Dan Strömberg

Oxidation of atomic mercury by hydroxyl radicals and photoinduced decomposition of methylmercury in the aqueous phase

Abstract The rate constant for Hg 0 + . OH , k Hg 0 + . OH =(2.4±0.3)×10 9 M −1 s −1 , in the aqueous phase was determined using a relative rate technique with methyl mercury as reference compound. The .OH initiated mercury reaction proceeds via the molecular Hg(I) radical which is oxidised to Hg(II) by dissolved O2. The reaction can be of importance under certain atmospheric circumstances, such as when the aqueous phase capacity of forming OH radicals is significant and the gas phase concentration of ozone drops. The same end product, i.e. Hg(II) was observed from the photodegradation of methylmercury hydroxide. In this case, molecular Hg(I) radicals are again likely to be formed after photodegradation of the Hg–C bond with subsequent oxidation. A lifetime of 230 h of methylmercury at outdoor conditions was estimated due to this reaction. The action of .OH on methylmercury species also involves breaking of organometallic bonds and formation of Hg(II). Speciation of these reaction products from methylmercury is important for the estimation of biogeochemical cycling of mercury.

A kinetic study on the abiotic methylation of divalent mercury in the aqueous phase

The mechanism and kinetics of the formation of methylmercury from an experimental solution containing divalent mercury and acetic acid has been investigated. The experiments were performed in a 2-dm(3) Teflon reactor. The organic mercury was measured with time resolutions varying between minutes and hours, after derivatisation, gas chromatography separation and Cold vapour atomic fluorescence detection. (GC)-CVAFS. CVAFS technique was used for determination of inorganic mercury in the aqueous phase using an automated mercury analyser. The experiments were carried out in concentrations relevant for natural waters. Our result shows that the reaction proceeds via mercury acetate complexes. A first order reaction coefficient has been calculated at various pH values, and was found to be (9.0+/-0.9) x 10(-7) s(-1) at pH 3.6-3.7. The rate was not found to be enhanced by UV-light when taking into account the photolytical degradation of methylmercury. The reaction rate at various pH values, the influences of some other relevant reaction parameters, and implications for atmospheric and terrestrial waters are discussed.

Determination of Henry’s law constant for elemental mercury

The assessment of the global mercury cycle involves estimations of the evasion of mercury form oceanic waters. In such estimations Henrys law constant is often used. In this study the Henrys law constant for elemental mercury has been re-determined in MQ water and artificial sea water. Moreover, for the first time it has been determined for 1.5M sodium chloride (NaCl) solution which is of relevance for modeling of atmospheric waters at coastal locations. For all solutions, experiments has been conducted at five different temperatures between 278 and 308K, using a novel technique, for mercury, based on direct measurements of the portioning of mercury between the aqueous and gaseous phase. Elemental mercury was extracted from the water column and the logarithm of the mass of extracted mercury was plotted against time. A dimensionless Henrys law constant, defined as: [see text] was obtained from the slope of the curve. Almost no difference was observed in the values comparing the Milli-Q water and artificial sea water, however for the 1.5M NaCl solution a salting-out effect was seen, i.e. the solubility of mercury in the water phase decreased. The decreased solubility will generate an increase in the value of Henrys law constant.

Distribution of atmospheric mercury species in Northern Europe: Final results from the MOE-project

The mercury species over Europe (MOE) project was aimed at identifying sources, occurrence and atmospheric behaviour of atmospheric Hg species. Within MOE, emission measurements, ambient air measurements, process and regional-scale modelling and laboratory measurements were conducted. In this work, a summary of some of the main results is given. From the emission measurements, information on stack gas concentrations and emission factors for five coal fired power plants and three waste incinerators are presented. Results from field measurements of mercury species in ambient air at five locations in Northern Europe are presented. Examples from regional-scale atmospheric modelling are also given. The results emphasise the importance of information on Hg species for instance in emission inventories and measurement data from background sites. Furthermore, the importance of considering the role of the global cycling of mercury in future control strategies is emphasised.

N2O adsorption and decomposition at a CaO(100) surface, studied by means of theory

The adsorption and decomposition of an N2O molecule at different sites on a CaO(s) surface are investigated by means of ab initio quantum chemistry. The calcium, Ca2+, and oxygen, Os2−, sites at a perfect (100) surface and at a corner position, Oc2−, are considered. Adsorption energies at different sites are calculated and the largest value, 6 kcal/mol, is obtained for a corner site. The barrier for dissociation is calculated to 26 and 27 kcal/mol at the Oc2− and Os2− sites, respectively. These values are some 10 kcal/mol lower than the experimental estimate, and the discrepancy is understood from methodological difficulties to describe the free N2O molecule. A mechanism for the dissociation over an O2− site is proposed, whereby the transfer of the O atom goes via a linear N-N … O … O2− transition state.

The bonding and migration of an O atom on a CaO(100) surface: a theoretical study

Abstract Ab initio quantum calculations of the dissociation energies and bond distances for the bond between an O atom situated at different sites on a CaO(100) surface have been performed, using surface clusters imbedded in an array of point charges. The point charges were evaluated by a modified Ewald technique. The calculations were carried out using the CASSCF and CCI methods. Different clusters were tested. The “peroxy” bond between the O atom and a surface O 2− site was found to be the strongest, ≈ 50 kcal/mol at the CCI level compared to a triplet oxygen atom far away from the surface (∼ 100 kcal/mol compared to a singlet oxygen atom). The bond on top of a surface Ca 2+ ion was very weak, ∼ 5 kcal/mol. The surface migration of an O atom from one O 2− site to another O 2− was investigated. The barrier for the migration was calculated to ∼ 66 kcal/mol.

Non-relativistic and relativistic calculations on some Zn, Cd and Hg complexes

Abstract A number of species containing Hg, Cd and Zn have been studied at the CI level using non-relativistic and relativistic techniques. Effective core potentials were used for Hg and Cd, while an all-electron description was used for Zn. Spectroscopic constants have been calculated for the positive hydride ions and the dichlorides of the three metals. The agreement with experiment was good in all cases. Calculations have also been carried out on Hg (OH) 2 , HgO and ZnO. The Hg (OH) 2 molecule is predicted to be covalently bound by two σ bonds formed between a metal sp hybrid with small but significant 5d contribution and the OH. The binding energy of Hg(OH) 2 is significant although smaller than the binding energy of HgCl 2 by 20 kcal/mol. The near UV spectra of HgCl 2 and Hg(OH) 2 are calculated to be quite similar except for excitations involving the OH bond. The calculated dissociation energies of HgO and ZnO are severely underestimated.

First-order relativistic calculations on Au2 and Hg22+

Abstract Spin-free first-order relativistic calculations have been carried out on Au, Hg at the SCF level and on Au 2 and Hg 2 2+ at the CI level using the size-consistent CPF procedure. The atomic SCF results agree with previous relativistic atomic calculations using the Cowan-Griffin procedure to within 0.5 eV. The spectroscopic constants calculated for Au 2 agree fairly well with experiment and with previously published relativistic effective core potential results. For Hg 2 2+ we obtain, in contrast to previous theoretical results, a relativistic destabilization of the binding energy.

Relativistic quantum calculations on some mercury sulfide molecules

Relativistic quantum calculations at the CASSCF- and CCI-levels were performed on the Hg(SH)2, HgSH and HgS molecules. The relativistic effects were taken into account by a relativistic effective core potential method. Dissociation energies and optimal geometries were calculated for these three molecules, which are plausible atmospheric Hg compounds. The Hg(SH)2 and HgSH molecules (in the gaseous phase) have never been studied before, neither experimentally nor theoretically, i.e. the existence of these molecules are uncertain. The theoretical dissociation energies, Des, of Hg(SH)2 and HgSH (at the CCI-level) were 59 kcal·mol−1 and 3 kcal·mol−1, respectively, indicating that Hg(SH)2 could be stable in the atmosphere but probably not HgSH. The theoretical De of HgS differs very much from the experimental one, but the reason for this is not clear. The Hg-S distances for Hg(SH)2, HgSH and HgS were found to be 2.38, 2.63 and 2.30 Å, respectively. The Hg-S-H angle in Hg(SH)2 was optimized to 93°. The excitation energies of Hg(SH)2, Hg(SH)2(H2O)4 and (HSHg)2S were calculated in order to see whether these species can absorb photons with wavelengths longer than 290 nm (the sunlight limit) and subsequently be photolyzed. The Hg(SH)2(H2O)4 complex is intended as a model for Hg(SH)2(aq). Photoreduction of Hg-sulfide species in sea water, yielding Hg0, could be an important source of Hg in the atmosphere. Excitation energies lower than the sunlight limit (4.3 eV≈290 nm) were found for Hg(SH)2 and Hg(SH)2(H2O)4, although the lowest spin and dipole allowed excitations probably lie slightly (0.2 to 0.3 eV) above this limit. Therefore a photodecomposition of Hg(SH)2(g) and Hg(SH)2(aq) by sunlight seems likely to occur.

Theoretical calculations on the structure of the hexahydrated divalent zinc, cadmium and mercury ions

SCF calculations have been performed on the title compounds in order to study the possible reasons for the anomalously large spread in the mean HgO bond distance previously obtained for hydrate mercury (II) ions in solution. An energy minimum is found for all three complexes, [M(H2O)6]2+, M  Zn, Cd or Hg, for a regular Th ground-state nuclear configuration. The larger spread of the HgO distances can be explained in terms of a weak second-order Jahn—Teller effect. An enhanced vibronic coupling leads in the mercury case to larger vibrational amplitudes of the coupling mode without invoking a static distortion. The longer mean HgO distance found for the hydrated mercury (II) ion in solution than in a solid hexahydrate can be explained by assuming a larger asymmetry in the distribution of the HgO bonds. Calculations on the [Hg(H2S)6]2+ complex also show an energy minimum for the Th configuration, although in this case the adiabatic potential surface is very flat, and more refined methods of calculation could yield a distored ground-state configuration.