Co-reporter:Carole Brown, Adrian Lita, Yuchuan Tao, Nathan Peek, Mark Crosswhite, Melissa Mileham, J. Krzystek, Randall Achey, Riqiang Fu, Jasleen K. Bindra, Matthew Polinski, Youhong Wang, Lambertus J. van de Burgt, David Jeffcoat, Salvatore Profeta Jr., A. E. Stiegman, and Susannah L. Scott
ACS Catalysis November 3, 2017 Volume 7(Issue 11) pp:7442-7442
Publication Date(Web):September 14, 2017
DOI:10.1021/acscatal.7b02677
The structure and mechanism of the formation of sites which initiate ethylene polymerization in the atomically dispersed Phillips catalyst (Cr/SiO2) are two of the great unsolved mysteries of heterogeneous catalysis. After CO or C2H4 reduction of silica-supported CrVI ions to CrII ions in the precatalyst, exposure to ethylene results in the formation of organoCrIII sites that are capable of initiating polymerization without recourse to an external alkylating cocatalyst. In this work, a Phillips catalyst prepared, via sol–gel chemistry, as a mesoporous, optically transparent monolith was reduced with CO to the spectroscopically determined CrII end point. Ethylene causes rapid reoxidation of these CrII sites to CrIII, even at low temperatures. Solid-state 13C CP-MAS NMR, IR, and Raman spectroscopies reveal that the resulting sites contain a vinyl ligand, described as (≡SiO)2CrIII–CH═CH2 although likely with a higher coordination number, which are capable of initiating polymerization. The formation of these vinyl sites is an incommensurate redox reaction involving one-electron oxidation of CrII via ethylene disproportionation. The accompanying formation of organic radical intermediates and their characteristic reaction products suggest that the key step is homolysis of a Cr–ethyl bond. Plausible pathways for the initiation mechanism are suggested.Keywords: active site; bond homolysis; ethylene polymerization; initiation mechanism; Phillips catalyst; sol−gel;
Co-reporter:Long Qi, Ricardo Alamillo, William A. Elliott, Amity Andersen, David W. Hoyt, Eric D. Walter, Kee Sung Han, Nancy M. Washton, Robert M. Rioux, James A. Dumesic, and Susannah L. Scott
ACS Catalysis May 5, 2017 Volume 7(Issue 5) pp:3489-3489
Publication Date(Web):April 3, 2017
DOI:10.1021/acscatal.7b01045
In the liquid-phase catalytic processing of molecules using heterogeneous catalysts—an important strategy for obtaining renewable chemicals from biomass—many of the key reactions occur at solid–liquid interfaces. In particular, glucose isomerization occurs when glucose is adsorbed in the micropores of a zeolite catalyst. Since solvent molecules are coadsorbed, the catalytic activity depends strongly and often nonmonotonically on the solvent composition. For glucose isomerization catalyzed by NaX and NaY zeolites, there is an initial steep decline when water is mixed with a small amount of the organic cosolvent γ-valerolactone (GVL), followed by a recovery as the GVL content in the mixed solvent increases. Here we elucidate the origin of this complex solvent effect using operando solid-state NMR spectroscopy. The glucopyranose tautomers immobilized in the zeolite pores were observed and their transformations into fructose and mannose followed in real time. The microheterogeneity of the solvent system, manifested by a nonmonotonic trend in the mixing enthalpy, influences the mobility and adsorption behavior of the carbohydrates, water, and GVL, which were studied using pulsed-field gradient (PFG) NMR diffusivity measurements. At low GVL concentrations, glucose is depleted in the zeolite pores relative to the solution phase, and changes in the local structure of coadsorbed water serve to further suppress the isomerization rate. At higher GVL concentrations, this lower intrinsic reactivity is largely compensated by strong glucose partitioning into the pores, resulting in dramatic (up to 32×) enhancements in the local sugar concentration at the solid–liquid interface.Keywords: carbohydrate isomerization; interfacial reaction; operando spectroscopy; selective adsorption; solid-state NMR; solvent nonideality;
Co-reporter:Thomas Bligaard, R. Morris Bullock, Charles T. Campbell, Jingguang G. Chen, Bruce C. Gates, Raymond J. Gorte, Christopher W. Jones, William D. Jones, John R. Kitchin, and Susannah L. Scott
ACS Catalysis 2016 Volume 6(Issue 4) pp:2590
Publication Date(Web):March 7, 2016
DOI:10.1021/acscatal.6b00183
Benchmarking is a community-based and (preferably) community-driven activity involving consensus-based decisions on how to make reproducible, fair, and relevant assessments. In catalysis science, important catalyst performance metrics include activity, selectivity, and the deactivation profile, which enable comparisons between new and standard catalysts. Benchmarking also requires careful documentation, archiving, and sharing of methods and measurements, to ensure that the full value of research data can be realized. Beyond these goals, benchmarking presents unique opportunities to advance and accelerate understanding of complex reaction systems by combining and comparing experimental information from multiple, in situ and operando techniques with theoretical insights derived from calculations characterizing model systems. This Perspective describes the origins and uses of benchmarking and its applications in computational catalysis, heterogeneous catalysis, molecular catalysis, and electrocatalysis. It also discusses opportunities and challenges for future developments in these fields.Keywords: benchmarking; catalytic performance; computational catalysis; electrocatalysis; heterogeneous catalysis; molecular catalysis
Co-reporter:Anthony Fong, Baron Peters, and Susannah L. Scott
ACS Catalysis 2016 Volume 6(Issue 9) pp:6073
Publication Date(Web):July 29, 2016
DOI:10.1021/acscatal.6b01728
In ethylene polymerization by the Phillips catalyst, inorganic Cr(II) sites are believed to be activated by reaction with ethylene to form (alkyl)CrIII sites, in a process that takes about 1 h at ca. 373 K. The detailed mechanism of this spontaneous self-initiation has long remained unknown. It must account both for the formation of the first Cr–C bond and for the one-electron oxidation of Cr(II) to Cr(III). In this study, we used density functional theory to investigate a two-step initiation mechanism by which ethylene oxidative addition leads first to various (organo)CrIV sites, and subsequent Cr–C bond homolysis gives (organo)CrIII sites capable of polymerizing ethylene. Pathways involving spin crossing, C–H oxidative addition, H atom transfer, and Cr–C bond homolytic cleavage were explored using a chromasiloxane cluster model. In particular, we used classical variational transition theory to compute free energy barriers and estimate rates for bond homolysis. A viable route to a four-coordinate bis(alkyl)CrIV site was found via spin crossing in a bis(ethylene)CrII complex followed by intramolecular H atom transfer. However, the barrier for subsequent Cr–C bond homolysis is a formidable 209 kJ/mol. Increasing the Cr coordination number to 6 with additional siloxane ligands lowers the homolysis barrier to just 47 kJ/mol, similar to reported homolysis paths in molecular [CrR(H2O)53+] complexes. However, siloxane coordination also raises the barrier for the prior oxidative addition step to form the bis(alkyl)CrIV site. Thus, we suggest that hemilability in the silica “ligand” may facilitate the homolysis step while still allowing the oxidative addition of ethylene.Keywords: bond homolysis; chromium; density functional theory; ethylene polymerization; initiation mechanism; Phillips catalyst
Co-reporter:Thuy-Ai D. Nguyen, Zachary R. Jones, Domenick F. Leto, Guang Wu, Susannah L. Scott, and Trevor W. Hayton
Chemistry of Materials 2016 Volume 28(Issue 22) pp:8385
Publication Date(Web):October 23, 2016
DOI:10.1021/acs.chemmater.6b03879
The copper hydride nanocluster (NC) [Cu29Cl4H22(Ph2phen)12]Cl (2; Ph2phen = 4,7-diphenyl-1,10-phenanthroline) was isolated cleanly, and in good yields, by controlled growth from the smaller NC, [Cu25H22(PPh3)12]Cl (1), in the presence of Ph2phen and a chloride source at room temperature. Complex 2 was fully characterized by single-crystal X-ray diffraction, XANES, and XPS, and represents a rare example of an N* = 2 superatom. Its formation from 1 demonstrates that atomically precise copper clusters can be used as templates to generate larger NCs that retain the fundamental electronic and bonding properties of the original cluster. A time-resolved kinetic evaluation of the formation of 2 reveals that the mechanism of cluster growth is initiated by rapid ligand exchange. The slower extrusion of CuCl monomer, its transport, and subsequent capture by intact clusters resemble elementary steps in the reactant-assisted Ostwald ripening of metal nanoparticles.
Co-reporter:Bryan R. Goldsmith; Taeho Hwang; Stefan Seritan; Baron Peters
Journal of the American Chemical Society 2015 Volume 137(Issue 30) pp:9604-9616
Publication Date(Web):July 2, 2015
DOI:10.1021/jacs.5b03750
Olefin epoxidation catalyzed by methyltrioxorhenium (MTO, CH3ReO3) is strongly accelerated in the presence of H2O. The participation of H2O in each of the elementary steps of the catalytic cycle, involving the formation of the peroxo complexes (CH3ReO2(η2-O2), A, and CH3ReO(η2-O2)2(H2O), B), as well as in their subsequent epoxidation of cyclohexene, was examined in aqueous acetonitrile. Experimental measurements demonstrate that the epoxidation steps exhibit only weak [H2O] dependence, attributed by DFT calculations to hydrogen bonding between uncoordinated H2O and a peroxo ligand. The primary cause of the observed H2O acceleration is the strong co-catalytic effect of water on the rates at which A and B are regenerated and consequently on the relative abundances of the three interconverting Re-containing species at steady state. Proton transfer from weakly coordinated H2O2 to the oxo ligands of MTO and A, resulting in peroxo complex formation, is directly mediated by solvent H2O molecules. Computed activation parameters and kinetic isotope effects, in combination with proton-inventory experiments, suggest a proton shuttle involving one or (most favorably) two H2O molecules in the key ligand-exchange steps to form A and B from MTO and A, respectively.
Co-reporter:Thuy-Ai D. Nguyen; Zachary R. Jones; Bryan R. Goldsmith; William R. Buratto; Guang Wu; Susannah L. Scott;Trevor W. Hayton
Journal of the American Chemical Society 2015 Volume 137(Issue 41) pp:13319-13324
Publication Date(Web):September 30, 2015
DOI:10.1021/jacs.5b07574
Atomically precise copper nanoclusters (NCs) are of immense interest for a variety of applications, but have remained elusive. Herein, we report the isolation of a copper NC, [Cu25H22(PPh3)12]Cl (1), from the reaction of Cu(OAc) and CuCl with Ph2SiH2, in the presence of PPh3. Complex 1 has been fully characterized, including analysis by X-ray crystallography, XANES, and XPS. In the solid state, complex 1 is constructed around a Cu13 centered-icosahedron and formally features partial Cu(0) character. XANES of 1 reveals a Cu K-edge at 8979.6 eV, intermediate between the edge energies of Cu(0) and Cu(I), confirming our oxidation state assignment. This assignment is further corroborated by determination of the Auger parameter for 1, which also falls between those recorded for Cu(0) and Cu(I).
Co-reporter:Carole Brown, J. Krzystek, Randall Achey, Adrian Lita, Riqiang Fu, Robert W. Meulenberg, Matthew Polinski, Nathan Peek, Youhong Wang, Lambertus J. van de Burgt, Salvatore Profeta Jr., A. E. Stiegman, and Susannah L. Scott
ACS Catalysis 2015 Volume 5(Issue 9) pp:5574
Publication Date(Web):August 13, 2015
DOI:10.1021/acscatal.5b00927
The detailed mechanism by which ethylene polymerization is initiated by the inorganic Phillips catalyst (Cr/SiO2) without recourse to an alkylating cocatalyst remains one of the great unsolved mysteries of heterogeneous catalysis. Generation of the active catalyst starts with reduction of CrVI ions dispersed on silica. A lower oxidation state, generally accepted to be CrII, is required to activate ethylene to form an organoCr active site. In this work, a mesoporous, optically transparent monolith of CrVI/SiO2 was prepared using sol–gel chemistry in order to monitor the reduction process spectroscopically. Using in situ UV–vis spectroscopy, we observed a very clean, stepwise reduction by CO of CrVI first to CrIV, then to CrII. Both the intermediate and final states show XANES consistent with these oxidation state assignments, and aspects of their coordination environments were deduced from Raman and UV–vis spectroscopies. The intermediate CrIV sites are inactive toward ethylene at 80 °C. The CrII sites, which have long been postulated as the end point of CO reduction, were observed directly by high-frequency/high-field EPR spectroscopy. They react quantitatively with ethylene to generate the organoCrIII active sites, characterized by X-ray absorption and UV–vis spectroscopy, which initiate polymerization.Keywords: active site; ethylene polymerization; heterogeneous catalyst; in situ spectroscopy; Phillips catalyst; reaction mechanism; redox activation; sol−gel
Co-reporter:Anthony Fong, Ye Yuan, Sam L. Ivry, Susannah L. Scott, and Baron Peters
ACS Catalysis 2015 Volume 5(Issue 6) pp:3360
Publication Date(Web):April 9, 2015
DOI:10.1021/acscatal.5b00016
The mechanism of ethylene polymerization by the widely used Phillips catalyst remains controversial. In this work, we compare initiation, propagation, and termination pathways computationally using small chromasiloxane cluster models for several previously proposed and new mechanisms. Where possible, we consider complete catalytic cycles and compare predicted kinetics, active site abundances, and polymer molecular weights to known properties of the Phillips catalyst. Prohibitively high activation barriers for propagation rule out previously proposed chromacycle ring expansion and Green–Rooney (alternating alkylidene/chromacycle) mechanisms. A new oxachromacycle ring expansion mechanism has a plausible propagation barrier, but initiation is prohibitively slow. On sites with adjacent bridging hydroxyls, either ≡Si(OH)CrII-alkyl or ≡Si(OH)CrIII-alkyl, initiated by proton transfer from ethylene, chain growth by a Cossee–Arlman-type mechanism is fast. However, the initiation step is uphill and extremely slow, so essentially all sites remain trapped in a dormant state. In addition, these sites make only oligomers because when all pathways are considered, termination is faster than propagation. A monoalkylchromium(III) site without an adjacent proton, (≡SiO)2Cr-alkyl, is viable as an active site for polymerization, although its precise origin remains unknown.Keywords: chain termination; chromium; density functional theory; ethylene polymerization; initiation; Phillips catalyst
Co-reporter:Chenghao Deng, Xuezhi Duan, Jinghong Zhou, Xinggui Zhou, Weikang Yuan and Susannah L. Scott
Catalysis Science & Technology 2015 vol. 5(Issue 3) pp:1540-1547
Publication Date(Web):24 Nov 2014
DOI:10.1039/C4CY01285B
In this work, bimetallic Ir–Re catalysts supported on KIT-6 are prepared by tuning the thermal treatment procedures, i.e., conventional calcination and reduction (Ir–Re/KIT-6-CR) and modified direct reduction (Ir–Re/KIT-6-R) after impregnation of two metal precursors. The structure of both catalysts is intensively characterized by H2-TPR, STEM-HAADF-EDX, XPS and CO-DRIFTS. Results indicate that an Ir–Re alloy forms on the KIT-6 support when direct reduction is employed, which exhibits excellent catalytic performance in hydrogenolysis of glycerol. The formation rate of 1,3-propanediol over Ir–Re/KIT-6-R reaches 25.6 mol1,3-PD molIr−1 h−1 at 63% glycerol conversion with the addition of amberlyst-15 under 8 MPa H2, 393 K and 20 wt% glycerol aqueous solution, almost twice that over Ir–Re/KIT-6-CR. It is revealed that Re species without prior calcination treatment could be fully reduced and therefore couple with Ir to form an Ir–Re alloy structure with enhanced resistance against particle aggregation, while the calcination and subsequent reduction leads to the formation of an Ir–ReOx structure since the rhenium oxide species generated during the calcination is difficult to be reduced.
Co-reporter:Susannah L. Scott
Topics in Catalysis 2014 Volume 57( Issue 17-20) pp:1303-1305
Publication Date(Web):2014 November
DOI:10.1007/s11244-014-0307-3
Co-reporter:Taeho Hwang ; Bryan R. Goldsmith ; Baron Peters
Inorganic Chemistry 2013 Volume 52(Issue 24) pp:13904-13917
Publication Date(Web):November 21, 2013
DOI:10.1021/ic401343m
The formation of peroxorhenium complexes by activation of H2O2 is key in selective oxidation reactions catalyzed by CH3ReO3 (methyltrioxorhenium, MTO). Previous reports on the thermodynamics and kinetics of these reactions are inconsistent with each other and sometimes internally inconsistent. New experiments and calculations using density functional theory with the ωB97X-D and augmented def2-TZVP basis sets were conducted to better understand these reactions and to provide a strong experimental foundation for benchmarking computational studies involving MTO and its derivatives. Including solvation contributions to the free energies as well as tunneling corrections, we compute negative reaction enthalpies for each reaction and correctly predict the hydration state of all complexes in aqueous CH3CN. New rate constants for each of the forward and reverse reactions were both measured and computed as a function of temperature, providing a complete set of consistent activation parameters. New, independent measurements of equilibrium constants do not indicate strong cooperativity in peroxide ligand binding, as was previously reported. The free energy barriers for formation of both CH3ReO2(η2-O2) (A) and CH3ReO(η2-O2)2(H2O) (B) are predominantly entropic, and the former is much smaller than a previously reported value. Computed rate constants for a direct ligand-exchange mechanism, and for a mechanism in which a water molecule facilitates ligand-exchange via proton transfer in the transition state, differ by at least 7 orders of magnitude. The latter, water-assisted mechanism is predicted to be much faster and is consequently in much closer agreement with the experimentally measured kinetics. Experiments confirm the predicted catalytic role of water: the kinetics of both steps are strongly dependent on the water concentration, and water appears directly in the rate law.
Co-reporter:Mark H. Tucker, Anthony J. Crisci, Bethany N. Wigington, Neelay Phadke, Ricardo Alamillo, Jinping Zhang, Susannah L. Scott, and James A. Dumesic
ACS Catalysis 2012 Volume 2(Issue 9) pp:1865
Publication Date(Web):July 9, 2012
DOI:10.1021/cs300303v
The activity, selectivity, and stability of several supported acid catalysts were evaluated in tubular reactors designed to produce 5-hydroxymethylfurfural (HMF) continuously from fructose dissolved in a single-phase solution of THF and H2O (4:1 w/w). The reactors, packed with the solid catalysts, were operated at 403 K for extended periods, up to 190 h. The behaviors of three propylsulfonic acid-functionalized, ordered porous silicas (one inorganic SBA-15-type silica, and two ethane-bridged SBA-15-type organosilicas) were compared with that of a propylsulfonic acid-modified, nonordered, porous silica. The HMF selectivity of the catalysts with ordered pore structures ranged from 60 to 75%, whereas the selectivity of the nonordered catalyst under the same reaction conditions peaked at 20%. The latter was also the least stable, deactivating with a first-order rate constant of 0.152 h–1. The organosilicas are more hydrothermally stable and maintained a steady catalytic activity longer than the inorganic SBA-15-type silica. The organosilica with an intermediate framework ethane content of 45 mol % was more stable, with a first-order deactivation rate constant of only 0.012 h–1, than the organosilica containing 90 mol % ethane linkers in the framework. The catalysts were recovered and characterized after use by 13C and 29Si solid-state NMR, elemental analysis, nitrogen adsorption/desorption, X-ray diffraction, and SEM/TEM. Deactivation under flow conditions is caused primarily by hydrolytic cleavage of acid sites, which can be (to some) extent recaptured by the free surface hydroxyl groups of the silica surface.Keywords: 5-hydroxymethylfurfural; catalyst deactivation rate; continuous dehydration; packed-bed reactor; periodic mesoporous organosilicas; propylsulfonic acid; SBA-15;
Co-reporter:Bethany N. Wigington;Dr. Michael L. Drummond;Dr. Thomas R. Cundari;Dr. David L. Thorn;Dr. Susan K. Hanson;Dr. Susannah L. Scott
Chemistry - A European Journal 2012 Volume 18( Issue 47) pp:14981-14988
Publication Date(Web):
DOI:10.1002/chem.201202499
Abstract
The first step in the catalytic oxidation of alcohols by molecular O2, mediated by homogeneous vanadium(V) complexes [LVV(O)(OR)], is ligand exchange. The unusual mechanism of the subsequent intramolecular oxidation of benzyl alcoholate ligands in the 8-hydroxyquinolinato (HQ) complexes [(HQ)2VV(O)(OCH2C6H4-p-X)] involves intermolecular deprotonation. In the presence of triethylamine, complex 3 (X=H) reacts within an hour at room temperature to generate, quantitatively, [(HQ)2VIV(O)], benzaldehyde (0.5 equivalents), and benzyl alcohol (0.5 equivalents). The base plays a key role in the reaction: in its absence, less than 12 % conversion was observed after 72 hours. The reaction is first order in both 3 and NEt3, with activation parameters ΔH≠=(28±4) kJ mol−1 and ΔS≠=(−169±4) J K−1 mol−1. A large kinetic isotope effect, 10.2±0.6, was observed when the benzylic hydrogen atoms were replaced by deuterium atoms. The effect of the para substituent of the benzyl alcoholate ligand on the reaction rate was investigated using a Hammett plot, which was constructed using σp. From the slope of the Hammett plot, ρ=+(1.34±0.18), a significant buildup of negative charge on the benzylic carbon atom in the transition state is inferred. These experimental findings, in combination with computational studies, support an unusual bimolecular pathway for the intramolecular redox reaction, in which the rate-limiting step is deprotonation at the benzylic position. This mechanism, that is, base-assisted dehydrogenation (BAD), represents a biomimetic pathway for transition-metal-mediated alcohol oxidations, differing from the previously identified hydride-transfer and radical pathways. It suggests a new way to enhance the activity and selectivity of vanadium catalysts in a wide range of redox reactions, through control of the outer coordination sphere.
Co-reporter:Trenton M. Tovar;S. Michael Stewart
Topics in Catalysis 2012 Volume 55( Issue 7-10) pp:530-537
Publication Date(Web):2012 July
DOI:10.1007/s11244-012-9831-1
The origin of the promoting effect of ZnCl2 on CH3ReO3 supported on γ-Al2O3, resulting in a dramatic increase in its olefin metathesis activity, was investigated by elemental analysis, IR, NMR, EXAFS, and active site counting. Dispersed ZnCl2 enables the formation of more numerous Re active sites, which themselves have higher per-site activities. These sites also possess Re–Cl interactions.
Co-reporter:Ying-Jen Wanglee ; Jerry Hu ; Rosemary E. White ; Ming-Yung Lee ; S. Michael Stewart ; Philippe Perrotin
Journal of the American Chemical Society 2011 Volume 134(Issue 1) pp:355-366
Publication Date(Web):November 22, 2011
DOI:10.1021/ja207838j
A supported, single-site Lewis acid, ≡SiOB(C6F5)2, was prepared by water-catalyzed grafting of B(C6F5)3 onto the surface of amorphous silica, and its subsequent use as a cocatalyst for heterogeneous olefin polymerization was explored. Although B(C6F5)3 has been reported to be unreactive toward silica in the absence of a Brønsted base, we find that it can be grafted even at room temperature, albeit slowly. The mechanism was investigated by 1H and 19F NMR, in both the solution and solid states. In the presence of a trace amount of H2O, either added intentionally or formed in situ by borane-induced dehydration of silanol pairs, the adduct (C6F5)3B·OH2 hydrolyzes to afford C6F5H and (C6F5)2BOH. The latter reacts with the surface hydroxyl groups of silica to yield ≡SiOB(C6F5)2 sites and regenerate H2O. When B(C6F5)3 is present in excess, the resulting grafted boranes appear to be completely dry, due to the eventual formation of [(C6F5)2B]2O. The immobilized, tri-coordinate Lewis acid sites were characterized by solid-state 11B and 19F NMR, IR, elemental analysis, and C5H5N-TPD. Their ability to activate two molecular C2H4 polymerization catalysts, Cp2ZrMe2 and an (α-iminocarboxamidato)nickel(II) complex, was explored.
Co-reporter:Samuel D. Fleischman
Journal of the American Chemical Society 2011 Volume 133(Issue 13) pp:4847-4855
Publication Date(Web):February 22, 2011
DOI:10.1021/ja108905p
Molecules encountering silica interfaces interact primarily with the hydroxyl groups that terminate the bulk structure. When the nominal surface density is very low, these “silanols” are presumed to be isolated. Nevertheless, silicas that are highly dehydroxylated by pretreatment at 800 °C react with Ga(CH3)3 at room temperature to give primarily disilanolate-bridged digallium sites, [(CH3)2Ga(μ-OSi≡)]2. The EXAFS at the Ga K-edge shows a prominent Ga−Ga scattering path, regardless of whether an excess or a limiting amount of Ga(CH3)3 is used. Some dimers are formed by the concerted reaction of Ga(CH3)3 with an “isolated” silanol and an adjacent siloxane bond. These grafting sites are proposed to be hydroxyl-substituted 2-rings, formed by condensation within a vicinal Q2−Q3 pair. Other dimers are formed by reaction of Ga(CH3)3 with vicinal Q3−Q3 pairs which have not condensed, even at 800 °C. In a computational model for the dimer sites, the O−O distance is <2.6 Å, which is far shorter than the calculated mean interhydroxyl separation for the thermally treated silicas (12.2 Å). This highly nonrandom distribution of surface silanols, in combination with the coupled reaction of “isolated” silanols and strained siloxane bonds, accounts for the preferential formation of grafted site pairs rather than isolated grafted sites when silica surfaces are chemically modified.
Co-reporter:Anthony J. Crisci, Mark H. Tucker, Ming-Yung Lee, Se Gyu Jang, James A. Dumesic, and Susannah L. Scott
ACS Catalysis 2011 Volume 1(Issue 7) pp:719
Publication Date(Web):May 5, 2011
DOI:10.1021/cs2001237
A bifunctional silane, 3-((3-(trimethoxysilyl)propyl)thio)propane-1-sulfonic acid (TESAS), was designed for incorporation into SBA-15-type silica by co-condensation. To achieve mesopore ordering in the functionalized silica, the standard SBA-15 synthetic protocol was modified, resulting in well-formed hexagonal particles. In a variation, the thioether group of TESAS was oxidized by H2O2 to the sulfone during the synthesis of the modified SBA-15. The presence, concentration, and location of functional groups were assessed by elemental analysis, potentiometric titration, and 13C and 29Si CP/MAS NMR spectroscopy. Pore structure and particle morphology were observed by scanning electron microscopy (SEM), transmission electron microscopy (TEM), and powder X-ray diffraction (XRD). The materials were studied and compared in the selective dehydration of fructose to 5-hydroxymethylfurfural (HMF). Interestingly, the thioether-containing TESAS-SBA-15 shows higher activity in the dehydration of aqueous fructose, as well as higher selectivity toward HMF (71% at 84% conversion) than its sulfone derivative, possibly because of its more hydrophobic nature.Keywords: 5-hydroxymethylfurfural; fructose dehydration; particle morphology; SBA-15; solid acid catalyst; solid-state NMR
Co-reporter:Philippe Perrotin, Jenny S. J. McCahill, Guang Wu and Susannah L. Scott
Chemical Communications 2011 vol. 47(Issue 24) pp:6948-6950
Publication Date(Web):10 May 2011
DOI:10.1039/C1CC00095K
A well-defined, homogeneous catalyst, [(Ph)(2-(2′,6′-(OMe)2-C6H3)-C6H4)P(2-SO3-C6H4)]Ni(Ph)PPh3, in which a single, bulky ortho-biphenyl substituent on the chelating phosphine blocks one axial position, is very active for formation of linear polyethylene (Mn = 403000 g mol−1, Mw/Mn = 1.87).
Co-reporter:Joshua A. Kurzman, Jun Li, Thomas D. Schladt, César R. Parra, Xiaoying Ouyang, Ryan Davis, Jeffrey T. Miller, Susannah L. Scott, and Ram Seshadri
Inorganic Chemistry 2011 Volume 50(Issue 17) pp:8073-8084
Publication Date(Web):July 29, 2011
DOI:10.1021/ic200455a
Complex oxides—containing at least two different cations on crystallographically distinct sites—have recently been shown to display redox cycling of platinum group metals (PGMs), such as Pd; for example, Pd-substituted complex oxides can reversibly extrude metallic Pd under reducing conditions and then reincorporate Pd2+ ions into the lattice under oxidizing conditions. The title compounds, YMn0.5Fe0.5–xPdxO3−δ (0 ≤ x ≤ 0.07) crystallizing in the noncentrosymmetric YMnO3 structure, were prepared using a sol–gel process at 800 °C, and the structures were refined from high-resolution synchrotron X-ray powder diffraction data. Their redox cycling behavior was monitored using synchrotron X-ray diffraction and EXAFS studies. In contrast to the previously studied complex oxide host compounds, YMn0.5Fe0.5–xPdxO3−δ is only modestly tolerant to cycling: repeated redox cycling leads to the formation of PdO, which, on the time-scale of the oxidation cycles, does not reincorporate in the complex oxide lattice. Both oxidized and reduced samples were tested for the oxidation of CO to CO2 under CO-lean conditions. YMn0.5Fe0.5–xPdxO3−δ performs essentially as well as previously studied YFe1–xPdxO3−δ. The CO oxidation light-off characteristics of the hexagonal hosts are very similar to finely dispersed PdO. Despite evidence that Pd is almost fully dispersed as divalent ions in the host lattice, which is presumably accompanied by the concurrent creation of oxygen vacancies (2 Pd2+:1 VO2–), the as-prepared hexagonal materials do not display any significant improvement in catalytic activity as a function of Pd substitution level. This suggests that the corner-connected trigonal bipyramids that characterize this structural family do not enable the transport of oxygen through the bulk of the lattice. The study casts light on factors in the solid-state chemistry of precious metal-substituted complex oxides that influence the efficacy of redox cycling of the precious metal, and catalytic performance.
Co-reporter:Brian C. Vicente, Zheng Huang, Maurice Brookhart, Alan S. Goldman and Susannah L. Scott
Dalton Transactions 2011 vol. 40(Issue 16) pp:4268-4274
Publication Date(Web):08 Mar 2011
DOI:10.1039/C0DT01369B
When a pincer-ligated iridium complex with a phosphinite substituent in the para-position of the aromatic backbone is immobilized on γ-alumina, it becomes a highly effective supported catalyst for the transfer-dehydrogenation of alkanes. The nature of the interaction between the organometallic complex and the support was investigated using solid-state 31P MAS NMR spectroscopy, solution-state 1H and 31P{1H} NMR spectroscopy, IR and GC/MS analysis of extracted reaction products. The phosphinite substituent is cleaved from the pincer ligand by its reaction with hydroxyl groups on the γ-alumina surface, resulting in covalent anchoring of the complex via the aryl ring. A similar reaction occurs on silica, allowing for ready grafting onto this support as well. A strategy for anchoring homogeneous catalysts on hydroxyl-terminated oxide supports though the selective cleavage of [POR]-containing ligand substituents is suggested.
Co-reporter:Bryanna M. Kunkel;Brian C. Peoples;Cathleen M. Yung
Macromolecular Materials and Engineering 2011 Volume 296( Issue 12) pp:1075-1080
Publication Date(Web):
DOI:10.1002/mame.201100124
Co-reporter:Ming-Yung Lee
Chemistry - A European Journal 2011 Volume 17( Issue 16) pp:4632-4639
Publication Date(Web):
DOI:10.1002/chem.201002576
Abstract
MgCl2-modified silica is an important component of some Ziegler–Natta catalysts used in the manufacture of polyethylene. Information about the structure of the dispersed magnesium sites formed by the reaction of di-n-butylmagnesium (nBu2Mg) with silica was sought to provide a basis for understanding their subsequent interactions with transition-metal or co-catalyst components. From infrared spectra and elemental analysis, we deduced that nBu2Mg reacts with porous silica in two ways: about half (47 %, 0.99 mmol g−1) is grafted through protonolysis by surface hydroxyl groups (SiOH), whereas the other half (53 %, 1.11 mmol g−1) reacts directly with siloxane bonds (SiOSi). In the 29Si and 13C CP/MAS NMR spectra of Sylopol-2100 silica pretreated at 500 °C then modified with nBu2Mg at room temperature, both alkylsilicon and alkylmagnesium sites are evident. The alkylmagnesium-modified silica surface is proposed to contain dimers and/or tetramers with the empirical formula [SiOMg(nBu)]n. Upon exposure of nBu2Mg-modified silica to anhydrous HCl, alkanes are liberated, hydroxyl groups are regenerated, and water is formed. The appearance of water suggests condensation of hydroxyl group pairs, induced by the coordinatively unsaturated nanoclusters (MgCl2)n that arise by ligand exchange on the silica-supported n-butylmagnesium oligomers.
Co-reporter:Brian C. Vicente ; Ryan C. Nelson ; Anthony W. Moses ; Swarup Chattopadhyay
The Journal of Physical Chemistry C 2011 Volume 115(Issue 18) pp:9012-9024
Publication Date(Web):April 14, 2011
DOI:10.1021/jp109929g
In oxide-supported perrhenate catalysts, isolated Re(VII) becomes anchored to the support at surface hydroxyl sites. However, oxides that also possess Lewis acidity offer the possibility of additional bonding interactions with perrhenate, which may account for their enhanced activity, for example, in olefin metathesis. Evidence for such interactions was sought by X-ray absorption spectroscopy at the Re L1- and L3-edges, in combination with DFT modeling. Only perrhenate grafted onto silica using an anhydrous method shows a simple ≡SiOReO3 structure, with three equivalent Re═O multiple bonds. Silica-supported perrhenate prepared by aqueous impregnation is not grafted, but interacts only weakly with the silica support as a tetrahedral ReO4– ion. On silica–alumina and γ-alumina, the anchored perrhenates also interact with adjacent Lewis acid sites. In particular, a terminal oxo ligand interacts with a coordinatively unsaturated Al atom, creating a well-defined Re–Al EXAFS path. In addition, an oxygen atom from the support is coordinated to Re, increasing the metal coordination number from four to five. Simulated vibrational spectra of DFT-optimized models indicate that these pentacoordinate Re sites are consistent with reported Raman spectra for perrhenate supported on γ-alumina.
Co-reporter:Haian Xia, Samuel D. Fleischman, Can Li, and Susannah L. Scott
The Journal of Physical Chemistry Letters 2011 Volume 2(Issue 3) pp:190-195
Publication Date(Web):January 11, 2011
DOI:10.1021/jz101507s
The effect of introducing extra-framework Ga on the local structure of the metal sites in Fe/ZSM-5, resulting in enhanced reactivity toward N2O, was investigated using a combination of Raman and X-ray absorption spectroscopies. The Raman spectra indicate an increased abundance of oxo- and/or hydroxo-bridged diiron sites, whereas the Fe K-edge XANES reveals more extensive reduction of Fe(III) to Fe(II). Curvefits of the EXAFS at both the Ga and Fe K-edges are consistent with heterometallic oxo-clusters containing both Ga−Fe and Fe−Fe paths. The spectroscopic evidence suggests a tetranuclear [Fe2Ga2O42+] core, possessing an open dicubane structure.Keywords (keywords): mixed metal oxide clusters; N2O decomposition; Raman spectroscopy; X-ray absorption; zeolite catalysts;
Co-reporter:Joshua A. Kurzman ; Xiaoying Ouyang ; Won Bin Im ; Jun Li ; Jerry Hu ; Susannah L. Scott ;Ram Seshadri
Inorganic Chemistry 2010 Volume 49(Issue 10) pp:4670-4680
Publication Date(Web):April 15, 2010
DOI:10.1021/ic100486g
La4LiAuO8 and La2BaPdO5, two previously known oxides, are presented as model compounds for examining the role of isolated and immobilized Au3+ and Pd2+ ions in heterogeneous catalysis. Structural characterization, stability, surface composition, and electronic structure of these compounds are presented. These are examined in studies ranging from synchrotron X-ray scattering, including pair distribution function (PDF) and maximum entropy method (MEM) analysis, to density functional calculations of the electronic structures. The exceptional stability displayed by these compounds as verified by thermogravimetric analysis can be attributed to the presence of covalent Au−O and Pd−O interactions revealed in MEM studies, which suggests a criterion for stabilizing these highly oxophobic transition metals in oxide environments. Catalytic testing of the two compounds as heterogeneous catalysts in the oxidation of CO to CO2 are presented. La2BaPdO5 appears to be an effective catalyst for CO oxidation, despite the low surface area of the oxide being used. This is the first time that a fully ordered (rather than doped) Pd2+ oxide had been used to catalyze CO oxidation. La4LiAuO8 on the other hand, is much less effective at catalyzing CO oxidation. Differences in the reactivities of the two compounds are discussed with respect to differences in their density functional electronic structures.
Co-reporter:Robert O. Savinelli and Susannah L. Scott
Physical Chemistry Chemical Physics 2010 vol. 12(Issue 21) pp:5660-5667
Publication Date(Web):29 Apr 2010
DOI:10.1039/B926474D
The local structure of molybdate ions in the crystalline materials MgMoO4 and MgMo2O7 and in Mo/HZSM-5 (2 wt% Mo, Si/Al = 15) was explored by EXAFS at the Mo K-edge, using both Fourier and wavelet transforms. For the two model compounds, curve fitting analysis of the FT-EXAFS reveals the need to include several single- and multiple-scattering paths in order to locate the Mo–Mo paths accurately. Some of these overlapping paths can be identified using WT-EXAFS analysis. Indeed, the positions of WT maxima can be predicted fairly accurately, allowing for qualitative identification of the paths required for curve fitting in samples of unknown structure and providing additional criteria for distinguishing statistically similar curve fits. The application of the WT analysis technique to the heterogeneous catalyst Mo/HZSM-5 clearly shows the presence of dimolybdate ions, although with structures perturbed by their interactions with the zeolite framework.
Co-reporter:Miyako Hisamoto, Ryan C. Nelson, Ming-Yung Lee, Juergen Eckert and Susannah L. Scott
The Journal of Physical Chemistry C 2009 Volume 113(Issue 20) pp:8794-8805
Publication Date(Web):April 28, 2009
DOI:10.1021/jp810057v
The deposition of volatile cis-(CH3)2Au(O,O′-acac) onto silica partially dehydroxylated at 400 °C leads to molecular dispersion of the organogold complex. X-ray absorption near-edge structure demonstrates that the gold retains its oxidation state upon binding. IR and 1H magic-angle spinning NMR spectra suggest that the molecular framework also remains intact. Computational models involving hydroxyl-terminated octasilsesquioxane cube clusters to represent silica predict the most energetically favorable interaction to be hydrogen-bonding between two surface hydroxyls and the oxygen donor atoms of the chelated acetylacetonate ligand. This mode of adsorption was confirmed by analysis of the IR and the Au LIII edge extended X-ray absorption fine structure. The surface hydroxyls most likely to participate in the attachment of the gold complex to silica are vicinal silanol pairs that are not involved in hydrogen bonding to other silanols.
Co-reporter:Miyako Hisamoto;Swarup Chattopadhyay
Journal of Chemical Crystallography 2009 Volume 39( Issue 3) pp:
Publication Date(Web):2009 March
DOI:10.1007/s10870-008-9449-0
The structure of cis-(CH3)2Au(O,O′-acac) was investigated by single-crystal X-ray diffraction. It crystallizes in the orthorhombic space group Cmca with unit cell dimensions a = 6.7831(13) Å, b = 15.080(3) Å, c = 18.073(3) Å and α = β = γ = 90°. The molecule is planar but lacks a C2 axis in the solid-state, with two Au–C bond lengths that differ by 0.078(21) Å due to crystal packing effects. The inequivalence of the crystallographic carbon positions is evident in the solid-state 13C CP-MAS NMR spectrum, which was recorded at low temperature due to the thermal sensitivity of the compound during magic-angle spinning. Head-to-tail stacking, which creates short intermolecular paths, was reproduced computationally in a dimer model.Solid-state Spectroscopic and Structural Investigation ofcis-(CH3)Au2(O,O′-acac)In the solid state, cis-(CH3)2Au(O,O′-acac) has no C2 axis, two different Au–C bond lengths (1.960, 2.038 Å; Δd = 0.078(21) Å) and a close Au–C…C–Au contact; these features give rise to additional features in the 13C CP-MAS NMR spectrum.
Co-reporter:Jun Li, Udayshankar G. Singh, Thomas D. Schladt, Judith K. Stalick, Susannah L. Scott and Ram Seshadri
Chemistry of Materials 2008 Volume 20(Issue 20) pp:6567
Publication Date(Web):September 25, 2008
DOI:10.1021/cm801534a
Metastable YFeO3 with the hexagonal YAlO3 structure was obtained by a sol−gel process at 700 °C, using metal nitrate precursors with pH control and the appropriate citric acid to nitrate ratio. Under similar conditions, YFe1−xPdxO3−δ (0 < x ≤ 0.1) compositions were also prepared. The substitution of Fe by Pd stabilizes the YAlO3 structure at higher temperatures. The crystal structures of YFe1−xPdxO3−δ (0 ≤ x ≤ 0.1) were refined by Rietveld analysis of X-ray and neutron powder diffraction data. The parent hexagonal YFeO3 (x = 0) crystallizes in the space group P63/mmc with a = 3.5099(3) Å and c = 11.759(2) Å. The redox-driven mobility of Pd to integrate into the oxide host as ions and to dissociate from it as fcc-Pd nanoparticles was monitored by a combination of X-ray diffraction and X-ray photoelectron spectroscopy. Pd nanoparticles in the reduced samples were detected by scanning backscattered electron microscopy and transmission electron microscopy. The Pd2+-containing materials showed significant low-temperature (near 100 °C) catalytic activity for CO oxidation, comparable to that of highly dispersed PdO/Al2O3, despite their relatively low surface areas.
Co-reporter:Ritu Ahuja, Sabuj Kundu, Alan S. Goldman, Maurice Brookhart, Brian C. Vicente and Susannah L. Scott
Chemical Communications 2008 (Issue 2) pp:253-255
Publication Date(Web):01 Nov 2007
DOI:10.1039/B712197K
Tandem dehydrogenation–olefin-metathesis catalyst systems, comprising a pincer-ligated iridium-based alkane dehydrogenation catalyst and a molybdenum-based olefin-metathesis catalyst, are reported to effect the metathesis-cyclooligomerization of cyclooctane and cyclodecane to give cycloalkanes with various carbon numbers, predominantly multiples of the substrate carbon number, and polymers.
Co-reporter:Susannah L. Scott, Anqiu Fu, Leonard A. MacAdams
Inorganica Chimica Acta 2008 Volume 361(Issue 11) pp:3315-3321
Publication Date(Web):27 July 2008
DOI:10.1016/j.ica.2008.01.012
The grafting of tetraneopentylchromium(IV) onto the surface hydroxyls of silica has been investigated by varying both the method of deposition and the nature of the silica support. Reaction of the volatile organometallic with the silica surface in vacuo is compared with reaction in solution. A faster reaction but a lower ultimate Cr loading was obtained using the solution deposition technique. The reactions of the organometallic compound with Aerosil 200, a fumed silica often used to model catalyst supports, and Sylopol 952, a silica gel used as a carrier for chromium-based ethylene polymerization catalysts, are compared. After thermal pretreatment at 200 °C, grafting on both supports yields bis(neopentyl)chromium(IV) fragments regardless of the grafting method used, suggesting that a paired arrangement of the surface hydroxyl groups exists on both types of silicas. The higher Cr loading achieved on the silica gel is attributed to its higher surface area. Thermally-induced neopentane elimination from the grafted bis(neopentyl)chromium(IV) fragments occurs at the same rate and with the same stoichiometry for both Aerosil- and Sylopol-supported materials. Consequently, interactions of the grafted organometallic fragments with nearby siloxanes appear to be unimportant in the early stages of the transformation of bis(alkyl)chromium(IV) to the alkylidene.Grafting of Cr(CH2C(CH3)3)4 onto silicas of different origin (fumed versus gel), prepared as powders or as self-supporting disks, results in the same bis(neopentyl)chromium(IV) fragment that undergoes α-H elimination to neopentylidenechromium(IV). However, much less Cr is fixed on the silica when the grafting occurs from solution compared to the vapor phase.
Co-reporter:Katharine Page, Jun Li, Robert Savinelli, Holly N. Szumila, Jinping Zhang, Judith K. Stalick, Thomas Proffen, Susannah L. Scott, Ram Seshadri
Solid State Sciences 2008 Volume 10(Issue 11) pp:1499-1510
Publication Date(Web):November 2008
DOI:10.1016/j.solidstatesciences.2008.03.018
As possible substitute materials for platinum group metal heterogeneous catalysts, high surface area carbides of the early transition metals Mo and W are of great interest. Here we report nanostructured, high surface area Mo2C and WC prepared by decomposing and carburizing ammonium paramolybdate [(NH4)6Mo7O24·4H2O] and ammonium paratungstate [(NH4)10W12O41·5H2O] in flowing 50%CH4/50%H2. Surface areas as high as 52 m2/g for Mo2C and 24 m2/g for WC were obtained, with both structures crystallizing in structures appropriate for catalytic activity. We have studied these materials using a combination of neutron diffraction Rietveld refinement, X-ray photoelectron spectroscopy, surface area measurements, and scanning transmission electron microscopy. In addition, we have used pair-distribution function (PDF) analysis of the neutron total scattering data as a means of establishing the presence of graphitic carbon in the as-prepared materials.
Co-reporter:Miyako Hisamoto, Susannah L. Scott
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 2008 Volume 71(Issue 3) pp:969-974
Publication Date(Web):1 December 2008
DOI:10.1016/j.saa.2008.02.020
The IR spectrum of cis-(CH3)2Au(O,O′-acac) has been reassigned by comparing frequencies for cis-(CH3)2Au(O,O′-acac) and cis-(CD3)2Au(O,O′-acac), and by analysis of the DFT-calculated normal modes and their frequencies for the isolated molecules. The vibrational intensity in the C–H stretching region arises almost entirely from the cis-(CH3)2Au fragment, while the methyl deformation intensity is largely of acetylacetonato ligand origin. A low frequency mode in the C–H stretching region is the first overtone of the δa(CH3) mode of cis-(CH3)2Au. The Au–C stretching modes are affected by deuteration of the cis-(CH3)2Au fragment, while the Au–O stretching modes are not.
Co-reporter:Lei Zhong, Ming-Yung Lee, Zhen Liu, Ying-Jen Wanglee, Boping Liu, Susannah L. Scott
Journal of Catalysis (September 2012) Volume 293() pp:1-12
Publication Date(Web):1 September 2012
DOI:10.1016/j.jcat.2012.05.014
Compositionally and structurally uniform chromate sites grafted onto silica via the ambient temperature reaction of CrO2Cl2 with silica pretreated at either 500 or 800 °C, followed by mild heating at 300 °C, were reduced by CO to the corresponding Cr(II) sites. The resulting structures were investigated by IR spectroscopy of adsorbed CO as well as X-ray absorption spectroscopy and compared with model structures predicted by ONIOM calculations (B3PW91*/BS1: B3PW91*/STO-3G). The IR spectra suggest that there are two distinct and non-interconverting Cr(II) sites whose relative amounts differ on the two types of pretreated silica. The calculated υCO frequencies for two model structures, (SiO)2Cr(CO)2 and (SiO)2(Si2O)Cr(CO), agree well with experimental data, although these structures do not necessarily reflect the coordination of Cr(II) in the absence of CO. The X-ray absorption spectra reveal that the CO-free Cr(II) sites are coordinated by additional siloxane ligands, which may influence their ability to initiate ethylene polymerization.Graphical abstractDownload high-res image (103KB)Download full-size imageHighlights► Cr(II)/SiO2 sites were prepared by the CO reduction of CrO2Cl2-modified silica. ► Variable siloxane coordination to Cr is evident in the EXAFS. ► Sites with a non-displaceable siloxane appear to be the direct precursors of the ethylene polymerization sites.
Co-reporter:Xiaoying Ouyang, Susannah L. Scott
Journal of Catalysis (28 July 2010) Volume 273(Issue 2) pp:83-91
Publication Date(Web):28 July 2010
DOI:10.1016/j.jcat.2010.04.020
Ba(Ce,Pd)O3 perovskite is an unusual catalyst for CO oxidation by O2, given its low surface area. The rate law for the catalytic reaction was evaluated under CO-lean and CO-rich conditions, as well as near-stoichiometric conditions. When O2 is present in excess, the kinetics show CO inhibition, consistent with a Langmuir–Hinshelwood mechanism in which both reactants compete for the same adsorption sites. The Arrhenius activation energy for this mechanism is surprisingly low, (7.8 ± 0.3) kcal/mol. It is attributed to weak adsorption of CO on ionic surface Pd(II) sites. When O2 is limiting, the reaction orders for both CO and O2 show a strong dependence on P(CO)/P(O2), and eventually become independent of both P(CO) and P(O2) at high P(CO). This suggests a new BaCeO3-mediated mechanism which dominates the reaction at high P(CO)/P(O2). Its Arrhenius activation energy is (9.7 ± 0.3) kcal/mol, probably reflecting the barrier to oxygen migration in the oxide phase. Both mechanisms contribute significantly for P(CO)/P(O2) ≈ 1. The catalytic activity of Pd-substituted BaCeO3 is attributed to the increased bulk oxygen mobility in the presence of square-planar Pd(II) ions that are located on the perovskite B-sites, each adjacent to an oxygen vacancy.Low surface area, crystalline Ba(Ce,Pd)O3 catalyzes CO oxidation using either oxygen adsorbed from the gas phase or labile lattice oxygen, and the rate law depends strongly on the P(CO)/P(O2) ratio. Catalytic activity is attributed to the presence of square-planar Pd(II) ions located on the perovskite B-sites, each adjacent to an oxygen vacancy.Download high-res image (159KB)Download full-size image
Co-reporter:Udayshankar G. Singh, Jun Li, Joseph W. Bennett, Andrew M. Rappe, Ram Seshadri, Susannah L. Scott
Journal of Catalysis (25 July 2007) Volume 249(Issue 2) pp:349-358
Publication Date(Web):25 July 2007
DOI:10.1016/j.jcat.2007.04.023
Perovskite BaCeO3 materials with low levels of substitution of Pd(II) on the Ce site and a corresponding number of oxygen vacancies were prepared by a high-temperature synthesis method. Although their surface areas are low (ca. 1.0 m2 g−1), their low-temperature (<200 °C) activity for CO oxidation is comparable to that of highly dispersed PdO/Al2O3. When the doped perovskites are reduced extensively in H2, causing extrusion of Pd(0) from the lattice, their catalytic activity declines dramatically. Consequently, activity is attributed to the presence of cationic Pd(II) in the perovskite lattice. Density functional theory was used to investigate the atomic and electronic character of the structures containing oxygen vacancies. Both experimental and theoretical evidence support a catalytic mechanism involving labilization of lattice and surface oxygen by cationic Pd(II).
Co-reporter:Cori A. Demmelmaier, Rosemary E. White, Jeroen A. van Bokhoven, Susannah L. Scott
Journal of Catalysis (15 February 2009) Volume 262(Issue 1) pp:44-56
Publication Date(Web):15 February 2009
DOI:10.1016/j.jcat.2008.11.024
The ambient temperature reaction of CrO2Cl2 with silica, followed by mild heating to induce formation of uniform grafted silylchromate diesters, was studied as a function of the silica pretreatment temperature. The reactivity of the resulting chromate sites toward ethylene is qualitatively different: those formed on the silica pretreated at 200 °C are incapable of initiating polymerization, while those formed on silicas pretreated at 450 and 800 °C spontaneously induce polymerization with kinetic profiles closely resembling that of the calcined Phillips catalyst (CrOx/SiO2). Comparison of their X-ray absorption spectra suggests subtle differences in the chromate–support interactions, which can be interpreted in terms of changes in the chromasiloxane ring size distribution. The unstrained sites favored on the highly hydroxylated silica surface are consistent with 8-membered chromasiloxane rings formed by attachment of the CrO2 fragment to non-vicinal hydroxyls, while the strained sites on highly dehydroxylated silica surfaces are suggested to be 6-membered chromasiloxane rings created from vicinal hydroxyls located on adjacent silicon atoms. Simple computational models for these sites predict changes in the vibrational spectra and the XANES that are consistent with experimental observations.XANES suggests that the uniform chromate sites formed by grafting chromyl chloride onto silicas pretreated at low and high temperatures differ in the size of their chromasiloxane rings, and in the ability of these sites to initiate ethylene polymerization.Download high-res image (73KB)Download full-size image
Co-reporter:Robert O. Savinelli and Susannah L. Scott
Physical Chemistry Chemical Physics 2010 - vol. 12(Issue 21) pp:NaN5667-5667
Publication Date(Web):2010/04/29
DOI:10.1039/B926474D
The local structure of molybdate ions in the crystalline materials MgMoO4 and MgMo2O7 and in Mo/HZSM-5 (2 wt% Mo, Si/Al = 15) was explored by EXAFS at the Mo K-edge, using both Fourier and wavelet transforms. For the two model compounds, curve fitting analysis of the FT-EXAFS reveals the need to include several single- and multiple-scattering paths in order to locate the Mo–Mo paths accurately. Some of these overlapping paths can be identified using WT-EXAFS analysis. Indeed, the positions of WT maxima can be predicted fairly accurately, allowing for qualitative identification of the paths required for curve fitting in samples of unknown structure and providing additional criteria for distinguishing statistically similar curve fits. The application of the WT analysis technique to the heterogeneous catalyst Mo/HZSM-5 clearly shows the presence of dimolybdate ions, although with structures perturbed by their interactions with the zeolite framework.
Co-reporter:Brian C. Vicente, Zheng Huang, Maurice Brookhart, Alan S. Goldman and Susannah L. Scott
Dalton Transactions 2011 - vol. 40(Issue 16) pp:NaN4274-4274
Publication Date(Web):2011/03/08
DOI:10.1039/C0DT01369B
When a pincer-ligated iridium complex with a phosphinite substituent in the para-position of the aromatic backbone is immobilized on γ-alumina, it becomes a highly effective supported catalyst for the transfer-dehydrogenation of alkanes. The nature of the interaction between the organometallic complex and the support was investigated using solid-state 31P MAS NMR spectroscopy, solution-state 1H and 31P{1H} NMR spectroscopy, IR and GC/MS analysis of extracted reaction products. The phosphinite substituent is cleaved from the pincer ligand by its reaction with hydroxyl groups on the γ-alumina surface, resulting in covalent anchoring of the complex via the aryl ring. A similar reaction occurs on silica, allowing for ready grafting onto this support as well. A strategy for anchoring homogeneous catalysts on hydroxyl-terminated oxide supports though the selective cleavage of [POR]-containing ligand substituents is suggested.
Co-reporter:Ritu Ahuja, Sabuj Kundu, Alan S. Goldman, Maurice Brookhart, Brian C. Vicente and Susannah L. Scott
Chemical Communications 2008(Issue 2) pp:NaN255-255
Publication Date(Web):2007/11/01
DOI:10.1039/B712197K
Tandem dehydrogenation–olefin-metathesis catalyst systems, comprising a pincer-ligated iridium-based alkane dehydrogenation catalyst and a molybdenum-based olefin-metathesis catalyst, are reported to effect the metathesis-cyclooligomerization of cyclooctane and cyclodecane to give cycloalkanes with various carbon numbers, predominantly multiples of the substrate carbon number, and polymers.
Co-reporter:Philippe Perrotin, Jenny S. J. McCahill, Guang Wu and Susannah L. Scott
Chemical Communications 2011 - vol. 47(Issue 24) pp:NaN6950-6950
Publication Date(Web):2011/05/10
DOI:10.1039/C1CC00095K
A well-defined, homogeneous catalyst, [(Ph)(2-(2′,6′-(OMe)2-C6H3)-C6H4)P(2-SO3-C6H4)]Ni(Ph)PPh3, in which a single, bulky ortho-biphenyl substituent on the chelating phosphine blocks one axial position, is very active for formation of linear polyethylene (Mn = 403000 g mol−1, Mw/Mn = 1.87).
Co-reporter:Chenghao Deng, Xuezhi Duan, Jinghong Zhou, Xinggui Zhou, Weikang Yuan and Susannah L. Scott
Catalysis Science & Technology (2011-Present) 2015 - vol. 5(Issue 3) pp:NaN1547-1547
Publication Date(Web):2014/11/24
DOI:10.1039/C4CY01285B
In this work, bimetallic Ir–Re catalysts supported on KIT-6 are prepared by tuning the thermal treatment procedures, i.e., conventional calcination and reduction (Ir–Re/KIT-6-CR) and modified direct reduction (Ir–Re/KIT-6-R) after impregnation of two metal precursors. The structure of both catalysts is intensively characterized by H2-TPR, STEM-HAADF-EDX, XPS and CO-DRIFTS. Results indicate that an Ir–Re alloy forms on the KIT-6 support when direct reduction is employed, which exhibits excellent catalytic performance in hydrogenolysis of glycerol. The formation rate of 1,3-propanediol over Ir–Re/KIT-6-R reaches 25.6 mol1,3-PD molIr−1 h−1 at 63% glycerol conversion with the addition of amberlyst-15 under 8 MPa H2, 393 K and 20 wt% glycerol aqueous solution, almost twice that over Ir–Re/KIT-6-CR. It is revealed that Re species without prior calcination treatment could be fully reduced and therefore couple with Ir to form an Ir–Re alloy structure with enhanced resistance against particle aggregation, while the calcination and subsequent reduction leads to the formation of an Ir–ReOx structure since the rhenium oxide species generated during the calcination is difficult to be reduced.
.jpg)
![Benzenesulfonic acid, 2-[(2',6'-dimethoxy[1,1'-biphenyl]-2-yl)phenylphosphino]-](/data/chemimg/163700/1312097-12-3.png)
![Benzenesulfonic acid, 2-[(2',6'-dimethoxy[1,1'-biphenyl]-2-yl)phenylphosphino]-](/data/chemimg/163700/1312097-12-3_b.png)
![Benzenesulfonic acid, 2-[bis(2',6'-dimethoxy[1,1'-biphenyl]-2-yl)phosphino]-](/data/chemimg/110000/926936-59-6.png)
![Benzenesulfonic acid, 2-[bis(2',6'-dimethoxy[1,1'-biphenyl]-2-yl)phosphino]-](/data/chemimg/110000/926936-59-6_b.png)
