Importance of formaldehyde in cloud chemistry

A physical-chemical model which is an extension of that of Hong and Carmichael (1983) is used to investigate the role of formaldehyde in cloud chemistry. This model takes into account the mass transfer of SO₂, O₃, NH₃, HNO₃, H₂O₂, CO₂, HCl, HCHO, O₂, OH and HO₂ into cloud droplets and their subsequent chemical reactions. The model is used to assess the importance of S(IV)-HCHO adduct formation, the reduction of H₂O₂ by HCHO, HCHO-free radical interactions, and the formation of HCOOH in the presence of HCHO in cloud droplets.Illustrative calculations indicate that the presence of HCHO inhibits sulfate production rate in cloud droplets. The direct inhibition of sulfate production rate in cloud water due to nucleophilic addition of HSO⁻₃ to HCHO_(aq) to form hydroxymethanesulfonate (HMSA) is generally low for concentrations of HCHO typical of ambient air. However, inhibition of sulfate production due to formaldehyde-free radical interactions in solution can be important. These formaldehyde-free radical reactions can also generate appreciable quantities of formic acid.     

A model of sulphate production in a cap cloud and subsequent turbulent deposition onto the hill surface

A model has been constructed of the dynamics and microphysics of a hill cap cloud. This has been used to investigate the aqueous phase oxidation of SO₂ in the cloud droplets and the subsequent turbulent deposition of chemical species onto the hill surface. It is suggested that the dominant oxidant is H₂O₂ in these clouds and that therefore the process is likely to be oxidant limited. The amount of sulphate produced is comparable to that found in cloud condensation nuclei typically found over the U.K. and elsewhere away from strong local sources of sulphate aerosol. Ammonia concentrations are very important as they alter the cloud water pH and hence the solubility of SO₂.Turbulent or ‘occult’ deposition is very sensitive to wind speed, the stability profile of the atmosphere and to the surface roughness. In a supercritical flow regime the occult deposition is a maximum just on the lee of the hill.     

Numerical studies of acidification processes within and below clouds with a flow-through chemical reactor model

A flow-through chemical reactor model has been exercised to assess the importance of various oxidation reactions and cloud processes on wet removal and redistribution of atmospheric pollutants and to investigate the effect of in-cloud acidification on precipitation chemistry at the surface. Preliminary results indicate that in-cloud acidification accounts for more than 60% of the wet deposited acids derived from acidification of initial SO₂, that 42–57% of water-soluble, non-reactive NH₃ and HNO₃ are removed by wet deposition. The pseudo-first-order conversion rate of SO₂ to SO₄²⁻ ranges from 3 to 25% h ⁻¹ depending on initial and boundary conditions.Sensitivity studies have been carried out to test the importance of time evolution of clouds on partitioning of pollutants in the atmosphere and to investigate the variability of precipitation chemistry due to changes in rate constants. The distributions of NH₃ and HNO₃ are found to be dependent largely on the cloud microphysical parameters, while the distributions of H₂O₂ and SO₂ depend largely on initial conditions of both species. Individual physical and chemical mechanisms can determine the overall rate of sulfate wet deposition at different stages of cloud evolution.     

A second generation model for regional-scale transport/chemistry/deposition

The regional-scale transport, chemistry and deposition of acidifying compounds, photochemical oxidants, and their precursors are analyzed using a second-generation Eulerian model. The important atmospheric processes are incorporated using chemical, dynamical and thermodynamical parameterizations having sufficient detail to accommodate boundary layer-free troposphere exchange in cloudy and cloud-free environments, and in-cloud and below-cloud wet removal and chemistry. Forty-one species are considered, many of which are also present in the liquid-drop phases. In the regional scale transport, the advected species are NO, NO₂, SO₂, SO⁻²₄, O₃, HNO₃, NH₃, PAN, H₂O₂, HCHO, alkanes, C₂H₄, other olefins, aromatics, RCHO, ROOH, HNO₂, RONO₂ and RO₂NO₂. The model capabilities are illustrated by showing simulations in which non-precipitating clouds are present to absorb gas-phase species, chemically alter these, and then release them to the atmosphere.     

Sources of hydrogen peroxide in cloudwater

Results of a theoretical investigation of H₂O₂ formation in cloud droplets arising from gaseous HO₂ radical scavenging are presented. It is shown that this process is pH dependent with the maximum rate of H₂O₂ production occurring below pH 3. This dependence arises as a result of the dissociation of HO₂ in water (pK_a = 4.9) and the subsequent disproportionation reaction of HO₂ and O₂⁻ to form hydrogen peroxide. O₂⁻ is also removed by reaction with O₃ to produce OH radicals and this process becomes more competitive as both the pH and $\text{O}{}_2{}^{\mathrm{-}}\text{HO}{}_2$ ratio increase. The presence of soluble organic species, such as aldehydes, in cloudwater counteracts the effect of ozone by converting OH back to HO₂. For low pHs (< 3) the net contribution of organic solutes of H₂O₂ production is predicted to be relatively small, being limited by the availability of OH radicals scavenged from the gas phase. Existing cloud chemistry models may overestimate the rate of aqueous oxidation of formaldehyde by OH radicals.Under conditions where scavenging of gas-phase free radicals by cloud droplets is efficient, uptake of HO₂ radicals may be reversible. The aqueous concentration of OH is unlikely to approach thermodynamic equilibrium with the gas phase (H ∼-30 M atm⁻¹ and can be treated as irreversible. In clouds with a small mean droplet radius, efficient scavenging of precursor OH radicals should result in a decrease in gas-phase HO₂ production with a reduction in the yield of aqueous H₂O₂, although this is offset by the presence of soluble organic species. A similar effect is predicted for clouds with a high liquid water content.The supply of HO₂ and OH radicals to cloud droplets is controlled by gas-phase ozone chemistry which is in turn dependent on the solar u.v. radiation intensity. The u.v. density in clouds may be higher than in clear air when the solar zenith angle is small, thus enhancing H₂O₂ production, but falls off markedly as the solar zenith angle becomes larger. Predicted rates of H₂O₂ formation in clouds based on midday conditions are likely to be considerably higher than the average daytime value, particularly in summer. Diurnal and seasonal effects on H₂O₂ generation are expected to be more marked in clouds than in clear air.     

Measurements of the scavenging of sulfate and nitrate in clouds

Measurements are presented on the scavenging of sulfate and nitrate by cumulus, stratus and strato-cumulus clouds. Assuming that all of the particulate sulfate was in the size range 0.1–1.0 μm radius and that sulfate was scavenged with the same efficiency as sub-micrometer particles in general, the nucleation scavenging coefficient of sulfate in these clouds was deduced to be 0.7 ±0.2 and evidence for sulfate production (1.0±0.3 μgm⁻³) within cloud water was also obtained. Evidence for nitrate scavenging, by nitrates serving as cloud condensation nuclei or by the absorption of HNO₃ by cloud droplets, is also presented. The data suggest that either gaseous nitrogen compounds in the air other than HNO₃ can dissolve and contribute to the nitrate concentration in cloud water or that nitrate can be produced within cloud droplets.     

A model for tropospheric multiphase chemistry: application to one cloudy event during the CIME experiment

A multiphase box model for a remote environment of the troposphere has been developed with an explicit chemistry for both gas and aqueous phase. The model applied to a set of measurements performed by Voisin et al. (2000) during the European CIME experiment for a cloud event on 13th December 1997 at the top of the Puy de Dôme (France). The results of the simulation are compared to the measurements in order to follow the evolution of the ambient chemical composition as a function of the pH and of the varying water content. After verifying that the model retrieves the main features observed in the behavior of species in the cloud droplets, a detailed analysis of the simulated chemical regime is performed. It essentially discusses the sources and sinks of radical in aqueous phase, the relative importance of the oxidation pathways of volatile organic compounds by the main radicals and the conversion of S(IV) into S(VI) which seems to be influenced by the presence of peroxonitric acid, HNO₄, in aqueous phase in the environmental conditions that are considered with low H₂O₂. These numerical results are then compared with the theoretical study from Herrmann et al. (2000), who proposed a slightly different mechanism, including C2 chemistry and transition metal chemistry whereas they neglect some reaction pathways, such as the one involving OHCH₂O₂ radical. This double confrontation between model results and both real experimental data and numerical results from Herrmann et al. (1999c) underlines limitations of such modeling approach that does not include any dynamical or microphysical coupling but also demonstrates its capability to identify the main oxidants or reactants in aqueous phase in real environmental conditions more realistic than a purely theoretical approach. The originality of this study resides in the explicit and exhaustive ways the chemical reactions are treated in aqueous phase and in a first attempt to compare such a detailed chemical scheme to real environmental conditions.     

Numerical simulation of chemical and physical properties of cumulus clouds

Numerical simulations have been carried out with a model consisting of aqueous chemical reactions, and dynamic and microphysical processes of cumulus clouds. The model computes temporal and spatial variations in dynamic parameters, chemical species concentrations, particle spectral evolution, and pH values for drops of different sizes. A preliminary simulation produces dynamic characteristics typical of fairweather cumulus. The in-cloud decrease in SO₂ concentration occurs mainly in the upper part of the cloud, where the decrease approaches 10%, while the corresponding SO₂ decrease in the near-cloud environment is about 2% during the 30 min of real time simulation. The dominant SO₂ oxidation pathway is kinetic oxidation by H₂O₂. For the case simulated, the droplet pH ranges between 3.5 and 5.3 for drops larger than about 20 μm in radius, while the pH of sub-μm evaporating droplets can decrease below 2.0.     

Redistribution of free tropospheric chemical species over West Africa: radicals (OH and HO₂), peroxide (H₂O₂) and acids (HNO₃ and H₂SO₄)

The purpose of this paper is to study the redistribution of chemical species (OH, HO₂, H₂O₂, HNO₃ and H₂SO₄) over West Africa, where the cloud cover is ubiquitously present, and where deep convection often develops. In this area, because of these cloud systems, chemical species are redistributed by the ascending and descending flow, or leached if they are soluble. So, we carry out a mesoscale study using the Regional Atmospheric Modelling System (RAMS) coupled to a code of gas and aqueous chemistry (RAMS_Chemistry). It takes into account all processes under mesh. We examine several cases following the period (November and July), with inputs emissions (anthropogenic, biogenic and biomass burning). The radicals OH and HO₂ are an indicator of possibilities for chemical activity. They characterize the oxidizing power of the atmosphere and are very strong oxidants. The acids HNO₃ and H₂SO₄ are interesting in their transformation into nitrates and sulfates in precipitation. In November, when photochemistry is active during an event of biomass burning, concentrations of chemical species are higher than those of November in the absence of biomass burning. The concentrations of nitric acid double and sulfuric acid increases 70 times. In addition, the concentrations are even lower in July if there is a deep convection. Compared to measures of the African Monsoon Multidisciplinary Analysis (AMMA), the results and observations of radicals OH and HO₂ are the same order of magnitude. Emissions from biomass burning increase the concentrations of acid and peroxide, and a deep convection cloud allows the solubility and the washing out of species, reducing their concentration. Rainfalls play a major role in solubility and washing out acids, peroxides and radicals in this region.     

Coupling quasi-spectral microphysics with multiphase chemistry: a case study of a polluted air mass at the top of the Puy de Dôme mountain (France)

An explicit multiphase chemistry model (Atm. Environ. 34 (29/30) (2000) 5015) has been coupled with quasi-spectral microphysics, based upon Berry and Reinhardt's parameterizations (1974a, b). This coupled model has been initialized with polluted conditions as observed at the Puy de Dôme mountain in the center of France and for a maritime cloud.The presence of clouds results in two effects on multiphase chemistry: a direct effect through mass transfer, solubility and reactivity, and an indirect effect through microphysical transfer from cloud water into rainwater and redistribution of reactive soluble species among interstitial air, cloud droplets and raindrops.Results demonstrate that microphysical processes are necessary to sketch out the complex, nonlinear multiphase chemistry in a real cloud. In addition to the direct exchange through mass transfer, incorporation of reactive oxidants such as HO_x in droplets can arise and consequently make those species no longer available for reacting in the gas-phase. Moreover, microphysical coalescence conversions favor NO_x destruction and enhance the chemical nitric acid production. Coalescence of cloud drops to form rain transfers dissolved species into drops that are undersaturated compared to Henry's law equilibrium. The rain becomes a reservoir for these species, allowing aqueous chemistry to produce more nitric acid than would be possible without the presence of rain.Finally, for the different cloud types, the fate of those intermediate and reactive species is investigated, looking at their budget in clear sky situation versus cloudy and/or rainy situations.     

The dynamics of nitric acid production and the fate of nitrogen oxides

A mathematical model is used to study the fate of nitrogen oxides (NO_x) emissions and the reactions responsible for the formation of nitric acid (HNO₃). Model results indicate that the majority of the NO_x inserted into an air parcel in the Los Angeles basin is removed by dry deposition at the ground during the first 24 h of travel, and that HNO₃ is the largest single contributor to this deposition flux. A significant amount of the nitric acid is produced at night by N₂O₅ hydrolysis. Perturbation of the N₂O₅ hydrolysis rate constant within the chemical mechanism results in redistribution of the pathway by which HNO₃ is formed, but does not greatly affect the total amount of HNO₃ produced. Inclusion of NO₃-aerosol and N₂O₅-aerosol reactions does not affect the system greatly at collision efficiencies, α, of 0.001, but at α = 0.1 or α = 1.0, a great deal of nitric acid could be produced by heterogeneous chemical processes.Ability to account for the observed nitrate radical (NO₃) concentrations in the atmosphere provides a key test of the air quality modeling procedure. Predicted NO₃ concentrations compare well with those measured by Platt et al. (Geophys. Res. Lett.7, 89–92, 1980). Analysis shows that transport, deposition and emissions, as well as chemistry, are important in explaining the behavior of NO₃ in the atmosphere.     

Measurements of the chemical composition of stratiform clouds

Measurements are reported of the chemical composition of the liquid water and interstitial air in warm (> 0°C), non-precipitating stratus and strato-cumulus clouds at various locations in the eastern United States. Inorganic ionic composition of the cloud water was generally dominated by H⁺, NH₄⁺, NO₃⁻ and SO₄²⁻, similar to the composition of precipitation in this region of the U.S. Concentrations of the corresponding interstitial aerosol species and gaseous HNO₃ were invariably low in comparison to concentrations of the respective ionic species in cloudwater. In contrast, the concentration of NO_x (i.e. NO + NO₂ + organic nitrates) was invariably comparable to or in excess of that of cloudwater nitrate. Sulfur dioxide was found at varying concentrations relative to cloudwater sulfate. In many cases, the SO₂ concentration was quite low (< 0.2 ppb) even in the presence of substantial quantities of cloudwater SO₄²⁻ (> 1 ppb equivalent gas-phase concentration), suggesting large fractional conversion and incorporation into cloudwater. In other cases, in which dilute SO₂ plumes (pso, > 5 ppb) were observed in the cloud interstitial air, the gaseous SO₂ concentration substantially exceeded the cloudwater sulfate concentration.Concentrations of H₂O₂ in cloudwater were found to exhibit strong inverse correlation with interstitial SO₂. Appreciable concentrations of SO₂ in cloud interstitial air and H₂O₂ in cloudwater were only rarely observed to coexist, for the most part only one or the other being present above the limit of detection. These observations are consistent with aqueous-phase oxidation of SO₂ by H₂O₂, as has been inferred previously on the basis of laboratory kinetic studies, and with the hypothesis that depending on relative concentrations, either of these species can be a limiting reagent for in-cloud SO₂ oxidation. The uptake of NO_x as cloudwater nitrate is less complete than the uptake of SO₂ as sulfate, and evidence for the occurrence of similar in-cloud processes causing the conversion of NO or NO₂ to cloudwater nitrate has not been found.     

On the difference between SO2−4 and NO−3 in wintertime precipitation

Winter rains have lower NO⁻₃ levels but higher SO²⁻₄ levels than snows in the NE United States. In this study, four years of winter precipitation data from SE Michigan were examined to help understand these differences. Although NO⁻₃ levels were indeed higher in snow than winter rain, the higher concentrations could be attributed to the generally lower precipitation depths associated with snow events than with rain events. The NO⁻₃ concentrations are inversely correlated with precipitation depth. There was no evidence that snow scavenged HNO₃ in the air more efficiently than rain.Conversely, SO²⁻₄ was far higher in winter rain than in snow. This could not be explained in terms of ground-level ambient S concentrations or the wind direction from which the storm originated. However, the cloud temperatures were high enough in the case of rain to suggest that the cloud hydrometeors could have been present as liquid droplets rather than ice crystals. The SO²⁻₄ concentrations of the precipitation were highly correlated with the temperatures of the cloud layers. The data suggest that SO₂ is incorporated and oxidized to SO²⁻₄ in clouds most efficiently when the hydrometeors are present as liquid droplets. The fact that NO⁻₃does not show the same relationship suggests that incorporation of N species into cloud water followed by oxidation is not as important a process for N as for S.     

The contribution of in-cloud oxidation of SO2 to wet scavenging of sulfur in convective clouds

The sensitivity of in-cloud oxidation of SO₂ in corrective clouds to a number of chemical and physical parameters is examined. The parameterization of precipitation growth processes is based on the work of Scott (1978) and Hegg (1983). A chemical model predicts gas and aqueous phase distributions of soluble gases and in-cloud uncatalyzed oxidation of SO₂ by O₃ and H₂O₂. Sulfate aerosol and SO₂, CO₂, NH₃, H₂O₂ and O₃ gases and their aqueous phase dissociation products are treated.The results indicate that in-cloud conversion is an important removal mechanism for SO₂ and accounts for a significant fraction of the precipitation sulfate. However, except at low SO₂ concentrations, the precipitation sulfate concentration is insensitive to the initial SO₂ concentration; the sulfate concentration is most sensitive to the initial H₂O₂ and NH₃ concentrations. At low SO₂ concentrations, the precipitation sulfate concentration is determined primarily by the initial sulfate aerosol concentration. The feedback between sulfate production and pH is important in limiting SO₂ oxidation by O₃. If gas phase H₂O₂ of order 1 ppb is the major source of aqueous phase H₂O₂ for S(IV) oxidation, it is likely that the oxidation reaction is oxidant limited. The sulfate concentration is a decreasing function of the precipitation rate. At low rainfall rates (< 1 mm h⁻¹), ice phase growth decreases the sulfate concentration. However, the results are insensitive to an ice phase origin at moderate and high rainfall rates.     

Simulation of trace species production by lightning and corona discharge in moist air

Numerical simulations of chemical processes initiated in moist air by lightning strokes and lightning coronas indicate that the amounts of major new species generated are not very different from the amounts previously found in dry air. Thus, the major new species produced globally in moist air are estimated to be: 2.4 megatonnes NO y⁻¹, 0.28 megatonnes NO₂ y⁻¹ and 220 tonnes N₂O y⁻¹. The minor species, ozone and those containing hydrogen released by lightning from water in moist air are: 1.2 kilotonnes O₃ y⁻¹, 25 kilotonnes HNO₂ y⁻¹, 1.2 kilotonnes HNO₃ y⁻¹, 0.32 kilotonnes H₂O₂ y⁻¹ and 0.32 kilotonnes HO₂ y⁻¹.     

Cloud physics and cloud water sampler comparison during FEBUKO

Optical methods for counting and sizing cloud droplets and a wide range of cloud water sampling methods were used to characterize the atmospheric liquid phase during the FEBUKO cloud experiments. Results near cloud base as well as more than 300 m inside the hill cap clouds are presented, reflecting their inhomogeneous nature. The cloud droplet number varies from 50 to 1000 cm⁻³ and drop sizes between 1 and 20 μm diameter are most frequent. Variations in the liquid water content (LWC) and in the total ion content (TIC) are much smaller when the measurement position is deeper in the cloud. Near cloud base variability in updraft strength and, near cloud top, entrainment processes (droplet evaporation by mixing with drier air, aerosol and gas scavenging) disturb the adiabatic conditions and produce large variations in LWC and chemical composition. Six different active cloud water collectors and impactors were running side by side; they differ in the principle of sampling, in the throughput of cloudy air per unit time and in the calculated 50% cutoff diameter, which influence also their sampling efficiency. Two of them are designed to collect cloud water in two droplet size fractions. Three cloud events were selected by the FEBUKO team for detailed cloud physical and chemical analyses because they serve best the modelling demands concerning connected flow between the upwind, summit and downwind sites for process studies. Frequency distributions of the LWC and, also of the cloud base height are given as statistical parameters for both FEBUKO experiments.     

Halogen production from aqueous tropospheric particles

Box model studies have been performed to study the role of aqueous phase chemistry with regard to halogen activation for marine and urban clouds and the marine aerosol as well. Different chemical pathways leading to halogen activation in diluted cloud droplets and highly concentrated sea salt aerosol particles are investigated. The concentration of halides in cloud droplets is significantly smaller than in sea-salt particles, and hence different reaction sequences control the overall chemical conversions. In diluted droplets radical chemistry involving OH, NO(3), Cl/Cl(2)(-)/ClOH(-), and Br/Br(2)(-)/BrOH(-) gains in importance and pH independent pathways lead to the release of halogens from the particle phase whereas the chemistry in aerosol particles with high electrolyte concentrations is controlled by non-radical reactions at high ionic strengths and relatively low pH values.For the simulation of halogen activation in tropospheric clouds and aqueous aerosol particles in different environments a halogen module was developed including both gas and aqueous phase processes of halogen containing species. This module is coupled to a base mechanism consisting of RACM (Regional Atmospheric Chemistry Mechanism) and the Chemical Aqueous Phase Radical Mechanism CAPRAM 2.4 (MODAC-mechanism). Phase exchange is described by the resistance model by Chemistry of Multiphase Atmospheric Systems, NATO ASI Series, 1986.It can be shown that under cloud conditions the bromine atom is mainly produced by OH initiated reactions, i.e. its concentration maximum is reached at noon. In contrast, the concentration level of chlorine atoms is linked to NO(3) radical chemistry leading to a smaller amplitude between day and night time concentrations.The contribution of radical processes to halogen atom formation in the particle phase is evident, e.g. by halogen atoms which undergo direct phase transfer. Furthermore, the application of the multiphase model for initial concentrations for sea-salt aerosols shows that the particle phase can act as a main source of halogen containing molecules (Cl(2), BrCl, Br(2)) which are photolysed in the gas phase to yield halogen atoms (about 70% of all Cl sources and more than 99% for Br).     

An Eulerian model for scavenging of pollutants by raindrops

An Eulerian model for simulating the coupled processes of gas-phase depletion and aqueousphase accumulation of the pollutant species during a rain event has been formulated. The model is capable of taking into account any realistic vertical profile of pollutant species concentrations and time-dependent initial aqueous-phase concentrations at the cloud base. The model considers the processes of single species absorption and dissociation in the aqueous phase. The coupled partial differential equations constituting the model are discretized into a set of ordinary differential equations by using the Galerkin method with chapeau functions as the basis functions. These equations are solved to obtain the pollutant concentrations of the gas phase and raindrops as well as the pH of raindrops as a function of time and distance below cloud-base.Simulations are performed for scavenging of gaseous HNO₃, H₂O₂, SO₂, formaldehyde and NH₃. For the case of highly soluble HNO₃ and H₂O₂, raindrops are far from equilibrium with the gas phase and their capacity for absorption of these gases is undiminished even as they reach ground level. The gas-phase concentrations for these species decrease exponentially with time and the washout is determined primarily by the rain intensity and mass-transfer coefficient of the gaseous species to the raindrops. The pollutant species concentrations in raindrops are an almost linear function of the distance below the cloud base. For the simulation conditions considered in this study, the half-life periods of these gases for removal from the atmosphere range from 15 to 40 min.For SO₂ and formaldehyde, the aqueous-phase concentrations approach equilibrium as the drops fall to ground level and the gas-phase concentrations show large gradients in the vertical. Half-life periods for SO₂ range from 1.3 to 13 h depending on the initial raindrop pH and rain intensity. For formaldehyde, the half-life ranges from 19 to 63 min.Solubility of NH₃ is a strong function of the raindrop pH. As NH₃ is absorbed, the raindrop pH increases and NH₃ solubility decreases. For pre-acidified drops (pH = 4.6), ammonia solubility is very high and the drops are far from equilibrium with the gas phase throughout the falling period. The half-life for ammonia ranges from 11 min to over 3 h in our simulations.     

Aqueous h2o2 production from o3 in glass impingers

The aqueous generation of H₂O₂ from bubbling ozonized air through water in glass impingers is investigated. O₃ loss is monitored throughout each experiment. Aqueous H₂O₂ is measured at the end of each experiment. The mole quantities of O₃ lost (ΔO₃) and H₂O₂ formed (ΔH₂O₂) are calculated. There is a stoichiometric relationship between ΔO₃ and ΔH₂O₂ which is independent of bubbling time, solution acidity and initial O₃ concentration but is dependent upon the liquid water content of each impinger. For 10 ml of water in each impinger, the AO₃ to ΔH₂O₂ mole ratio is between 2:1 and 3:1 and for 20 ml is between 1:1 and 1:2. Increasing the solution acidity increases the rates of O₃ loss and H₂O₂ generation. This result contradicts mechanisms in which a O₃ + HO⁻ reaction is used to initiate radical formation and H₂O₂ production.These results are interpreted through a combination of glass surface chemistry and bulk aqueous chemistry. The O₃-H₂O₂ impinger mechanisms involve O₃ adsorption and/or reaction on the impinger surface and its surface products subsequent reaction with water or trace impurities in the water. Experiments using tetrafluorethylene (TFE) impingers in place of glass impingers are conducted to examine the role of surface chemistry. The TFE impingers are found to substantially increase ΔO₃ and ΔH₂O₂ without changing their mole ratio. This is attributed to an enhanced O₃-H₂O surface chemistry or O₃ reaction with unsaturated Teflon linkages or impurities in the Teflon from machining.     

Modeling study of the potential importance of heterogeneous surface reactions for NOx transformations in plumes

A parametric model of gas-particle surface reactions is incorporated into a reactive plume model and used to assess the potential importance of heterogeneous surface reactions on gas phase plume chemistry. Heterogeneous loss of the following species is found to be potentially significant: H₂O₂, PAN, NO₃, N₂O₅, OH, HO₂ and in cold weather HO₂NO₂. This simple model is unable to account for equilibrium/capacity effects in the condensed phase and therefore cannot be used for SO₂ and HNO₃ reaction with aerosol surface.