Article pubs.acs.org/est
Surface Water Enhances the Uptake and Photoreactivity of Gaseous Catechol on Solid Iron(III) Chloride Julia Tofan-Lazar and Hind A. Al-Abadleh* Chemistry Department, Wilfrid Laurier University, Waterloo, Ontario N2L 3C5, Canada S Supporting Information *
ABSTRACT: Uptake and photoreactivity of catechol−Fe complexes are investigated at the gas/solid interface under humid and dry conditions, along with the nature of the hydrogen-bonding network of adsorbed water. Catechol was chosen as a simple model for organics in aerosols. Iron chloride was used to distinguish ionic mobility from binding to coordinated iron(III) in hematite. Studies were conducted using diffuse reflectance infrared Fourier transform spectroscopy as a function of irradiation time. Results show that adsorbed water at 30% relative humidity (RH), not light, increases the concentration of adsorbed catechol by a factor of 3 over 60 min relative to dry conditions. Also, our data show that, at 30% RH and under light and dark conditions, growth factors describing the concentration of adsorbed catechol are very similar suggesting that light does not significantly enhance the uptake of catechol vapor on FeCl3. Surface water also enhances the initial photodecay kinetics of catechol−Fe complexes at 30% RH by a factor of 10 relative to control experiments (RH < 1%, or no FeCl3 under humid conditions). Absorptions assigned to carbonyl groups were not observed with irradiation time, which was explained by the dominance of FeCl2+ species relative to FeOH2+ in the highly acidic “quasi-liquid” phase at 30% RH. Clear differences in the hydrogen-bonding network upon gaseous catechol uptake are observed in the dark and light and during the photodecay of adsorbed catechol. The implications of these results on our understanding of interfacial processes in aged iron-containing surfaces are discussed.
■
INTRODUCTION The chemical composition and phase of aerosols influence their direct and indirect effects on the climate.1−4 Depending on their phase state and diffusivity properties, chemical transformation of aerosols due to condensed phase and surface chemistry proceeds differently.5−14 Because aerosols contain a number of important chromophores such as humic-like substances (HULIS), nitrite, nitrate, and some transition metals such as iron (Fe), photochemical reactions also play a role in changing their bulk and surface composition, in addition to the surrounding gas phase.12 Photochemical reactions of Fe in the bulk aqueous phase have been studied extensively using bulk techniques, which may take place in large cloud droplets that can also be influenced by gas-phase chemistry.15 Formation of reactive oxygen species (ROS) such as superoxide (O2−), hydrogen peroxide (H2O2), and hydroxyl (•OH) radicals has been reported in aerosols containing trace metals.16−21 For example, the dark Fenton reaction Fe(II) + H2O2 → Fe(III) + OH + OH− is famous for oxidizing organic molecules as a result of forming •OH radicals. In this reaction, the recycling of Fe(III) to Fe(II) is the rate-limiting step22 and can be enhanced by light. Hydroxyl radicals produced from the Fenton reaction contribute to the oxidation of water-soluble organic matter. In addition, Fe(III) can be easily complexed by ligands such as sulfate and carboxylate anions, and the photolysis of these complexes produces reactive radicals that can carry on further reactions.15 In small droplets or at the interface of systems containing chromophores such as Fe, an observed overall increase in quantum yield is expected for radical formation. This increase is © 2013 American Chemical Society
explained by the enhancement in surface concentration of reactants, increases in light intensity due to morphologydependent resonances and/or refraction, and/or decreased solvent-cage effects.23 Nissenson et al. demonstrated through experiment and modeling that the third factor is the major contributing factor after accounting for surface enrichment of the organics and light distribution throughout the droplet.23,24 Very little experimental work has explored the extent to which bulk Fenton and photo-Fenton chemistry takes place at the gas/solid interface in the presence of a few layers of adsorbed water. Previous laboratory work on model mineral dust13 and sea salt11 aerosols demonstrated the role of surface water in enhancing or inhibiting surface reactivity and in inducing ionic mobility. Our approach is to investigate the role of Fe in driving photodegradation of organic matter in model multicomponent aerosol systems containing organic matter, iron, and halide ions using a surface sensitive spectroscopic technique under atmospherically relevant conditions. Catechol was chosen as a simple model for HULIS in aerosols, aged polyaromatic hydrocarbons (PAH), and a precursor for secondary organic aerosols (SOA)25 formed from biomass burning26,27 or reaction of benzene with hydroxyl radicals.28 We recently reported the role that surface water plays in the complexation of catechol to solid FeCl3 under humid conditions in the dark using diffuse Received: Revised: Accepted: Published: 394
September 27, 2013 November 26, 2013 December 2, 2013 December 2, 2013 dx.doi.org/10.1021/es404321s | Environ. Sci. Technol. 2014, 48, 394−402
Environmental Science & Technology
Article
prior to introducing catechol vapor while the sample was being irradiated. Spectral collection was done in a similar fashion as in the experiments at 30% RH. The second type of experiments aimed at investigating the effect of light in photodegrading surface catechol under dry and humid conditions. The sample was prepared and kept under dry air flow for about 12 h prior to flowing catechol vapor. The flow of dry catechol vapor was started in the dark. Spectra were collected every 10 min for 1 h followed by dry air for 1 h in the dark to remove any weakly adsorbed catechol, with spectral collection every 10 min. After that, the light was turned on, and spectra were collected every 1 min for the first 10 min and then every 5 min for 2 h. For experiments conducted at 30% RH, samples were first exposed to dry air right after preparation to remove any residual CO2. Then, the flow of humid air at 30% RH started for 30 min, followed by the flow of catechol vapor at 30% RH for 1 h, and then, humid air only. After 30 min of flowing humid air, the light was turned on, and spectra were collected in the same fashion as described above for dry experiments. More experimental details are provided in the Supporting Information. To estimate the photon flux at the sample in the DRIFTS reaction chamber under dry and humid conditions, solid 2-NB was used as an actinometer.32,33 To minimize light screening and obtain low light absorbing conditions as done in similar experiments in solution,34 a 0.05% (w/w) sample in diamond powder was prepared. Prior to irradiating the sample, dry air was flowed for 1 h in one case and in the other case humid air at 30% RH for 30 min to collect a background spectrum. Once the light was turned on, spectra were collected every 4 s for 10 min by averaging 3 scans at 8 cm−1 resolution.
reflectance infrared Fourier transform spectroscopy (DRIFTS).29 We found that, upon increasing relative humidity (RH) to a value below the deliquescence of FeCl3, surface water facilitates ionic mobility resulting in the formation of monodentate catechol−Fe complexes. These complexes are stable at the gas/solid interface and do not undergo any further degradation in the dark as observed from bulk studies. This is in contrast to dry conditions where the data showed that gasphase catechol adsorbs molecularly and is fully protonated on samples containing FeCl3 with no evidence of complexation to Fe(III). The objective of this work was to investigate the photoreactivity of catechol−Fe complexes under humid and dry conditions and the nature of the hydrogen-bonding network of adsorbed water at the gas/solid interface for these systems.
■
EXPERIMENTAL SECTION Chemicals. All chemicals were used as received without further purification: catechol (1,2-benzendiol, >99%, CAS: 12080-9, Sigma−Aldrich), iron(III) chloride (FeCl3, sublimed grade, >99.9%, 7705-08-0, Sigma−Aldrich), 2-nitrobenzaldehyde (2-NB, 98%, Sigma−Aldrich), and diamond powder (6 μm, Lands Superabrasives LST600T). Given that catechol is a semivolatile organic with a vapor pressure of 1.3 Pa at 23 °C,30 a relatively small amount of catechol powder was placed at the bottom of a glass bubbler, which was connected to a flow of dry air in the gas handling system described below. Humid air was obtained by flowing dry air into a bubbler containing milli-Q water (18.5 MΩ cm). Estimation of the concentration of gasphase catechol (32 ppb) is provided in the Supporting Information. DRIFTS Experiments. Detailed description of the DRIFTS setup was reported earlier.31 Briefly, DRIFTS spectra were collected using a Praying Mantis diffuse reflectance accessory (Harrick, DRK-4N18) with a stainless steel high-temperature reaction chamber treated with a special SilcoSteel-CR coating (HVC, cup 6 × 3 mm2, dome cover with two Ge windows and one quartz viewing window). A total mass of 0.23 g filled the sample cup of the HVC reaction chamber. The gas handling system described in ref 31 was used to introduce catechol vapor and humid air into the reaction chamber. The humidity of the air flow is controlled by mass flow controllers of wet and dry air, and the accuracy is ±2%. Similar results were obtained within ±3% in RH. The DRIFTS accessory was installed into a Nicolet 8700 FTIR spectrometer (Thermo Instruments) equipped with a purge gas generator and a liquid N2 cooled MCT detector. Each experiment started by preparing a fresh sample containing either 1% (w/w) FeCl3 in diamond powder or diamond powder only for control experiments. The light used in the DRIFTS experiments was directed at an angle at the reaction chamber using a liquid light guide (LLG) from a 150 W Xe lamp (solar simulator, Newport Corp.) fitted with a heatabsorbing filter (2 in. × 2 in., FSQ-KG5, Newport Corp.). Figure S1 (Supporting Information) shows the irradiance spectrum of the solar simulator. Two types of experiments were conducted using DRIFTS. In the first type of experiments, the goal was to examine the role of light and surface water on the uptake of gas-phase catechol on samples containing solid FeCl3. For experiments at 30% RH, humid air was flowed for 30 min in the dark. Then, the light was turned on, while flowing humid air, for 30 min. This was followed by starting the flow of catechol vapor under irradiation and at 30% RH. For experiments under dry conditions, the sample was dried overnight, followed by irradiation for 15 min
■
RESULTS AND DISCUSSION Solid 2-NB Actinometry using DRIFTS. To quantify the absolute and relative light intensities in DRIFTS samples under humid (30% RH) and dry (RH < 1%) conditions, samples containing 0.05% solid 2-NB in diamond powder (w/w) were prepared and exposed to light from the solar simulator in the same fashion as samples containing solid 1% FeCl3. Figure 1 shows DRIFTS absorbance spectra before and after irradiation under dry conditions. Experiments were also conducted under humid conditions and spectra are shown in Figure S2 (Supporting Information). 2-NB is a reliable solid-phase chemical actinometer for UV-A (315−400 nm) and UV-B (280−315 nm) components of the electromagnetic spectrum and has been utilized for estimating light intensity in halide matrices and snowpacks in the field.32−35 Upon exposure to light, photoisomerization of 2-NB to 2-nitrosobenzoic acid takes place, whether in solution or in the solid phase. The reaction can be followed using transmission mode or total internal reflection IR spectroscopy as illustrated by Pitts et al.33 for samples prepared in a KBr disc and by Rowland and Grannas for samples dispersed in a polymethylmethacrylate (PMMA) matrix.32 This is done by following the decrease in the absorbance of the peak at 1531 cm−1 assigned to the asymmetric stretching mode of the −NO2 group in 2-NB. The absorbance of this peak is proportional to the 2-NB concentration accessed by the IR light. Spectra shown in Figures 1a and S2 (Supporting Information) were obtained by referencing to a dry diamond powder only sample spectrum and show that this peak is well defined and has no interference from the product that forms with increasing irradiation time. Also, data collected at 30% RH show little water uptake 395
dx.doi.org/10.1021/es404321s | Environ. Sci. Technol. 2014, 48, 394−402
Environmental Science & Technology
Article
and co-workers for 2-NB in water at 298 K and wavelength range 290−405 nm,34 we obtain a value for E that equals 4.9(1) × 1015 photons cm−2 s−1. For comparison, the average solar radiant output in the same wavelength range for noontime in Waterloo, Ontario, on July 1 is approximately 6 × 1014 photons cm−2 s−1.2 Role of Light and Humidity Level on Catechol Uptake. To examine the role that UV−vis light plays in the uptake of gas-phase catechol on samples containing 1% FeCl3(s), experiments were conducted such that the solid sample was irradiated before the introduction of catechol vapor under humid or dry conditions for 30 min to collect a background spectrum (Figure S3, Supporting Information). This was followed by introducing catechol vapor under continuous irradiation, and spectra were collected as a function of time. There are no reactions in the gas phase that will destroy catechol vapor because the spectral output of the solar simulator we use (λ > 290 nm) does not overlap with the UV absorbance of gas-phase catechol, which absorbs at λ less than 280 nm. The solid lines in Figure 2a show spectra of surface catechol as a function of irradiation time at 30% RH. The dashed lines represent spectra collected under dry conditions ( 2 min. cCalculated by multiplying the apparent photodecay rate with the light irradiance ratio of E(noontime Waterloo)/E(solar simulator) described in the first section of the Results and Discussion section.
This analysis clearly shows that surface water enhances the initial photodecay kinetics of catechol−Fe complexes at the solid/humid air interface by a factor of 10. The photodecay continues at a slower rate under humid conditions at longer irradiation times. The formation of Fe(II) species from reactions 1−3 and its slow oxidation by dissolved oxygen to Fe(III) is the likely cause for the decrease in the photodecay kinetics. The kinetic data also shows that, in the absence of Fe, 399
dx.doi.org/10.1021/es404321s | Environ. Sci. Technol. 2014, 48, 394−402
Environmental Science & Technology
Article
the presence of organics. These trends are inline with our interpretation above on the effect of irradiation on the structure of the hydrogen-bonding network. The high frequency ν(OH) loss feature at 3602 cm−1 can be assigned to the −Fe(OH) group involved in weak hydrogen bonding with neighboring sites. The broad loss feature at 3070 cm−1 arises most likely from hydrogen-bond breaking among strongly H-bonded water molecules due to ionic mobility. The positive ν(OH) features arise from the rearrangement of water molecules and the extent of their hydrogen bonding due to light absorption, from relatively weak and liquid-like (at 3471 and 3413 cm−1) to strong and ice-like (at 3225 and 3182 cm−1). The combination band at 2190 cm−1 becomes more pronounced with increasing irradiation time, which further suggests that, overall, the structure of the hydrogen-bonding network in our system resembles that in liquid water. All in all, the presence of catechol−Fe complexes, solvated Fe species, and surface water at RH as low as 30% renders samples more photoreactive than in the absence of Fe. In summary, the chemistry presented herein is relevant to iron-containing surfaces aged due to acid processing,68 which increases the availability of solvated Fe(III) cations and adsorption of organics from the gas phase. These surfaces include mineral dust particles12 and urban aerosols69 containing iron oxides and hydroxides as well as buildings in urban centers. Chloride content in coastal70 and urban aerosols (from HCl71) and burning of waste containing polyvinyl chloride72 can influence the speciation of iron and pH in the quasi-liquid phase, with direct consequences on photolysis rates of iron species,39 and the type of condensed and gas-phase products.40 As estimated above, the lifetime of organics due to this heterogeneous pathway is shorter by 3 orders of magnitude relative to the bulk aqueous phase. Overall, given the large surface area of aerosols and buildings, these results highlight the role that transition metals play in driving photochemistry during daytime under atmospherically relevant conditions, which eventually can lead to modifying their chemical composition and morphology.
■
ASSOCIATED CONTENT
S Supporting Information *
Experimental details; figures showing additional spectra for control experiments; fitting equation for data in Figure 2c; and addition information concerning estimations of radical formation and estimations of lifetime of organics from reaction with Cl radicals in the aqueous phase. This material is available free of charge via the Internet at http://pubs.acs.org.
Figure 4. Difference DRIFTS absorbance spectra showing the 2000− 4000 cm−1 spectral range as a function of time for catechol uptake on samples containing 1% FeCl3(s) (a) under irradiation and (b) in the dark, and (c) during photoreactivity of surface catechol. Solid and dashed lines represent spectra collected under humid (30% RH) and dry (
