Patrice Coddeville

A study of the source-receptor relationships influencing the acidity of precipitation collected at a rural site in France

In order to examine the qualitative and quantitative source–receptor relationships responsible for acid rains at a background site in France, a receptor-oriented model was applied to the precipitation data collected from 1992 to 1995. Origins of acidic and alkaline species in precipitations have been investigated. The methodology combines precipitation chemical data with air parcel backward trajectories to establish concentration field maps of likely contributing sources. Highest acidities and concentrations of sulfate and nitrate in precipitation were associated with transport from the high emission areas of central Europe. Alkaline events were associated with air masses originating from Mediterranean basin or northern Africa. The quantitative relationships between the maps of potential sources and the European emissions of SO2 and NOx were examined performing a correlation analysis. Good correlations were found between computed concentrations of acidic species and emissions of SO2 and NOx. Substantial seasonal variations of acidic species were revealed. The highest concentrations occurred during the warm season. These seasonal variations are the effect of change of meteorological conditions and of the strength atmospheric processes according to the season.

Intercomparison between three receptor-oriented models applied to acidic species in precipitation

Abstract Three different receptor-oriented models have been tested on the precipitation chemical data of a background station in France from 1992 to 1995. These models which allowed the identification of source–receptor relationships for acidic species (non-sea salt SO 4 2− , NO 3 − and NH 4 + ) in rain have been compared. The receptor-oriented methodologies are based on the same construction scheme. Precipitation data and air parcel backward trajectories were combined in the model calculations to produce probability or concentration maps of likely contributing sources. The regions of Europe identified by the three models generally coincide well with the location of known emission sources. Using non-parametric Spearman rank-order correlation, the computed concentrations provided by these three models were compared with the European emissions established by EMEP. A good correlation was found between computed ionic concentrations and the emissions of SO 2 , NO x and NH 3 . Each model shows its own advantages to examine the source regions that give high concentrations of acidic species at a receptor site.

Spatial variability and source identification of rural precipitation chemistry in France

Abstract The precipitation chemistry in France has been examined for large-scale spatial variability using the MERA/WMO-GAW network of 13 rural sites equipped with wet-only collectors. Three classes of mean chemical compositions of MERA sites, corresponding to three different geographical regions in France have been identified by applying hierarchical clustering analysis. Factor analysis has also been performed on the deposition and concentration data sets for the three MERA sites. All the factors could be interpreted as falling into one of four categories: Acid, Sea, Neutralization, or an Agriculture/Soil association. Both sulfate and nitrate were found to be significant components of the acid factor for two stations (Morvan and Donon). These individual site analyses indicate that the sea has a strong influence on the precipitation chemistry at these continental stations. A comparison between factor analysis of concentrations and factor analyses of depositions suggests on the one hand that the acid forms, H 2 SO 4 and HNO 3 , seem preferentially present in cloud and on the other hand that the neutralization of precipitation acidity by soil-derived particles and by ammonia emissions seems to depend on the geographical and climatic characteristics of a site (altitude, proximity to human activity and precipitation amount).

Atmospheric Chemistry of 2,3-Pentanedione: Photolysis and Reaction with OH Radicals

The kinetics of the overall reaction between OH radicals and 2,3-pentanedione (1) were studied using both direct and relative kinetic methods at laboratory temperature. The low pressure fast discharge flow experiments coupled with resonance fluorescence detection of OH provided the direct rate coefficient of (2.25 ± 0.44) × 10(-12) cm(3) molecule(-1) s(-1). The relative-rate experiments were carried out both in a collapsible Teflon chamber and a Pyrex reactor in two laboratories using different reference reactions to provide the rate coefficients of 1.95 ± 0.27, 1.95 ± 0.34, and 2.06 ± 0.34, all given in 10(-12) cm(3) molecule(-1) s(-1). The recommended value is the nonweighted average of the four determinations: k(1) (300 K) = (2.09 ± 0.38) × 10(-12) cm(3) molecule(-1) s(-1), given with 2σ accuracy. Absorption cross sections for 2,3-pentanedione were determined: the spectrum is characterized by two wide absorption bands between 220 and 450 nm. Pulsed laser photolysis at 351 nm was used and the depletion of 2,3-pentanedione (2) was measured by GC to determine the photolysis quantum yield of Φ(2) = 0.11 ± 0.02(2σ) at 300 K and 1000 mbar synthetic air. An upper limit was estimated for the effective quantum yield of 2,3-pentanedione applying fluorescent lamps with peak wavelength of 312 nm. Relationships between molecular structure and OH reactivity, as well as the atmospheric fate of 2,3-pentanedione, have been discussed.

A Comparison of Precipitation Sensors Used on the Wet-Only Collectors

The variability in performance of three precipitation sensor types, mounted on three identical wet-only collectors, has been compared for eight months at a single site. The catch efficiencies, determined from the sample volumes, are relatively high for these three apparatuses, since they exceed 89% on average. The recorded data of openings and closings of collector lids reveal that the highest losses of collection efficiency, observed for rainfalls of low intensity and low amount, are largely imputable to design and running characteristics of each sensor. Ionic compositions between the samples of these three collectors are relatively close, although significant differences are especially found for H+, Ca2+, Mg2+ and K+, suggesting that these apparatuses are differently exposed to the dry deposition of soil particles. The RS 85 sensor seems to be the most suitable one of the three, since it ensures a high catch efficiency for all rain types and tends to limit the exposures of funnel to the dry deposition.

First direct detection of HONO in the reaction of methylnitrite (CH3ONO) with OH radicals.

We report on the development of a new environmental simulation chamber coupled with an in situ continuous wave cavity ring-down spectrometer operating in the near IR (∼1.5 μm). The first application reported in this paper dealt with the chemical mechanism of UV photolysis of methyl nitrite (CH(3)ONO) in air. HONO has been detected for the first time and shown to be formed in the OH + CH(3)ONO reaction. A dense spectrum of cis-HONO absorption lines has been observed near 1.5 μm, in agreement with a previous study (Guilmot et al.). CH(2)O has been measured as primary product with good sensitivity and time resolution. In contrast to Zhao et al., we did not detect any NO(2) absorption features in this wavelength range. Calibration experiments provided very low NO(2) absorption cross sections in this region (∼10(-25) cm(2)), leading to conclude that NO(2) cannot be observed in this wavelength range in the presence of equal amounts of CH(2)O.

Low-Pressure Photolysis of 2,3-Pentanedione in Air: Quantum Yields and Reaction Mechanism

Dicarbonyls in the atmosphere mainly arise from secondary sources as reaction products in the degradation of a large number of volatile organic compounds (VOC). Because of their sensitivity to solar radiation, photodissociation of dicarbonyls can dominate the fate of these VOC and impact the atmospheric radical budget. The photolysis of 2,3-pentanedione (PTD) has been investigated for the first time as a function of pressure in a static reactor equipped with continuous wave cavity ring-down spectroscopy to measure the HO2 radical photostationary concentrations along with stable species. We showed that (i) Stern-Volmer plots are consistent with low OH-radical formation yields in RCO + O2 reactions, (ii) the decrease of the photodissociation rate due to pressure increase from 26 to 1000 mbar is of about 30%, (iii) similarly to other dicarbonyls, the Stern-Volmer analysis shows a curvature at the lower pressure investigated, which may be assigned to the existence of excited singlet and triplet PTD states, (iv) PTD photolysis at 66 mbar leads to CO2, CH2O and CO with yields of (1.16 ± 0.04), (0.33 ± 0.02) and (0.070 ± 0.005), respectively, with CH2O yield independent of pressure up to 132 mbar and CO yield in agreement with that obtained at atmospheric pressure by Bouzidi et al. (2014), and (v) the PTD photolysis mechanism remains unchanged between atmospheric pressure and 66 mbar. As a part of this work, the O2 broadening coefficient for the absorption line of HO2 radicals at 6638.21 cm(-1) has been determined (γO2 = 0.0289 cm(-1) atm(-1)).

Investigation of the Gas-Phase Photolysis and Temperature-Dependent OH Reaction Kinetics of 4-Hydroxy-2-butanone

Hydroxyketones are key secondary reaction products in the atmospheric oxidation of volatile organic compounds (VOCs). The fate of these oxygenated VOCs is however poorly understood and scarcely taken into account in atmospheric chemistry modeling. In this work, a combined investigation of the photolysis and temperature-dependent OH radical reaction of 4-hydroxy-2-butanone (4H2B) is presented. The objective was to evaluate the importance of the photolysis process relative to OH oxidation in the atmospheric degradation of 4H2B. A photolysis lifetime of about 26 days was estimated with an effective quantum yield of 0.08. For the first time, the occurrence of a Norrish II mechanism was hypothesized following the observation of acetone among photolysis products. The OH reaction rate coefficient follows the Arrhenius trend (280-358 K) and could be modeled through the following expression: k4H2B(T) = (1.26 ± 0.40) × 10(-12) × exp((398 ± 87)/T) in cm(3) molecule(-1) s(-1). An atmospheric lifetime of 2.4 days regarding the OH + 4H2B reaction was evaluated, indicating that OH oxidation is by far the major degradation channel. The present work underlines the need for further studies on the atmospheric fate of oxygenated VOCs.

First Cavity Ring-Down Spectroscopy HO2 Measurements in a Large Photoreactor

Abstract The HO2 radical is one of the most important intermediate species in atmospheric chemistry. We report on the development of a new photoreactor with first in-situ measurement of HO2 radical photostationary concentrations using continuous wave cavity ring-down spectrometry (cw-CRDS). Characterization of the actinic photon flux was carried out by NO2 actinometry. Photolysis of Cl2/methanol mixtures in air under UV light allowed the measurement of HO2 photostationary concentrations of a few 1010 molecules cm-3 with an HO2 detection limit of 1.5 × 1010 molecules cm-3 at 6638.207 cm-1. The feasibility of HO2 direct measurement in a reaction chamber is demonstrated through the measurement of the HO2 overall loss at different pressures showing the importance of HO2 diffusion and wall loss in such low pressure quartz reactor. The rate coefficient for the HO2+HO2 reaction has been measured at 6.6, 24 and 118 mbar and found to be in good agreement with the recommended value.

Dynamic Validation Procedure for the Mera and French WMO-GAW Precipitation Chemistry Networks

This article presents the setting-up of a validation procedure, especially adapted to the rainwater samples of the MERA and French WMO-GAW networks. A first study reveals that the ion and conductivity balances, usually used as quality criteria, depend on salinity. The statistical tool or “physico-chemical filter” proposed in this paper, includes two mathematical models based on 97th percentile of the ion and conductivity balances, from the 1991 and 1992 data taken as references. This dynamic validation uses the rejection limits which increase with salinity. Its application shows a significant decrease of the total rejection rate between the first and third year of running. This seems to be due to the improvement in quality of the analytical line. The rejected sample inspection does not reveal apparent bias errors for the filter. This kind of validation seems to be well adapted to screen the samples in a precipitation collection network such as MERA and WMO-GAW.