No abstract

The preparation and study of gas-phase transition-metal complexes in their higher oxidation states, i.e., Cu(II), Cr(III), Fe(II), etc., presents a considerable technical challenge. Charge transfer prevents such species from being "grown" as cluster ions and techniques, such as electrospray, do not always produce the desired charge state or allow for experiments to be performed on a broad range of ligands. Discussed here are new results from a technique which promises to overcome some of these problems, and appears capable of producing complexes from a wide variety of metals and ligands. Data are presented for complexes based on silver(II) in association with a broad range of ligands, including pyridine, tetrahydrofuran, and benzene. For each [AgL n ] 2+ system, two important quantities are identified: (i) the minimum number of ligands required to form a stable unit and (ii) the value of n for which the intensity distribution reaches a maximum. For nitrogencontaining ligands these numbers are 2 and 4, respectively, and for oxygen-containing ligands 4 and 5. A series of aromatic ligands all exhibit coordination numbers of 2. For several of the nitrogen-based ligands the most stable combinations correspond to those identified in the condensed phase, and [Ag(pyridine) 4 ] 2+ is a very good example of such behavior. In the case of the oxygen-containing ligands, there are no direct condensedphase analogues, but some of the more stable combinations identified may offer prospects for future preparative work. Within the latter group, not only was the presence of stable silver(II)/CO 2 complexes very unexpected, but with [Ag(CO 2 ) 4 ] 2+ being the most stable combination, the pattern of behavior is markedly different from that of other oxygen-containing ligands. The composition and charge states of many of the stable complexes were confirmed via collisional activation, where both ligand loss and charge-transfer processes could be identified. Only one example of a chemical reaction could be clearly identified as being initiated by the presence of silver(II).

A technique has been developed that provides a solution to the very considerable technical problem of preparing gas-phase complexes from transition metals in their higher oxidation states, i.e., Cu(II), Cr(III), Fe(II), etc. Charge transfer prevents complexes, such as [Cu•(H 2 O) n ] 2+ , from being prepared via nucleation about an ion core, and yet these ions are pivotal to an understanding of transition metal chemistry. Discussed here are new results from a technique that appears capable of producing complexes from a wide variety of metals and ligands. Data are presented for copper(II) in association with 20 different ligands, including water, ammonia, pyridine, tetrahydrofuran, and benzene. For each [Cu•L n ] 2+ system, two important quantities are identified: (i) the minimum number of ligands required to form a stable unit and (ii) the value of n for which the intensity distribution reaches a maximum. The data show considerable variation as a function of the composition and size of solvent molecule, with evidence of stable coordination shells containing between 2 and 8 molecules. In most instances, coordination shells containing more than four molecules can be attributed to the formation of an extended network of hydrogen bonds. Collisional activation of size-selected clusters reveals the presence of extensive ligand-to-metal electron transfer in the smaller complexes, and in several cases, charge transfer is also accompanied by chemical reactivity. The extent of charge transfer is frequently observed to be determined by the stability of the singly charged metal-containing product.

Microwave spectra of the complexes KrAuF and KrAgBr have been measured for the first time using a cavity pulsed jet Fourier transform microwave spectrometer. The samples were prepared by laser ablation of the metal from its solid and allowing the resulting plasma to react with an appropriate precursor (Kr, plus SF6 or Br2) contained in the backing gas of the jet (usually Ar). Rotational constants; geometries; centrifugal distortion constants; vibration frequencies; and 197Au, 79Br, and 81Br nuclear quadrupole coupling constants have all been evaluated. The complexes are unusually rigid and have short Kr-Au and Kr-Ag bonds. The 197Au nuclear quadrupole coupling constant differs radically from its value in an AuF monomer. In addition 83Kr hyperfine structure has been measured for KrAuF and the previously reported complex KrAgF. The geometry of the latter has been reevaluated. Large values for the 83Kr nuclear quadrupole coupling constants have been found for both complexes. Both the 197Au and 83Kr hyperfine constants indicate a large reorganization of the electron distribution on complex formation. A thorough assessment of the nature of the noble gas-noble metal bonding in these and related complexes (NgMX; Ng is a noble gas, M is a noble metal, and X is a halogen) has been carried out. The bond lengths are compared with sums of standard atomic and ionic radii. Ab initio calculations have produced dissociation energies along with Mulliken populations and other data on the electron distributions in the complexes. The origins of the rigidity, dissociation energies, and nuclear quadrupole coupling constants are considered. It is concluded that there is strong evidence for weak noble gas-noble metal chemical bonding in the complexes.

An experimental study of the stability and coordination of oxygen- and nitrogen-containing ligands in association with Mg2+ in the gas phase has been undertaken. The ligands chosen exhibit a wide range of physical properties in terms of their ionization energies, dipole moments, and polarizabilities, and a simple electrostatic model reveals a semiquantitative trend between these properties and the ability of each ligand to stabilize Mg2+. The model clearly demonstrates why water is extremely effective at stabilizing Mg2+, and in this respect, CO2 also proves to be a good ligand. Evidence of a discrete first solvation shell is apparent only for those ligands which do not display hydrogen bonding. For water, methanol, and ethanol, hydrogen bonding leads to extended solvation units for which the boundaries are less obvious. However, for more complex alcohols, steric interactions appear to negate the influence of hydrogen bonding. Discrete solvation shells are observed for most aprotic ligands, and the optimum coordination number is 4. However, there is some slight variation in this value, mainly as a consequence of ligand size. Assuming Mg2+ to be a hard Lewis acid, the results are used to order the ligands in terms of how effective they are at stabilizing Mg2+ in their role as hard Lewis bases. Evidence of the first gas-phase Mg2+ bidentate metal complex is also provided.

No abstract

Ni(+)(CO(2))(n), Ni(+)(CO(2))(n)Ar, Ni(+)(CO(2))(n)Ne, and Ni(+)(O(2))(CO(2))(n) complexes are generated by laser vaporization in a pulsed supersonic expansion. The complexes are mass-selected in a reflectron time-of-flight mass spectrometer and studied by infrared resonance-enhanced photodissociation (IR-REPD) spectroscopy. Photofragmentation proceeds exclusively through the loss of intact CO(2) molecules from Ni(+)(CO(2))(n) and Ni(+)(O(2))(CO(2))(n) complexes, and by elimination of the noble gas atom from Ni(+)(CO(2))(n)Ar and Ni(+)(CO(2))(n)Ne. Vibrational resonances are identified and assigned in the region of the asymmetric stretch of CO(2). Small complexes have resonances that are blueshifted from the asymmetric stretch of free CO(2), consistent with structures having linear Ni(+)-O=C=O configurations. Fragmentation of larger Ni(+)(CO(2))(n) clusters terminates at the size of n=4, and new vibrational bands assigned to external ligands are observed for n> or =5. These combined observations indicate that the coordination number for CO(2) molecules around Ni(+) is exactly four. Trends in the loss channels and spectra of Ni(+)(O(2))(CO(2))(n) clusters suggest that each oxygen atom occupies a different coordination site around a four-coordinate metal ion in these complexes. The spectra of larger Ni(+)(CO(2))(n) clusters provide evidence for an intracluster insertion reaction assisted by solvation, producing a metal oxide-carbonyl species as the reaction product.

V+(CO2)n and V+(CO2)nAr complexes are generated by laser vaporization in a pulsed supersonic expansion. The complexes are mass-selected within a reflectron time-of-flight mass spectrometer and studied by infrared resonance-enhanced (IR-REPD) photodissociation spectroscopy. Photofragmentation proceeds exclusively through loss of intact CO2 molecules from V+(CO2)n complexes or by elimination of Ar from V+(CO2)nAr mixed complexes. Vibrational resonances are identified and assigned in the region of the asymmetric stretch of free CO2 at 2349 cm(-1). A linear geometry is confirmed for V+(CO2). Small complexes have resonances that are blueshifted from the asymmetric stretch of free CO2, consistent with structures in which all ligands are bound directly to the metal ion. Fragmentation of the larger clusters terminates at the size of n=4, and a new vibrational band at 2350 cm(-1) assigned to external ligands is observed for V+(CO2)5 and larger cluster sizes. These combined observations indicate that the coordination number for CO2 molecules around V+ is exactly four. Fourfold coordination contrasts with that seen in condensed phase complexes, where a coordination number of six is typical for V+. The spectra of larger complexes provide evidence for an intracluster insertion reaction that produces a metal oxide-carbonyl species.