials. In this work, we have shown by atom probe tomographic, the existence of nanoparticles whose the average size increases with the annealing temperature. The chemical mapping of the nanoparticles shows the presence of the Fe-rich phases for the annealed samples. Magnetic study (Mössbauer spectrometry and Squid) shows the ferromagnetic contribution is due to the magnetic nanoparticles and/or the diluted Fe atoms in the matrix. The correlation between structural and magnetic properties allowed showing that diluted Fe atoms and substitute to Si sites contribute to the ferromagnetic contribution below 300 K. In coupling many characterization techniques in order to give a detailed description of the different studied samples, we have shown that the size and nature of the phase present in the nanoparticles depend on the implantation conditions and the annealing temperatures and consequently it is necessary to anneal our samples at high temperature to reveal ferromagnetic order., Ce travail de thèse porte sur l’étude du carbure de silicium, dopé avec du fer dans le but de réaliser un semi-conducteur magnétique dilué à température ambiante pour des applications à la spintronique. Le dopage en fer a été réalisé par implantation ionique de type multi-énergie (30 - 160 keV) à différentes fluences, conduisant à une concentration atomique constante de 2 % de 20 à 100 nm. Il a été suivi d’un recuit à haute température dans le but d’homogénéiser la concentration en dopants. Les implantations se sont déroulées à une température de 550 °C. L’optimisation des propriétés magnétiques et électroniques du SiC–Fe, de même que la compréhension des mécanismes physiques à l’origine du magnétisme induit, ont nécessité une caractérisation poussée de la microstructure des matériaux implantés. Les objectifs de ce travail ont été d’une part, de réaliser une étude à l’échelle atomique de la nanostructure en fonction des conditions d’implantations (température, fluence) et des traitements thermiques post-implantation, et d’autre part, de déterminer les propriétés magnétiques des matériaux implantés. Dans ce travail, nous avons montré par Sonde Atomique Tomographique, la présence de nanoparticules dont la taille moyenne augmente avec la température de recuit. La cartographie chimique des nanoparticules a permis de révéler l’existence de phases riches en Fe pour les échantillons recuits. L’étude magnétique (spectrométrie Mössbauer et Squid) a montré que la contribution ferromagnétique est due principalement aux nanoparticules magnétiques et/ ou aux atomes de fer magnétiques dilués dans la matrice. La corrélation entre les propriétés structurale et magnétique a permis de montrer que les atomes de fer dilués dans la matrice et substitués sur sites de silicium contribuent au signal ferromagnétique en dessous de 300 K. Nous avons donc montré dans ce travail, que la taille et la nature des phases présentes dans les nanoparticules dépendent des conditions d’implantation et des températures de recuit et qu’il est nécessaire de recuire les échantillons à haute température pour faire apparaître un ordre ferromagnétique.

The aim of the thesis is to develop a methodology for the investigation of Mg which acts as p-type doping in GaN. We relate the spatial distribution of the dopants with their electrical activity which is achieved by coupling two complementary approaches, Atom Probe Tomography (APT) and Off-axis electron holography. These measurements have also been combined with high-resolution electron microscopy (HR-(S)TEM) for the structural characterization. APT is a unique characterization technique, based on the field effect evaporation of individual atoms of a needle shape sample, allowing the analysis of nano-devices both in terms of morphology and composition in three dimensions at the atomic scale. Off-axis electron holography uses an electron biprism to form an interference pattern from which the electrostatic potential arising from the active dopants can be determined. In this work the experimental procedure has been optimized for both techniques including specimen preparation, the microscope parameters and data treatment to recover accurate information about the position and activity of the dopants. For the holography measurements, a careful analysis of the artifacts that are present in these specimen has been performed to understand the effects of specimen preparation and charging under electron irradiation. We have performed these experiments at high temperature in-situ in the TEM (400 °C) as this increases the ionized dopant concentrations and reduces the artifacts that are present in our measurements. Having developed the methodology, these two techniques are then used to study the effect of temperature and dopant concentrations on the growth of Mg-doped GaN by MOCVD. We have been able to show by APT the existence of precipitates of Mg which are present from a concentration of 3E19 cm-3 whose size and density depends on the growth temperature and the total nominal dopant concentration. Their presence reduces the concentration of dopants that are potentially active in the specimens. However, the measurements of active dopants by holography combined with simulations suggest that the presence of these precipitates do not dominate the electrical properties of the material and that even in very highly doped specimens up to 2E20cm-3 the total active dopant concentrations are still higher than expected from previously published studies. The correlation between these techniques will provide valuable information to improve the Mg activation GaN which is currently a big issue for device manufacture., La thèse porte sur l’étude du dopage de type p, à base de Mg, dans des nanostructures de GaN, dans le but de relier la distribution spatiale du dopant à son activité électrique grâce à la corrélation entre sonde atomique tomographique (APT) et holographie électronique hors axe optique réalisée dans un microscope électronique en transmission (TEM). L'APT est une technique de caractérisation qui repose sur l'évaporation par effet de champ des atomes de surface d'un échantillon, permettant l'analyse en termes de morphologie et de composition, en trois dimensions et à l'échelle atomique. L'holographie électronique hors axe optique fournit des cartographies du potentiel électrostatique introduit par les dopants actifs électriquement. Dans un premier temps, les conditions expérimentales ont dues être optimisées pour chacune des techniques, incluant la préparation des échantillons, les conditions de mesure ainsi que le traitement des données, de façon à obtenir des données fiables et les plus quantitatives possibles. Une analyse soigneuse et détaillée des artefacts et des erreurs qu’ils introduisent est rapportée. Il a en particulier été montré que réaliser les expériences d’holographie in-situ à haute température (400 °C) grâce à un porte-objet chauffant permettait d’augmenter très significativement le signal lié au dopage et ainsi accroitre la sensibilité de la mesure. Dans un deuxième temps, ces deux méthodes d’analyse ont été corrélées pour étudier d’une part l’influence de la température de croissance en MOCVD, d’autre part celle de la concentration nominale en dopants dans des nanostructures dédiées GaN. Nous avons pu confirmer grâce à l’APT l’existence de précipités riches en Mg dès que la concentration nominale excède environ 3E19 cm-3, dont la densité augmente avec la concentration nominale et diminue avec la température de croissance. Leur présence diminue la concentration en dopants potentiellement actifs situés dans la matrice en dehors de ces précipités. Néanmoins, les résultats obtenus par holographie, appuyés par des simulations numériques, indiquent que ces précipités n’auraient pas un rôle prépondérant dans la variation du potentiel électrostatique en fonction de la concentration nominale en dopants même pour des concentrations en Mg qui s’élèvent à 2E20cm-3.

Cast austenitic-ferritic stainless steels are used in primary circuit of 2nd generation nuclearpower plants. At operating temperature (285 °C - 325 °C), evolution of mechanical properties isobserved due to the phase transformations occurring within the ferrite: spinodal decomposition (SD)leading to the formation on a Fe rich phase (α) and a Cr rich phase (α ') and the precipitation of the G-phase at α/α' interfaces. This evolution of the mechanical properties can be prohibitive for thecomponents. If it is well known that the steel composition plays an important role on the evolution ofthe properties (steels less rich in Ni and Mo are less sensitive to aging), the role of solute elements asNi, Mo and Mn on the aging kinetics is not yet known so as the contribution of the G-phase on thehardening during the thermal aging. In this study, the aging kinetics of the ferrite of some austenitic-ferritic stainless steels with or without Mo and model alloys with tuned compositions have been studied by atom probe tomography (APT) and by micro hardness measurements. This works answered the three following questions: - Quantification of the contribution of both spinodal decomposition and G-phase precipitation on hardening of the ferrite: combination of hardening models and data obtained with APT permitted to show that G-phase precipitation is clearly the main contributor to ferrite hardness increase at early stage of ageing in Mo-bearing steels. This is due to the high number density of G-phase particles. In Mo-free steels which have ten times less G-phase particles, contributions of both spinodal decomposition and G-phase precipitation are similar. In both cases, when coarsening of G-phase particles occurs and SD is well developed, SD contribution becomes larger. - Influence of Ni, Mo and Mn on aging kinetics: The study of model alloys with tuned composition has shown that only Ni plays a role on SD by enhancing the decomposition. Mn is a key element for the precipitation of G-phase particles at α/α' interfaces. - Efficiency of regeneration heat treatment at 550 °C: an alternative to component replacement is to perform a heat treatment at higher temperature in order to restore the properties of the components. The heat treatments performed permitted to entirely restore the mechanical properties of Mo free steels and partially the properties of Mo bearing ones. This is due to the presence of undissolved G-phase particles in the case of Mo bearing alloys. In each case, SD was totally dissolved.; Les aciers inoxydables austéno-ferritiques moulés sont utilisés pour certains composants ducircuit primaire des centrales nucléaires de génération II. Aux températures de service (285 °C -325 °C), des modifications de propriétés mécaniques sont observées. Elles sont imputables auxtransformations de phases au sein de la ferrite de ces aciers : la décomposition spinodale (DS) en phaseα (riche en Fe) et α’ (riche en Cr) et la précipitation de la phase G aux interfaces α/α'. S'il est admisque la composition de l’acier influe sur l’évolution des transformations de phase de la ferrite (lesaciers moins riches en Ni et Mo sont moins sensibles au vieillissement), aucune étude, à ce jour, n'apermis de mettre en évidence l'influence du Ni, Mo, Mn et des synergies éventuelles sur levieillissement des aciers ni de déterminer la contribution de la phase G à l'évolution des propriétésmécaniques. Dans cette étude, les cinétiques des transformations de phase de la ferrite d'aciers inoxydable austéno-ferritiques pauvres en Mo et riches en Mo ainsi que de celle d'alliages modèles decomposition ciblée ont été étudiées par sonde atomique tomographique (SAT) et par mesure demicrodureté. Les travaux ont répondu aux trois questions suivantes : - Quantification de la contribution des différentes phases au durcissement : L'utilisation conjointe de modèles de durcissement et des données obtenues par la sonde a montré que, contrairement à ce qui était dit dans la littérature, la phase G est le contributeur majoritaire au durcissement pour les aciers avec Mo. Ce n'est qu'aux temps longs, lorsque la coalescence des particules de phase G intervient et que la DS est plus développée que la contribution de la DS devient prépondérante. Ceci est dû à la forte densité de particules de phase G dans ces aciers. Dans le cas des aciers sans Mo qui contiennent dix fois moins de particules en début de cinétique, la phase G et la DS ont des contributions équivalentes. – Influence du Ni, Mo et Mn : L’étude d’alliage modèles de compositions ciblées a montré que seul le Ni accélère la décomposition spinodale et que le Mn a un rôle prépondérant dans la formation des particules de phase G aux interfaces α/α’. - Efficacité d'un traitement thermique de régénération à 550 °C: Une alternative au remplacement des composants les plus vieillis pourrait être un traitement thermique dit de régénération. Les recuits à 550 °C permettent de restaurer entièrement les aciers sans Mo et partiellement les aciers avec Mo. Ceci est dû au fait que les particules de phase G ne sont pas entièrement dissoutes dans le cas des aciers avec Mo, induisant un durcissement résiduel. Dans tous les cas, la DS est entièrement dissoute.

Ce mémoire porte sur mes activités de recherche dans la période 2009-2018, et il est organisé decette façon :Le chapitre 1 illustre une sélection de travaux menés durant mon post-doc à l’IEF Orsay, avec une attention particulière aux aspects de microscopie corrélative, i.e. aux analyses effectuées en spectroscopie optique et microscopie électronique sur des nano-objets uniques.Le chapitre 2 est une introduction à la physique de l’évaporation par effet de champ et à la technique de la SAT. On y introduit des notions qui seront développées dans les chapitres successifs.Le chapitre 3 traite des biais de mesure de composition dans l’analyse des semiconducteurs composés par SAT. Les résultats marquants sont ici la démonstration que ces biais sont liés au champ électrostatique de surface, l’estimation des rendements de détection des différentes espèces dans le GaN et dans l’AlGaN et l’interprétation d’une partie des mécanismes physiques conduisant à la nondétection de certaines espèces. Ce volet d’études ne faisait pas à l’origine partie de mon projet de maître de conférences, mais au cours de mes activités je me suis rendu compte de l’importance des aspects métrologiques pour les analyses corrélatives basées sur la SAT. Aujourd’hui, la métrologie compositionnelle en SAT constitue un de mes centres d’intérêt principaux, ouvert à de nombreuses hypothèses d’approfondissement.Le chapitre 4 illustre les approches de microscopie corrélative basées sur la SAT, en particulier celle que j’appelle l’approche statistique, et l’approche séquentielle sur un nano-objet unique. Un exemple d’approche statistique est illustré dans le cas de l’analyse de l’AlGaN, où les informations spectroscopiques ont été exploitées pour corriger le biais de mesure de composition dans cet alliage. L’approche à nano-objet unique est développée pour les puits quantiques InGaN/GaN et pour les boites quantiques GaN/AlGaN. L’importance de la microscopie électronique comme technique complémentaire y est particulièrement soulignée. Le chapitre se termine sur les possibles développements de la microscopie corrélative basée sur la SAT.Le chapitre 5 décrit la spectroscopie de µPL in situ dans une SAT, à travers les études menées sur le diamant et sur des hétérostructures à base d’oxydes. Le premier résultat marquant obtenu grâce à cette approche est la mesure de la contrainte mécanique induite par l’application du champ au sommet de nano-aiguilles en diamant. Le chapitre et le mémoire se concluent sur les nombreuses perspectives ouvertes par ce nouvel instrument.

Laser-assisted Atom Probe Tomography (La-APT) is a powerful tool for investigating the 3D atomic distribution of the chemical species in a wide variety of semiconductor materials. However, important compositional biases affect atom probe analyses revealing a non-stoichiometric composition. In the thesis a systematic study of selected binary (GaN, GaAs, ZnO) and ternary (AlGaN, MgZnO) semiconductors of high technological interest was developed in order to: (i) obtain a coherent description of the compositional biases in APT; (ii) identify the physical mechanisms leading to these biases; (iii) assess the experimental conditions for which the compositional analysis is reliable. In order to interpret the results, the hypothesis of preferential evaporation of metallic species (Ga, Al, Zn, Mg) at high field and emission of neutral non-metallic molecules (N2, O2) at low field has been proposed. Another important aim of this thesis is materials physics-oriented. It is indeed of utmost importance to study both composition and morphology of some devices of technological interest, such as in multi-quantum-well systems. In this perspective, the knowledge of 3D composition field and morphology is essential because these features determine the optical and electrical properties of the systems. In order to do it, a correlative microscopy approach can be adopted. This approach was successfully applied to the study of ZnO/MgZnO multi-quantum wells designed for quantum cascade lasers. Structural, compositional and optical properties were investigated performing correlative La-APT - Electron Tomography (ET) - micro-PhotoLuminescence (µ-PL) on the same atom probe tip specimens. The complementary APT and ET analyses yield a clear picture of the structure and composition of the system investigated, revealing important decomposition phenomena in the MgZnO alloy. In particular, La-APT proved to be a unique technique for a direct assessment of local composition. Moreover, µ-PL also proved to be extremely useful in order to get information related the composition, supporting La-APT results. Finally, a new correlative in-situ approach in which La-APT and µ-PL are simultaneously performed is presented. Thanks to the development of a specially designed tomographic atom probe, it is shown that µ-PL can be successfully performed on a single Zno/MgZnO atom probe tip during La-APT. This is extremely attractive and challenging because allows to strictly correlating the variation photoluminescence signal with nano-metric scale volumes of the tip evaporated during APT. In principle, the emission of single quantum light emitters (i.e. single QW or QD) can be revealed. The new approach presented can be extended to a wide range of materials, opening new perspectives for correlative studies of single atom probe tips., La Sonde Atomique Tomographique (SAT) assistée par laser (La-APT) est un outil puissant pour étudier la distribution atomique 3D des espèces chimiques dans une grande variété de matériaux semi-conducteurs. Cependant, des biais de composition importants affectent les analyses de sondes atomiques révélant une composition non stœchiométrique. Dans cette thèse, une étude systématique de certains semi-conducteurs binaires (GaN, GaAs, ZnO) et ternaires (AlGaN, MgZnO) a été menée afin de: (i) obtenir une description cohérente des biais de composition en APT; (ii) identifier les mécanismes physiques à l'origine de ces biais; (iii) évaluer les conditions expérimentales pour lesquelles l'analyse compositionnelle est fiable. Afin d’interpréter les résultats, l’hypothèse de l’évaporation préférentielle d’espèces métalliques (Ga, Al, Zn, Mg) à haut champ et d’émission de molécules neutres non métalliques (N2, O2) à champ bas a été proposée. Un autre objectif important de cette thèse est orienté vers la physique des matériaux. L’étude de la composition et la morphologie de certains dispositifs d’intérêt technologique, tels que les systèmes à multi-puits quantiques, est très important. Dans cette perspective, la connaissance du champ de composition 3D et de la morphologie de ces systèmes est essentielle car ces caractéristiques déterminent leurs propriétés optiques et électriques. Pour ce faire, une approche par microscopie corrélative peut être adoptée. Cette approche a été appliquée avec succès à l'étude des multi-puits quantiques ZnO/MgZnO conçus pour les lasers à cascade quantique. Les propriétés structurales, compositionnelles et optiques ont été étudiées en effectuant la tomographie par électrons (ET) - micro-photoluminescence (µ-PL) corrélative sur les mêmes échantillons de sonde atomique. Les analyses complémentaires APT et ET donnent une image claire de la structure et de la composition du système étudié, révélant d'importants phénomènes de décomposition dans l'alliage MgZnO. En particulier, La SAT s’est révélé une technique unique pour une évaluation directe de la composition locale. De plus, la µ-PL apparait extrêmement utile pour obtenir des informations relatives à la composition, en lien avec les résultats de La-APT. Enfin, nous présentons une nouvelle approche in-situ corrélative dans laquelle les mesures APT et µ-PL sont exécutées simultanément. Grâce au développement d'une sonde atomique tomographique spécialement conçue, on démontre que la µ-PL peut être mesurée avec succès sur une pointe de sonde atomique Zn :/MgZnO pendant nos analyse. Ceci est extrêmement attrayant car cela permet de corréler strictement le signal de photoluminescence avec les volumes explorés à l'échelle nanométrique. En principe, émission depuis des d'émetteurs de lumière quantiques uniques (c'est-à-dire un seul QW ou QD) peut être révélée. La nouvelle approche présentée peut être étendue à un large éventail de matériaux, ouvrant de nouvelles perspectives pour les études corrélatives.

Understanding the composition evolution of grain boundaries and grain boundary precipitation at near-atomic scale in aluminum alloys is crucial to tailor mechanical properties and to increase resistance to corrosion and stress corrosion cracking. Here, we elucidate the sequence of precipitation on grain boundaries in comparison to the bulk in a model Al-Zn-Mg-Cu alloy. We investigate the material from the solution heat treated state (475{\textdegree}C), through the very early stages of aging to the peak aged state at 120{\textdegree}C and further into the overaged regime at 180{\textdegree}C. The process starts with solute enrichment on grain boundaries due to equilibrium segregation accompanied by solute depletion in their vicinity, the formation of Guinier--Preston (GP) zones in the solute-enriched grain boundary regions, and GP zones growth and transformation. The equilibrium segregation of solutes to grain boundaries during aging accelerates this sequence compared to the bulk. Analysis of the ~10 nm wide precipitate-free zones (PFZs) adjacent to the solute-enriched grain boundaries 2 shows that the depletion zones are determined by (i) interface equilibrium segregation; (ii) formation and coarsening of the grain boundary precipitates and (iii) the diffusion range of solutes in the matrix. In addition, we quantify the difference in kinetics between grain boundary and bulk precipitation. The precipitation kinetics, as observed in terms of volume fraction, average radius, and number density, is almost identical next to the depletion zone in the bulk and far inside the bulk grain remote from any grain boundary influence. This observation shows that the region influenced by the grain boundaries does not extend beyond the PFZs.

The redistribution of C and N during tempering of X4CrNiMo16-5-1 martensitic stainless steel containing 0.034 wt% C and 0.032 wt% N was studied using in-situ synchrotron X-ray diffraction (XRD) and atom probe tomography (APT). The unit cell volume of martensite decreased continuously during tempering. APT showed that this volume decrease is accounted entirely for by segregation of the interstitial atoms, implying that in low interstitial martensitic stainless steel stress relaxation only contributes negligibly to changes in the martensite unit cell volume.

International audience; Origin of Pr 3+ luminescence in hafnium silicate films: combined atom probe tomography and TEM investigations To cite this article: Rémi Demoulin et al 2018 Nano Futures 2 035005 View the article online for updates and enhancements. Related content Thermal stability of high-k Si-rich HfO2 layers grown by RF magnetron sputtering L Khomenkova, X Portier, J Cardin et al.-High-k Hf-based layers grown by RF magnetron sputtering L Khomenkova, C Dufour, P-E Coulon et al.-XRD and EXAFS studies on the structure of Er3+-doped SiO2–HfO2 glass-ceramic waveguides: Er3+-activated HfO2 Abstract Structural, chemical, and luminescence properties of Pr 3+-doped HfSiO x layers fabricated by radio-frequency magnetron sputtering were examined as a function of annealing temperature. Phase separation between SiO 2 and HfO 2 as well as the location of Pr 3+ dopants were investigated using atom probe tomography and transmission electron microscopy while optical properties of Pr 3+ ions were studied using photoluminescence measurements. As a result, (i) we evidenced the location of the Pr 3+ dopants in the HfO 2 phase while the SiO 2 phase was discovered to be free of these dopants, (ii) the HfO 2 phase was identified to crystallize in the cubic phase until 1050 °C annealing, (iii) no Pr clusters were detected as function of annealing, and (iv) luminescence properties were discussed in regard to the location of Pr in the HfO 2 cubic phase.

International audience; Often the experimentally-observed, single-phase high entropy alloy (HEA) is the result of second-phase precipitation constrained by thermodynamic and kinetic factors. Using Al0.3CoCrFeNi as a candidate HEA, this paper demonstrates the strong influence of thermo-mechanical processing on the transformation pathway adopted for isothermal second-phase precipitation. A traditional thermo-mechanical processing route comprised of homogenization cold-rolling solution treatment in the single fcc phase region, followed by a precipitation anneal at a lower temperature, results in a homogeneous distribution of nanometer scale-ordered L12 (gamma prime-like) precipitates within the fcc matrix. In contrast, if cold-rolling is followed directly by annealing at the precipitation temperature, then the resulting microstructural evolution pathway changes completely, with concurrent recrystallization of the matrix fcc grains and precipitation of B2 and sigma phases, largely at the grain boundaries. These experimentally observed variations in transformation pathway have been rationalized via the competition between the thermodynamic driving force and activation barrier for second-phase nucleation in this alloy, coupled with the kinetics of the process. The microstructural variations that result from these dramatically different phase transformation pathways can lead to some rather exceptional mechanical properties that can be varied over a large range even for a single Al0.3CoCrFeNi HEA composition.

International audience; Er-doped silica or rich-silicon oxide has been widely studied as 1.54 mm emitters. The incorporation of Si-nanoclusters is known for improving luminescence yield of Er 3þ ions through an efficient sensitization of the neighboring rare earth ions. The aim of this work is to investigate the influence of Silicon excess and Erbium concentration on the formation of silicon nanoclusters and Er-rich phase responsible of the quenching of 1.54 mm emission. Atom probe tomography and photoluminescence spectroscopy were used to explore the nanostructure and optical activity of Er-doped silicon rich silica. We present a direct evidence that both silicon excess and Er concentration influence the growth of Si nanoclusters, the formation of Er-Si-O phase and the luminescence of both Si nanoclusters and Er 3þ ions. We explain these finding in relation with the growth mechanism of nanoparticles and the presence of a Snowman-like Janus morphology between silicon and Erbium silicates nanoparticles. In this contribution, we deciphered the nanoscale structure spatial correlations between Er and Si atoms in Er doped SiO X which remained unclear for a long time.

International audience; Using atom probe tomography, it is demonstrated that Mg doping of GaN nanowires grown by Molecular Beam Epitaxy results in a marked radial inhomogeneity, namely a higher Mg content in the periphery of the nanowires. This spatial inhomogeneity is attributed to a preferential incorporation of Mg through the m-plane sidewalls of nanowires and is related to the formation of a Mg-rich surface which is stabilized by hydrogen. This is further supported by Raman spectroscopy experiments which give evidence of Mg-H complexes in the doped nanowires. A Mg doping mechanism such as this, specific to nanowires, may lead to higher levels of Mg doping than in layers, boosting the potential interest of nanowires for light emitting diode applications.

International audience; Atom probe tomography (APT) has emerged as a valuable tool in the study of nitride semiconductors, despite the challenges involved in achieving controlled field evaporation. In optoelectronics, it has provided insights into the nanostructure of light emitting diodes, laser diodes and microwires. In electronics, it has allowed insights into impurity doping and alloying effects in transistors. Coupled with direct correlative studies using other techniques and theoretical modelling based on the APT data, the availability of three dimensional compositional information on nitride heterostructures has had (and will continue to have) a profound impact on the design and development of devices.

International audience; The segregation of solutes at dislocations in a polycrystalline and a single crystal nickel-based superalloy is studied. Our observations confirm the often assumed but yet unproven diffusion along dislocations via pipe diffusion. Direct observation and quantitative, near-atomic scale segregation of chromium and cobalt at dislocations within γ' precipitates and at interfacial dislocations leading to the partial or complete dissolution of γ' precipitates at elevated temperatures is presented. Our results allow us to elucidate the physical mechanism by which pipe diffusion initiates the undesirable dissolution of γ' precipitates.

International audience; From direct observations and measurements at the atomic scale by atom probe tomography coupled with calculations using a mean-field approach, the kinetics of low temperature carbon redistribution in martensite of a Fe-Ni-C alloy was investigated for duration up to 4.5 years. The results suggest that the ordered phase α″-Fe16C2 forms via a spinodal decomposition mechanism, in the range of a few days at room-temperature. The approach proposed allows us to retrieve the diffusivity of carbon in a large composition range during decomposition kinetics of martensite into α-Fe and α″-Fe16C2.