Photochemical Fate of Amicarbazone in Aqueous Media: Laboratory Measurement and Simulations

Chemicals

All solutions were prepared using ultrapure water (Milli-Q). Technical-grade amicarbazone (AMZ) (4-amino-N-tert-butyl-4, 5-dihydro-3-isopropyl-5-oxo-1H-1, 2, 4-triazol-1-carboxamide, C₁₀H₁₉N₅O₂, CAS 129909-90-6) of minimum purity 95.4% was supplied by Arysta LifeScience Corp. and used as a standard in liquid chromatography (HPLC) analysis and in photochemical experiments. Suwannee River Natural Organic Matter (SRNOM) was obtained from the International Humic Substances Society (Catalog No. 1R101N); after dissolving SRNOM in water, the stock solution was diluted to a final concentration of 10 mg/L in a phosphate buffer solution (10 mmol/L, pH 7.2). Hydrogen peroxide, methylene blue, and bovine catalase (Sigma Aldrich); benzene (Sigma Aldrich); phenol (Sigma Aldrich); furfuryl alcohol (FFA) (TCI America); para-chlorobenzoic acid (pCBA) (ICN Biomedicals, Inc.); and sodium nitrate and sodium phosphate (Fisher Scientific) were all of reagent-grade purity. For HPLC analysis, acetonitrile (HPLC quality), methanol (HPLC quality), and phosphoric acid (100%) were purchased from Honeywell.

Solar simulator system

All experiments were performed using a Model 94041 Oriel solar simulator equipped with a 1,000 W lamp and air mass 1.5 global filter (Newport Corp.; Oriel Instruments), providing 68 W/m² in the wavelength range 290–400 nm. During the experiments, samples were contained in 2-mL Pyrex vials with no headspace and exposed to light while in a water bath maintained at 20°C.

Experimental methods

AMZ photodegradation in the presence of SRNOM

A series of solar light-driven photodegradation experiments were conducted to determine the contribution of direct and indirect photolysis (through \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} and ¹O₂) on AMZ (20.7 μmol/L) degradation in the presence of SRNOM (10 mg/L). In these experiments, pH was 7.2 and exposure to simulated solar light lasted 28 hours.

Analytical methods

Dissolved organic carbon (DOC) was measured using a TOC-VSCH (Shimadzu Corp.) analyzer. The method detection level was 0.2 mg C/L. An HPLC–UV (Agilent 1200) system equipped with an Agilent Eclipse XDB-C18 column (250×4.6 mm inner diameter, 5 μm) was used for monitoring AMZ, pCBA, FFA, and phenol. AMZ was determined at 224 nm by gradient elution with phosphate buffer (pH 2.8)/acetonitrile solution. The following gradient elution was used: 0–3 minutes (15% acetonitrile); 3–20 minutes (15–40% acetonitrile); 20–22 minutes (40% acetonitrile); 22–23 minutes (40–15% acetonitrile); and 23–25 minutes (15% acetonitrile).

The mobile phase used to determine pCBA was 45% phosphate buffer (pH 2.8)/55% methanol, with a detection wavelength of 234 nm. An isocratic elution with 70% phosphate buffer (pH 2.8)/30% methanol was used to monitor FFA at 219 nm. Finally, phenol was monitored at 210 nm, using 60% phosphate buffer (pH 2.8)/40% methanol as the mobile phase.

Photochemical modeling

The model used in this study to describe the environmental fate of AMZ has been previously described and validated for different pollutants (Vione et al., 2011; Marchetti et al., 2013). The software APEX (Aqueous Photochemistry of Environmentally occurring Xenobiotics) was derived from this model at the Department of Chemistry of the University of Torino and can be freely downloaded (Laurentiis et al., 2013; Bodrato and Vione, 2014). Briefly, it is used to calculate pseudo first-order kinetic constant of pollutant degradation in surface water and its half-life based on the chemical composition of water, irradiation depth, and on the spectral photon flux density of sunlight. Required input data are the photolysis quantum yield of the pollutant under sunlight and the second-order rate constants of its reaction with ROS, which we measured experimentally and are reported for the first time in the present study. The model uses the energy reaching the ground in a summer sunny day corresponding to 10 h of continuous irradiation at 22 W/m² irradiance to define the time unit. Further details, including model equations, are given at http://chimica.campusnet.unito.it/do/didattica.pl/Show?_id=4pyh.

Light absorption properties of AMZ solutions

The overlap between the absorption spectrum of AMZ in aqueous solution and the spectrum of light emitted by the solar simulator is a necessary condition for the direct photolysis of the pollutant. AMZ exhibits low values of the molar absorption coefficient above 300 nm (ɛ<10 L/mol·cm) (Supplementary Fig. S1), whereas the solar irradiance on the earth surface ranges from 290 nm to higher wavelengths. Our experiments showed that no AMZ degradation occurred when the herbicide was spiked into ultrapure water and irradiated for 8 hours in the solar simulator. This suggests low quantum yields for direct photolysis associated with the low values of ɛ above 300 nm. As a result, AMZ degradation in natural surface waters is expected to be due to indirect photodegradation associated with NOM and inorganic nitrogen species, and to a lesser extent possibly to any biotic action.

Generation and action of ROS toward AMZ degradation

The rate of formation of ROS (¹O₂, \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} ) is controlled by the amount of photons absorbed by NOM and the associated quantum yields (Φ_1O2 and Φ_OH, respectively), defined as the number of molecules of ¹O₂ or \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} formed per photon absorbed. The rate of formation divided by the sum of physical quenching and reactions with other species is equal to the steady-state concentration of the ROS, according to Equations (1) and (2) (Mostafa and Rosario-Ortiz, 2013). \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}[ ^1O_2 ] _ { ss } = \frac { k_a \Phi_ { 102 } [ { \rm S } ] } { \Sigma k_x [ X ] + k_d } \tag { 1 } \end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}[ \bullet OH ]_{ ss } = \frac { k_a \Phi_ { 0 } [ { \rm S } ] } { \Sigma k_x [ X ] } , \tag { 2 } \end{align*} \end{document}

where k_a is the specific light absorbance of the water matrix (assumed to be dominated by NOM); S represents a photosensitizing molecule; Φ_1O2 and Φ_OH are the quantum yields; k_X represents the reaction rate between the ROS and any species of X present in the water matrix (including but not limited to NOM); k_d is the rate of decay of ROS to its ground state due to physical quenching without undergoing a reaction. One should observe that only ¹O₂ has a significant k_d value (Mostafa and Rosario-Ortiz, 2013) and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} radicals are scavenged quickly by constituents of the water matrix. Table 1 shows the steady-state concentrations, the formation rates, and the quantum yields of formation of singlet oxygen and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} radicals in a solution containing 10 mg/L of SRNOM under simulated sunlight.

The steady-state concentrations shown in Table 1 are in close agreement with those found in surface water bodies, which have been reported to be in the order of 10⁻¹⁷ to 10⁻¹⁶ and 10⁻¹⁵ to 10⁻¹⁴ mol/L, for \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} radicals and ¹O₂, respectively (Vione et al., 2010b).

The second-order kinetic rate constants of AMZ with ¹O₂ and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} were determined by using methylene blue and H₂O₂ as the ROS sources, respectively. Competition kinetics was exploited between AMZ and pCBA in the case of hydroxyl radicals (Reactions [R1] and [R2], respectively) and between AMZ and FFA for ¹O₂ (Reactions [R3] and [R4], respectively): \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}{\rm AMZ} + ^ \bullet { \rm OH} \rightarrow { \rm P}1 \ k_{{ \rm AMZ , OH} \bullet}\ ( { \rm R}1 )\end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}{\rm pCBA} + ^ \bullet { \rm OH} \rightarrow { \rm P}2 \ k_{{ \rm pCBA , OH} \bullet}\ ( { \rm R}2 )\end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}{\rm AMZ} + ^1 { \rm O}_2 \rightarrow { \rm P}3 \ k_{{ \rm AMZ} , 1 { \rm O}2}\ ( { \rm R}3 )\end{align*} \end{document} \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}{\rm FFA} + ^1 { \rm O}_2 \rightarrow { \rm P}4 \ k_{{ \rm FFA} , 1 { \rm O}2 \bullet}\ ( { \rm R}4 )\end{align*} \end{document}

The values of \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$k_{{ \rm AMZ , OH}\bullet}$$ \end{document} and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$k_{{ \rm AMZ} , 1{ \rm O}2}$$ \end{document} were determined using Equation (3): \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}k_ { { \rm AMZ } , { \rm ROS } } = \frac { k_ { AMZ ( obs ) } } { k_ { ref ( obs ) } } \times k_ { { \rm ref } , { \rm ROS } } \tag { 3 } \end{align*} \end{document}

where \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$k_{{ \rm AMZ , ROS}}$$ \end{document} is the rate constant of the reaction between AMZ and the ROS, ¹O₂, or \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} ; k_AMZ(obs) is the measured degradation rate constant for amicarbazone; k_ref(obs) is the measured degradation rate constant for the reference compound (FFA or pCBA); k_ref,ROS is the rate constant of the reaction between the reference compound and ROS, k_FFA,1O2=1.2×10⁸ L/mol·s (Mostafa and Rosario-Ortiz, 2013), and k_{pCBA, OH•}=5×10⁹ L/mol·s (Elovitz and von Guten, 1999). The measured value of k_{AMZ, OH•} was 2.05×10¹⁰ L/mol·s; in contrast, k_AMZ,1O2 was found to be much lower (∼5×10⁶ L/mol·s), which indicates that singlet oxygen plays a less important role in AMZ degradation. Conversely, the effective diffusion controlled value of k_{AMZ, OH•} compared with the value of the rate constant for the reaction between NOM and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} radical (about 10⁸–10⁹ mol/L·s) suggests that this pathway may be more important (Buxton et al., 1988; Westerhoff et al., 2007; McKay et al., 2013).

AMZ photochemical degradation

The experiments performed with [AMZ]₀=20.7 μmol/L in the water matrix containing 10 mg/L of SRNOM indicated that AMZ removal achieved only 7.0% (±0.6) after 28 hours of exposure to simulated solar light. Therefore, in the absence of any other species (CO₃⁻², HCO₃⁻, NO₂⁻, NO₃⁻, etc.) the half-life time of AMZ in surface waters is expected to be about 10 days, which suggest that the herbicide, once leached from soils may remain in surface water bodies for long periods. Possamai et al. (2013) identified the inhibitory effects of AMZ on the growth of the bioindicator Cucumis sativus in soils with clayey and sandy textures, after 100 and 51 days after application, respectively, which indicates the persistence of AMZ in soils. Besides, the authors showed that herbicide leaching was favored in sandy soils, owing to the high AMZ solubility in water.

Photochemical modeling

The APEX model considers direct and indirect photolysis. In our case, since the herbicide degradation in DI water was not observed after 8 hours of exposure to simulated solar light, AMZ direct photolysis could be neglected, owing to the very low values of the molar absorption involved. Considering that the reaction between AMZ and ¹O₂ is also unlikely, as shown by the low value of k_AMZ,1O2 we obtained, the main pathway of AMZ photodegradation was assumed to be the reaction with hydroxyl radicals, since here the effect of ³CDOM* was not considered. It is worth observing that photodegradation would be faster if additional processes were involved. Based on this experimental information and on the second-order rate constant of the reaction between AMZ and hydroxyl radicals, the half-life time and pseudo first-order rate constant (k_tot) of AMZ degradation was modeled as a function of environmental variables, such as the concentration of particular chemical species and water column depth, by using the software APEX.

To achieve the relative importance of the model variables on AMZ half-life time (t_1/2, in days), a factorial design at two levels of each variable (2⁵) (Box et al., 1978) was used to perform the simulations using APEX (Supplementary Table S1). The ranges of values of the model variables (nitrate, nitrite, bicarbonate, and dissolved organic matter concentrations) were selected based on an extensive review of published data on Brazilian rivers (Table 2); the water depth was varied between 0.5 and 5 m. It is worth remembering that bicarbonate and carbonate concentrations do not vary independently because of the acid–base equilibrium involved. As a consequence, the composition of the HCO₃⁻/CO₃²⁻ system in water is strongly dependent on the pH (Oppenländer, 2003). The APEX model was, therefore, adapted to account for this fact and simulations were performed at four different pH values (6.5, 7.5, 8.5, and 9.5) for a range of bicarbonate concentrations found in the literature (Table 2). The pKa at 25°C for the HCO₃⁻/CO₃²⁻ system was calculated according to Clark and Fritz (1997). The corresponding carbonate concentrations were calculated by Equation (4): \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} \begin{align*}[ { CO_3 } ^ { 2 - } ] = \frac { 10^ { pka } [ { HCO_3 } ^ { - } ] } { 10^ { pH } } . \tag { 4 } \end{align*} \end{document}

According to the simulations, the variables that showed greater importance on amicarbazone degradation for pH 6.5 and 7.5 were DOC concentration (50.3%), water depth (43.4%), and nitrite concentration (2.8%); the numbers in parentheses correspond to the relative impact of each variable upon AMZ half-life time. In fact, for the values of bicarbonate concentrations shown in Table 2, in this pH range [CO₃²⁻] is ∼0.1–1 μmol/L; also, the second-order rate constants for reactions of hydroxyl radicals with bicarbonate and carbonate are \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$k_{{\rm HCO}3 - ,\ {\rm OH}_{\bullet}}$$ \end{document} =8.5×10⁶ L/mol·s and \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$k_{{\rm CO}32 - ,\, {\rm OH}_{\bullet}}$$ \end{document} =3.9×10⁸ L/mol·s (Oppenländer, 2003), respectively. As a consequence, scavenging of \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} species by CO₃²⁻ and HCO₃⁻ is small. On the other hand, for pH 8.5 and 9.5 carbonate anions gain importance in such a way that simulations indicated DOC concentration (50.0%), water depth (43.8%), and bicarbonate/carbonate concentration (2.7%) as the most influential variables in AMZ half-life time for pH 8.5; for pH 9.5, the higher effects on AMZ half-life are due to DOC concentration (43.8%), water depth (43.8%), and bicarbonate/carbonate concentration (10.5%); again, the number in parentheses correspond to the relative impact of each variable on t_1/2. Figures 2–7 summarize the main trends observed.

OPEN IN VIEWER

FIG. 2.

Effect of nitrite and dissolved organic carbon (DOC) concentrations on technical-grade amicarbazone (AMZ) half-life as predicted by the Aqueous Photochemistry of Environmentally occurring Xenobiotics (APEX) model for the pH range 6.5–7.5. All other variables fixed at their average values: [NO₃⁻]=50.5 μmol/L, [HCO₃⁻]=1.1×10³ μmol/L, and 2.75 m water depth. (a) 300-fold increase in nitrite for minimum and maximum DOC levels and (b) 20-fold increase in DOC concentration for minimum and maximum nitrite levels.

OPEN IN VIEWER

FIG. 3.

Effect of water depth and nitrite concentration on AMZ half-life as predicted by the APEX model for the pH range 6.5–7.5. All other variables fixed at their average values: [NO₃⁻]=50.5 μmol/L, [HCO₃⁻]=1.1×10³ μmol/L, and [DOC]=5.25 mg/L. (a) 300-fold increase in nitrite for minimum and maximum water depth and (b) 10-fold increase in water depth for minimum and maximum nitrite levels.

OPEN IN VIEWER

FIG. 4.

Effect of water depth and DOC concentration on AMZ half-life as predicted by the APEX model for the pH range 6.5–7.5. All other variables fixed at their average values: [NO₃⁻]=50.5 μmol/L, [HCO₃⁻]=1.1×10³ μmol/L, and [NO₂⁻]=7.5 μmol/L. (a) Ten-fold increase in water depth for minimum and maximum DOC levels. (b) Ten-fold increase in DOC concentration for minimum and maximum water depth.

OPEN IN VIEWER

FIG. 5.

Effect of bicarbonate and DOC concentrations on AMZ half-life as predicted by the APEX model for the pH range 8.5–9.5. All other variables fixed at their average values: [NO₃⁻]=50.5 μmol/L, [NO₂⁻]=7.5 μmol/L, and 2.75 m water depth. (a) Ten-fold increase in bicarbonate concentration for minimum and maximum DOC levels. (b) Ten-fold increase in DOC concentration for minimum and maximum bicarbonate levels.

OPEN IN VIEWER

FIG. 6.

Effect of water depth and bicarbonate concentration on AMZ half-life as predicted by the APEX model for the pH range 8.5–9.5. All other variables fixed at their average values: [NO₃⁻]=50.5 μmol/L, [NO₂⁻]=7.5 μmol/L, and [DOC]=5.25 mg/L. (a) Ten-fold increase in bicarbonate concentration for minimum and maximum water depth. (b) Ten-fold increase in water depth for minimum and maximum bicarbonate levels.

OPEN IN VIEWER

FIG. 7.

Effect of water depth and DOC concentration on AMZ half-life as predicted by the APEX model for the pH range 8.5–9.5. All other variables fixed at their average values: [NO₃⁻]=50.5 μmol/L, [NO₂⁻]=7.5 μmol/L, and [HCO₃⁻]=1.1×10³ μmol/L. (a) Ten-fold increase in water depth for minimum and maximum DOC levels. (b) Twenty-fold increase in DOC concentration for minimum and maximum water depth.

For the pH range 6.5–7.5, the predicted behavior of AMZ half-life time (t_1/2) as a function of nitrite and DOC concentrations is shown in Fig. 2, the water depth set at its average value (2.75 m) in the range 0.5–5.0 m. As expected, AMZ degradation becomes slower with increasing DOC concentration, which means higher DOM and CDOM contents. While CDOM is a photochemical source of hydroxyl radicals, DOM often acts as an important \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} scavenger (Passananti et al., 2014). Figure 2 also shows that a 300-fold increase in nitrite concentration at low DOC levels contributes to a 3-fold increase in t_1/2. In contrast, for the same increase in [NO₂⁻], a slight decrease in AMZ half-life is expected at higher DOC concentrations. These tendencies show that the impact of nitrite anions on AMZ degradation, as an additional source of hydroxyl radicals is limited, depending strongly on the composition of the key-role species (DOC, HCO₃⁻, and CO₃²⁻) in natural aqueous systems (see also Supplementary Figs. S2 and S3).

In comparison with the effect of nitrite, Fig. 3 shows that the model predicts a much more pronounced influence of water depth on AMZ persistence in natural systems. Actually, the model predicts an average increase in half-life of about 6 days to more than 1 month for a 10-fold increase in water depth. It is worth noticing that the joint effect of nitrite and water depth on t_1/2 does not depend on nitrate and bicarbonate levels (Supplementary Figs. S4 and S5), with average increments on t_1/2 following the same patterns as those shown in Fig. 3.

The predicted effects of water depth and DOC on AMZ half-life time, according to the APEX model calculations, are shown in Fig. 4. As expected, AMZ photodegradation will be faster in shallow and DOC-poor water bodies. The increase in t_1/2 with path length is accounted for by the fact that deeper water bodies are less thoroughly illuminated by sunlight, with lower \documentclass{aastex}\usepackage{amsbsy}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{bm}\usepackage{mathrsfs}\usepackage{pifont}\usepackage{stmaryrd}\usepackage{textcomp}\usepackage{portland, xspace}\usepackage{amsmath, amsxtra}\pagestyle{empty}\DeclareMathSizes{10}{9}{7}{6}\begin{document} $$^\bullet{\rm OH}$$ \end{document} formation rate as a consequence. Figure 4 shows that the effect of increasing DOC on t_1/2 would be lower than the effect of increasing water depth. Instead, the APEX model predicts average four- and seven-fold increases in half-life with depth for the minimum and maximum DOC levels, respectively. Supplementary Figs S6 and S7 reveal that the effects of water depth and DOC on AMZ persistence, when jointly examined, are not influenced by nitrate and bicarbonate. In fact, considering all levels of these variables, the average predicted increases in t_1/2 with water depth and DOC are equal to 5.7±1.5 and 3.1±0.8 times, respectively.

For the pH range 8.5–9.5, Fig. 5 shows the calculated values of AMZ half-life time as a function of bicarbonate and DOC concentrations. For a 10-fold increase in HCO₃⁻ concentration, t_1/2 increases only 14% and 7% for the minimum and maximum DOC levels, respectively. In contrast, average 3.5- and 3-fold increases in t_1/2 are expected for a 20-fold increase in DOC concentration. These tendencies show a much more pronounced scavenging of hydroxyl radicals by dissolved organic matter in comparison with bicarbonate/carbonate anions in natural aqueous systems (see also Supplementary Figs. S8 and S9). In addition, for bicarbonate/carbonate levels usually found in surface waters at higher pH values, the effect of water depth on AMZ persistence is evident (Fig. 6 and Supplementary Figs. S10 and S11).

Finally, Fig. 7 shows that the predicted effect of water depth and DOC concentration on AMZ persistence in the pH range 8.5–9.5 are very similar to that already shown in Fig. 4 (pH 6.5–7.5): the model predicts an average increase in AMZ half-life from 4 to 9 days (0.5 m) and from 17 to more than 2 months (5 m) when DOC increases from 0.5 to 10 mg/L. Considering all possible levels of the model input variables, the average predicted increases in t_1/2 with water depth and carbonate concentration are equal to 5.4±1.4 and 3.6±1.4 times, respectively (Supplementary Figs. S12 and S13), following the same trend presented in Fig. 7. This indicates that the predicted joint effects of water depth and DOC concentration on AMZ persistence in surface waters exposed to sunlight dominate and depend only slightly on nitrate/nitrite and bicarbonate/carbonate concentrations.