Aucun !

The goal of this PhD was to elaborate supported Ru-NHC catalytic materials based on hybrid organic-inorganic materials having imidazolium units perfectly distributed within a silica matrix. Passivation of these imidazolium materials followed by formation of NHC-carbene and reaction with [RuCl2(p-cymene)]2 provided these well-defined surface sites of general structures RuCl2(NHC)(L), where L was para-cymene (p-cymene) or THF depending on the reaction conditions, which could be further replaced by PMe3. These systems were then tested in the hydrogenation of CO2 in presence of amine to give formamides. The mono-NHC systems were highly active only in the presence of PMe3 ligands, but suffered from Ru leaching, evidencing the low stability of the NHC-Ru bond under the reaction conditions. On the other hand, dinuclear bis-NHC Ru systems were also developed, and they displayed much improved activity and stability in the hydrogenation of CO2 in the presence of PMe3 compared to the mono-NHC systems. This allowed the use of much higher reaction temperatures (200 °C) and provided heterogeneous catalysts with performances close to those obtained with the best homogeneous catalysts, Cl2Ru(dppe)2.

The direct catalytic hydrogenation of nitrile butadiene rubber (NBR) in latex form was studied as a model system for the development of a new latex hydrogenation process for the modification of unsaturated diene-based polymers. NBR is a synthetic rubber of copolymerized acrylonitrile and butadiene produced in latex form by emulsion polymerization. The catalytic hydrogenation of NBR is an important post-polymerization process resulting in a more stable and tougher derivative, hydrogenated NBR (HNBR), which has been widely used in the automotive and oil drilling industry. The present commercial process involves a number of cumbersome steps to obtain solid NBR from the latex and subsequent dissolution of the solid NBR in a large amount of organic solvent followed by solvent recovery after coagulation of the hydrogenated NBR. Since NBR is produced in latex form, it is very desirable to directly hydrogenate NBR in the latex form which will significantly simplify the hydrogenation process and facilitate subsequent applications. As an economical and environmentally benign alternative to the commercial processes based on the hydrogenation of NBR in organic solution, this direct latex hydrogenation process is of special interest to industry. The objective of this project is to develop an efficient catalytic system in order to realize the direct catalytic hydrogenation of NBR in latex form. OsHCl(CO)(O2)(PCy3)2 was initially used as the catalyst to investigate the possibility of hydrogenation of NBR in latex form and to understand the major factors which affect the hydrogenation operation. It was found that an organic solvent which is capable of dissolving or swelling the NBR was needed in a very small amount for the latex hydrogenation using the Os catalyst, and gel occurred in such a catalytic system during hydrogenation. Wilkinson’s catalyst, RhCl(PPh3)3, was then used for the latex hydrogenation in the presence of a small amount of solvent successfully without gel formation. Further investigation found that Wilkinson’s catalyst has a high activity for NBR latex hydrogenation without the use of any organic solvent. The influences of various operation conditions on hydrogenation rate, such as catalyst and polymer concentrations, latex system composition, agitation, reaction temperature and hydrogen pressure, have been investigated. It was found that the addition of triphenylphosphine (TPP) has a critical effect for the hydrogenation of NBR latex, and the hydrogenation rate was mainly controlled by the amount of catalyst which diffused into the polymer particles. In the presence of TPP, NBR latex can be hydrogenated to more than 95% degree of hydrogenation after about 30 hours at 160oC using Wilkinson’s catalyst with a catalyst to NBR rubber ratio of 1 wt%, without the addition of any organic solvent. The apparent activation energy for such NBR latex hydrogenation over the temperature range of 152oC to 170oC was found to be 57.0 kJ/mol. In the present study, it was also found that there are some impurities within the NBR latex which are detrimental to the hydrogenation reaction and are suspected to be water-soluble surfactant molecules. Deliberately designed solution hydrogenation experiments were conducted to study the impurity issue, and proper latex treatment methods have been found to purify the latex before hydrogenation. To improve the hydrogenation rate and to optimize the latex hydrogenation system, water soluble RhCl(TPPMS)3 catalyst (TPPMS: monosulphonated-triphenylphosphine) was used for the latex hydrogenation of NBR. The latex hydrogenation using the water soluble catalyst with TPP can achieve more than 90% degree of hydrogenation within 20 hours at 160oC. Further experiments using RhCl3 with TPP proved that the water soluble RhCl3 can be directly used as a catalyst precursor to generate the catalytic species in situ for the latex hydrogenation, and a stable NBR latex with 96% degree of hydrogenation can be produced without any gel problem within 19 hours of reaction at 160oC. The catalyst mass transport processes for these Rh based catalysts in the latex system were investigated in order to further optimize the solvent-free latex hydrogenation process. While maintaining the emulsified state of the original latex, the direct catalytic hydrogenation of NBR latex can be carried out efficiently without any cross-linking problem to more than 92% degree of hydrogenation within 8 hours at 160oC. As a result of this research project, new latex hydrogenation technologies were successfully developed to fulfill all major requirements for a solvent-free polymer latex hydrogenation route, which is a significant milestone for the improvement of this polymer modification technology. The finding of TPP’s role as the “catalyst mass transfer promoter” is a breakthrough for the research field related to the hydrogenation of unsaturated diene-based polymers in latex form.

Hydrogenation of nitrile butadiene rubber (NBR) has been carried out industrially for a number of years, producing a material with exceptional resilience to high temperatures and oxidative conditions. Current processes involve a batch reactor which is difficult to optimize further for larger scale production. A continuous process for this particular process is required in order to provide a large volume of production with consistent qualities. The integration of heat balance could be realized in a continuous process. A novel continuous process for hydrogenation of NBR has been developed in the present work. A multistage agitated contactor (MAC) was proposed as a gas liquid reactor for this process. Comprehensive hydrodynamic data have been acquired under various process conditions. The hydrodynamic behaviour under different operating variables such as stirring speed, liquid flow rate and gas flow rate has been understood through experimental study. It is found that an increase in stirring speed intensifies liquid backmixing while an increase liquid flow rate decreases liquid backmixing. The presence of gas flow helps in reducing liquid back mixing by two coupled effects: liquid entrainment effect due to a cocurrent operation manner and a strengthening effect of liquid flow rate due to its reduction of liquid hold-up. Contradictory conclusions regarding the effect of liquid viscosity on liquid backmixing in a MAC have been resolved through experimental investigation and computational fluid dynamics (CFD) simulations. It is shown that an increase in liquid velocity dampens turbulence which contributes to liquid phase backmixing within the reactor. The established hydrodynamic understanding of MACs in the present work widens its potential application for gas liquid process. Based on comprehensive understanding of the proposed reactor, a bench-scale prototype was designed and constructed in order to demonstrate hydrogenation performance. One more efficient catalyst for NBR hydrogenation, an osmium-based catalyst, was used in the present work. Hydrogenation degree of NBR in the continuous unit was investigated at operating conditions relevant to industrial applications. It is indicated from the experimental results that a desired hydrogenation degree of over 95% in 2.5% and 5% NBR solutions can be achieved at the conditions investigated. It is also shown that both system pressure and catalyst loading increase hydrogenation conversion. Mathematical modeling of the designed process was established by coupling the intrinsic catalytic hydrogenation from batch studies and flow behavior of the reactor. A cascade of stirred tanks with back flow (CTB) model was used to characterize the dynamic hydrogenation performance in a MAC. The comparison of experimental results and numerical prediction indicates that the established model could satisfactorily predict the hydrogenation in the designed process with consideration of approximately 30%-50% catalyst deactivated due to impurities and oxygen contamination in the polymer solution. A revised n CSTRs-in-series model was proposed to predict the hydrogenation degree at steady state and a good agreement was found when comparing the predicted results with the experimental data. A continuous process for hydrogenation at a pilot scale was designed based on the primary results from the bench scale process. A process with a capacity of 50 tons/year was targeted and the hydrogenation efficiency provided by the pilot scale unit has been estimated through the established reactor model.

The direct catalytic hydrogenation of nitrile butadiene rubber (NBR) in latex form was studied as a model system for the development of a new latex hydrogenation process for the modification of unsaturated diene-based polymers. NBR is a synthetic rubber of copolymerized acrylonitrile and butadiene produced in latex form by emulsion polymerization. The catalytic hydrogenation of NBR is an important post-polymerization process resulting in a more stable and tougher derivative, hydrogenated NBR (HNBR), which has been widely used in the automotive and oil drilling industry. The present commercial process involves a number of cumbersome steps to obtain solid NBR from the latex and subsequent dissolution of the solid NBR in a large amount of organic solvent followed by solvent recovery after coagulation of the hydrogenated NBR. Since NBR is produced in latex form, it is very desirable to directly hydrogenate NBR in the latex form which will significantly simplify the hydrogenation process and facilitate subsequent applications. As an economical and environmentally benign alternative to the commercial processes based on the hydrogenation of NBR in organic solution, this direct latex hydrogenation process is of special interest to industry. The objective of this project is to develop an efficient catalytic system in order to realize the direct catalytic hydrogenation of NBR in latex form. OsHCl(CO)(O2)(PCy3)2 was initially used as the catalyst to investigate the possibility of hydrogenation of NBR in latex form and to understand the major factors which affect the hydrogenation operation. It was found that an organic solvent which is capable of dissolving or swelling the NBR was needed in a very small amount for the latex hydrogenation using the Os catalyst, and gel occurred in such a catalytic system during hydrogenation. Wilkinson’s catalyst, RhCl(PPh3)3, was then used for the latex hydrogenation in the presence of a small amount of solvent successfully without gel formation. Further investigation found that Wilkinson’s catalyst has a high activity for NBR latex hydrogenation without the use of any organic solvent. The influences of various operation conditions on hydrogenation rate, such as catalyst and polymer concentrations, latex system composition, agitation, reaction temperature and hydrogen pressure, have been investigated. It was found that the addition of triphenylphosphine (TPP) has a critical effect for the hydrogenation of NBR latex, and the hydrogenation rate was mainly controlled by the amount of catalyst which diffused into the polymer particles. In the presence of TPP, NBR latex can be hydrogenated to more than 95% degree of hydrogenation after about 30 hours at 160oC using Wilkinson’s catalyst with a catalyst to NBR rubber ratio of 1 wt%, without the addition of any organic solvent. The apparent activation energy for such NBR latex hydrogenation over the temperature range of 152oC to 170oC was found to be 57.0 kJ/mol. In the present study, it was also found that there are some impurities within the NBR latex which are detrimental to the hydrogenation reaction and are suspected to be water-soluble surfactant molecules. Deliberately designed solution hydrogenation experiments were conducted to study the impurity issue, and proper latex treatment methods have been found to purify the latex before hydrogenation. To improve the hydrogenation rate and to optimize the latex hydrogenation system, water soluble RhCl(TPPMS)3 catalyst (TPPMS: monosulphonated-triphenylphosphine) was used for the latex hydrogenation of NBR. The latex hydrogenation using the water soluble catalyst with TPP can achieve more than 90% degree of hydrogenation within 20 hours at 160oC. Further experiments using RhCl3 with TPP proved that the water soluble RhCl3 can be directly used as a catalyst precursor to generate the catalytic species in situ for the latex hydrogenation, and a stable NBR latex with 96% degree of hydrogenation can be produced without any gel problem within 19 hours of reaction at 160oC. The catalyst mass transport processes for these Rh based catalysts in the latex system were investigated in order to further optimize the solvent-free latex hydrogenation process. While maintaining the emulsified state of the original latex, the direct catalytic hydrogenation of NBR latex can be carried out efficiently without any cross-linking problem to more than 92% degree of hydrogenation within 8 hours at 160oC. As a result of this research project, new latex hydrogenation technologies were successfully developed to fulfill all major requirements for a solvent-free polymer latex hydrogenation route, which is a significant milestone for the improvement of this polymer modification technology. The finding of TPP’s role as the “catalyst mass transfer promoter” is a breakthrough for the research field related to the hydrogenation of unsaturated diene-based polymers in latex form.

Catalytic naphthalene hydrogenation to tetralin in water/hydrocarbon emulsions with simultaneous water gas shift as the hydrogen source was performed in a 300 ml batch autoclave as a model for aromatic hydrogenation in water/bitumen emulsions. The catalyst utilized was an unsupported and dispersed type based on molybdenum sulfide (MoS2). Distinguishing the fate of hydrogen from water as opposed to molecular hydrogen in hydrogenation and water gas shift was accomplished by utilizing deuterium oxide (D2O) with NMR spectroscopy. The use of D2O allowed determination of isotope effects when compared with H2O. Diffuse Reflectance Infrared Fourier Transform Spectroscopy was performed to observe CO adsorption on the MoS2 sulfide surface. Ruthenium was tested as a potential candidate to enhance the activity of the Mo catalyst. Iron, nickel and vanadium were utilized in combination with molybdenum to test promotional/inhibitive activity during naphthalene hydrogenation and water gas shift since Ni and V are found in significant quantities in real bitumen feed. Finally, a multifactorial experiment was performed to test the hydrogenation and water gas shift activity of a binary VNiMo-sulfide catalyst towards H2S partial pressure, temperature and H2 versus CO atmospheres. Deuterium from D2O was incorporated into both saturated and aromatic hydrogen positions in tetralin products. Calculation of a Hydrogenation Index and Exchange Index indicated the extent of H-exchange is greater than hydrogenation. Exchange between D2O and organic products was enhanced with the MoS2 catalyst under H2 or CO compared to N2. A kinetically measured isotope effect of 1.58 was in agreement with a quasi-equilibrium thermodynamic isotope effect for O-H dissociations measured in the literature. A true kinetic isotope effect may be masked by transient surface concentrations occurring under batch conditions. Two strong vibrational bands associated with adsorbed CO were observed over MoS2 above 160 °C. Activation of the MoS2 surface with CO produces COS, suggesting an analgous mechanism to the production of H2S during reduction in H2. In the presence of H2S, Ru displayed low catalytic activity for both water gas shift and naphthalene hydrogenation, attributed to incomplete sulfidation to active RuS2. FeMo and VMo exhibited lower hydrogenation activity than Mo, but the water gas shift activity of VMo was high. A ternary VNiMo displayed lower hydrogenation activity than NiMo and Mo but was higher than VMo, implying Ni could offset the inhibition caused by V. Recycle of V and Ni rich asphaltene residues in catalytic slurry upgrading may therefore be feasible. An analysis of the effect of H2S pressure, temperature and type of reduction gas (CO vs. H¬2) concluded that temperature had the greatest positive effect on rate, followed by a small interaction effect of temperature/gas type and PH2S/gas type. The proximity to equilibrium conversions in WGS limited the analysis, while equilibrium limited the conversion of naphthalene at 380 °C in the batch reactor.

Catalytic naphthalene hydrogenation to tetralin in water/hydrocarbon emulsions with simultaneous water gas shift as the hydrogen source was performed in a 300 ml batch autoclave as a model for aromatic hydrogenation in water/bitumen emulsions. The catalyst utilized was an unsupported and dispersed type based on molybdenum sulfide (MoS2). Distinguishing the fate of hydrogen from water as opposed to molecular hydrogen in hydrogenation and water gas shift was accomplished by utilizing deuterium oxide (D2O) with NMR spectroscopy. The use of D2O allowed determination of isotope effects when compared with H2O. Diffuse Reflectance Infrared Fourier Transform Spectroscopy was performed to observe CO adsorption on the MoS2 sulfide surface. Ruthenium was tested as a potential candidate to enhance the activity of the Mo catalyst. Iron, nickel and vanadium were utilized in combination with molybdenum to test promotional/inhibitive activity during naphthalene hydrogenation and water gas shift since Ni and V are found in significant quantities in real bitumen feed. Finally, a multifactorial experiment was performed to test the hydrogenation and water gas shift activity of a binary VNiMo-sulfide catalyst towards H2S partial pressure, temperature and H2 versus CO atmospheres. Deuterium from D2O was incorporated into both saturated and aromatic hydrogen positions in tetralin products. Calculation of a Hydrogenation Index and Exchange Index indicated the extent of H-exchange is greater than hydrogenation. Exchange between D2O and organic products was enhanced with the MoS2 catalyst under H2 or CO compared to N2. A kinetically measured isotope effect of 1.58 was in agreement with a quasi-equilibrium thermodynamic isotope effect for O-H dissociations measured in the literature. A true kinetic isotope effect may be masked by transient surface concentrations occurring under batch conditions. Two strong vibrational bands associated with adsorbed CO were observed over MoS2 above 160 °C. Activation of the MoS2 surface with CO produces COS, suggesting an analgous mechanism to the production of H2S during reduction in H2. In the presence of H2S, Ru displayed low catalytic activity for both water gas shift and naphthalene hydrogenation, attributed to incomplete sulfidation to active RuS2. FeMo and VMo exhibited lower hydrogenation activity than Mo, but the water gas shift activity of VMo was high. A ternary VNiMo displayed lower hydrogenation activity than NiMo and Mo but was higher than VMo, implying Ni could offset the inhibition caused by V. Recycle of V and Ni rich asphaltene residues in catalytic slurry upgrading may therefore be feasible. An analysis of the effect of H2S pressure, temperature and type of reduction gas (CO vs. H¬2) concluded that temperature had the greatest positive effect on rate, followed by a small interaction effect of temperature/gas type and PH2S/gas type. The proximity to equilibrium conversions in WGS limited the analysis, while equilibrium limited the conversion of naphthalene at 380 °C in the batch reactor.

The chemisorption and electrochemical hydrogenation of hydroquinone (H2Q) at polycrystalline (pc) Pd, well-ordered Pd(100), and Pd-modified Au(hkl) electrodes were studied using a combination of ultra-high vacuum (UHV) surface spectroscopy, electrochemistry (EC), and electrochemical mass spectrometry (EC-MS). H2Q was found to form a slightly tilted flat-oriented quinone (Q) adlayer, when adsorbed from low concentrations; when chemisorbed from high concentrations, an edgewise-oriented H2Q adlayer was indicated. The hydrogenation of the chemisorbed layer is initiated at potentials before the onset of the hydrogen evolution region. As expected, the kinetics increases as the applied potential is increased, but the hydrogenation pathway appears to be independent of the potential. Hydrogenation in the absence of absorbed hydrogen (sub-surface) was studied at ultra-thin Pd films on Au single-crystal substrates. Hydrogenation and/or potential induced desorption were established, although non-volatile and/or hydrophobic products were detected. In comparison, negative excursions with benzene-coated electrodes resulted in nothing more than potential-induced desorption of the starting material. Negative-potential electro-desorption was more facile at terraces than at steps. Vibrational spectroscopic measurements suggested that hydrogenation occurs one molecule at a time to the fullest extent that resulted in desorption of product; that is, partially hydrogenated species do not exist on the surface.

The use of microporous materials with ordered porous structures as the hosts to stabilize metal nanoclusters has attracted particular interest in the catalysis because the pore size restriction could confine the growth of nanoclusters and lead to an increase in the percentage of catalytically active surface atoms. In this dissertation we report the preparation, characterization and the investigation of the catalytic performance of zeolite framework stabilized ruthenium(0) nanoclusters in the hydrolysis of sodium borohydride and the hydrogenation of aromatics. The zeolite framework stabilized ruthenium(0) nanoclusters were prepared by borohydride reduction of ruthenium(III)-exchanged zeolite-Y in aqueous solution at room temperature and isolated as black powders. Their characterization by using ICP-OES, XRD, TEM, ZC-TEM, HR-TEM, TEM-EDX, SEM, XPS, DR-UV-vis, far-IR, mid-IR, Raman spectroscopy, N2 adsorption-desorption technique and (P(C6H11)3)/(PC6H11O3) poisoning experiments reveal the formation of ruthenium(0) nanoclusters within the zeolite cages as well as on the external surface of zeolite without causing alteration in the framework lattice or loss in the crystallinity. The catalytic performance of zeolite framework stabilized ruthenium(0) nanoclusters depending on the various parameters was tested in the hydrolysis of sodium borohydride and the hydrogenation of aromatics. The important results obtained from these experiments can be listed as follows: (i) the zeolite framework stabilized ruthenium(0) nanoclusters provide a record total turnover number (103200 mol H2/mol Ru) and turnover frequency (33000 mol H2/mol Ru&bull / h) in the hydrolysis of sodium borohydride at room temperature, (ii) they also catalyze the same reaction in the basic medium (in 5 wt % NaOH solution) at room temperature with the unprecedented catalytic activity (4000 mol H2/mol Ru&bull / h) and lifetime (27200 mol H2/mol Ru), (iii) the isolated and vacuum dried samples of zeolite framework stabilized ruthenium(0) nanoclusters are active catalysts in the hydrogenation of cyclohexene, benzene, toluene and o-xylene in cyclohexane, they provide TOF values of 6150, 5660, 3200, and 1550 mol H2/mol Ru&bull / h, respectively under mild conditions (at 22.0 &plusmn / 0.1 &deg / C, and 40 &plusmn / 1 psig of initial H2 pressure), (iv) more importantly, the zeolite framework stabilized ruthenium(0) nanoclusters are the lowest temperature, most active, most selective (100 % selectivity with complete conversion) and longest lifetime catalyst hitherto known for the hydrogenation of benzene to cyclohexane in the solvent-free system (TTON of 2420 and TOF of 1040 mol benzene/mol Ru&bull / h) under mild conditions (at 22.0 &plusmn / 0.1 &deg / C, and 40 &plusmn / 1 psig of initial H2 pressure), (v) moreover, the resultant ruthenium(0) nanoclusters exhibit high durability throughout their catalytic use against agglomeration and leaching. This significant property makes them reusable catalyst without appreciable loss of their inherent activity.

Ethylene adsorption was studied by use of DFT/B3LYP with basis set 6-31G(d,p) in Gaussian&lsquo / 03 software. It was found that ethylene adsorbs molecularly on the Ni13 nanocluster with &amp / #960 / adsorption mode. &amp / #960 / adsorption mode is studied for the Ni10 (1 1 1), Ni13 (1 0 0) and Ni10 (1 1 0) surface cluster as well. Relative energy values were calculated as &amp / #8722 / 50.86 kcal/mol, &amp / #8722 / 20.48 kcal/mol, &amp / #8722 / 32.44 kcal/mol and &amp / #8722 / 39.27 kcal/mol for Ni13 nanocluster, Ni10 (1 1 1), Ni13 (1 0 0) and Ni10 (1 1 0) surface cluster models, respectively. Ethylene adsorption energy was found inversely proportional to Ni coordination number when Ni10 (1 1 1), Ni13 (1 0 0) and Ni10 (1 1 0) cluster models and Ni13 nanocluster were compared with each other. DFT/B3LYP and basis set of 86-411(41d)G in Gaussian&lsquo / 03 was used to investigate Ni55 nanocluster. Ethylene adsorption on Ni55 nanocluster was studied by means of equilibrium geometry calculations with &amp / #960 / adsorption modes for two different coordination numbers as 6 and 8. The related adsorption energies were approximately found as -22.07 and -14.82 kcal/mol for these coordination numbers of surfaces, respectively. In addition, the binding energies stated in literature that are for Ni2 dimer and Ni13 nanoclusters were considered together with our binding energy results for Ni55 nanocluster. Accordingly, when a correlation line was drawn and the intercept of binding energies was obtained against the value of &amp / #8213 / n&amp / #8722 / 1/3&amp / #8214 / where n is the number of atoms in the cluster / the result of interception gives a good estimation for bulk nickel binding energy at infinite &amp / #8213 / n&amp / #8214 / . This interception result was found as 4.58 eV/atom where the experimental value is reported as 4.45 eV/atom for bulk in the literature. Ehtylene hydrogenation mechanisms were also investigated in terms of the resultant geometries and total energy required for the related mechanism steps.