UNIVERSITÉ DE PROVENCE LABORATOIRE CHIMIE PROVENCE CNRS / UMR 6264
ÉQUIPE INSTRUMENTATION ET RÉACTIVITÉ ATMOSPHÉRIQUE
ETUDE DE LA REACTIVITE HETEROGENE DES COMPOSES ORGANIQUES ISSUS DE LA COMBUSTION DE LA BIOMASSE
Sopheak NET
Thèse de doctorat de l’Université de Provence Spécialité Chimie de l’Environnement École doctorale des Sciences de l’Environnement
Soutenue le 03 décembre 2010 devant le jury composé de : Denis Petitprez Abdelwahid Mellouki Sylvia Pietri Barbara D’Anna Henri Wortham Sasho Gligorovski
Professeur (PC2A, Lille) Directeur de recherche (ICARE, Orléans) Directrice de recherche (LCP, Marseille) Charge de recherche (IRCELYON, Lyon) Professeur (LCP-IRA, Marseille) MCF (LCP-IRA, Marseille)
Rapporteur Rapporteur Examinateur Examinateur Directeur de thèse Directeur de thèse
REMERCIEMENTS
Mes remerciements s’adressent d'abord aux rapporteurs de ce travail : Messieurs Petitprez Denis et Mellouki Abdelwahid et aux membres du jury : Mesdames Pietri Syvia et Barbara D’Anna pour le temps passé à examiner ma thèse et pour avoir accepté de participer à ma soutenance.
J’adresse mes plus vifs remerciements à Monsieur le Professeur Henri Wortham et au Docteur Sasho Gligorovski, pour avoir encadré ces travaux avec confiance, enthousiasme, disponibilité et rigueur. Ainsi, j’ai pu réaliser ce travail dans des conditions matérielles excellentes et présenter mes résultats dans plusieurs congrès, nationaux et internationaux. Je leur suis particulièrement reconnaissante d’avoir rendu ces trois années enrichissantes, épanouissantes et surtout passionnantes.
J’adresse également mes remerciements à l’Agence National de la Recherche (ANR) et à l’Université de Provence pour avoir financé une partie ces travaux de thèse.
Je suis particulièrement reconnaissante envers mon parrain, le Docteur Pascal David, ancien chercheur au CNRS et chargé de relations internationales, pour ses précieux conseils, ses encouragements dans mes activités universitaires et pour m’avoir soutenue financièrement et moralement dans ma vie quotidienne lors de mes trois années de thèse.
Je voudrais aussi exprimer ma gratitude envers le Docteur Brice Temime-Roussel et le Docteur Nicolas Marchand pour leurs nombreux conseils avisés, leurs nombreux dépannages au laboratoire et pour avoir su me faire profiter de leurs grandes expériences en chimie analytique et expérimentale. Je les remercie très vivement pour leur grande disponibilité et leur bonne humeur.
Je remercie également tous les membres du Laboratoire Chimie Provence, les étudiants, les permanents et les stagiaires qui ont rendu l’atmosphère de travail agréable, chaleureuse et propice à des discussions intéressantes et innovantes.
1
Je tiens également à manifester toute ma reconnaissance à mes amis doctorants. Il ne m’est pas possible de rapporter, ici, tous leurs noms, mais je citerai parmi eux, Mademoiselle Yao Liu, Monsieur Imad El Haddad, Mademoiselle Audrey Manoukian, Mademoiselle Sabrine Tlili, Mademoiselle Ehgere Abidi et Mademoiselle Aude Vesin. Avec eux j’ai pu avoir des discussions passionnantes et des échanges d’idées à la fois professionnelles et culturelles. Je les remercie d’avoir partagé mes joies et mes humeurs.
Je voudrais aussi exprimer ma gratitude envers mes amis bretons qui m’ont acceuillie chaleureusement depuis mon arrivée en France et qui m’ont toujours soutenue. Parmis eux, je citerai Monsieur le Professeur Pierre Le Corre et sa femme Paule, le Docteur Pascal Morin, le Docteur Essyllt Louarn, Mademoiselle Virginie Tanguy, Mademoiselle Natalie BeauMonvoisin, Mademoiselle Virginie Delagrandrie et Mademoiselle Raissa Jeanne Zang.
Enfin, j’aimerais également remercier très sincèrement mes parents, mon grand frère et toute ma famille pour m’avoir soutenue et encouragée dans tous les domaines et m’avoir témoignée leur confiance et leur fierté. J’adresse aussi mes vifs remerciements à ma marraine, le Docteur Sylvie De Boyer et ma marraine « Doctoriales », le Docteur Michèle Durand Pinchard, d’avoir été à mes côtés pour ces moments difficiles…
2
Résumé La combustion de la biomasse est une des principales sources de matières organiques fines ( 300 nm) during the heterogeneous ozonolysis on organic coated particles. The reaction products identified in this study (3,4,5-trimethoxybenzoic acid, syringic acid, methyl 3,4,5-trimethoxybenzoate) absorb light in the spectral window (l > 300 nm) which implies that lightinduced heterogeneous ozone processing can have an influence on the aerosol surfaces by changing their physico-chemical properties. The main identified product of the heterogeneous reactions between gas-phase ozone and 3,4,5-trimethoxybenzaldehyde under dark conditions and in presence of light was 3,4,5-trimethoxybenzoic acid. For this reason we estimated the carbon yield of 3,4,5-trimethoxybenzoic acid. Carbon yields of 3,4,5trimethoxybenzoic acid decreased with increasing ozone mixing ratio; from 40% at 250 ppb to 15% at 2.5 ppm under dark conditions. At ozone mixing ratio (250 ppbe1 ppm), carbon yields of 3,4,5-trimethoxybenzaldehyde are relatively higher in the experiment under dark condition than under simulated solar light. Ó 2010 Elsevier Ltd. All rights reserved.
Keywords: Heterogeneous chemistry Photodegradation Aromatic carbonyl compounds Aerosol Surface reaction GCeMS
1. Introduction Wood combustion has been identified as a non fossil fuel source (biofuel) that can contribute significantly to the deterioration of both outdoor and indoor air quality (Perzon, 2010; Hawthorne et al., 1992; Honicky et al., 1985; Bari et al., 2009; Standley and Simoneit, 1987). Biomass combustion is one of the major sources of fine organic materials ( 290 nm) (Anastasio et al., 1997) so the direct and indirect photolysis process can be a significant factor of its aging. Recently, there has been a surge of interest towards heterogeneous ozone and OH processing on atmospheric aerosol surfaces (Kwamena et al., 2007; Mmereki and Donaldson, 2003; McNeill et al., 2007; McIntire et al., 2005; Kahan et al., 2006; Vlasenko et al., 2008; Bertram et al., 2001; Mmereki et al., 2004; Perraudin
S. Net et al. / Atmospheric Environment 44 (2010) 3286e3294
et al., 2005, 2007; Pflieger et al., 2009) but only few studies were focused on the light-induced heterogeneous reactions (Park et al., 2006; Gomez et al., 2006; Jammoul et al., 2008; Nieto-Gligorovski et al., 2008; Net et al., 2009, 2010a, 2010b). In the present work, the kinetic measurements for the heterogeneous reactions of gas-phase ozone with 3,4,5-trimethoxybenzaldehyde adsorbed on silicon oxide were performed for a time period between 1 and 8 h both in dark and under simulated solar light. The condensed-phase products which emerged in such heterogeneous reactions were analyzed by gas chromatographymass spectrometry (GC/MS). 2. Materials and methods 2.1. Chemicals and reagents The chemicals used were supplied by SigmaeAldrich; their name and their stated purities were as follows: 3,4,5-trimethoxybenzaldehyde (98%), methyl 3,4,5-trimethoxybenzoate (98%), 3,4,5-trimethoxybenzoic acid (99%), syringic acid (97%). The SiO2 powder AEROSILÒR812 with purity 99.8%, average size of 7 nm and specific surface of 260 30 m2 g1 were supplied by Evonik (France).
3287
2.3. GCeMS analysis The filtered solutions were analyzed by GCeMS using electron impact ionization mode (70 eV) according to the following parameters: column THERMO TR-5MS (internal diameter 0.25 mm, length 30 m, film thickness 0.25 mm), injection volume: 1 ml, inlet temperature: 250 C, interface temperature: 330 C, with the following temperature program: hold 1 min at 80 C; increase temperature to 220 at a rate 15 C min1; increase temperature to 300 at 25 C min1; hold 15 min at 300 C. A Combi PAL autosampler was used to allow automated analysis. For product identification and kinetic study, the samples were injected in the splitless and in the split10 injection mode, respectively. Spectra were taken before and after the processing of the organic coatings. These spectra were obtained using GCeMS coupled to a direct analysis and a derivatization analytical technique as reported in our previous paper (Net et al., 2009). The organic compounds formed during the heterogeneous reactions of gas-phase ozone with 3,4,5-trimethoxybenzaldehyde adsorbed on silica particles were identified. The later was done preliminary based on characteristic fragments and on library mass spectra (NIST MS search version 1.7, Ó 1998). In such way identified products were confirmed by double standard analysis (direct and after derivatization with BSTFA analysis).
2.2. Experimental 3. Results and discussion To investigate the photodegradation of 3,4,5-trimethoxybenzaldehyde, this aromatic carbonyl compound as was coated silicon oxide particles via liquidesolid adsorption. A brief description is given bellow. 38 mg of 3,4,5-trimethoxybenzaldehyde in 100 ml of dichloromethane was mixed with 1.5 g of SiO2 powder, in a pyrex bulb with a volume of 500 cm3. This bulb was wrapped with the aluminium foil and left in the ultrasonic bath during 30 min to get the homogeneous particles and then attached to a rotary evaporator where the particles were dried approx. 60 min at 40 C and 850 85 mbar. The prepared particles were then additionally dried by flowing nitrogen gas about 15 min prior to experiments. All experiments were performed at 297 K. About 300 mg of obtained particles were transferred into other pyrex bulb for exposition to different ozone mixing ratios i.e. 0 ppb, 250 ppb, 1 ppm, 3 ppm and 6 ppm. The latter was done under dark conditions and in presence of simulated solar light. The increase of ozone mixing ratio up to 6 ppm was essential to obtain a significant ozonolysis of 3,4,5-trimethoxbenzaldehyde in the time frame of the experiment which enabled us to clearly identify the reaction products. The dried coated particles (300 mg) were exposed to the simulated sunlight emitted from a broadband continuous light source such as a xenon lamp (700 W m2 for 315e400 nm and 160 W m2 for 400e700 nm) at distance of 10 cm from the bulb. The rotation of the bulb ensured a homogeneous illumination of the particle’s surface during the whole experiment. The production of ozone was performed by use of commercial ozone generator (UVP, LLC Upland, UK) which generate ozone via photolysis of O2 performed by mercury penray lamp. Ozone diffusion limitations within the reactor and powder sample could result in a nonuniform exposure of the particles to ozone. However, we have ruled this possibility out (see the later section Method validation). After a reaction time experiment ranging from 1 to 8 h, the organic compounds adsorbed on the surface of silica particles were desorbed in dichloromethane for 30 min by sonication (Branson 3510, USA). Then the prepared suspension of the coated particles was centrifuged in order to separate the phases (Sorvall LEGEND MICRO17, Electron Corp., TermoFisher). Finally, the clear solutions were analyzed by GCeMS for the quantification and product identification.
3.1. GCeMS experiment data 3.1.1. Reaction products The reaction samples were analyzed by GCeMS directly and after derivatization with bis(trimethylsilyl)-trifluoro-acetamide (BSTFA). The GCeMS analytical technique linked with BSTFA as a derivatization agent allows identification of organic compounds which contain eOH and/or eCOOH moieties. Indeed, polar compounds can develop strong interactions with the stationary phase in the chromatographic column, leading to lower sensitivity and resolution. Furthermore, the quantities of some reaction products formed during the reaction are very low so they could not be detected by direct analysis. The applied derivatization decreased the polarity of the compounds and enhanced their volatility. The total ion chromatograms obtained after 8 h of ozonolysis of 3,4,5trimethoxybenzaldehyde (dotted line) and 8 h of simultaneous ozone and light exposure of 3,4,5-trimethoxybenzaldehyde (full line) are illustrated in Fig. 1. Numbering in Fig. 1B corresponds to the organic compounds listed in Table 1. Table 1 depicts the identified organic compounds following the derivatization of the reacted samples. When samples of 3,4,5-trimethoxybenzaldehyde were derived with BSTFA, acetal artifacts were observed. This type of artifacts which were already reported and discussed in our previous articles (Net et al., 2010b) are most probably formed by the reaction of BSTFA with the gem-diol (hydrate) of the aldehyde function to form a bis(trimethylsiloxy)acetal (Little, 2003). During the heterogeneous reactions of ozone with 3,4,5-trimethoxybenzaldehyde adsorbed on silica particles, many oxidation products were detected. Two of these products i.e. methyl 3,4,5trimethoxybenzoate and syringic acid were identified and confirmed with their original standards. These two products resulted from the gain or loss of methyl group. It is noteworthy that syringic acid was detected only in the experiments performed with simultaneous ozonolysis and light illumination of the organic coated particles. Fig. 2 shows the mass spectra of methyl 3,4,5-trimethoxybenzoate (panel A) and mass spectra of trimethylsilyl (TMS) derivative of syringic acid (panel B).
3288
S. Net et al. / Atmospheric Environment 44 (2010) 3286e3294
A 7e+7
Total Ion Current
6e+7 5e+7
Reagent
4e+7 3e+7 2e+7 Mw=212 without OH
1e+7 0 9,0
9,5
10,0 10,5 Retention Time (min)
B
11,0
5
1,2e+8
Total Ion Current
1,0e+8 4
8,0e+7 2
6,0e+7 4,0e+7 2,0e+7 1
3
6
0,0
9,0
9,5
10,0
10,5
11,0
11,5
Retention Time (min) Fig. 1. The total ion chromatogram taken after 8 h of ozonolysis in dark (dotted line) and after 8 h of simultaneous ozone and light (solid line) exposure of the particles coated with 3,4,5-trimethoxybenzaldehyde. A) Corresponds to direct GCeMS analysis and B) shows the total ion chromatogram after derivatization with BSTFA. The numbering of the compounds corresponds to the numbers in Table 1.
Although many reaction products were detected, the principle way of (photo)oxidation is the transformation of the aldehyde’s function to acid function. In this manner, two major products of 3,4,5-trimethoxybenzaldehyde containing the same molecular mass, 212 g mol1 were detected. One of the products, was identified as 3,4,5-trimethoxybenzoic acid (RT ¼ 10.75 min) and confirmed with the standard compound. Fig. 3 shows the mass spectra of 3,4,5-trimethoxybenzoic acid obtained by direct analysis (panel A) and via derivatization (panel B). The other detected product which emerged at retention time 9.25 min is unknown compound. The mass spectrum of this unknown compound with its suggested structure is illustrated in Fig. 4. Indeed, the GCeMS direct analysis revealed a compound with m/z ¼ 212. Its fragment ion 184 (Mþ-28) in the mass spectra (Fig. 4)
Fig. 2. Mass spectra of (A) methyl 3,4,5-trimethylbenzoate and (B) of TMS derivative of syringic acid.
represented the probable loss of CO. This unidentified product contains 16 amu higher than the molecular weight of 3,4,5-trimethoxybenzaldehyde. This surplus mass corresponds to one atom of oxygen which is added to 3,4,5-trimethoxybenzaldehyde. The characteristic fragment of the GCeMS signal exhibited that this compound did not contain OH function (eOH or eCOOH) i.e. its intensity signal seen before and after derivatization with BSTFA remained unchanged. In the chromatogram (Fig. 1, Panel B) it can be seen that the peak at RT ¼ 10.80 min which corresponds to compound with m/z 284. It seems that this peak (RT ¼ 10.80) is the corresponding mono-TMS derivative of the unidentified compound detected in the direct analysis at RT ¼ 9.27 min (Fig. 1, Panel A). Indeed, Little (2003) has reported that ketones with a-hydrogens react to form artifacts through their enol-form. So this unknown compound (m/z 212) which contains CO, CH3 and C]O can be suggested as 4-hydroxy-3-methoxy-6-oxocyclohexa-1,4-diene-1carbaldehyde but this tentative suggestion cannot be confirmed by the standard analysis because its corresponding standard is not available.
Table 1 The identified oxidation products during the reactions between ozone and 3,4,5-trimethoxybenzaldehyde. Number
RT (min)
Mw
1 2 3 4 5 6
9.26 9.55 10.39 10.85 11.01 11.11
212 196 226 196 212 198
Mw-TMS
Functions
Identification
Standard confirmed
358 284 342
no OH no OH no OH 2 OH COOH OH þ COOH
Unknown 3,4,5-Trimethoxybenzaldehyde (reagent) Methyl 3,4,5-trimethoxybenzoate Artefact (gem-diol of reagent) 3,4,5-trimethoxybenzoic acid Syringic acid
no yes yes yes yes yes
S. Net et al. / Atmospheric Environment 44 (2010) 3286e3294
Abundance (arbitrar y unit)
A 2,1e+6
m/z 212
1,8e+6 m/z 197
1,5e+6 1,2e+6 9,0e+5
3,0e+5
m/z 141
m/z 93
6,0e+5
m/z 65
m/z 11
m/z 154 m/z 169
0,0 50
100
150 200 250 Mass/Charge ratio
300
350
400
B 2,5e+7 Abundance (arbitrary unit)
m/z 269
2,0e+7
1,5e+7
m/z 225
1,0e+7
3.1.2. Carbon balance of the reaction The total carbon yield is the sum of the carbon yields of each quantified reaction product. The calculation of the carbon yield of primary reaction product was reported previously (Liu et al., 2009; Net et al., 2010b). For each primary reaction product (i), the carbon yield can be determined as follows:
5,0e+6 m/z 243
m/z 151
0,0 50
100
150 200 250 Mass/Charge ratio
300
350
400
carbon yeildðiÞ ¼
Fig. 3. Mass spectra of 3,4,5-trimethoxybenzoic acid: A) direct analysis, B) after BSTFA derivatization analysis.
In the past, many studies were focused on ozonation mechanism of aromatic compounds in solution (Mvula and von Sonntag, 2003; Komissarov and Zimin, 2006) but the ozonolysis concerning the heterogeneous reactions are very scarce. Perraudin et al. (2007) proposed a general mechanism for the heterogeneous reactions of gas-phase ozone with anthracene and phenanthrene adsorbed on silica particles. Perraudin et al. (2007) reported that ozone reacts with aromatic compounds leading either to a substitution (atom attack) or to a ring opening (bond attack). Atkinson and Arey (2003) proposed the reaction mechanism of ozone with biogenic volatile organic compounds which contain C]C bonds. They suggested that ozone initially added to the C]C bond, to form a primary ozonide
1,6e+6 m/z 169
1,4e+6 1,2e+6 1,0e+6
nC ðiÞ my ðiÞ nC ðRÞ
(1)
where: nC (i) is the number of the carbon atoms of the product i, nC (R) is the number of the carbon atoms of the reagent and my (i) is the molar yield of the product i. The total carbon yield is the sum of the carbon yields of each quantified primary reaction products. As mentioned in the section above we have detected several oxidation products but their quantities were very close to the detection or quantification limit. The 3,4,5-trimethoxybenzoic acid is the main identified product and therefore its carbon yield under dark condition and in presence of light has been calculated. A dependence of the carbon yield of 3,4,5-trimethoxybenzoic acid with the ozone mixing ratios is illustrated in Fig. 5. The experimental data were fitted with the single exponential decay parameter. At an ozone mixing ratio of 250 ppb, the carbon yield of 3,4,5-trimethoxybenzoic acid is 40% for the ozonolysis of 3,4,5-trimethoxybenzaldehyde in dark condition and 28% for the ozonolysis under light irradiation of the organic coated particles (Fig. 5). In both cases, the carbon yield of 3,4,5-trimethoxybenzoic acid exponentially decrease with increasing ozone mixing ratio, leading to plateau at 2 ppm of ozone. Following the light irradiation of adsorbed 3,4,5-trimethoxybenzaldehyde on silica particles in absence of ozone the carbon yield of 3,4,5-trimethoxybenzoic acid is approx. 30% (Fig. 5).
m/z 151
8,0e+5
m/z 212
6,0e+5 4,0e+5
m/z 69
3.2. Kinetic results
m/z 126
Abundance (arbitrary unit)
which rapidly decomposes via two pathways to a carbonyl plus a “Criegee intermediate”. Atkinson and Arey (2003) claimed that third pathway via a direct reaction resulting epoxides as a product was presumably possible. Considering the above proposed mechanism for the ozonolysis of aromatic compounds, two possibilities can be suggested for the unknown compound i.e. 3,4,5-trimethoxy6-oxocyclohexa-2,4-diene-1-carbaldehyde resulting from C atom attack as reported by Perraudin et al. (2007) or epoxides product (3,4,5-trimethoxy-7-oxabicyclo[4.1.0]hepta-2,4-diene-1-carbaldehyde) resulting from direct reaction as reported by Atkinson and Arey (2003). Little et al. (2003) reported that epoxides could react with trimethylchlorosilane during the derivatization (BSTFA þ 10% trimethylchlorosilane) but the corresponding derivative have not been observed and the intensity signal of this unknown compound before and after derivatization with BSTFA remained unchanged. Therefore, the unknown compound with m/z ¼ 212 tentatively can be suggested as 3,4,5-trimethoxy-6-oxocyclohexa-2,4-diene-1carbaldehyde resulting from the ozone attack on the C atom but again this suggestion cannot be confirmed by standard analysis because its corresponding standard is not commercially available.
m/z 284
m/z 195
m/z 73
3289
m/z 184
m/z 109
2,0e+5 0,0 50
100
150
200 250 Mass/Charge ratio
300
350
400
Fig. 4. Mass spectrum of the detected unknown oxidation product with its suggested structure.
3.2.1. Method validation For the kinetic measurements, 15 expositions were performed to verify the possible loss of 3,4,5-trimethoxybenzaldehyde induced by flow gas N2/O2/O3 mixing during the exposition. Different experimental conditions were carried out: blank without exposition, exposition to N2, exposition to O2, exposition to O3 at 6 ppm under dark and under irradiation with the simulated solar light. Each exposition was performed with the same flow rate
3290
S. Net et al. / Atmospheric Environment 44 (2010) 3286e3294
The flow rate used in this study slightly modified the quantity of 3,4,5-trimethoxybenzaldehyde adsorbed on silica particles in the reaction chamber. A significant loss of 3,4,5-trimethoxybenzaldehyde was observed only when the coated particles were exposed to O3 in presence and in absence of light. Under dark conditions, the quantities of 3,4,5-trimethoxybenzaldehyde remained the same when the particles coated with 3,4,5-trimethoxybenzaldehyde were exposed to O2 comparing to N2. These results demonstrate that the adsorption procedure was reproducible and the volatilization phenomenon was negligible for the experimental conditions used in this study. Therefore, the blank expositions to N2 were used as a reference for the kinetic measurements.
Fig. 5. Plot of the carbon yield of 3,4,5-trimethoxybenzoic acid against the ozone mixing ratio: B) under dark condition, C) under simulated solar light.
(150 ml min1) and it was repeated three times for an exposition time of 8 h. Fig. 6 shows the normalised concentration of 3,4,5trimethoxybenzaldehyde adsorbed on silica particles for five different exposition conditions and the error bars obtained within three repeated experiments. The errors bars (Fig. 6) represent the 1s uncertainty levels based on the average of at least three repeated experiments.
3.2.2. First-order rate constants The kinetics for the reaction between gaseous ozone and 3,4,5trimethoxybenzaldehyde adsorbed on silica particles were investigated by monitoring the GCeMS signals as a function of time. When adsorbed 3,4,5-trimethoxybenzaldehyde was exposed to ozone, in absence and/or in presence of light, an irreversible loss of 3,4,5-trimethoxybenzaldehyde was observed. The exponential decays of reactive normalised concentrations of 3,4,5-trimethoxybenzaldehyde as a function of time obtained at five ozone mixing ratios i.e. 0 ppb, 250 ppb, 1 ppm, 3 ppm and 6 ppm, are presented in Fig. 7. The normalised concentrations of 3,4,5-trimethoxybenzaldehyde are derived from the chromatographic peak area, as [react]t/[react]0
Normalised 3,4,5-trimethoxybenzaldehyde concentration (%)
A 100
80
60
40 0ppb 250ppb 1ppm
20
3ppm 6ppm
0 0
1
2
3 4 5 6 Time of reaction (hours)
7
8
7
8
Normalised 3,4,5-trimethoxybenzaldehyde concentration (%)
B 100
80
60
40 0ppb 250ppb
20
1ppm 3ppm 6ppm
0 Fig. 6. A) Normalised concentration of 3,4,5-trimethoxybenzaldehyde adsorbed on silica particles for five different exposition conditions; Blank: 3,4,5-trimethoxybenzaldehyde adsorbed on silicon oxide without exposition; N2, O2, O3 and O3 þ hv exposure of 3,4,5-trimethoxybenzaldehyde adsorbed on silicon oxide. The applied ozone mixing ratio is 6 ppm; B) Comparison between the N2 and O2 exposures of the coated particles at different times ranging from 0 to 8 h.
0
1
2
3 4 5 Time of reaction (hours)
6
Fig. 7. Decays of normalised concentration of 3,4,5-trimethoxybenzaldehyde adsorbed on silicon oxide for five different ozone mixing ratios. >) 0 ppb, C) 250 ppb, B) 1 ppm, ;) 3 ppm, 6) 6 ppm. A) Ozonolysis under dark conditions B) Ozonolysis in presence of light.
S. Net et al. / Atmospheric Environment 44 (2010) 3286e3294
3291
where [react]t is the peak area corresponding to the concentration of 3,4,5-trimethoxybenzaldehyde at time t and [react]0 is the peak area respect to the initial concentration of 3,4,5-trimethoxybenzaldehyde. Experimental data points were fitted by the first0 order exponential functions (Eq. (2)), where kI (pseudo-first-order rate constant) is the exponential coefficient derived from the fit.
0 ½reactt ¼ exp kI t ½react0
(2)
The errors bars (Fig. 7) represent the 1s uncertainty levels based on the average of at least three repeated experiments. In general, under simulated sunlight irradiation, the ozonolysis was faster than the reaction under dark conditions. More importantly, at atmospherically relevant ozone mixing ratio of 250 ppb, the pseudo-first-order rate constant (k1) is z29 times higher under light illumination of the surface than under dark conditions. Indeed, the ozone mixing ratio of 250 ppb corresponds to highly ozone polluted regions such as the megacities. Very recent paper reported that during the plume of pollution in Cotonou the ozone mixing ratio reached 300 ppb (Minga et al., 2010). However, this huge difference between the reaction rates is much smaller at extremely high and not relevant for the atmosphere ozone mixing ratios; 7 times higher at 1 ppm of ozone and only z2 times higher at 3 ppm and 6 ppm of ozone. 3.2.3. Treatment of the kinetic data by the immediate “gas-surface reactions” and via modified LangmuireHinshelwood mechanism Gas-surface interactions which lead to the formation or rupture of chemical bonds in the involved gas-phase ozone or surface molecules of interest can be described either as quasi-elementary “gas-surface reactions” (Pöschl et al., 2007) or as bimolecular surface reactions which can be treated with modified LangmuireHinshelwood mechanism (e.g. Pöschl et al., 2001; Kwamena et al., 2004, 2006, 2007; Brigante et al., 2008; Reeser et al., 2009; Pflieger et al., 2009; Net et al., 2010a, 2010b). Under dark conditions the ozonolysis of 3,4,5-trimethoxybenzaldehyde proceeds as immediate reaction of gas-phase ozone colliding with 3,4,5-trimethoxybenzaldehyde adsorbed on solid surface. The latter can be seen in Fig. 8 from the displayed linear dependence of the first-order rate constants respect to the ozone mixing ratios. The probable explanation for such behaviour of the kinetic data under dark conditions is that when the reaction proceeds slow the limiting factor is the kinetic rate constant. In this case in order to observe LeH mechanism higher ozone mixing ratios ought to be applied. On the other hand, when the reaction is fast as we observed in this study under light irradiation of the surface, the limiting factor is the ozone concentration and the LeH mechanism is observed. From the regression line plotted in Fig. 8 a second-order rate constants was derived k2nd ¼ (3.1 1.0) $ 1019 cm3 molecule1 s1 for the ozonolysis experiments carried out under dark conditions. The simulated sunlight irradiation of the surface induces an enhancement of about one order of magnitude in the reaction rate and a change to a modified LangmuireHinshelwood mechanism on ozone mixing ratio (Fig. 8). Such enhancement of heterogeneous reaction rates induced by solar light has already been reported for different reaction systems on solid or liquid surface (George et al., 2005; Jammoul et al., 2008; Brigante et al., 2008; Styler et al., 2009; Reeser et al., 2009; Net et al., 2010a, 2010b). The modified LangmuireHinshelwood mechanism was in details discussed in our previous papers (Net et al., 2010a, 2010b), hence, here only brief description is given for the treatment of the kinetic data that do follow the non-linear dependence with the applied ozone mixing ratios. According to this mechanism, the
Fig. 8. Pseudo-first-order rate constants for the reactions of gas-phase ozone with 3,4,5-trimethoxybenzaldehyde adsorbed on silicon oxide as a function of the ozone concentrations. B) under dark conditions, C) under solar light irradiation of the surface. The long dash line represents the linear regression for the immediate “gassurface” reaction and the solid line represents the fit to the LangmuireHinshelwood mechanism expressed by Eq. (3). Using the RSD on the scattering of the experimental data, the error on the slope k was calculated by the algorithm available in the Sigma Plot software.
observed first-order rate constant (k1obs) for degradation of organic compounds on the surface can be expressed as:
k2nd ½SSKO3 O3g ¼ k1obs 1 þ KO3 O3g
(3)
where k2nd is the second-order rate constant, [SS] is the number of adsorption sites available for ozone, KO3 is the ozone gas-surface equilibrium constant and [O3g] is the gas-phase ozone concentration. In Eq. (3) the first-order rate constant is expressed as a secondorder rate constant multiplied by the number of the adsorption sites. Hence, Eq. (3) can be simplified as:
k1max KO3 O3g ¼ k1obs 1 þ KO3 O3g
(4)
where k1max is the maximum rate constant experimentally obtained at high ozone mixing ratios (Kwamena et al., 2007). Eq. (4) was used to fit the kinetic data that do obey the LangmuireHinshelwood mechanism under assumption that the adsorbed 3,4,5-trimethoxybenzaldehyde has a very long residence time on the silica surface by which means it is a part of the surface. According to this mechanism, the reaction rate becomes independent of gas-phase ozone mixing ratios when all surface sites are occupied. According to Eq. (4), KO3 can be obtained by fitting the experimental results, obtained from the plot of k1obs versus [O3g], to a non-linear least-square. By fitting the observed kinetic data to Eq. (4) the following value of KO3 (4.5 1.4) $ 1014 cm3 is obtained. This result is in good agreement with the observed literature data for various solid substrates (Pöschl et al., 2001; Kwamena et al., 2004, 2006; Net et al., 2010a, 2010b). 3.2.4. Mechanistic proposition When the light illuminates the organic coated silica particles the interaction between gas-phase ozone and the surface can be considered as structureesensitive reaction (Masel, 1996) which implies that excited triplets of 3,4,5-trimethoxybenzaldehyde (3P*) may activate the sites whereas the number of reactive sites at the surface is proportional to the number of the photoactivated
3292
S. Net et al. / Atmospheric Environment 44 (2010) 3286e3294
species (P*) (Monge et al., 2010). In this case the kinetics of the bimolecular surface mediated reaction that occurs principally on a few especially active sites can follow LangmuireHinshelwood mechanism (Adamson and Gast, 1997; Masel, 1996; Emeline et al., 2005). The electronically activated species (P*), as the aromatic compounds with carbonyl and carboxyl moieties can undergo electron transfer with O2, via triplet (3P*) and singlet excited states (1P*) (Mizuno and Otsuji, 1994). Such electron transfer reaction between (3P*) and molecular oxygen can lead to the formation of excited singlet state of molecular oxygen, O2(1Sg) (Styler et al., 2009). However, Bodesheim and Schmidt (1997) who performed a series of reactions between the O2(1Sg) and organic compounds in aqueous solution have shown that the rates of reaction are extremely slow and thus negligible. In addition, Cope and Kalkwarf (1987) irradiated PAH compounds (coated on glass) by UVeVIS light in absence of ozone and they did not detect any photooxidation product. Schutt et al. (1996) have shown that the light (UVeVIS) induced degradation of naphthalene adsorbed on silica gel was much slower in oxygen flow compared to that in an ozone stream. We performed several experiments of photolysis of adsorbed 3,4,5trimethoxybenzaldehyde on silica particles in absence of ozone (c.f Method validation). In our experiments the photoinduced loss of 3,4,5-trimethoxybenzaldehyde under oxygen flow in absence of ozone was three times slower (kobs ¼ 1.07 $ 105 s1) compared to the rate of degradation (kobs ¼ 2.9 $ 105 s1) obtained under simultaneous ozone and light exposure of the coated particles over the timescale of the ozonation reaction. Therefore, we are suggesting that the reason for the enhanced degradation of 3,4,5-trimethoxybenzaldehyde upon the light illumination could be the formation of an excited complex between ground state of ozone and (3P*) of 3,4,5-trimethoxybenzaldehyde via electron transfer mechanism. The latter suggestion is supported by the displayed non-linear LangmuireHinshelwood dependence of the light-induced first-order loss as a function of ozone. The formation of excited complex [3P*.O3] we have already proposed in our previous paper (Net et al., 2010b) and elsewhere (Styler et al., 2009). However, the existence of such exciplex remains to be verified by additional laboratory experiments and theoretical calculations. 3.2.5. Atmospheric implications We have studied the degradation of 3,4,5-trimethoxybenzaldehyde on silica particles by ozone oxidation, by photolysis and by simultaneous ozonolysis and photolysis in order to evaluate which process is the most important for atmospheric chemistry. Based on the experimental finding presented in this study, the light-induced heterogeneous ozonolysis of 3,4,5-trimethoxybenzaldehyde exceeds the rate of photolysis by factor of three and the dark reaction by even more than one order of magnitude (see Fig. 8) under our experimental conditions. We could estimate the second-order rate constants for the ozonolysis reactions under dark conditions k2nd ¼ (3.1 1.0) $ 1019 cm3 molecule1 s1 and accordingly the atmospheric lifetimes of 3,4,5-trimethoxybenzaldehyde adsorbed on silica particles. Assuming average ozone concentrations of 40 ppb (Vingarzan, 2004) the lifetime of 3,4,5-trimethoxybenzaldehyde adsorbed on silica particles is about five weeks. In addition to the ozone related lifetimes of particulate 3,4,5-trimethoxybenzaldehyde we could calculate the lifetime of 3,4,5-trimethoxybenzaldehyde adsorbed on silica particles related to the direct photolysis. Considering the photolysis rate constant obtained with simulated solar light (6.0 4.5) 106 s1 the lifetime of 3,4,5-trimethoxybenzaldehyde was estimated to be approximately two days.
Having in mind the obtained first-order loss under simultaneous ozonolysis and light irradiation it can be envisaged that the lifetime of 3,4,5-trimethoxybenzaldehyde is even shorter than two days. Unfortunately, the second-order rate constants cannot be determined for the kinetic data that do follow the LangmuireHinshelwood mechanism because the active site surface concentration is not known for the silica particles used in this study. 4. Conclusions Since the earth’s atmosphere is driven by the input of solar radiation, the direct and indirect photolysis processes play a crucial role in determining its composition. This work demonstrated that light accelerate the aging (ozonolysis) of the particles coated with 3,4,5-trimethoxybenzaldehyde. Influence of light on rate constants of 3,4,5-trimethoxybenzaldehyde was observed at all applied ozone mixing ratios from 0 ppb to 6 ppm and especially at low ozone mixing ratios. At atmospherically relevant ozone mixing ratio of 250 ppb, an enhancement was observed in the rates of degradation of 3,4,5trimethoxybenzaldehyde under illuminated versus dark conditions (28 times higher); 1.01 $ 106 s1 in the dark versus 2.90 $ 105 s1 under simulated sunlight. On the other hand, the degree of enhancement was only 3 times higher at elevated ozone mixing ratios (3e6 ppm) for the simultaneous ozonolysis and light irradiation of the coated particles in comparison to the straight ozonolysis. Moreover, the observed loss of 3,4,5-trimethoxybenzaldehyde exhibited a linear dependence with the ozone under dark condition which changes to a non-linear LangmuireHinshelwood dependence under simultaneous ozone and light illumination of the coated particles. The identified reaction products (3,4,5-trimethoxybenzoic acid, syringic acid) in this study are particularly important since they absorb in the tropospheric actinic window, which means they can easily initiate further photochemistry either by direct photolysis or indirect photochemical processes. These results exhibited that organic fraction within aerosols may change with time due to the oxidation and photochemical transformations predominantly taking place at the aerosol’s surface. Hence, such chemical aging processes may affect the hygroscopic properties of particles and hence their growth to cloud droplets, their optical properties and finally their lifetime. In addition, such change of the aerosol properties induced by the reaction products which remain on the surface may impact the atmospheric radiative balance (Nieto-Gligorovski et al., 2009). Light can affect not only the formation of the reaction products but the product yields, as well, since the carbon yield of 3,4,5-trimethoxybenzoic acid is different for the same mole of reagent conversion. In both cases, under dark and under simulated solar light, the carbon yield of 3,4,5-trimethoxybenzoic acid exponentially decayed to a level of approximately 15% at very high ozone mixing ratio (2 ppm). Finally, it can be suggested that light-induced heterogeneous ozone processing can have an influence on the aerosol surfaces by changing their physico-chemical properties. We believe that the results of this study may contribute to better understanding of the role of the light-induced heterogeneous processing which can be of relevance in the dry regions impacted by biomass burning processes and strong ozone pollution. Acknowledgement The authors are gratefully acknowledged, for the financial support of this work, to the “Agence Nationale de Recherche”
S. Net et al. / Atmospheric Environment 44 (2010) 3286e3294
within the framework “Contaminant, Ecosystème et Santé”, through the integrated project “INTOX” and to the European Community via “Fonds Européens de Développement Régional” (FEDER).
References Adamson, A.W., Gast, A.P., 1997. Physical Chemistry of Surfaces. John Wiley & Sons, Inc., New York. Anastasio, C., Faust, B.C., Janakiram Rao, C., 1997. Aromatic carbonyl compounds as aqueous-phase photochemical sources of hydrogen peroxide in acidic sulphate aerosols, fogs, and clouds. I. Non-phenolic methoxybenzaldehydes and methoxyacetophenones with reductants (phenols). Environment Science and Technology 31, 218e232. Atkinson, R., Arey, J., 2003. Gas-phase tropospheric chemistry of biogenic volatile organic compounds: a review. Atmospheric Environment 37 (Suppl. 2), S197eS219. Bari, M.A., Baumbach, G., Kuch, B., Scheffknecht, G., 2009. Wood smoke as a source of particle-phase organic compounds in residential areas. Atmospheric Environment 43, 4722e4732. Bertram, A.K., Ivanov, A.V., Hunter, M., Molina, L.T., Molina, M.J., 2001. The reaction probability of OH on organic surfaces of tropospheric interest, the reaction probability of OH on organic surfaces of tropospheric interest. Journal of Physical Chemistry A 105, 9415e9421. Bodesheim, M., Schmidt, R., 1997. Chemical reactivity of sigma singlet oxygen O2(1Sg). Journal of Physical Chemistry A 101, 5672e5677. Brigante, M., Cazoir, D., D’Anna, B., George, C., Donaldson, D.J., 2008. Photoenhanced uptake of NO2 by pyrene solid films. Journal of Physical Chemistry A 112, 9503e9508. Cope, V.W., Kalkwarf, D.R., 1987. Photooxidation of selected polycyclic aromatic hydrocarbons and pyrenequinones coated on glass surfaces. Environment Science and Technology 21, 643e648. Emeline, A.V., Ryabchuk, V.K., Serpone, N., 2005. Dogmas and misconceptions in heterogeneous photocatalysis. Some enlightened reflections. Journal of Physical Chemistry B 109, 18515e18521. George, C., Strekowski, R.S., Kleffmann, J., Stemmler, K., Ammann, M., 2005. Photoenhanced uptake of gaseous NO2 on solid-organic compounds: a photochemical source of HONO? Faraday Discussion 130, 195e210. Gomez, A.L., Park, J., Walser, M.L., Lin, A., Nizkorodov, S.A., 2006. UV photodissociation spectroscopy of oxidized undecylenic acid films. Journal of Physical Chemistry A 110, 3584e3592. Hawthorne, S.B., Miller, D.J., Lagenfeld, J.J., Krieger, M.S., 1992. PM-10 high volume collection and quantification of semi- and nonvolatile phenols, methoxylated phenols, alkanes, and polycyclic aromatic hydrocarbons from winter urban air and their relationship to wood smoke emissions. Environment Science and Technology 26, 2251e2262. Honicky, R.E., Osborne, J.S., Akpom, C.A., 1985. Symptoms of respiratory illness in young children and the use of wood burning stoves for indoor heating. Pediatrics 75, 587e593. Jammoul, A., Gligorovski, S., George, C., D’Anna, B., 2008. Photosensitized heterogeneous chemistry of ozone on organic films. Journal of Physical Chemistry A 112, 1268e1276. Kahan, T.F., Kwamena, N.O.A., Donaldson, D.J., 2006. Heterogeneous ozonation kinetics of polycyclic aromatic hydrocarbons on organic films. Atmospheric Environment 40, 3448e3459. Komissarov, V.D., Zimin, Y.S., 2006. On the mechanism of phenol ozonolysis. Kinetics and Catalysis 47, 850e854. Kwamena, N.O.A., Earp, M.E., Young, C.J., Abbatt, J.P.D., 2006. Kinetic and product yield study of the heterogeneous gas-surface reaction of anthracene and ozone. Journal of Physical Chemistry A 110, 3638e3646. Kwamena, N.O.A., Thornton, J.A., Abbatt, J.P.D., 2004. Kinetics of surface-bound benzo[a]pyrene and ozone on solid organic and salt aerosols. Journal of Physical Chemistry A 108, 11626e11634. Kwamena, N.O.A., Staikova, M.G., Donaldson, D.J., George, I.J., Abbatt, J.P.D., 2007. Role of the aerosol substrate in the heterogeneous ozonation reactions of surface-bound PAHs. Journal of Physical Chemistry A 111, 11050e11058. Little, J.L., 2003. Artifacts in Trimethylsilyl Derivatization Reactions and Ways to Avoid Them. Eastman Chemical Company, Kingsport, TN 37662-5150, USA. Liu, Y., El Haddad, I., Scarfogliero, M., Nieto-Gligorovski, L., Temime-Roussel, B., Quivet, E., Marchand, N., Picquet-Varrault, B., Monod, A., 2009. In-cloud processes of methacrolein under simulated conditions e part 1: aqueous phase photooxidation. Atmospheric Chemistry and Physics 9, 5093e5105. Masel, R.I., 1996. Principles of Adsorption and Reaction on Solid Surfaces. John Wiley & Sons. McIntire, T.M., Scott Lea, A., Gaspar, D.J., Jaitly, N., Dubowski, Y., Li, Q., FinlaysonPitts, B.J., 2005. Unusual aggregates from the oxidation of alkene self-assembled monolayers: a previously unrecognized mechanism for SAM ozonolysis? Physical Chemistry Chemical Physics 7, 3605e3609. McNeill, V.F., Wolfe, G.M., Thornton, J.A., 2007. The oxidation of oleate in submicron aqueous salt aerosols: evidence of a surface process. Journal of Physical Chemistry A 111, 1073e1083.
3293
Minga, A., Thouret, V., Saunois, M., Delon, C., Serça, D., Mari, C., Sauvage, B., Mariscal, A., Leriche, M., Cros, B., 2010. What caused extreme ozone concentrations over Cotonou in December 2005? Atmospheric Chemistry and Physics 10, 895e907. Mizuno, K., Otsuji, Y., 1994. Addition and Cycloaddition Reactions via Photoinduced Electron Transfer in Topics in Current Chemistry, vol. 169. Springer-Verlag, Berlin, Heidelberg. Mmereki, B.T., Donaldson, D.J., 2003. Direct observation of the kinetics of an atmospherically important reaction at the aireaqueous interface. Journal of Physical Chemistry A 107, 11038e11042. Mmereki, B.T., Donaldson, D.J., Gilman, J.B., Eliason, T.L., Vaida, V., 2004. Kinetics and products of the reaction of gas-phase ozone with anthracene adsorbed at the aireaqueous interface. Atmospheric Environment 38, 6091e6103. Monge, M.E., D’Anna, B., Mazri, L., Giroir-Fendler, A., Ammann, M., Donaldson, D.J., George, C., 2010. Light changes the atmospheric reactivity of soot. Atmospheric Chemistry Special Feature. doi:10.1073/pnas.0908341107. Mvula, E., von Sonntag, C., 2003. Ozonolysis of phenols in aqueous solution. Organic and Biomolecular Chemistry 1, 1749e1756. Net, S., Nieto-Gligorovski, L., Gligorovski, S., Temime-Rousell, B., Barbati, S., Lazarou, G.Y., Wortham, H., 2009. Photosensitized heterogeneous ozone processing on the organic coatings in the atmosphere. Atmospheric Environment 43, 1683e1692. Net, S., Nieto-Gligorovski, L., Gligorovski, S., Wortham, H., 2010a. Heterogeneous ozonation kinetics of 4-phenoxyphenol in presence of photosensitizer. Atmospheric Chemistry and Physics 10, 1545e1554. Net, S., Gligorovski, S., Pietri, S., Wortham, H., 2010b. Photoenhanced degradation of veratraldehyde upon the heterogeneous ozone reactions. Physical Chemistry Chemical Physics. doi:10.1039/b922957. Nieto-Gligorovski, L., Net, S., Gligorovski, S., Zetzsch, C., Jammoul, A., D’Anna, B., George, C., 2008. Interactions of ozone with organic surface films in the presence of simulated sunlight: impact on wettability of aerosols. Physical Chemistry Chemical Physics 10, 2964e2971. Nieto-Gligorovski, L., Net, S., Gligorovski, S., Wortham, H., Grothe, H., Zetzsch, C., 2009. Spectroscopic study of organic coatings on fine particles, exposed to ozone and simulated sunlight. Atmospheric Environment. doi:10.1016/j. atmosenv.2009.10.043. Nolte, C.G., Schauer, J.J., Cass, G.R., Simoneit, B.R.T., 2001. Highly polar organic compounds present in wood smoke in the ambient atmosphere. Environment Science and Technology 35, 1912e1919. Parham, R.A., Gray, R.L., 1984. In: Rowell, R. (Ed.), The Chemistry of Solid Wood. American Chemical Society, Washington, DC, pp. 3e56 (Chapter 1). Park, J., Gomez, A.L., Walser, M.L., Lin, A., Nizkorodov, S.A., 2006. Ozonolysis and photolysis of alkene-terminated self-assembled monolayers on quartz nanoparticles: implications for photochemical aging of organic aerosol particles. Physical Chemistry Chemical Physics 8, 2506e2512. Perraudin, E., Budzinski, H., Villenave, E., 2005. Kinetic study of the reactions of NO2 with polycyclic aromatic hydrocarbons adsorbed on silica particles. Atmospheric Environment 39, 6557e6567. Perraudin, E., Budzinski, H., Villenave, E., 2007. Identification and quantification of ozonation products of anthracene and phenanthrene adsorbed on silica particles. Atmospheric Environment 41, 6005e6017. Perzon, M., 2010. Emissions of organic compounds from the combustion of oats e a comparison with softwood pellets. Biomass and Bioenergy. doi:10.1016/j. biombioe.2010.01.027. Pettersen, R.C., 1984. In: Rowell, R. (Ed.), The Chemistry of Solid Wood. American Chemical Society, Washington, DC, pp. 57e126 (Chapter 2). Pflieger, M., Monod, A., Wortham, H., 2009. Kinetic study of heterogeneous ozonolysis of alachlor, trifluralin and terbuthylazine adsorbed on silica particles under atmospheric conditions. Atmospheric Environment 43, 5597e5603. Pope, C.A., 2000. Review: epidemiological basis for particulate air pollution health standards. Aerosol Science and Technology 32 (1), 4e14. Pöschl, U., Letzel, T., Schauer, C., Niessner, R., 2001. Interaction of ozone and water vapour with spark discharge soot aerosol particles coated with benzo[a]pyrene: O3 and H 2O adsorption, benzo[a]pyrene degradation, and atmospheric implications. Journal of Physical Chemistry A 105, 4029e4041. Pöschl, U., Rudich, Y., Ammann, M., 2007. Kinetic model framework for aerosol and cloud surface chemistry and gas-particle interactions e part 1: general equations, parameters, and terminology. Atmospheric Chemistry and Physics 7, 5989e6023. Reeser, D.I., Jammoul, A., Clifford, D., Brigante, M., D’Anna, B., George, C., Donaldson, D.J., 2009. Photoenhanced reaction of ozone with chlorophyll at the seawater surface. Journal of Physical Chemistry C 113, 2071e2077. Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R., Simoneit, B.R.T., 1998. Source of fine organic aerosol: 9 pine, oak and synthetic log combustion in residential fireplaces. Environment Science and Technology 32, 13e22. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2001. Measurement of emissions from air pollution sources. 3. CleC29 organic compounds from fireplace combustion of wood. Environment Science and Technology 35, 1716e1728. Schutt, W.S., Sigman, M.E., Li, Y., 1996. Fluorimetric investigation of reactions between ozone and exited state aromatic compounds on silica gel. Analytica Chimica Acta 319, 369e377.
3294
S. Net et al. / Atmospheric Environment 44 (2010) 3286e3294
Standley, L.J., Simoneit, B.R.T., 1987. Composition of extractable plant wax, resin and thermally matured components in smoke particles from prescribed burns. Environment Science and Technology 21, 163e169. Styler, S.A., Brigante, M., D’Anna, B., George, C., Donaldson, D.J., 2009. Photoenhanced ozone loss on solid pyrene films. Physical Chemistry Chemical Physics 11, 7876e7884. Vingarzan, R., 2004. A review of surface ozone background levels and trends. Atmospheric Environment 38, 3431e3442.
Vione, D., Maurino, V., Minero, C., Pelizzetti, E., Harrison, M.A.J., Olariu, R.I., Arene, C., 2006. Photochemical reactions in the tropospheric aqueous phase and on particulate matter. Chemical Society Reviews 35, 441e453. Vlasenko, A., George, I.J., Abbatt, J.P.D., 2008. Formation of volatile organic compounds in the heterogeneous oxidation of condensed-phase organic films by gas-phase OH. Journal of Physical Chemistry A 112, 1552e 1560.
IV.3. Présentation des articles 6 – 7 « Réactivités hétérogènes entre l’ozone gazeux et des méthoxyphénols particulaires : étude des produits et des cinétiques des réactions »
La famille des méthoxyphénols a fait l'objet de deux publications. La première est très générale puisqu'elle étudie la réactivité des sept méthoxyphénols, diversement substitués, les plus abondamment émis lors de la combustion des bois durs et tendres. Ce travail montre que la majorité de ces molécules sont peu réactives vis-à-vis de l'ozone même en présence de la lumière. Le second article est spécifiquement dédié à l'étude de la réactivité du coniferyaldéhyde. L'attention particulière portée à cette molécule s'explique par les résultats de l'étude toxicologique menée par les partenaires biologistes du projet ANR qui a financé ce travail. En effet, les tests de toxicités ont montré que le coniferyaldéhyde était l'espèce la plus toxique de tous les produits de combustion du bois étudiés dans le cadre de ce projet. Les résultats des travaux de toxicologie ne seront évidemment pas présentés dans cette thèse qui se focalise sur la réactivité chimique mais il est important de connaître dans le détail la réactivité de cette espèce toxique.
L’article 6 est principalement dédié à l’étude de l’identification des produits de dégradation des sept méthoxyphénols. Ces sept méthoxyphénols sont la vanillin, l’acide vanillique, le syringaldéhyde, l’acide syringique, l’acétovanillone, l’acétosyringone et le coniferyl alcool. Les réactivités de ces méthoxyphénols dépendent fortement de leurs fonctions chimiques qui particularisent la molécule. Parmi les sept méthoxyphénols étudiés, deux ont une fonction carboxylique (acide vanillique et acide syringique) et deux autres ont une fonction cétone (acétovanillone et acétosyringone). Aucun produit de réaction n’est détecté lors de l’ozonolyse de chacun de ces quatre méthoxyphénols, ni en l’absence ni en présence de lumière. Le principal chemin réactionnel de méthoxyphénol qui porte la fonction aldéhyde sur leur cycle aromatique est la conversion la fonction aldéhyde de la molécule en fonction acide carboxylique. Ainsi, l’acide vanillique et l’acide syringique sont respectivement les principaux produits de vanillin et de syringaldéhyde. Enfin, le dernier méthoxyphénol, et le coniferyl alcool se dégradent très rapidement en donnant naissance à de nombreux produits d’oxydation. Ces produits résultent essentiellement de l’attaque d’ozone sur la double liaison C=C de coniferyl alcool.
160
Enfin, l’article 7 est focalisé sur l’étude de la réactivité de coniferyl aldéhyde à différentes concentrations d’ozone (0 ppb, 250 ppb, 1 ppm, 3 ppm et 6 ppm) dans l’obscurité et en présence de la lumière. Le coniferyl aldéhyde se dégrade très rapidement en donnant naissance à de nombreux produits d’oxydation. Des produits identifiés ont permis de proposer un mécanisme réactionnel pour l’ozonolyse de coniferyl aldéhyde. Les résultats montrent que l’ozone réagit avec le coniferyl aldéhyde en attaquant la double liaison C=C de la molécule. Contrairement à ce qui est présenté précédemment dans l’article 4-6, la conversion de la fonction aldéhyde en fonction acide carboxylique n’a pas été observée car l’acide ferulique n’a pas été détecté. Le coniferyl aldéhyde absorbe la lumière jusqu'à 400 nm et la photolyse de cette molécule est observée. Les résulats montrent clairement que la réactivité de coniferyl aldéhyde sous irradiation est plus rapide par rapport à sa réactivité dans l'obscurité. Cependant, une variation linéaire des constantes cinétiques du pseudo-premier ordre en fonction de la concentration d’ozone a été observée pour les deux cas, en l’absence et en présence de lumière. Inversement de ce qui a été présenté précédemment dans les cas de méthoxybenzaldéhydes dans les articles 4 et 5, cette photolyse est indépendante et elle n’influence pas la dégradation de coniferyl aldéhyde par l’ozone. Pour le coniféryl aldéhyde, il n’y a pas de phénomène d’autophotosensibilisation. En effet, l'écart des deux droites est constant et il est équivalent à la constante de photolyse. C’est la raison pour laquelle, nous proposons qu’au cours de la réaction de coniferyl aldéhyde, sous irradiation, il se produit deux phénomènes réactionnels indépendants simultanés : la photolyse et l'ozonolyse.
161
- Article 6 -
Heterogeneous reactions of ozone with methoxyphenols, in presence and absence of light Sopheak Net, Sasho Gligorovski* and Henri Wortham
Université d’Aix-Marseille I, II, III-CNRS UMR 6264 : Laboratoire Chimie Provence Equipe Instrumentation et Réactivité Atmosphérique Case courrier 29, 3 place Victor Hugo, F - 13331 Marseille Cedex 3, France
Submitted to Atmospheric Environment 10/08/2010
Corresponding author: Sasho Gligorovski Email:
[email protected] Tel: +33 4 13 55 10 52 Fax: +33 4 91 10 63 77
162
Keywords: Heterogeneous reaction, photodegradation, methoxyphenols, silica particles, irradiation, GC-MS.
Abstract In this work, we investigated the heterogeneous reactions between gaseous ozone and seven particulate methoxyphenols, biomass tracers. The ozonolysis of silica particles coated with vanillin, vanillic acid, syringaldehyde, syringic acid, acetovanillone, acetonsyringone and coniferyl alcohol was studied successively and was carried out both in total darkness and under illumination with simulated solar light at 297 K. The condensed-phase products which emerged in such heterogeneous reactions were analyzed by gas chromatography-mass spectrometry (GC/MS). No reaction product was detected during the ozonolysis of vanillic acid, syringic acid, acetovanillone and acetosyringone under our experimental conditions. The main tranformation partway of vanillin and syringaldehyde was the conversion of aldehyde group to carboxylic fonction. Thus, syringic acid and vanillic acid were respectively the main oxidation products of syringaldehyde and vanillin. The oxidation of coniferyl alcohol was relatively fast and the total degradation was observed after 16 hours of ozone exposure. Five oxidation products: glycolic acid, oxalic acid, vanillin, vanillic acid and 3,4-dihydroxybenzoic acid, were identified and confirmed by their corresponding standards. It is interesting to note that 3,4-dihydroxybenzoic acid was detected only in the experiment performed under combined ozone and light exposure of the coated with coniferyl alcohol particles. Vanillin and vanillic acid absorb light in the region λ > 300nm by which means they can further influence the photolysis (direct and indirect) driven chemistry in the troposphere. A mechanistic pathway was proposed in order to elucidate the ozonolysis reaction of coniferyl alcohol and to explain the identified reaction products.
1. Introduction
Biomass combustion is one of the major sources of fine organic materials (
