Search

Accurate modelling of friction coefficient is of primary importance in efficiency, vibration and failure analyses of enclosed gear drives. After showing the influence of surface/lubricant interactions on friction, the authors used a semi-empirical model which can take all these aspects into account. Lubricant is modelled as an Eyring–Reynolds fluid and rough surfaces are described with two parameters via a stochastic approach. A specific two-disc machine is used to perform series of friction measurements on smooth and rough discs. Smooth discs allow to operate under full film lubrication and to measure a reference shear stress of lubricant, whereas rough discs reproduce gear tooth roughness and generate a representative value of friction on asperities. The purpose of this present paper is to describe calculations of this physical-based friction coefficient model and to present the experimental process. On the basis of new results, the impact of a surface finishing process is assessed as well as the consequence of calculating friction coefficient based on oil injection instead of local bulk temperature.

Replacing the CdS buffer layer with a ZnSnO one in Cu2ZnSnS4-based solar cells allows both to improve the device performances and to avoid using toxic Cd. Additionally, using a sputtered buffer layer is a major asset for solar cells fabricated by physical vapor deposition processes. In this study, ZnSnO layers are deposited by sputtering of a single metal-oxide target. Structural and optical properties of the layers deposited on Si or glass are first described. The possibility of modifying the ZnSnO metallic composition by adjusting the deposition power is demonstrated. Attempts to improve the optoelectronic properties of the ZnSnO layers with Ar:O2 or Ar:SF6 reactive sputtering are shown as well. These ZnSnO buffer layers are transferred in Mo/CZTS/ZnSnO/ZnO:Al solar cells. After post-deposition thermal treatment and optimization of the deposition condition (notably with the use of Ar:O2 or Ar:SF6 reactive sputtering), a solar cells with a power conversion efficiency of 5.2% is demonstrated. It is 0.6% absolute higher than the reference solar cell with a CdS buffer layer. To avoid absorber damaging and achieve high performances, deposition power must be as low as possible. A two stages sputtering process is used to conciliate both the absorber surface preservation and a reasonable deposition time. Last, photovoltaic properties of optimized CZTS- and CZTSe-based solar cells with ZnSnO buffer layers are compared.

The impact of epitaxial strain on the structural, electronic, and thermoelectric properties of p-type transparent Sr-doped LaCrO3 thin films has been investigated. For this purpose, high-quality fully-strained La0.75Sr0.25CrO3 (LSCO) epitaxial thin films were grown by molecular beam epitaxy on three different (pseudo)cubic (001)-oriented perovskite-oxide substrates: LaAlO3, (LaAlO3)0.3(Sr2AlTaO6)0.7, and DyScO3. The lattice mismatch between the LSCO films and the substrates induces in-plane strain ranging from -2.06% (compressive) to +1.75% (tensile). The electric conductivity can be controlled over two orders of magnitude, σ ranging from ~0.5 S cm-1 (tensile strain) to 35 S cm-1 (compressive strain). Consistently, the Seebeck coefficient S can be finely tuned by a factor of almost two from ~127 μV K-1 (compressive strain) to 208 μV K-1 (tensile strain). Interestingly, we show that the thermoelectric power factor (PF = S2 σ) can consequently be tuned by almost two orders of magnitude. The compressive strain yields a remarkable enhancement by a factor of three for 2% compressive strain with respect to almost relaxed films. These results demonstrate that epitaxial strain is a powerful lever to control the electric properties of LSCO and enhance its thermoelectric properties, which is of high interest for various devices and key applications such as thermal energy harvesters, coolers, transparent conductors, photo-catalyzers and spintronic memories., Financial support from the European Commission through the project TIPS (H2020-ICT-02-2014-1-644453), the French national research agency (ANR) through the projects MITO (ANR-17-CE05-0018), LILIT (ANR-16-CE24-0022), DIAMWAFEL (ANR-15-CE08-0034-02), the CNRS through the MITI interdisciplinary programs (project NOTE), IDEX Lyon-St-Etienne through the project IPPON, the Spanish Ministerio de Ciencia e Innovación, through the “Severo Ochoa” Programme for Centres of Excellence in R&D (SEV2015-0496) and the MAT2017-85232-R (AEI/FEDER, EU), PID2019-107727RB-I00 (AEI/FEDER, EU), and from Generalitat de Catalunya (2017 SGR 1377) is acknowledged. The China Scholarship Council (CSC) is acknowledged for the grant of Dong Han. Ignasi Fina acknowledges Ramón y Cajal contract RYC-2017-22531. Seebeck measurements at ILM were made within the ILMTech transport platform. The authors are also grateful to Jean-Baptiste Goure, Philippe Regreny, Aziz Benamrouche, and Bernat Bozzo for their technical support and the reviewers for their valuable and constructive comments that have improved the quality of the manuscript.

We report on the precise determination of both the band gap Eg and the characteristic energy U of the band tail of localized defect states, for monocrystalline Ag2ZnSnSe4. Both photoluminescence excitation and time-resolved photoluminescence studies lead to Eg = 1223 ± 3 meV, and U = 20 ± 3 meV, at 6 K. The interest of the methodology developed here is to account quantitatively for the time-resolved photoluminescence and photoluminescence excitation spectra by only considering standard textbook density of states, and state filling effects. Such an approach is different from the one most often used to evaluate the energy extent of the localized states, namely by measuring the energy shift between the photoluminescence emission and the excitation one-the so-called Stokes shift. The advantage of the present method is that no arbitrary choice of the low power excitation has to be done to select the photoluminescence emission spectrum and its peak energy.

International audience; We show that the photonic dispersion of a two coupled-mode system can be actively tuned using graphene-induced non-radiative loss. Our implementation exploits the spatial modulation of graphene's absorption via patterned oxide substrates.

We report on non equilibrium field effect in insulating amorphous NbSi thin films having different Nb contents and thicknesses. The hallmark of an electron glass, namely the logarithmic growth of a memory dip in conductance versus gate voltage curves, is observed in all the films after a cooling from room temperature to 4.2 K. A very rich phenomenology is demonstrated. While the memory dip width is found to strongly vary with the film parameters, as was also observed in amorphous indium oxide films, screening lengths and temperature dependence of the dynamics are closer to what is observed in granular Al films. Our results demonstrate that the differentiation between continuous and discontinuous systems is not relevant to understand the discrepancies reported between various systems in the electron glass features. We suggest instead that they are not of fundamental nature and stem from differences in the protocols used and in the electrical inhomogeneity length scales within each material., Comment: Submission SciPost

International audience

Cu(In, Ga)Se2 (CIGS) based thin film solar cells have been extensively studied and today, power conversion efficiencies higher than 20% have been demonstrated on both rigid and flexible substrates. However, very little is known about the mechanical resistance of flexible CIGS solar cells under flexion. Here we report an original study on the mechanical properties of CIGS solar cells fabricated on 100 µm-thick ultra-thin glass substrates. The Young's modulus and hardness of Mo and CIGS thin films are measured by nanoindentation, a technique well adapted to the characterization of thin film materials. Young's modulus values of 289 GPa and 70 GPa are obtained for the Mo and the CIGS layers respectively, as well as a CIGS hardness of 3.4 GPa. These values, combined with an analytical model, allow calculating the strain induced in thin film during the flexion of solar cells fabricated on ultra-thin glass substrate as well as on polyimide substrate. Thereby, we show that using a substrate with a low thickness and a low Young's modulus enables to lower the thin films strain during the flexion of cells.

International audience; Inelastic background analysis of HAXPES spectra was recently introduced as a powerful method to get access to the elemental distribution in deeply buried layers or interfaces, at depth up to 60 nm below the surface. However the accuracy of the analysis highly relies on suitable scattering cross-sections able to describe effectively the transport of photoelectrons through overlayer structures consisting of individual layers with potentially very different scattering properties. Here, we show that within Tougaard’s practical framework as implemented in the Quases-Analyze software, the photoelectron transport through thick (25–40 nm) multi-layer structures with widely different cross-sections can be reliably described with an effective cross-section in the form of a weighted sum of the individual cross-section of each layer. The high-resolution core-level analysis partly provides a guide for determining the nature of the individual cross-sections to be used. We illustrate this novel approach with the practical case of a top Al/Ti bilayer structure in an AlGaN/GaN power transistor device stack before and after sucessive annealing treatments. The analysis provides reliable insights on the Ti and Ga depth distributions up to nearly 50 nm below the surface.

About 30 nm thick (001)-oriented (Sr0.25La0.75)CrO3 (SLCO) epitaxial thin films were grown by solid-source oxide molecular beam epitaxy on four different single-crystalline cubic or pseudocubic (001)-oriented oxide substrates, LaAlO3, (LaAlO3)0.3(Sr2AlTaO6)0.7, SrTiO3, and DyScO3, which result in lattice mismatch ranging from –2% to +1.7%. All the films are of high-quality, flat, and strained by the substrates. By assessing the evolution of the out-of-plane lattice parameter as a function of the in-plane lattice parameter of the samples, we determine both the Poisson ratio (ν = 0.32) and the bulk lattice constant (ab = 3.876 A) of SLCO. The Poisson ratio significantly differs from LaCrO3 (ν = 0.23), and the (SrxLa1−x)CrO3 solid solution appears to obey structural Vegard's law. Since SLCO is the only one p-type transparent conductive oxide of perovskite structure and has promising thermoelectric properties, integrating SLCO in heterostructures and devices is, therefore, of paramount importance, which confers on our results their strong interest. Besides, the method used here can be straightforwardly applied to other complex oxides.About 30 nm thick (001)-oriented (Sr0.25La0.75)CrO3 (SLCO) epitaxial thin films were grown by solid-source oxide molecular beam epitaxy on four different single-crystalline cubic or pseudocubic (001)-oriented oxide substrates, LaAlO3, (LaAlO3)0.3(Sr2AlTaO6)0.7, SrTiO3, and DyScO3, which result in lattice mismatch ranging from –2% to +1.7%. All the films are of high-quality, flat, and strained by the substrates. By assessing the evolution of the out-of-plane lattice parameter as a function of the in-plane lattice parameter of the samples, we determine both the Poisson ratio (ν = 0.32) and the bulk lattice constant (ab = 3.876 A) of SLCO. The Poisson ratio significantly differs from LaCrO3 (ν = 0.23), and the (SrxLa1−x)CrO3 solid solution appears to obey structural Vegard's law. Since SLCO is the only one p-type transparent conductive oxide of perovskite structure and has promising thermoelectric properties, integrating SLCO in heterostructures and devices is, therefore, of paramount importance, which confe...

In this paper the early stages of the self-catalyzed Vapor-Liquid-Solid (VLS) growth of GaAs nanowires on Si substrates by Molecular Beam Epitaxy (MBE) are studied. The interaction of Ga nanodroplets (NDs) with the silica overlayer is investigated by X-ray Photoemission Spectroscopy (XPS) and Atomic Force Microscopy (AFM). We show how Ga NDs drill the silica overlayer and make contact with bulk Si to allow GaAs nanowires (NWs) epitaxial growth by studying each of the three steps of the NW growth process sequentially: Ga NDs pre-deposition, post-deposition annealing to reach the NW growth temperature, and finally NW growth itself. The pre-deposition temperature allows to control the density and morphology of the NDs. A high enough annealing temperature enhances the reduction of SiOx by Ga oxidation and leads to the formation of holes which is seen by a new component in XPS spectra, assumed to be the consequence of the Ga Si interaction. Finally, these nano-holes act as nucleation sites for the epitaxial GaAs VLS growth. During the GaAs growth, three different chemical environments of Ga are identified in Ga2p core level: metallic Ga due to the droplets at the top of the NWs, Ga oxide in contact with the SiOx overlayer and Ga arsenide from NWs.

Cu2ZnSnS4-based solar cells suffer from limited power conversion efficiency (PCE) and relative small grain size compared to selenium containing absorbers. Introduction of Na in Cu2ZnSnS4 absorbers either during the synthesis or after this step is used to improve device performances and to determine whether its effect is based on structural properties improvement (grain size enhancement, better crystallization) or on opto-electronic properties improvement (defect passivation). In both cases, presence of Na in the absorber notably improves current and voltage of the solar cells, but the effect is more pronounced when Na is present during synthesis. Quantum efficiency analysis shows that these improvements can be related to longer minority carrier diffusion length and reduced absorber/buffer interface recombination. Introducing Na in the process mostly leads to preferential (112) orientation of the crystal which is clearly correlated with better device performances. Otherwise, the performance limitation due to small grain size is discarded by the joint use of Sb and Na, which has a significant impact on grain size but does not affect solar cells efficiency.

Epitaxially strained La1+δCrO3 (LCO) thin films, with δ ranging between −0.25 (Cr-rich) and 0.25 (La-rich), were grown on SrTiO3(001) substrates by solid-source molecular beam epitaxy. The effect of the cationic-stoichiometry deviation (δ) on the structural properties (surface structure, morphology, terminations, and unit-cell parameters) is reported. Whereas deviation from stoichiometry does not significantly affect the structural quality (all films keep a perovskite structure, are epitaxial, atomically flat, fully strained, and present mosaicity below 0.1°); increasing |δ| leads to an increase of the out-of-plane lattice parameter and of the unit-cell volume. These results are of importance for further studies on structure–property relationships in the view to build enhanced LCO-based devices.Epitaxially strained La1+δCrO3 (LCO) thin films, with δ ranging between −0.25 (Cr-rich) and 0.25 (La-rich), were grown on SrTiO3(001) substrates by solid-source molecular beam epitaxy. The effect of the cationic-stoichiometry deviation (δ) on the structural properties (surface structure, morphology, terminations, and unit-cell parameters) is reported. Whereas deviation from stoichiometry does not significantly affect the structural quality (all films keep a perovskite structure, are epitaxial, atomically flat, fully strained, and present mosaicity below 0.1°); increasing |δ| leads to an increase of the out-of-plane lattice parameter and of the unit-cell volume. These results are of importance for further studies on structure–property relationships in the view to build enhanced LCO-based devices.

L'objectif de cette these est de realiser l'integration monolithique de nanofils (NFs) a base de l’alliage Al0.2Ga0.8As sur des substrats de Si par epitaxie par jets moleculaires via la methode vapeur-liquide-solide (VLS) auto-assistee et de developper une cellule solaire tandem (TSC) a base de ces NFs.Pour atteindre cet objectif, nous avons tout d'abord etudie la croissance de NFs GaAs, etape cle pour le developpement des NFs p-GaAs/p.i.n-Al0.2Ga 0.8As coeur/coquille, qui devraient constituer la cellule superieure de la TSC. Nous avons montre, en particulier, l'influence de l'angle d'incidence du flux de Ga sur la cinetique de croissance des NFs GaAs. Un modele theorique et des simulations numeriques ont ete realisees pour expliquer ces resultats experimentaux.Nous avons ensuite utilise le savoir-faire acquis pour faire croitre des NFs p-GaAs/p.i.n-Al0,2Ga0,8As coeur/coquille sur des substrats de Si prets pour l'emploi. Les caracterisations EBIC realisees sur ces NFs ont montre qu'ils sont des candidats potentiels pour la realisation d’une cellule photovoltaique. Nous avons ensuite fait croitre ces NFs sur des substrats de Si patternes afin d'obtenir des reseaux reguliers de ces NFs. Nous avons developpe un protocole, base sur un pre-traitement thermique, qui permet d'obtenir des rendements eleves de NFs verticaux (80-90 %) sur une surface patternee de 0,9 x 0,9 mm2.Enfin, nous avons consacre une partie de notre travail a definir le procede de fabrication optimal pour la TSC, en concentrant notre attention sur le developpement de la jonction tunnel de la TSC, l'encapsulation des NFs et le contact electrique superieur du reseau de NFs.

The electrical conductivity induced near the superconducting transition by thermal fluctuations was measured in different granular aluminum films. The seemingly anomalous behavior at high reduced temperatures and magnetic fields is explained by taking into account a total-energy cutoff in the superconducting fluctuation spectrum in both the direct (Aslamazov-Larkin) and the indirect (anomalous Maki-Thompson) contributions to the fluctuation effects. The analysis allowed a reliable determination of the coherence length amplitudes, which resulted to be much larger (20-48 nm) than the grains size (5-10 nm). This suggests that the grains are strongly Josephson-coupled, while the Tc value is still as high as twice the bulk value. These results could contribute to identify the mechanisms enhancing Tc in these materials., Comment: 9 pages, 8 figures

We have studied the growth of a SrTiO3 shell on self-catalyzed GaAs nanowires grown by vapor-liquid-solid assisted molecular beam epitaxy on Si(111) substrates. To control the growth of the SrTiO3 shell, the GaAs nanowires were protected using an arsenic capping/decapping procedure in order to prevent uncontrolled oxidation and/or contamination of the nanowire facets. Reflection high energy electron diffraction, scanning electron microscopy, transmission electron microscopy, and X-ray photoelectron spectroscopy were performed to determine the structural, chemical, and morphological properties of the heterostructured nanowires. Using adapted oxide growth conditions, it is shown that most of the perovskite structure SrTiO3 shell appears to be oriented with respect to the GaAs lattice. These results are promising for achieving one-dimensional epitaxial semiconductor core/functional oxide shell nanostructures.

Cu$_2$ZnSn(S,Se)$_4$ (CZTSSE)-based solar cell performances are limited by band tailing due to a large amount of CuZ$_n$ antisite defects. Partially replacing the Cu atoms by larger Ag ones can significantly reduce the prevalence of these defects, which are particularly detrimental close to the front interface. Herein, the possibility of synthesizing (Cu$_{1‐x}$Ag$_x$)$_2$ZnSnSe$_4$ absorbers with various Ag contents by vacuum-based processes is demonstrated. Although the synthesis of high-quality materials is demonstrated, their use in thin film photovoltaic devices does not exhibit performance improvement compared with efficient pure CZTSSE-based solar cells. Moreover, the comparison with literature data reopens the debate of the beneficial effect of homogeneous Ag alloying in kesterite. On the contrary, a new method is proposed to fabricate graded (Cu$_{1‐x}$Ag$_x$)$_2$ZnSnSe$_4$) absorbers with increased Ag content at the interfaces. The solar cells with graded absorbers exhibit better performances than the reference Ag-free ones. Particularly, improved current collection at the back contact and slight reduction of the front interface recombination are demonstrated

International audience; The recent progress in HAXPES combined with Inelastic Background Analysis makes this method a powerful, non-destructive solution to get quantitative information on deeply buried layers and interfaces at depths up to 70 nm. However, we recently highlighted the need for carefully choosing the scattering cross-sections in order to accurately describe the transport of photoelectrons through a complex overlayer structure with layers presenting very different scattering properties. It is found that the transport through such thick bi-layer structures can be described with an effective inelastic scattering cross-section in the form of a weighted sum of individual cross-sections of the pure layers. In this study, we have experimentally investigated this by analyzing Al/Ta/AlGaN stacks on a GaN substrate. We present a refined analytical method, based on the use of a reference spectrum, for determining the required input parameters, i.e. the inelastic mean free path and the effective inelastic scattering cross-section. The use of a reference sample gives extra constraints which make the analysis faster to converge towards a more accurate result. Based on comparisons with TEM, the improved method provides results determined with a deviation typically better than 5% instead of around 10% without reference. The case of much thicker overlayers up to 66 nm is also discussed, notably in terms of accounting for elastic scattering in the analysis.

Oligo ethylene glycol layers are widely used in biosensor applications, mainly for their anti-fouling abilities. Such surfaces are often characterized by X-ray photoelectron spectroscopy (XPS) as this method allows a precise determination of the surface chemical composition. We show herein that X-rays used during XPS characterization quickly and significantly degrade oligo ethylene glycol immobilized onto silica substrates. It is therefore necessary to introduce a correcting factor to assess the true (i.e. without degradation) corresponding ether contribution in the XPS spectrum of such organic layers. Eventually, fluorescence scans show the loss of anti-fouling properties of the degraded surface, leading to greater amounts of adsorbed (fluorescently labeled) protein. Copyright © 2015 John Wiley & Sons, Ltd.

International audience; Insulating granular aluminum is one of the proto-typical disordered insulators whose low temperature electrical conductance exhibits ubiquitous non-equilibrium phenomena. These include slow responses to temperature or gate voltage changes, characteristic field effect anomalies and ageing phenomena typical of a glass. In this system the influence of temperature on the glassy dynamics has remained elusive, leading to the belief that the slow relaxations essentially proceed via elastic quantum tunneling. A similar situation was met in insulating indium oxide and it was concluded that in high carrier density Anderson insulators, electrons form a quantum glass phase. In this work we experimentally demonstrate that thermal effects do play a role and that the slow dynamics in granular aluminum is subject to thermal activation. We show how its signatures can be revealed and activation energy distributions can be extracted, providing a promising grasp on the nature of the microscopic mechanism at work in glassy Anderson insulators. We explain why some of the experimental protocols previously used in the literature fail to reveal thermal activation in these systems. Our results and analyses call for a reassessment of the emblematic case of indium oxide, and question the existence of a quantum glass in any of the systems studied so far.

We have used the Scanning Kelvin probe microscopy technique to monitor the charging process of highly resistive granular thin films. The sample is connected to two leads and is separated by an insulator layer from a gate electrode. When a gate voltage is applied, charges enter from the leads and rearrange across the sample. We find very slow processes with characteristic charging times exponentially distributed over a wide range of values, resulting in a logarithmic relaxation to equilibrium. After the gate voltage has been switched off, the system again relaxes logarithmically slowly to the new equilibrium. The results cannot be explained with diffusion models, but most of them can be understood with a hopping percolation model, in which the localization length is shorter than the typical site separation. The technique is very promising for the study of slow phenomena in highly resistive systems and will be able to estimate the conductance of these systems when direct macroscopic measurement techniques are not sensitive enough., 8 pages, 7 figures

Analyses of current–voltage characteristics of Cu2ZnSn(S,Se)4-based solar cells reveal a non-superposition between dark and illuminated curves. Numerical simulations have been carried out to model the behavior of the Cu2ZnSn(S,Se)4/CdS heterojunction, and to extract from experimental data the minority carrier diffusion length in the absorber. A small value of 200 nm has been found, which induces collection losses in the absorber and is therefore a limiting factor to achieve high power conversion efficiencies. Additional experimental results, such as irradiance dependent photovoltaic properties and voltage dependent external quantum efficiency, are also presented and simulated with the same model. This confirms the validity of our approach and reveals interesting effects in Cu2ZnSn(S,Se)4-based solar cells.

International audience; This study shows the fabrication of CIGS solar cells with various absorber thicknesses synthesized with two different methods: the classical 3-stages coevaporation process and a 1-stage coevaporation process. Structural characteristics and photovoltaics properties as function of absorber thickness are described. 1-stage coevaporation process gives lower efficiencies for CIGS solar cells with nominal absorber thickness ($\sim$2000 nm) but performs similarly or even better when the absorber thickness is decreased down to 600 nm. Periodically textured glass substrates with different feature sizes have been fabricated and used to increase current in 1-stage coevaporated CIGS solar cells. A current improvement up to 4.1 mA·cm$^{−2}$ (6% relative) has been obtained leading to an increase in efficiency up to 5% relative. A maximum increase has been found when texturation size is similar to CIGS absorber thickness.

This work compares the equilibrium states reached by glassy selenium (g-Se) after several ageing durations at temperatures lower than the glass transition temperature T g, with the forecasts given by the configurational entropic model. The comparison is carried out through experimental data collected both on glassy samples after short-term ageing and on glassy samples older than 20 years, kept at room temperature. It is shown that the configurational entropic model is not able to describe the behaviour of g-Se, both at short- and long-ageing terms with the same set of fitting parameters. For short-term ageing, the forecasts given by the entropic model are in good agreement with the experimental data; the hypothesis that the heat capacities corresponding to the equilibrium state and to the state extrapolated from the liquid differ by an amount δ is therefore properly justified. For long-term ageing, the amount δ vanishes and the assumption does not hold any longer. Indeed, experimental results support the ...

This work deals with the influence of sodium on the properties of CZTSSe material and solar cells. For that purpose, two types of substrates are compared, one with low sodium content (borosilicate glass), the other one with higher sodium content (soda-lime glass). In each case the Na-content in the CZTSSe passing from the substrate through the Mo back contact is quantified by secondary ion mass spectroscopy analysis. Photoluminescence spectroscopy indicates that better quality material is achievable when increasing the Na-content in the CZTSSe. The material characterization results are compared to the photovoltaic properties. Index Terms — Cu2ZnSn(S1-xSex)4, CZTSSe, CZTS, CZTSe, Sodium, Kesterite, thin film, solar cell.

Memory is one of the unique qualities of a glassy system. The relaxation of a glass to equilibrium contains information on the sample's excitation history, an effect often refer to as ``aging.'' We demonstrate that under the right conditions a glass can also possess a different type of memory. We study the conductance relaxation of electron glasses that are fabricated at low temperatures. Remarkably, the dynamics are found to depend not only on the ambient measurement temperature but also on the maximum temperature to which the system was exposed. Hence the system ``remembers'' its highest temperature. This effect may be qualitatively understood in terms of energy barriers and local minima in configuration space and therefore may be a general property of the glass state.

International audience; We propose an arsenic-capping/decapping method, allowing the growth of an epitaxial shell around the GaAs nanowire (NW) core which is exposed to an ambient atmosphere, and without the introduction of impurities. Self-catalyzed GaAs NW arrays were firstly grown on Si(111) substrates by solid-source molecular beam epitaxy. Aiming for protecting the active surface of the GaAs NW core, the arsenic-capping/decapping method has been applied. To validate the effect of this method, different core/shell NWs have been fabricated. Analyses highlight the benefit of the As capping-decapping method for further epitaxial shell growth: an epitaxial shell with a smooth surface is achieved in the case of As-capped–decapped GaAs NWs, comparable to the in situ grown GaAs/AlGaAs NWs. This As capping method opens a way for the epitaxial growth of heterogeneous material shells such as functional oxides using different reactors.

Saint-Girons, Guillaume et al., SrTiO3 (STO) crystalline layers grown on Si open unique perspectives for the monolithic integration of functional oxides in silicon-based devices, but their fabrication by molecular beam epitaxy (MBE) is challenging due to unwanted interfacial reactions. Here we show that the formation of single-crystal STO layers on Si by MBE at the moderate growth temperature imposed by these interface reactions results from the crystallization of a partially separated amorphous mixture of SrO and TiO2 activated by an excess of Sr. We identify the atomic pathway of this mechanism and show that it leads to an antiphase domain morphology. On the basis of these results, we suggest and test alternative STO growth strategies to avoid antiphase boundary formation and significantly improve the STO structural quality. The understanding provided by these results offers promising prospects to crystallize perovskite oxides on semiconductors at moderate temperature and circumvent the issue of parasitic interface reactions., This work was partly supported by the European projects SITOGA (STREP FP7, grant number 619456) and TIPS (ICT H2020, grant number 107347) as well as by the French ANR programs HIRIS and DIAMWAFEL.

We present an optical system which integrates a plasmonic sensing surface and an angular tracking system to enable a compact refractive index measurement. A refractive index change at the surface of the sensing membrane causes a change in the angle at which monochromatic light is transmitted through the membrane. This transmission angle is measured by the angular tracking system. We show good theoretical and experimental agreement of the transmission of the plasmonic sensing surface at different angular illumination of the membranes. Using this compact optical setup the embedded angular tracking system has an accuracy of

International audience; We have used 3-D optical modelling to investigate light management concepts based on periodic textures and material optimization for photovoltaic devices based on Cu(In,Ga)Se$_2$ (CIGS) absorber material. At first, calibration of the software based on the characterization of a reference (1500-nm thick) CIGS device was carried out. The effects of 1-D and 2-D symmetric gratings on the cell were then investigated, showing significant improvement in anti-reflection effect and in absorptance in the active layer, achieved by excitation of guided modes in the absorber. In addition, device configurations endowed with alternative back reflector and front transparent conductive oxide (TCO) were tested with the goal to quench parasitic absorption losses at front and back side. The use of In$_2$O$_3$:H (IOH) as front and back TCO, combined with an optimized 2-D grating structure, led to a 25% increase of the optical performance with respect to an equally-thick flat device. Most of the performance increase was kept when the absorber thickness was reduced from 1500 nm to 600 nm.

In this article, a method for in situ synthesis of silver benzene-dithiolate hybrid films is presented. Silver nanoparticles, generated on ZrO 2 films, are transformed into silver benzene 1,4-dithiolate or, partially, into silver benzene 1,2-dithiolate after sample immersion in the corresponding thiol solutions. These transformations occur at room temperature owing to the catalytic action of ZrO 2 . It is also shown that TiO 2 in place of ZrO 2 is very efficient, both for the catalytic generation of silver nanoparticles and for their further transformation in benzene 1,4-dithiolate compound. This latter semiconductor has an optical bandgap of about 3 eV and the film is made of touching nanoparticles in an amorphous state. Our work has potential applications in the electronic and photovoltaic fields.

We study the role of a Ti surface treatment applied to the As-terminated GaAs (001) substrate surface prior to SrTiO3 (STO) epitaxial growth by comparing STO/GaAs samples prepared with and without Ti interlayers. Reflection high energy electron diffraction, transmission electron microscopy and x-ray photoelectron spectroscopy are used to assess the structural and chemical properties of the layers and interfaces. Without Ti interlayer, a polycrystalline TiGa compound is formed near the interface. It significantly degrades STO structural properties. A Ti interlayer efficiently prevents the formation of this unwanted compound by limiting As desorption from the GaAs substrate during STO growth. It improves significantly the structural quality of the oxide layer.

Electron-beam deposition of an insulating granular aluminium or of an off-stoichiometric amorphous alumina layer on a $SrTiO_3$ surface is a simple way to get a metallic interface from insulating materials. No heating nor specific preparation of the $SrTiO_3$ surface are needed. In this paper, we investigate how the electrical properties of this interface can be tuned by the use of a back gate voltage (electrical field through the $SrTiO_3$ substrate). We demonstrate that the slow field-effect observed at room temperature can be used to tune reversibly and in a controlled way the low temperature electrical properties of the interface. In particular, important parameters of a transistor such as the amplitude of the resistance response to gate voltage changes or the existence of an "on" or an "off" state at zero gate voltage and at low temperature can be adjusted in a single sample. This method should be applicable to any $SrTiO_3$-based interface in which oxygen vacancies are involved and might provide a powerful way to study the metal or superconductor insulator transition observed in such systems., 16 pages

International audience; Al2O3 thin films with thickness between 2 and 100 nm were synthetized at 250°C by thermal atomic layer deposition on silicon substrates. Characterizations of as-deposited and annealed layers were carried out using ellipsometry, X-ray reflectivity and X-ray photoelectron spectroscopy. A silicon-rich SiOx layer at the interface between Si and Al2O3 was introduced in the optical models to fit the experimental data. Surface passivation performances of Al2O3 layers deposited on n-type float-zone monocrystalline silicon were investigated as a function of thickness and post-deposition annealing conditions. Surface recombination velocity around 2 cm.s-1 was measured after the activation of the negative charges at the Si/ Al2O3 interface under optimized annealing at 400°C for 10 min. The evolution of the interface layer and of the material properties with the thermal treatment was studied.

Temperature Modulated Differential Scanning Calorimetry (TMDSC) is used to estimate Cooperative Rearranging Region (CRR) average sizes for polymer/clay nanocomposites, obtained by mixing polyethylene 1,4-cyclohexylenedimethylene terephthalate glycol (PETg) filled and organically modified nanoclay (C15A) following a master-batch process. Two different basal distances are obtained. It is shown that the greater the basal distance and the nanofiller content, the lower the heat capacity step at the glass transition temperature ΔCp(Tg), and the lower the CRR volume. It is also shown that the evolution of the CRR volume is consistent with the evolution of the fragility index obtained by DSC and Broadband Dielectric Spectroscopy (BDS) when the nanofiller content changes. The fragility index and the CRR size decreases can be correlated to nanofiller presence, hindering the molecular movements. From the Vollenberg and Heikens [34] approach, this behaviour can also be interpreted through the existence of an interfacial bilayer. This interfacial bilayer is composed by a zone, which is next to the nanofiller, with a density higher than the matrix one, followed by a more expanded zone with a density lower than the matrix one.

The recent advances in the inelastic background analysis of XPS spectra recorded with hard X-rays (HAXPES) make this method a powerful, non-destructive solution to retrieve the depth distribution of deeply buried elements. In this work, we apply this technique to study diffusion phenomena, upon annealing, in power transistor devices. We present a complete analysis of a sample under different stages of the fabrication process. We investigate the accuracy of the determination of the elemental depth distributions and concentrations with the inelastic background analysis. This is done by cross-checking with destructive techniques such as TEM/EDX and XPS depth profiling, and we shed light on the complementarities with HAXPES core-level analysis. We obtain consistent results which show the formation of an inhomogeneous structure, consisting of a blend of Al, Ga, and Ta, created during annealing. The inelastic background analysis permits us to successfully retrieve the depth distributions of Al, Ga, and Ta as well as the concentration of these elements in each layer, over 70 nm below the surface. This is a reliable solution to investigate diffusion phenomena and improve the fabrication processes of devices with critical and deeply buried interfaces.

In this study, we detail a Cu2ZnSnSe4 based solar cell fabrication process based on the selenization of metallic precursor stacks with elemental Se. 9.4% efficient devices without antireflection coating have been obtained. First, reproducibility issues of the process are carefully shown and discussed. It is demonstrated that device performances are strongly impacted by the precise control of the precursor composition. Then, starting from this robust process, a review of existing strategies to improve kesterite efficiencies is conducted. A significant increase in efficiency (+1.4% absolute efficiency and +50 mV VOC) is obtained with absorber surface treatment and post-annealing, while no effect of Ge incorporation in the precursor stack is observed. This contradictory result to most of the recent publications raises the question of the universality of this strategy to improve kesterite solar cell performance. Finding a universal activation step to boost kesterite efficiencies and bring it to the market remains a crucial need for the community.In this study, we detail a Cu2ZnSnSe4 based solar cell fabrication process based on the selenization of metallic precursor stacks with elemental Se. 9.4% efficient devices without antireflection coating have been obtained. First, reproducibility issues of the process are carefully shown and discussed. It is demonstrated that device performances are strongly impacted by the precise control of the precursor composition. Then, starting from this robust process, a review of existing strategies to improve kesterite efficiencies is conducted. A significant increase in efficiency (+1.4% absolute efficiency and +50 mV VOC) is obtained with absorber surface treatment and post-annealing, while no effect of Ge incorporation in the precursor stack is observed. This contradictory result to most of the recent publications raises the question of the universality of this strategy to improve kesterite solar cell performance. Finding a universal activation step to boost kesterite efficiencies and bring it to the market rem...

Highly tensile strained (up to 2.2%) thin monocrystalline silicon (mc-Si) films were fabricated by a simple and low-cost method based on the in-plane expansion of meso-porous silicon (PS) substrates upon low temperature oxidation. To control the film thickness below 100 nm, an original “two wafer” technique was employed during the porosification process. This method enables the fabrication of a 60 nm thick mc-Si films on 250 μm thick meso-porous silicon substrates over areas as large as 2 in. with a surface roughness and cleanliness comparable to that of standard Si wafers. Crack-free 60 nm thick Si films can be strained up to 1.2% by controlled low temperature oxidation of the PS substrate. Structural and strain analysis of the PS/mc-Si structures performed by transmission electron microscopy and micro-Raman scattering spectroscopy are reported.

We present a new set of electrical field effect measurements on granular aluminium insulating thin films. We have explored how the conductance relaxations induced by gate voltage changes depend on the age of the system, namely the time elapsed since its quench at low temperature. A clear age dependence of the relaxations is seen, qualitatively similar to ageing effects seen in other well studied glassy systems such as spin glasses or polymers. We explain how our results differ from the previous ones obtained with different protocols in indium oxide and granular aluminium thin films. Our experimental findings bring new information on the dynamics of the system and put new constraints on the theoretical models that may explain slow conductance relaxations in disordered insulators., 4 pages, 13th Transport in interacting disordered systems (TIDS13) conference, 31.08.2009 - 05.09.2009, Rackeve, Hungary

A simple and low-cost technological approach for the straining of thin crystalline silicon (Si) films using porous silicon (PS) as stress generating nanomaterial is reported. Firstly, a technological approach allowing fabrication of thin Si films on a bulk meso-PS substrate is described. Secondly, low temperature (

Temperature modulated differential scanning calorimetry (TMDSC) and dynamic mechanical analysis (DMA) are used to calculate cooperative rearranging region (CRR) average sizes for drawn poly(ethylene terephthalate) (PET) with different draw ratios (λ) ranging from λ=1 to 4, according to Donth’s approach. It is shown for both studies that the CRR size decreases when increases, due to the amorphous phase confinement by the crystals generated during the drawing. However, differences observed between the values calculated from TMDSC and DMA investigations are explained by the differences between a mechanical uniaxial dynamic solicitation (DMA) or a thermal solicitation (TMDSC) in terms of cooperative rearrangements at the glass transition.

International audience; Samples of polyethylene-1.4-cyclohexylenedimethylene terephthalate glycol (PETg) with different filler contents were prepared by a master batch process. The intercalated dispersion state of montmorillonite (MMT) was characterized using X-Ray Diffraction. Two different sample series are put in evidence with different basal distances (3.31 and 3.48 nm). The influence of nanofiller on ­ alpha and beta relaxations was studied by Dielectric Relaxation Spectroscopy and Differential Scanning Calorimetry. The use of these two techniques allowed us to determine accurately the fragility index m at the glass transition temperature Tg. For Tg, m, the Kauzmann temperature TK, and the relaxation time at Tg (Tg), we showed a decrease of the values more important for 3.48 nm basal distance than for 3.31 nm. The did not seem affected in its apparent activation energy Ea by the MMT addition.

Amorphous films of poly(ethylene terephthalate) were uniaxially drawn at temperature above the glass transition for different draw ratio λ in order to study samples with various structural states. When λ is lower than 2, the films remain practically amorphous while for higher λ values, a strain induced crystallization takes place and reaches a maximum. In regard to the Strong Fragile concept proposed by C.A. Angell, the different materials show a behaviour breaking when the draw ratio λ is greater than 3. Indeed, the weakly drawn samples show a fragile behavior, while the highly drawn samples show a strong one. Moreover, the determination of the average size of the Cooperative Rearranging Regions (CRR) proves the crystallite presence drastically reduces the CRR size in the remaining amorphous phase. This is attributed to an amorphous phase confinement effect by the crystallites and the rigid amorphous fraction.

Heteroepitaxial SrO films grown on Si(001) are characterized by reflection high energy electron diffraction and x-ray photoelectron spectroscopy. Special emphasis is put on the interface chemical, structural, and thermal stability because SrO films can be used as template layers for growing crystalline high-k oxides on Si(001). Ultrathin SrO layers of good crystalline quality with sharp interface with Si(001) can be grown at low temperature (50°C) and low partial oxygen pressure (

This study investigates the tensile-strained growth of LaAlO 3 on SrTiO 3 (0 0 1) substrates by molecular beam epitaxy (MBE). Reflection high energy electron diffraction (RHEED) analysis shows that for thicknesses smaller than 3.8 nm, LaAlO 3 films grow pseudomorphically. Plastic relaxation starts for films thicker than this critical thickness of 3.8 nm. The in- and out-of-plane lattice parameters of the strain-relaxed layers have been measured by ex situ X-ray diffraction (XRD) at grazing angles. The relaxation mechanism is discussed on the basis of these experiments.

Reactive compatibilization of recycled low- or high-density polyethylenes (LDPE and HDPE, respectively) and ground tire rubber (GTR) via chemical interactions of pre-functionalized components in their blend interface has been carried out. Polyethylene component was functionalized with maleic anhydride (MAH) as well as the rubber component was modified via functionalization with MAH or acrylamide (AAm) using chemically or irradiation (γ-rays) induced grafting techniques. The grafting degree and molecular mass distribution of the functionalized polymers have been measured via FTIR and Size Exclusion Chromatography (SEC) analyses, respectively. Thermoplastic elastomer (TPE) materials based on synthesized reactive polyethylenes and GTR as well as ethylene-propylene-diene rubber, EPDM were prepared by dynamic vulcanization of the rubber phase inside thermoplastic (polyethylene) matrix and their phase structure, and main properties have been studied using DSC and mechanical testing. As a final result, the high performance TPE with improved mechanical properties have been developed.

Thin films of a few layers of graphene obtained by solid-state graphitization from 6H-SiC(0001) substrates have been studied by X-ray photoelectron spectroscopy (XPS) and X-ray photoelectron diffraction (XPD). The C1s core-level was resolved into components, which were associated with carbon from bulk SiC, carbon from graphene and carbon at the graphene/6H-SiC(0001)interface. Then, the intensity of each of these components was recorded as a function of the polar (azimuth) angle for several azimuth (polar) angles. These XPD measurements provide crystallographic information which clearly indicates that the graphene sheets are organized in a graphite-like structure on 6H-SiC(0001), an organisation that results from the shrinking of the 6H-SiC(0001) lattice after Si depletion. Finally the decoupling of graphene from the 6H-SiC(0001) substrate by oxygen intercalation was studied from the XPS point of view.

The development of lightweight and flexible photovoltaic devices is highly desirable for integration in new applications and to reduce the manufacturing cost of modules. In this context, a lot of effort is put into the development of Cu(In,Ga)Se 2 (CIGS) based solar cells on flexible substrates as alternatives to the standard soda-lime glass substrates. Flexible metallic and polyimide foils are frequently used, but in this work we investigated another substrate material presenting advantageous properties: flexible ultra-thin glass. In this article we demonstrated CIGS solar cells with 11.2% efficiency grown on flexible glass as thin as 100 μm. It was shown that the differences between solar cells fabricated on ultra-thin glass and standard cells fabricated on 1-mm-thick soda-lime glass lie in the lower Na supply in the first case. We also studied the evolution of solar cell performances when they are bent. This last point is rarely discussed in studies on CIGS on flexible substrates in spite of its importance for flexible device applications.