The reaction of gaseous N 2 O 5 with sea salt and its components is a potential source of halogen atoms in the marine boundary layer. There are two possible reaction paths when water is present on the salt surface. Reaction with the chloride ion forms nitryl chloride (ClNO 2 ), a photolyzable compound: N 2 O 5 + NaCl ! ClNO 2 + NaNO 3 , while hydrolysis of N 2 O 5 generates HNO 3 that can react further with NaCl to form gaseous HCl: N 2 O 5 + H 2 O (on NaCl) ! 2 HNO 3 , HNO 3 + NaCl ! HCl + NaNO 3 . We report here Knudsen cell studies at 23 C of the reaction of N 2 O 5 with NaCl, using less than one layer of salt particles. A model, which takes into account the effective salt surface area exposed to the gas, was applied, allowing for the determination of uptake coefficients without introducing uncertainties associated with diffusion into multiple layers of salt particles. The net uptake coefficient for the sum of both channels for the N 2 O 5 reaction was measured to be g N 2 O 5 ¼ (2.9 AE 1.7) Â 10 À3, where the error cited is the 2s statistical error. The cumulative error is estimated to be better than a factor of three. Both ClNO 2 and HCl were observed as gaseous products from the N 2 O 5 -salt reaction and the branching ratio for ClNO 2 was 0.73 AE 0.28 (2s). A limited number of experiments were carried out for the reaction with synthetic sea salt, resulting in an uptake coefficient of about an order of magnitude larger than for NaCl, and a ClNO 2 yield of 100%. We propose a mechanism for this reaction in which surfaceadsorbed water plays a key role in the competition between hydrolysis of N 2 O 5 to generate HNO 3 and the reaction with NaCl to generate ClNO 2 . Reaction with NaCl is shown to be a potentially important source of ClNO 2 , and thus, of highly reactive chlorine atoms in urban marine regions at dawn. Application of our model to previous data from this laboratory for the reaction of chlorine nitrate (ClONO 2 ) with fractional layers of NaCl gives a corrected uptake coefficient of g ClONO 2 ¼ (2.4 AE 1.2) Â 10 À2 (2s), which suggests that the ClONO 2 -NaCl reaction may contribute significantly to the observed concentrations of Cl 2 in the marine boundary layer.

The number of heavy-duty vehicles using alternative fuels such as compressed natural gas (CNG) and new low-sulfur diesel fuel formulations and equipped with after-treatment devices are projected to increase. However, few peer-reviewed studies have characterized the emissions of particulate matter (PM) and other toxic compounds from these vehicles. In this study, chemical and biological analyses were used to characterize the identifiable toxic air pollutants emitted from both CNG and low-sulfur-diesel-fueled heavy-duty transit buses tested on a chassis dynamometer over three transient driving cycles and a steady-state cruise condition. The CNG bus had no after-treatment, and the diesel bus was tested first equipped with an oxidation catalyst (OC) and then with a catalyzed diesel particulate filter (DPF). Emissions were analyzed for PM, volatile organic compounds (VOCs; determined on-site), polycyclic aromatic hydrocarbons (PAHs), and mutagenic activity. The 2000 model year CNG-fueled vehicle had the highest emissions of 1,3-butadiene, benzene, and carbonyls (e.g., formaldehyde) of the three vehicle configurations tested in this study. The 1998 model year diesel bus equipped with an OC and fueled with low-sulfur diesel had the highest emission rates of PM and PAHs. The highest specific mutagenic activities (revertants/microg PM, or potency) and the highest mutagen emission rates (revertants/mi) were from the CNG bus in strain TA98 tested over the New York Bus (NYB) driving cycle. The 1998 model year diesel bus with DPF had the lowest VOCs, PAH, and mutagenic activity emission. In general, the NYB driving cycle had the highest emission rates (g/mi), and the Urban Dynamometer Driving Schedule (UDDS) had the lowest emission rates for all toxics tested over the three transient test cycles investigated. Also, transient emissions were, in general, higher than steady-state emissions. The emissions of toxic compounds from an in-use CNG transit bus (without an oxidation catalyst) and from a vehicle fueled with low-sulfur diesel fuel (equipped with DPF) were lower than from the low-sulfur diesel fueled vehicle equipped with OC. All vehicle configurations had generally lower emissions of toxics than an uncontrolled diesel engine. Tunnel backgrounds (measurements without the vehicle running) were measured throughout this study and were helpful in determining the incremental increase in pollutant emissions. Also, the on-site determination of VOCs, especially 1,3-butadiene, helped minimize measurement losses due to sample degradation after collection.

Abstract. The uptake of sulfur dioxide (SO2) on synthetic sea salt (SSS) and its components, NaC1 and MgCI2o6H20, was studied at 298 K using a Knudsen cell interfaced to a quadrupole mass spectrometer. Significant uptake on dried salts was not observed, placing upper limits on the uptake coefficients, ¾, of < 1 x 10-4 for NaCI, < 5 x 10-4 for MgC12-6H20, and < 8 x 10-5 for SSS. However, SSS and MgCI2o6H20 that had not been dried before use showed significant uptake of SO2. The magnitude of the uptake depended strongly on the exposure time and the amount of water desorbing. Initially, the measured uptake coefficients for SO2 on SSS were as high as 0.09, but they rapidly decreased below 10-2 with a t-l/2 dependence as expected for approach to the equilibrium saturation concentration in an aqueous solution. The decreasing uptake coefficient slowly approaches zero over hours, consistent with reactions in a water layer with species such as CaCO3. The products of the reaction were shown by diffuse reflectance IR Fourier transform spectroscopy (DRIFTS) to include low solubility metal sulfites. These studies show that uptake of SO2 on sea salt particles, even below their deliquescence/effiuorescence points, can be treated as if it is into an aqueous salt solution.

The uptake and reaction of chlorine nitrate (ClONO2) on NaCl and synthetic sea salt (SSS) powders was studied at 298 K using a Knudsen cell interfaced to a quadrupole mass spectrometer. A time-dependent uptake coefficient was observed, with a large initial uptake coefficient measured for most samples of γinit > 0.1, followed by a smaller and slowly declining uptake coefficient at longer reaction times. This behavior is shown to be consistent with uptake into, and reaction in, water on the salt surface. The steady-state uptake coefficient on NaCl was invariant over a range of ClONO2 concentrations from 1012 to 1013 molecules cm-3 but was dependent on the number of salt particle layers in a manner consistent with approximately two layers of particles being available for reaction. The results of experiments using monolayers and sub-monolayers of salts, where the available reactive surface is known, give an uptake coefficient at longer times after the rapid initial uptake for ClONO2 on NaCl of γ t = (6.5 ± 3.0) × 10-3 (2σ). The larger uptake coefficient obtained initially compared to longer reaction times helps to reconcile different values reported earlier by other research groups for this reaction. The uptake coefficient on SSS, which holds more water, is much larger than that for NaCl, with γinit = (0.4 ) and γ t = (0.1 ) (2σ). At the higher uptake coefficients measured for SSS, this reaction could be a significant source of Cl2 in the marine boundary layer if chlorine nitrate is available at a constant concentration of ∼5 ppt.