Schematic diagram of the O2(a) generator coupled to a fast flow with laser induced fluorescence detection for studying metal atom reactions (exemplified by Mg with O2(a1Δg)).
O2(a) emission current measured with the In-Ga-As detector at 1270 nm, as a function of [Cl2] in the generator. The corresponding calibrated [O2(a)] is shown on the right-hand ordinate. Data from a selection of experimental runs over several months show that the efficiency for O2(a) production ranged from 16%–26% of the Cl2.
Kinetic plots showing the first-order removal rate of Ca as a function of [O2(a)], at five different pressures of N2 in the flow tube.
Plot of the second-order rate coefficient for Ca + O2(a) as a function of N2 concentration. This reaction exhibits third-order (pressure) dependence demonstrating the formation of CaO2(1A1); the significant intercept indicates that the bimolecular channel to CaO + O is also active.
Kinetic plots showing the first-order removal rate of Fe as a function of [O2(a)], at two different pressures of N2 in the flow tube.
Plot of the second-order rate coefficient for Mg + O2(a) and Fe + O2(a) as a function of [N2]. The Mg reaction exhibits third-order kinetics forming MgO2(1A1). The Fe reaction shows no pressure dependence, indicating that the formation of FeO + O is the only reactive channel.
Potential energy curves (calculated at the B3LYP/6-311+g(2d,p) level of theory) for: Mg + O2(a) (top panel); and Ca + O2(a) (bottom panel). Singlet surfaces are shown by red lines and triplet surfaces by black lines. For Mg + O2(a), the only product is MgO2(1A1). Recombination of Ca + O2(a) produces mostly CaO2(1A1). However, there is a non-adiabatic crossing seam between OCaO(1A1) and OCaO(3B2), where there is a small probability of switching onto the triplet surface and generating the bimolecular products CaO + O(3P).
Potential energy surfaces for Mg + O2(a) (monochrome shading) and Mg + O2(X) (coloured shading), calculated at the B3LYP/6-311+g(2d,p) level of theory using a relaxed scan where the O–O bond distance was optimized at each point on the surface. The diagram illustrates that there are no intersections between the surfaces. Thus, the only possible reaction of Mg with O2(a) is recombination to MgO2(1A1).
Time-resolved concentration profiles predicted by MESMER. Top panel: Mg + O2(a), [O2(a)] = 1.0 × 1014 cm−3; [N2] = 3.2 × 1017 cm−3. Bottom panel: Ca + O2(a), [O2(a)] = 5.0 × 1012 cm−3; [N2] = 3.9 × 1016 cm–3.
π and σ orbitals used to carry out the CASSCF calculations described in the text. The left- and right-hand sides of the figure show the orbitals as they appear in OCaO and CaO2, respectively. Each orbital configuration is also identified by the corresponding symmetry label of its irreducible representation within the C2v point group. The last set of orbitals, a1 ′ and b2 ′, are labelled using a prime (′) only to distinguish them from the top set of orbitals, which have the same symmetries.
Relaxed CASSCF potential energy scans along the OCaO angle. As discussed in the text, 1B1 and 1B2 are not shown. There is a crossing between 1A1 and 3B2, and the triplet surfaces all increase rapidly to very high energies instead of evolving smoothly to a 3CaO2 superoxide type structure.
Modelled dependence of the rate coefficients for the reaction of Ca with O2(a) to form CaO2(1A1) (dash-dot line) or CaO + O (dash line) as a function of N2 concentration. The total rate coefficient (solid line) is compared with the rate coefficients measured in the present study.
Rate coefficients measured in the present study at 296 K. The quoted uncertainties are the standard errors from kinetics plots such as in Figures 3 and 5. The total uncertainty, which mostly arises from the systematic uncertainty in the O2(a) concentration, is estimated to be ±40%.
RRKM parameters for Mg + O2(a).
Computed relative energies and molecular properties for Ca + O2(a).
Article metrics loading...
Full text loading...