Rhodium nanoparticles stabilised by p-based ligands. Synthesis, characterisation and application in selective hydrogenations
- Autores: Jessica Llop Castelbou
- Directores de la Tesis: Cyril Godard (dir. tes.), Carmen Claver Cabrero (codir. tes.)
- Lectura: En la Universitat Rovira i Virgili ( España ) en 2014
- Idioma: español
- Tribunal Calificador de la Tesis: Bruno Chaudret (presid.), Paul Dyson (secret.), Piet W. N. M. Van Leeuwen (voc.)
- Materias:
- Enlaces
- Tesis en acceso abierto en: TDX
- Resumen
Metal nanoparticles are currently an area of intense scientific research, due to a wide variety of potential applications in biomedical, optical, electronic and catalytic fields. In catalysis, soluble nanoclusters are considered at the frontier between homogeneous and heterogeneous catalysts since they could present the advantages of both: high selectivity by modulation of their surface and catalyst recovery and reuse.
Soluble nanoparticles exhibit several advantages such as higher surfaces areas that confer higher activity and avoid internal mass transfer limitations. Moreover, these systems are freely rotational and three- dimensional in reaction systems. Therefore, their metal- surface active sites are much more accessible for the reactant molecules, which enhance their activity.
Transition-metal nanoclusters are only kinetically stable, since the thermodynamic minimum is the bulk metal. There is therefore a tendency for aggregation which leads to the loss of the properties associated with the colloidal state of these metallic particles. The use of a stabilising agent is required and opens a wide range of possibilities. Polymers, surfactants or ionic liquids are used for instance as stabilisers.
Phosphorus based ligands such as phosphines and phosphites are extensively used in homogeneous catalysis due to their broad coordination chemistry with transition metals and the possibility to fine-tune the electronic and steric properties of the catalysts through structural modifications of the ligands to obtain high activity and selectivity in catalytic processes. In the last decade, these ligands were also shown to efficiently stabilise metal nanoparticles that are catalysts in several catalytic reactions. However, to date, the fine tuning of the properties of this type of catalysts to achieve specific selectivities remains a challenge.
These systems could be applied as catalysts for several transformations such as C-C couplings, oxidations or hydrogenation reactions. The nature of the real catalyst when dealing with soluble nanoparticles could be in some cases controversial and could required parallel studies.
This thesis focus on the synthesis and characterisation of rhodium nanoparticles stabilised with P-based ligands and their application in selective hydrogenation reactions. A preliminary study on the anchor of these systems onto modified silica will also be explored.
The rhodium nanoparticles were stabilised using mono and bidentate phosphines and phosphites. Triphenylphosphine 1 and triphenylphosphite 5 were used as model for each family. In general terms, the most important difference between these two systems was the observation of ligand hydrogenation during the synthesis for the phosphine stabilised nanoparticles that was not detected in the case of Rh5 system.
The amount of stabilising ligand during the synthesis was explored for these stabilisers. No direct correlation between the amount of stabiliser used to synthesised the nanoparticles and the content revealed by TGA was observed for the phosphine systsystems, while the phosphite systems followed the coherent result of more stabilising ligand at the surface of the nanoparticles with more stabilised used during the synthesis.
Moreover, the steric and electronic properties of the ligands were also studied. For that purpose, tricyclohexyl 2 and trimethylphosphine 3 were employed for the phosphine family and triorthotertbutyl 6 and trimethylphosphite 7 for the phosphites. More crowded systems at the surface were observed using the less sterical demand stabilisers.
The hapticity of the ligands was also studied by the use of bidentate ligands. The expected bidentate stabilisation was observed using dppb 4 as stabilising ligand for the phosphine family. However, both triphenylphosphite 5 and the diphosphite 8 could be stabilising the nanoparticles in a monodentate way.
Rhodium nanoparticles stabilised with the PVP K-90 polymer and with the mixture of solvents THF/MeOH were also synthesised for comparative purposes.
IR spectroscopy was also applied to study the surface of the previously synthesised nanoparticles. Moreover, carbon monoxide was adsorbed onto the surface of these systems try to gain information of the nature and position of the stabilising ligands by correlation with the carbonyls observed and consequently of the active sites of the nanoparticles.
The different P- coverage modulated with the different ligands used a stabilising ligands seems to mainly affect the faces of the NPs, as large shifts were observed in the bridging CO region. Two distinct CO environments were detected for both bridging and terminal ligands, which indicates two distinct terminal positions for COs and two surroundings for these ligands on the faces of the NPs. The use of small ligands resulted in high mobility of the CO ligands at the NPs surface and more different active sites and consequently different CO adsorptions were observed with more sterical demanding ligands as stabilisers.
These systems were applied in chapter 3 as catalysts for the hydrogenation of arenes. In a first stage, the disubstituted arenes 20a-f were used as substrate. The nature and position of the substrates was evaluated using ortho, meta and para xylenes and methylanisoles. System Rh5 stabilised with triphenylphosphite 5 was not active, maybe due to the crowding of the surface that could inhibit the substrate approach. Higher activities were observed using the phosphine stabilised systems compared to the phosphites. However, system Rh4 stabilised by dppb 4 was the less active of the series. Partially hydrogenated products were observed with significant amounts in the reactions carried out with system Rh8 stabilised with the diphosphite 8 as stabiliser. Reactions parameters such as the temperature, pressure, solvent or the use of additives were studied without improving of the amount of these products. No clear effect of the nature of the stabiliser on the cis/ trans selectivity was observed for these substrates.
The substitution of the substrate was also studied. High activities at room temperature and partially hydrogenated products were systems, while the phosphite systems followed the coherent result of more stabilising ligand at the surface of the nanoparticles with more stabilised used during the synthesis.
Moreover, the steric and electronic properties of the ligands were also studied. For that purpose, tricyclohexyl 2 and trimethylphosphine 3 were employed for the phosphine family and triorthotertbutyl 6 and trimethylphosphite 7 for the phosphites. More crowded systems at the surface were observed using the less sterical demand stabilisers.
The hapticity of the ligands was also studied by the use of bidentate ligands. The expected bidentate stabilisation was observed using dppb 4 as stabilising ligand for the phosphine family. However, both triphenylphosphite 5 and the diphosphite 8 could be stabilising the nanoparticles in a monodentate way.
Rhodium nanoparticles stabilised with the PVP K-90 polymer and with the mixture of solvents THF/MeOH were also synthesised for comparative purposes.
IR spectroscopy was also applied to study the surface of the previously synthesised nanoparticles. Moreover, carbon monoxide was adsorbed onto the surface of these systems try to gain information of the nature and position of the stabilising ligands by correlation with the carbonyls observed and consequently of the active sites of the nanoparticles.
The different P- coverage modulated with the different ligands used a stabilising ligands seems to mainly affect the faces of the NPs, as large shifts were observed in the bridging CO region. Two distinct CO environments were detected for both bridging and terminal ligands, which indicates two distinct terminal positions for COs and two surroundings for these ligands on the faces of the NPs. The use of small ligands resulted in high mobility of the CO ligands at the NPs surface and more different active sites and consequently different CO adsorptions were observed with more sterical demanding ligands as stabilisers.
These systems were applied in chapter 3 as catalysts for the hydrogenation of arenes. In a first stage, the disubstituted arenes 20a-f were used as substrate. The nature and position of the substrates was evaluated using ortho, meta and para xylenes and methylanisoles. System Rh5 stabilised with triphenylphosphite 5 was not active, maybe due to the crowding of the surface that could inhibit the substrate approach. Higher activities were observed using the phosphine stabilised systems compared to the phosphites. However, system Rh4 stabilised by dppb 4 was the less active of the series. Partially hydrogenated products were observed with significant amounts in the reactions carried out with system Rh8 stabilised with the diphosphite 8 as stabiliser. Reactions parameters such as the temperature, pressure, solvent or the use of additives were studied without improving of the amount of these products. No clear effect of the nature of the stabiliser on the cis/ trans selectivity was observed for these substrates.
The substitution of the substrate was also studied. High activities at room temperature and partially hydrogenated products were systems, while the phosphite systems followed the coherent result of more stabilising ligand at the surface of the nanoparticles with more stabilised used during the synthesis.
Moreover, the steric and electronic properties of the ligands were also studied. For that purpose, tricyclohexyl 2 and trimethylphosphine 3 were employed for the phosphine family and triorthotertbutyl 6 and trimethylphosphite 7 for the phosphites. More crowded systems at the surface were observed using the less sterical demand stabilisers.
The hapticity of the ligands was also studied by the use of bidentate ligands. The expected bidentate stabilisation was observed using dppb 4 as stabilising ligand for the phosphine family. However, both triphenylphosphite 5 and the diphosphite 8 could be stabilising the nanoparticles in a monodentate way.
Rhodium nanoparticles stabilised with the PVP K-90 polymer and with the mixture of solvents THF/MeOH were also synthesised for comparative purposes.
IR spectroscopy was also applied to study the surface of the previously synthesised nanoparticles. Moreover, carbon monoxide was adsorbed onto the surface of these systems try to gain information of the nature and position of the stabilising ligands by correlation with the carbonyls observed and consequently of the active sites of the nanoparticles.
The different P- coverage modulated with the different ligands used a stabilising ligands seems to mainly affect the faces of the NPs, as large shifts were observed in the bridging CO region. Two distinct CO environments were detected for both bridging and terminal ligands, which indicates two distinct terminal positions for COs and two surroundings for these ligands on the faces of the NPs. The use of small ligands resulted in high mobility of the CO ligands at the NPs surface and more different active sites and consequently different CO adsorptions were observed with more sterical demanding ligands as stabilisers.
