Kenneth L. Watters

Evidence for production of surface formate upon direct reaction of CO with alumina and magnesia

Abstract The observation of adsorbed formate ion produced by the reaction of CO with MgO or γ-Al 2 O 3 surfaces at temperatures between 375 and 575 K is reported. Infrared bands observed near 1590, 1380, 1360, and 1050 are assigned to the fundamentals of adsorbed HCO 2 − ion on the basis of spectra that have been reported for formate ion in solution. The spectral bands and observed isotopic shifts for adsorbed formate produced by reaction of the oxide surface with CO agree with the spectra of formates produced by reaction of methanol or formic acid with the same surfaces. The intensity of IR bands due to formate produced by reaction of the oxide surfaces with CO depends upon the hydroxyl group concentration at the surface. A mechanism for conversion of CO to formate via a surface acylium ion is proposed.

Characterization of supported iridium catalysts prepared from Ir4(CO)12

Abstract The characterization of silica- and alumina-supported iridium catalysts prepared from Ir 4 (CO) 12 is described. The techniques employed are infrared spectroscopy, temperature-programmed desorption, and gravimetric chemisorption measurements. Supported Ir 4 (CO) 12 begins to decompose on heating in vacuo above 75 °C, and decomposition is complete at 350 °C. The major product of decomposition on silica is metallic iridium, whereas on alumina a fraction of the iridium becomes oxidized. Hydrogen reduction of both silica- and alumina-supported catalysts following decomposition of the parent complex in vacuo produces highly dispersed metallic iridium. From observed shifts in the infrared frequencies of adsorbed CO with coverage it is concluded that the iridium may be in the form of two-dimensional rafts, containing possibly 20 atoms or more.

A chemical and physical characterization of alumina-supported Rh6(CO)16

The decarbonylation and carbonylation reactions of the metal cluster carbonyl Rh6(CO)16 supported on Al2O3 were investigated using volumetric gas adsorption measurements, ir spectroscopy, isotopic measurements, transmission electron microscopy, and EPR spectroscopy. Complete decarbonylation by O2 and subsequent recarbonylation to Rh6(CO)16 could be achieved at room temperature provided sufficient physically adsorbed H2O was present. A two-step reaction scheme is presented which accounts for the experimental data; it is suggested that the Rh6 cluster remained intact during these reactions. When heated, in vacuo, above 250 °C, the catalyst lost its ability to undergo the reversible carbonylation/decarbonylation cycle, and the resulting material resembled a highly dispersed conventional Rh catalyst. TEM data showed that no larger crystallites (>10 A) of Rh are formed even when heated in vacuo to 450 °C.

Metal-flavin complexation. A resonance Raman investigation

Several groups have recently shown that high quality resonance Raman spectra can be obtained for flavin species in spite of their intense fluorescence. We are interested in obtaining the resonance Raman spectra of flavins in various chemical environments in order to determine whether the spectra are useful in probing the chemical interaction between flavins and protein in flavoenzymes. We have obtained the resonance Raman spectrum of a nonfluorescent Ag+ complex of FMN. Several large changes occur in the FMN resonance Raman spectrum upon Ag+ complexation; among these are changes in the 1580 cm-1 region of the FMN spectrum (assigned to nu C=N at N-5 and C-4a), the 1410 cm-1 region and the 1260 cm-1 region (associated with a vibration having some delta N-N-H character at N-3). Similar changes are observed in the same region of a Ru2+-FMN complex. Since these spectral changes occur in two metal flavin complexes with very different electronic spectra, they would seem to be due to vibrational changes induced by metal complexation at N-5 and the oxygen at C-4 of flavin rather than the details of the vibronic interactions which give rise to the resonance enhancement of the spectrum. A structure for the Ag+-FMN complex is suggested. This study has potential physiological significance, because it illustrates the possible role of resonance Raman spectroscopy as a tool for the determination of direct flavin metal interaction in dilute aqueous solution of metalloflavoproteins.

Evidence for reversible formation of rhodium tricarbonyl species in surface supported rhodium catalysts

The formation of an apparent tricarbonyl of rhodium was observed when [Rh(CO)2Cl]2 deposited on SiO2 was exposed to 600 Torr of CO at 300 K. The v(CO) bands assigned to the tricarbonyl species were at 2072 and 2107 cm−1, quite close in frequency to the two bands reported when [Rh(CO)2Cl]2 in solution was converted to a tricarbonyl complex of Rh by exposure to high pressures (13 atm) of CO. On SiO2 and in solution the conversion to the tricarbonyl was only partial with the bands of the initial dimer always remaining. The ability to add a third CO per Rh was observed subsequently for mildly reduced RhCl3SiO2 or Rh(NO3)3SiO2, but with these materials it was possible to prepare samples in which the conversion to the tricarbonyl was nearly complete at 600 Torr; a third band associated with the tricarbonyl emerged near 2030 cm−1 as the 2040-cm−1 band of the surface dicarbonyl disappeared. The spectral behavior observed in this study was quite similar to that reported for Rh1(CO)2 on Al2O3 when exposed to higher pressures of CO. The presence of three bands for the tricarbonyl eliminates a strict C3v symmetry for the tricarbonyl and suggests a square planar C2v or a distorted C3v structure. The increased ability to add a third CO is briefly discussed in terms of the effect of the surface on the geometry and electronic structure about Rh atoms.

Interactions of cobalt carbonyls with oxide surfaces: I. (CO)9Co3CCH3 on silicas, aluminas, and zeolites

Abstract The adsorption and subsequent reactions of the cobalt cluster compound μ 3 -ethylidine-tris-(tricarbonyl cobalt), I , on silica, alumina, and zeolite supports were investigated. The supports were pretreated, in vacuo , at temperatures between 295 and 700K. On silica, I appears to be physisorbed and reacts only slowly with O 2 to produce decarbonylated products. Adsorption on alumina produces physisorbed I and at least three cobalt carbonyl products of the reactions of I with the support. Aluminas pretreated at lower temperatures are the most reactive toward I , as evidenced by the greater tendency of I to react and form novel cobalt carbonyls on these aluminas. These reaction products are immediatley oxidized by O 2 to produce carbonate, bicarbonate, and oxidized cobalt. Compound I adsorbs at room temperature inside the supercages of a Na Y zeolite; it is subsequently converted, at elevated temperatures (350K), to the carbonyl species which were identified on alumina. Infrared and manometric data show that thermolysis and oxidation of I on zeolite ultimately lead to its total decarbonylation to produce carbonates, bicarbonates, and organic carbonyl fragments.

Studies of dinuclear metal carbonyl carbenes. Synthesis, characterization, and raman spectra of amino- and alkoxycarbenes

Abstract The new dinuclear metal carbene complexes (CO) 9 Re 2 C(OC 2 H 5 )R (R = CH 3 , C 6 H 5 ) and (CO) 9 M 2 C(NHCH 3 )CH 3 (M = Mn, Re) have been synthesized. All products were obtained with the carbene in the equatorial position (cis to the MM bond) and the methylaminomethyl carbene ligand in the syn configuration. The Re aminocarbene could be isomerized into a syn-anti mixture, but the anti form of the Mn aminocarbene could not be isolated. Low frequency Raman spectra for several dinuclear carbenes show an intense ν (MM) band near 120 cm −1 for M = Re and 150–160 cm −1 for M = Mn; the frequency of this band is relatively insensitive to substitution of carbene for carbonyl in the cis position and to changes in the cis carbene. The pattern of Raman bands in the rest of the Δν = 100-650 cm −1 region confirms the structural similarity of the series of compounds studied.

A study of substitution reactions in Rh6(CO)16 using simple and polymer bound phosphine ligands

Abstract Substitution of phosphines for CO on Rh 6 (CO) 16 as been studied. Controlled addition of triphenylphosphine or of tris-(2-diphenylphosphinoethyl)phosphine to Rh6(CO) 16 resulted in displacement of 1–3 carbonyl ligands to form compounds of general formula Rh 6 (CO) 16−n Ln. Increased substitution to n > 6 produced unstable materials which were difficult to characterize. These reactions are compared with the reaction of Rh 6 (CO) 16 with phosphine substituted polystyrene to produce polymer bound Rh 6 clusters. Spectroscopic evidence for a very close similarity between the polymer bound rhodium species and the homogeneous phase complexes is presented.

H2D2 equilibration over supported iridium catalysts prepared from Ir4(CO)12

X-Ray photoelectron spectroscopy (XPS) measurements have been performed on silica- and alumina-supported iridium catalysts prepared from Ir4(CO)12, and the activity of these catalysts for the H2:D2 equilibration reaction determined as a function of the activation treatment. The XPS results show that predominantly zerovalent indium is obtained upon decomposition of supported Ir4(CO)12 in vacuo. Maximum activity for H2:D2 equilibration is generated only after removal of all carbonyl ligands. The reaction is poisoned by readsorption of CO, and the recovery of activity upon desorption of CO parallels the original generation of activity.

A laser Raman study of lysozyme denaturation

Abstract The Raman spectrum of chemically denatured lysozyme was studied. The denaturants studied included dimethyl sulfoxide, LiBr, guanidine · HCl, sodium dodecyl sulfate, and urea. Previous studies have shown that the amide I and amide III regions of the Raman spectrum are sensitive to the nature of the hydrogen bond involving the amide group. The intensity of the amide III band at 1260 cm−1 (assigned to strongly hydrogen-bonded α-helix structure) relative to the intensity of the amide III band near 1240 cm−1 (assigned to less strongly hydrogen-bonded groups) is used as a parameter for comparison with other physical parameters used to assess denaturation. The correlation between this Raman parameter and denaturation as evidenced by enzyme activity and viscosity measurements is good, leading to the conclusion that the amide III Raman spectrum is useful for assessing the degree of denaturation. The Raman spectrum clearly depends on the type of denaturant employed, suggesting that there is not one unique denatured state for lysozyme. The data, as interpreted, place constraints on the possible models for lysozyme denaturation. One of these is that the simple two-state model does not seem consistent with the observed Raman spectral changes.