Regarding bio-interfaces, biodetection systems involving original syntheses of silica-coated gold nano-rods and also gold- and silver-based core-shell nanoparticles grafted with antibodies have been designed. These devices have enabled antigens to be detected visually, even in small quantities or in complex matrices such as milk. This work has enabled the laboratory to consolidate its expertise in the surface chemistry of plasmonic nanoparticles for the controlled association of biomolecules with the objective to design nanosensors.

In catalysis, the colloidal approach for the synthesis of gold nanoparticles has also been exploited, with the inherent difficulties of removing, often at high temperature, the stabilising agents still present once the particles have been deposited on a support. In this context, the use of plasma to treat Au/TiO₂ catalysts in the powder form was validated thanks to the design of a fluidised bed reactor (post-doctorate by K. Leiva, Labex Matisse). The elimination of polyvinyl alcohol was established without any significant change in the size of the gold nanoparticles. An important aspect of catalysis involving metallic nanoparticles is the characterisation of their structure and surface composition, particularly in the case of bimetallic particles, under conditions that are as close as possible to those of the reaction. Studies using double aberration-corrected environmental electron microscopy (A. Nassereddine PhD, guest PhD student, coll. J. Nelaya, Paris Cité) have revealed, for the first time, in the case of gold nanoparticles on TiO₂ with dimensions of less than 4 nm, a change in morphology accompanied by a loss of the face-centred cubic crystalline structure (between 200 and 400°C under H₂ at atmospheric pressure). An unknown stable surface molecular structure of hydrogenated gold atoms decorating a highly deformed core was identified. In the case of AuCu bimetallic nanoparticles, changes in morphology were observed in the presence of H₂ but, whatever the initial size of the nanoparticles, no structural modification was observed. However, atoms migration (from the core to the surface for Cu and vice versa for Au) was demonstrated by calculation.

Confinement within mesoporous systems enables highly reactive nickel nanoparticles to be stabilised under the conditions of dry methane reforming. However, to obtain nanoparticles that are particularly small and preferentially positioned within the porous network, the LRS has recently adopted original strategies based on incorporating nickel in its oxidised form into aluminium-based Metal Organic Frameworks (MOFs) or mesoporous silicas during the synthesis of the support (alumina or silica). In the latter cases, a suitable heat treatment followed by a reduction step enables very good dispersions of Ni to be obtained within the porosity. These catalysts have shown excellent performance in the dry reforming of methane (PhD L. Karam, O. Daoura, EraNet SolCare Project and ED 397) and in the methanation reaction. Nickel on silica materials are also used in the depolymerisation of lignin by hydrogenolysis to selectively obtain phenolic compounds (PhD R. Sassine, DIM RESPORE). As the decomposition of MOFs to produce alumina by calcination of organic ligands is not a priori interesting from an economic and environmental point of view, another approach based on the substitution of para-terephthalic acid by domestic polyethylene terephthalate waste has been developed.

Still with a focus on dispersing and stabilising metal in the form of nanoparticles, or even isolated atoms on oxide supports, work has recently been undertaken by J. Schnee. Her project involves exploiting other crystalline solids, for example Cu(II)-substituted hydroxyapatites (HAp) synthesised at the LRS, to generate the dispersed metal by exsolution using a suitable reducing treatment. Using HAp at 1.5% w/w, complete exsolution of the copper was achieved, leading to highly dispersed species active in the selective catalytic reduction of NOx by NH₃. The particularity of this approach lies in the use of hydroxyapatites, which then promote resistance to sintering. The use of co-precipitation methods that are well mastered at LRS means that it is possible to prepare catalysts in the form of isolated atoms with contents of more than 2% w/w.

Producing more while consuming less raw material and less energy often leads to the design of complex multifunctional catalysts. This means being able to locate functions close to each other or not, as early as during the synthesis stage. An example of an application developed in the laboratory concerns the hydrocracking of alkanes, which involves acid sites and metal sites. Original nanocomposite supports based on nanozeolites and Boehmite nanoparticles synthesised by heterocoagulation of the two precursor materials were used to selectively deposit metallic species on one of the two constituent phases and thus compartmentalise or not the acid and metallic sites. It should be noted that heterocoagulation has also been used to develop materials combining catalytic and up-conversion functions. Another form of compartmentation of active species, this time based solely on the use of porous supports, is currently being studied to develop heterogeneous chemo-enzymatic catalytic systems combining a metallic chemical catalyst integrated into MOFs and an enzyme interacting strongly with this structure (see also theme 2). For other applications, the bifunctional catalysts designed in the lab can be organic/inorganic hybrid materials obtained by covalent grafting of metal complexes and organocatalytic functions onto the same porous support. This was particularly illustrated by the synthesis of cyclic carbonates directly from alkenes, O₂ and CO₂ (M. Balas PhD, ANR OxCyCat). The idea was to couple oxygen epoxidation catalysts (manganese or chromium Salen complexes) with catalysts for the cycloaddition of CO2 onto epoxides (quaternary ammonium salts). During this work, which combined optimisation of the catalytic system in solution with transposition onto a support, it was demonstrated that gains in terms of activity were obtained for the cycloaddition step when this was carried out in the presence of supported quaternary ammonium salts due to the activation of the epoxides by the surface silanol groups. The introduction of tertiary amine functions, capable of interacting with CO₂, either as substituents on the metal complexes or directly on the support, enabled the overall reaction to be carried out under milder conditions, leading to a more selective epoxidation process and at the same time increasing the cyclic carbonate yield. This work is currently being continued by H. Sun (Chinese Scholarship Council PhD).

As part of the SolCare project (ERANETMed JC-EN-ERGY-2014), significant work has been carried out to produce highly dispersed Ni nanoparticles on a porous support that are stable at high temperatures and can therefore be used in a demanding catalytic test such as the dry reforming of methane. From the synthesis point of view, different and original methodologies had to be implemented depending on whether the support was alumina (PhD L. KARAM) or silica (PhD O. DAOURA).

For alumina, a mesoporous mixed oxide phase of the NiAl2O4 type was produced by thermal degradation of a porous coordination polymer (MOF) impregnated with Ni²⁺ ions.

The reduction of this intermediate material led to the formation of Ni(0) nanoparticles that were very well dispersed because they interacted strongly with the support. The performance of these catalysts for dry reforming and methanation was superior to that of commercial catalysts prepared conventionally on aluminas.

Regarding silica, two options were studied in parallel to optimise the dispersion of Ni (5% w/w). In one option ('One-pot' method), Ni²⁺ ions were incorporated into the synthesis gel of a mesoporous silica of the SBA-15 type prepared in the presence of a minimum quantity of water.

In the other, a pre-formed silica (Aerosil-380) was impregnated with an aqueous ammoniacal solution of Ni(II) ions. In both cases, significant Ni metal dispersion values for silica (of the order of 40% (measured by H₂ chemisorption) were achieved after reduction. The resulting catalysts proved to be efficient and stable in dry methane reforming, even with high GHSV values. Prior to the reduction step, X-ray scattering studies (PDF, collaboration with C. Sassoye, LCMCP) established the formation of nanometric Ni phyllosilicates for the impregnation method and NiO nanoclusters for the 'one-pot' method. In the latter case, the conversion of nickel oxide nanoclusters into Ni(0) nanoparticles integrated mainly into the mesopores network was demonstrated by tomography (collaboration with O. Ersen, IPCMS).

In the field of catalysis, colloidal nanocomposites of the core-crown type resulting from a combination of nanomaterials providing different catalytic functions or a catalytic function and an up-conversion function have been developed using electrostatic heterocoagulation, a technique that is still little used. It was thus possible to show that the performance of the systems obtained, involving detailed control of the respective locations of the two nanomaterials, was generally better than that of a simple mixture.

With regard to the combination of two different catalytic functions, work has focused on the development of materials with acid and metal functions, used in particular for the hydrocracking reaction. Nanocomposites with a core based on nanozeolite  and a shell made of boehmite nanoparticles have been developed by heterocoagulation (O. Ben Moussa PhD, Labex Matisse). After calcination, the resulting materials were used as Pt supports that could be positioned either on the zeolite part or on the alumina part. These nanocomposites were then tested as catalysts for the hydrocracking of heptane, demonstrating that the selectivity of the reaction can be affected by the location of the Pt and the degree of intimacy between the two components of the support.

A similar approach, also based on electrostatic hetero-coagulation, was used (D. Hu PhD, ANR NSERC UpPhotoCat) to combine a photocatalyst (ZnO nano-rods) with a Tm³⁺,Yb³⁺:LiYF₄ nanoparticle with up-conversion properties (i.e. capable of summing two or more IR radiations to produce radiations over a wide energy range (from the near IR to the UV)). The aim was to obtain nanocomposites that are photocatalytically active in the near IR range for use in wastewater treatment. Heterocoagulation was carried out between the silica-coated up-conversion particle (PZC≈2) and the ZnO nanorods (PZC≈9) and resulted in colloidal nanocomposites combining the two functions. Measurements of photocatalytic activity under UV and near-IR radiation (974 nm laser) established that combining the two components under these mild conditions perfectly preserves the properties of each of them, which is not the case when using conventional association routes (hydrothermal growth of ZnO on silica).

In the field of bio-detection, other nanocomposites, again of the gold-based core-shell type, have been developed for antigen detection based on plasmon resonance perturbation, the idea being to enable visual detection at thresholds in the ng/mL range.

Surface functionalisation using antibodies is a prerequisite. In the case of gold nanorods, the presence of cetyltrimethylammonium bromide (CTAB) is problematic. Work has therefore been undertaken in the Laboratory to surround the nano-rods with a film of silica, which is also more favourable to functionalisation (V. Pellas PhD, ED 397). It was thus possible to control the formation of the silica layer by adjusting the pH so as to separate the tetraethoxysilane hydrolysis and silica condensation stages. In particular, mastering synthesis conditions enabled us to control the orientation of the porosity (induced by the presence of CTAB), which can be either parallel or perpendicular to the gold surface. The CTAB could then be completely removed by washing, allowing the use of the nano-objects formed in this way for drug delivery or photo-thermal therapy. For the thinnest thicknesses, bio-detection experiments based on gold surface plasmon resonance were carried out. The presence of silica made it possible to functionalise the surface with rabbit anti-IgG antibody molecules using covalent or electrostatic approaches.

Once the non-specific sites had been eliminated by adsorption of bovine serum albumin, the nanosensors prepared in this way demonstrated their efficiency for rabbit IgG concentration ranges of between 25 and 500 ng/mL using a simple bench-top UV-Vis. spectrometer.

Other systems, based on Au@AgNP or Ag@AuNP core-shell nanoparticles, have been designed to control the position of the plasmon resonance band with nanometric precision (PhD L. Zhang, CMI Initiative). The reported detection limits for staphylococcal enterotoxin A (SEA) are in the range of 0.2 to 0.4 nM. It has also been shown that nano-objects obtained by incorporating Ag into hollow gold nanoparticles are the most suitable for visual bio-detection. Coupling gold nanoparticles functionalised by the staphylococcal enterotoxin A antibody with glass surfaces exposing the same antibody led to design devices for detecting SEA in milk with a limit of 1.5 ng/mL.