Search

International audience; Sn addition is known to affect the recrystallisation texture development of α-Fe. To date, even though the texture effect of Sn has been experimentally verified, although indirectly, the underlying mechanism has not been studied. In the present study, we focus on the static recrystallisation behaviour of Fe-Si-Sn alloys to elucidate the texture preferences assessing for the effect of Sn solute addition. The experimental approach is based on sequential electron back-scatter diffraction (EBSD) texture analysis complemented with a site-specific chemical analysis of grain boundaries by atom probe tomography (APT). It is directly observed through our experiments that boundary segregation phenomena, which occur at the onset and later stages of recrystallisation, provide crucial understanding of texture development. Especially in the presence of solute, the orientation dependence of the energy stored by deformation largely determines the course of the subsequent recrystallisation. In the Fe-Si-Sn alloy, recrystallisation at the highest stored energy grains is hindered. As a consequence of segregation, {001}〈110〉 orientations recrystallise at later stages of the process, forming predominantly grains with orientations of the {h11}〈1/h12〉 fibre. These grains are suggested to form following an oriented nucleation mechanism, while their growth could be interpreted with Ibe and Lücke's theory of selective growth of 26.5 〈110〉 boundaries in bcc metals.

International audience; The properties of engineering materials can be improved by optimising the microstructural developments during annealing processes. Here, we investigate the effect of Sn on the recovery annealing of cold rolled Fe-3%Si alloys. We use a multiscale approach combining micro hardness, electron back scattering diffraction (EBSD), and dark field X-ray microscopy (DFXM): a recent, nondestructive synchrotron-based technique that allows 3D mapping of orientation and lattice strain within individual grains embedded in bulk samples. Micro hardness results show that the Sn solute has a strong effect on the recovery kinetics. These results are compared to a physical kinetic model suggesting that Sn limits the softening. This observation is further discussed by a complementary atomistic modelling that demonstrates solute-dislocation interaction around edge dislocations. In situ DFXM experiments reveal the 3D microstructural evolution upon annealing at the grain level with high angular resolution. The DFXM observations show that Sn slows the recovery kinetics within individual grains, in agreement with the other microscopic investigations. Furthermore, the DFXM results provide a direct observation of strain fields around dislocation loops in an embedded single grain, which is argued to remain static due to solute effect during recovery.

International audience; Solute atoms play an important role in grain boundary migration phenomena, which are critical for understanding the microstructural evolution in metals. Here we investigate the segregation of Sn at migrating boundaries during recrystallisation in a ferritic Fe-Si-Sn alloy using atom probe tomography. Experimental evidence demonstrated a solute depletion zone ahead of the moving interface. The behaviour of Sn was rationalized with the Cahn solute drag theory. A difference in segregation between two investigated migrating interfaces was observed and could be explained by a variation in either binding energy, intrinsic velocity of each interface or in driving force.

The industrial production route of Grain Oriented Electrical Steels (GOES) is complex and fine-tuned for each grade. Its metallurgical process requires in all cases the abnormal grain growth (AGG) of the Goss orientation during the final high temperature annealing (HTA). The exact mechanism of AGG is not yet fully understood, but is controlled by the different inhibition systems, namely MnS, AlN and CuxS, their size and distribution, and the initial primary recrystallized grain size. Therefore, among other parameters, the initial heating stage during the HTA is crucial for the proper development of primary and secondary recrystallized microstructures. Cold rolled 0.3 mm Cu-bearing Grain Oriented Electrical Steel has been submitted to interrupted annealing experiments in a lab tubular furnace. Two different annealing cycles were applied:• Constant heating at 30°C/h up to 1000°C. Two step cycle with initial heating at 100°C/h up to 600°C, followed by 18 h soaking at 600°C and then heating at 30°C/h up to 1050°C. The materials are analyzed in terms of their magnetic properties, grain size, texture and precipitates. The characteristic magnetic properties are analyzed for the different extraction temperatures and Cycles. As the annealing was progressing, the coercivity values (Hc 1.7T [A/m]) decreased, showing two abrupt drops, which can be associated to the on-set of primary and secondary recrystallization. The primary recrystallized grain sizes and recrystallized fractions are fitted to a model using a non-isothermal approach. This analysis shows that, although the resulting grain sizes were similar, the kinetics for the two step annealing were faster due to the lower recovery. The on-set of secondary recrystallization was also shifted to higher temperatures in the case of the continuous heating cycle, which might end in different final grain sizes and final magnetic properties. In both samples, nearly all the observed precipitates are Al-Si-Mn nitrides, ranging from pure AlN to Si4Mn-nitride.

Sample thinning for TEM observation introduces large changes with respect to the initial strain state of the bulk sample and particularly relaxation via the free surfaces which leads to HOLZ lines splitting in the CBED pattern. This phenomenon has been simulated owing to extensive calculations either in the kinematical or the dynamical framework of electron diffraction mainly using displacement fields resulting from finite element modelling of the sample relaxation. HOLZ line splitting is well reproduced and numerical fits can be used to compare experimental and calculated curves. This paper proposes new analytical solutions for the kinematical equation of electron diffraction. Simple mathematical functions are used to approximate the deformation profiles. We showed that, under certain conditions, the rocking curve profile can be analytically calculated, thus providing some clue to separate different contributions to the rocking curves against deformation profile. These simplified analytical expressions are used to extract the maximum amplitude displacement within the sample with about 10% accuracy. This accuracy can even be improved to 1% with a short adjustement routine. The influence of the shape of the displacement profile on the rocking curves is demonstrated.

Carbon fiber composite (CFC) Sepcarb® N11 is used in the tokamak Tore Supra as plasma-facing components. To investigate the fuel retention capability of this material, a mobile sample holder was used to expose CFC N11 samples to direct irradiation by the scrape-off layer plasma of Tore Supra at fluences up to 1 × 1025 m−2. Deuterium (D) elemental mapping using nuclear reaction analysis for the most-exposed CFC sample showed that D retention occurs at depths greater than 8 μm due to the presence of deep (>3.5 μm) local retention sites. In this work, combining transmission electron microscopy (TEM) and electron energy-loss spectroscopy (EELS), we describe at a high spatial resolution where and how D atoms are trapped in these sites. TEM experiments performed on thin cross-sections of the plasma-modified surface show evidence of the presence of a 3.5 μm-thick deuterated amorphous carbon layer deposited on the CFC surface. We show that specific localized retention sites correspond to the filling of relatively large (∼3 μm.) and deep (at least 3 μm below the initial CFC surface) cracks between fibres and matrix by the deuterated amorphous carbon layer.

An Al-4 at.% Ni alloy was prepared by a melt spinning technique and characterized by X-ray diffraction, scanning electron microscopy, energy dispersive X-ray spectroscopy and high resolution transmission electron microscopy. The resulting ribbon microstructure consists of intermetallic Al{sub 9}Ni{sub 2} globular-like structures embedded within an aluminum matrix. Characteristic globules are nanometric ({approx} 100 nm) and are mainly located at the grain boundaries. The resulting effect on the mechanical properties is the enhancement of the alloy hardness from 58 to 371 HV.

The microstructure and corrosion behavior of commercial alloy ZE41 modified by surface laser cladding with Al–Si powder mixture was studied by SEM, TEM, X-ray diffraction and electrochemical methods. The coating is composed of an Al–Mg matrix and dendrite precipitates of Mg2Si. In function of the laser speed, the matrix is formed by a Mg solid solution in Al or by the intermetallic phase Mg17Al12. The presence of different matrixes is responsible for galvanic corrosion and decrease of corrosion resistance in interfacial area between coats. Isolated samples of the bulk coatings material showed similar corrosion potentials inspite of different matrix composition. This interpreted in terms of a mechanism involving two steps: (1) an initial dissolution of anodic Mg2Si particles followed by (2) pitting in the formed crevices. The proposed mechanism corresponds well with the experimental observations and the mechanisms of localized corrosion observed for aluminium alloys in the chloride media described in the literature. Improved corrosion resistance can be achieved by the microstructure homogenization through the optimization of laser parameters and/or following heat treatment.

A methodology is presented to select a consistent method, using electron energy loss spectroscopy (EELS), to extract the fraction of sp 2 -bonded carbon atoms in carbonaceous materials. According to this methodology, a reliable method has to conjointly fulfill two criteria. The first one consists in verifying, on a perfect graphite sample, that the experimental evolution of R = I π* /( I π* + I σ* )-ratios is in good agreement with the one theoretically predicted as a function of experimental settings. The second criterion consists in measuring sp 2 fractions in amorphous carbon samples with a minimum of fluctuation. We test three commonly used R -extraction techniques, and we show that they exhibit some failures. We thus implement a more accurate R -extraction process that accounts for the predicted graphite R -evolution and exhibits a low intrinsic 4%-noise, as determined from the sp 2 fraction fluctuation. Moreover, we check the transferability of our method on a wide range of EELS spectra, acquired with different experimental resolutions on samples exhibiting various sp 2 contents.

For safety reasons, the formation of carbon deposits in tokamaks must be limited. To understand how these deposits are formed, a comparative study of the characteristics of carbon deposits collected in TEXTOR and Tore Supra tokamaks has been performed. Both deposits exhibit self-similar morphologies like asparagus tips. Transmission electron microscopy performed on thin foils also reveals structure and texture similarities, with the existence of a regular network of parallel mesopores. Besides, methane adsorption isotherm measurements give evidence for microporosity (0.01 cm 3 g −1 for TEXTOR and 0.06 cm 3 g −1 for Tore Supra). Raman microspectrometry as well as temperature programmed desorption experiments demonstrate the presence of polyaromatic species, at the origin of the disordered graphitic nature of these carbons. The analysis of samples heat-treated at 2500 °C indicates that they graphitize only to a limited extent due to the existence of nanoparticles. Metallic and ferromagnetic inclusions are also evident inside the deposits. These results therefore suggest complex growth processes, mixing homogeneous nucleation and vapor deposition, similar to what can be observed for catalytic chemical vapor deposition processes.

During an anodization process, porous silicon (PS) consisting of pores with a diameter of about 40 nm and a depth from 5 µm to 40 µm has been produced. To achieve oriented channels in this mesoporous range, a p+-type Si wafer was electrochemically etched in an aqueous electrolyte of HF. We report the formation, after the anodization step, of a cobalt nanostructure in a porous silicon matrix. Co nanocrystals on and in a porous silicon layer have been prepared by the UHV evaporation technique and characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM) and electron energy loss spectroscopy (EELS). This technique was performed to show the chemical element distribution within the channels. It is found that the deposition condition is an important factor for obtaining nano-structures. Initial deposition leads to Co particle penetration in silicon pores whereas subsequent deposition results only in an increase of the thickness at the surface with no further penetration. Additional experiments were carried out by using the magneto-optical Kerr effect to obtain information about the magnetic properties. The first results show that the magnetic response for layers

To face with the continuous integrated circuit densification, passive components size has to be reduced, particularly for RF and analog applications where lots of them are needed. A Metal-Insulator-Metal (MIM) capacitor is integrated with a high developed area architecture to increase the capacitance density and limit encumbrance. The combination of this architecture with Ta"2O"5 dielectric with a permittivity of 25 allows capacitance densities of more than 15fF/@mm^2. As metal insulator interface is critical, two stacks TiN/Ta"2O"5/TiN and TiN/Ta"2O"5/Cu are integrated among copper interconnects, evaluated and compared.

The structural and optical properties of porous silicon (PS) layers prepared by Vapour-Etching (VE) of moderately and heavily boron-doped Si substrates are investigated. The VE technique produces rough PS layers that are essentially formed of interconnected cluster-like structures. Optical investigations indicate that this surface roughness enables the PS layers to be used as antireflection coatings in silicon based devices. These optical characteristics are investigated by optical reflectivity and light scattering. The local chemical state and the microstructure of the PS layers are studied by electron energy loss spectroscopy (EELS) in transmission electron microscopy (TEM), and are correlated to the red photoluminescence (PL). TEM studies point out that the cluster-like interconnected structures are composed of luminescent nanocrystallites. PL measurements display that both quantum confinement and surface passivation determine the electronic states of the silicon nanocrystallites. The complex dielectric function is calculated from the experimental single-scattering distribution spectrum using a Kramers Kronig analysis. The first resonance peak in the imaginary part is observed at 2.3 eV; two other broadened features appear at 4.7 and 8.8 eV. The latter is generally related to an interface plasmon resulting from the silicon-silicon oxide interface. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

The electron microscopy and microanalysis techniques available at the CP2M Centre have been used to study the complex precipitation phenomena in a titanium-containing dual phase steel. Nitrides, carbides and carbonitrides of titanium have been characterized at microscopic and nanoscopic scales using carbon extraction replicas and thin foils. Precipitation occurred mainly heterogeneously on existing particles (CaS, TiN). Thin foils have also been examined in a new scanning electron microscope with a field emission gun.

Analytical transmission electron microscopy was employed to characterize the precipitation at each step of the fabrication process and thermomechanical treatment of an industrial dual-phase steel microalloyed with titanium. Theoretical thermodynamic calculations as well as experimental analysis showed that more than half of the titanium carbosulfide (Ti4C2S2) precipitates would dissolve during reheating at 1240 °C. Despite this dissolution at 1240 °C, the remaining titanium carbonitrides and carbosulfides were effective in pinning austenitic grain boundaries, keeping the austenitic grain size at around 40 µm (at 1240 °C). It is also shown that, during hot rolling, there exist three regions of titanium carbide precipitation. The first is defined by an increase of titanium carbide precipitation due to deformation. The second region is marked by the insignificant change in precipitation. The third region is indicated by another increase in precipitation due to the austenite-to-ferrite transformation. The experimental and theoretical results on the contribution of TiC precipitation to hardening of ferrite (Orowan mechanism) were in excellent agreement, showing that TiC precipitates have the most important effect on increasing the yield strength, overshadowing the austenitic grain-boundary pinning contributions by Ti(C,N) and Ti4C2S2 precipitates.

Transmission electron microscopy was used to study the precipitation in four different dual-phase steels microalloyed with titanium. TiN, Ti4C2S2 and TiC were characterized. Detailed statistical analysis of hundreds of images of the various Ti compounds fit rather well with thermodynamical calculations and they show TiC particles (approximately 10 nm) to be the most dominant factor in strengthening.

Scanning electron microscopy as well as analytical transmission electron microscopy techniques such as high resolution, electron diffraction, energy dispersive X-ray spectrometry (EDX), parallel electron energy loss spectroscopy (PEELS) and elemental mapping via a Gatan Imaging Filter (GIF) have been used to study complex precipitation in commercial dual phase steels microalloyed with titanium. Titanium nitrides, titanium carbosulfides, titanium carbonitrides and titanium carbides were characterized in this study. Both carbon extraction replicas and thin foils were used as sample preparation techniques. On both the microscopic and nanometric scales, it was found that a large amount of precipitation occurred heterogeneously on already existing inclusions/precipitates. CaS inclusions (1 to 2 μ m), already present in liquid steel, acted as nucleation sites for TiN precipitating upon the steel's solidification. In addition, TiC nucleated on existing smaller TiN (around 30 to 50 nm). Despite the complexity of such alloys, the statistical analysis conducted on the non-equilibrium samples were found to be in rather good agreement with the theoretical equilibrium calculations. Heterogeneous precipitation must have played a role in bringing these results closer together.

In the present letter, Mg-doped GaN thin films grown by metalorganic vapor-phase epitaxy were studied using parallel electron energy-loss spectroscopy in a transmission electron microscope. A microstructural characterization of such thin films showed the presence of pyramidal defects (PDs) with a density of about 1018 cm−3. Comparison of energy-loss spectra recorded outside a PD and from the PD showed a significant change in the energy-loss near-edge structure strongly reflecting the presence of inclusions (Mg-based), the electronic properties of which differ from those of GaN. Considering, however, their relatively high density (∼1018 cm−3), one can expect that the optical properties of such inclusions may interfere with those of GaN and, therefore, be at the origin of the frequently obtained blue emission at 2.8–2.9 eV in heavily doped samples.

TRIP properties in four hot-rolled microalloyed steels were studied as a function of bainitic domain entrance temperature. It was found that this temperature affects the size and morphology of the retained austenite grains. Size and morphology was shown to play a role on the stability of the retained austenite. This stability was shown to be the major factor affecting the mechanical properties of these steels. In the case of the lowest bainitic temperature, the TRIP effect is hindered due to the premature transformation of austenite into martensite probably due to an insufficient carbon enrichment of the small austenite grains. Moreover the bainitic treatment performed at the highest temperature contained a very low volume fraction of retained austenite that does not transform even under deformation. Using TEM and the Kikuchi-line method, it was shown that the sample that did not undergo a phase transformation during FIB milling had a carbon concentration in its retained austenite grains of approximately 1.7%.

For many years, Convergent Beam Electron Diffraction (CBED) has been intensively used for lattice parameter determination because of the extremely high sensitivity of High Order Laue Zone (HOLZ) lines position to small lattice parameter changes. But, for highly stressed systems, the sample thinning down to electron transparency induces a relaxation. In such cases, HOLZ lines are splitted due to an inhomogeneous variation of lattice parameter (Figure 1.a). Conventional measurement of lattice parameter using quasi-kinematical approach of HOLZ lines shift is then impossible and new models have to be developed. Recently, several approaches have been proposed in order to reproduce the experimental profile of split HOLZ lines also called “rocking curves” I g(s), and deduce the stress distribution in the sample [1, 2, 3, 4]. But, several questions are still open and relation between the rocking curve features and the atomic displacement profile is not completely understood.

Carbon concentration in retained austenite grains is of great importance determining the mechanical properties of hot-rolled TRansformation Induced Plasticity steels. Among the different techniques available to measure such concentrations, Kikuchi lines obtained in Transmission Electron Microscopy provide a relatively easy and accurate method. The major problem however is to be able to locate an austenitic grain in the observed Transmission Electron Microscopy thin foil. Focused Ion Beam in combination with Scanning Electron Microscopy was used to successfully prepare a thin foil for Transmission Electron Microscopy and carbon concentration measurements from a 700 nm retained austenite grain.

Submicron and nano‐magnetic particles have been attracting interest both from academic and industrial fields. More information concerning their structure and magnetic properties are needed for extensive development of these small particles to high‐density recording media. In this work, the growth and the magnetic behaviour of cobalt on porous silicon are studied as a function of substrate preparation. We show that rinsing the porous silicon (PS) substrates with ethanol leads to an oxidation of the layer that offers a three dimensional growth. The magnetic behaviour of Co particles grown on PS substrates is rather different than the conventional one and may exhibit a superparamagnetic effect.

The aim of this paper is to check the effect of artefacts introduced by focused ion beam (FIB) milling on the strain measurement by convergent beam electron diffraction (CBED). We show that on optimized silicon FIB samples, the strain measurement can be performed with a sensitivity of about 2.5 × 10−4 which is very close to the theoretical one and we conclude that FIB preparation can be suitable for such measurements in microelectronic devices. To achieve this, we first used CBED and electron energy loss spectroscopy (EELS) which provide a procedure permitting an exact knowledge of the sample geometry, i.e. the thickness of both amorphous and crystalline layers. This procedure was used in order to measure the FIB-amorphized sidewall layer. It was found that if the FIB preparation is optimized one can reduce this amorphous layer down to around 7 nm on each side. Secondly different preparation techniques (cleavage, Tripod™ and FIB) permit to check if the surface damaged layer introduced by FIB influences the strain state of the sample. Finally, it was found that the damaged layer does not introduce measurable strain in pure silicon but reduces appreciably the quality of the CBED patterns.

We perform calculations of the fully relativistic, corrected geometrical weighting of the pi* and sigma* transitions measured from the 1s core loss electron energy loss spectroscopy (EELS) spectrum in any uniaxial specimen. We present a complete calculation of the differential scattering cross-section (DSCS), taking into account the collection angle, the illumination angle and the tilt of the sample over the optical axis. Owing to high electron velocity in an EELS experiment, the relativistic correction has to be considered. We thus, present a relativistic, corrected DSCS by using the theory recently developed by Jouffrey et al. [Ultramicroscopy 102 (2004) 61] and P. Schattschneider et al. [Phys. Rev. B 72 (2005) 045142]. The relativistic correction is first performed in the natural coordinate system of the scattering event. We then point out a straightforward method to introduce this correction in the microscopic coordinate system, where all calculations have to be done to be experimentally useful. Using the fully corrected DSCS, we present an expression predicting the evolution of the R=pi*/(pi*+sigma*) ratio (related to the ratio of sp2 and sp3 bondings) as a function of experimental settings. We show how the R-evolution can be predicted, for any experimental setting, by the knowledge of one unique reference value. We verify on graphite specimens, the validity of the R-calculation by comparing theoretical predictions presented in this work with experimental data published elsewhere [Daniels et al., Ultramicroscopy 96 (2003) 523 and Menon et al., Ultramicroscopy 74 (1998) 83].

In this work we present first results concerning the detailed structure of porous silicon (PS) layers prepared by a new method using a vapour-etching (VE)-based technique. Studies of the photoluminescence properties of VE-based PS show that the visible emission occurs at high energies as compared with PS prepared by conventional techniques. To understand the VE-based PS features, we need to point out the PS microstructure throughout its general morphology. For this purpose a microscopy multiscale study was done. Scanning, conventional transmission, and high-resolution transmission electron microscopes were employed. The investigations were made on PS films prepared from moderately and heavily doped n- and p-type silicon. SEM images show that VE-based PS layers are essentially formed of clusters like interconnected structures. TEM studies show that these clusters are composed of nanocrystallites with different shapes. The effect of the doping type of the starting Si substrate on the characteristics of the PS layers was examined (thickness, porosity, behavior). Pore propagation was found to depend on doping type. The crystallinity of the PS layers was also locally studied in depth.

We report further observations of graphitic carbon onions and polyhedra that are formed together with graphitic planes under the conditions of the chemical vapour deposition of diamond as previously described by Constant et al. (1997, Surf. Sci., 387, 28). The occurrence of such carbon onions is quite specific to the copper substrate. They were observed with various pre-treatments of the substrate. It is suggested that both the nucleation and the growth of such carbon onions occur on the copper surface rather than in the gas phase. It is believed that, instead of high-energy electron or ion beams as described elsewhere by Banhart (1997, J. appl. Phys., 81, 3440), the formation of carbon onions is due to the high reactivity of the atomic hydrogen present in the gas phase. These hydrogen radicals induce a curvature of the graphitic planes by preferential reactivity on the corner and edge sites. A dense phase in the core of few carbon onions was sometimes seen. We then measured an interplanar distan...

In Pt-doped SnO 2 thin films prepared by using a submicroscopic aerosol pyrolysis method where the metal elements are codeposited, the best CO sensitivity is obtained from low concentration Pt precursor solutions. Transmission electron microscopy (TEM) observations of SnO 2 thin films doped by small amounts of platinum have been performed to analyze in detail the structural properties as a function of Pt doping concentration. Strong evidence of porous structure has been found in the TEM images and an assessment of the porous volume fraction could be done. Although this approach clearly underestimates the assessed porous fraction, results for undoped SnO 2 films nevertheless indicate that the film density is close to half of the bulk value. It increases with the film doping concentration, [Pt]/([Sn] + [Pt] + [O]), as films become thinner. For doping concentrations higher than 2%, no significant number of pores could be detected and the thickness reaches a plateau likely corresponding to the maximum density that can be obtained using our synthesis conditions. The usually observed maximum in the film sensitivity at solution concentrations close to 0.6% film doping, i.e., 3% in terms of solution concentration, coincides with a maximum in the product of the porous volume and the density of small Pt aggregates. Additionally high-resolution electron microscopy observations of selected film samples have been performed to examine grain arrangements and microstructural effects of Pt and SnO 2 nanocrystallites.