Ecotoxicology and Environmental Safety 197 (2020) 110644

Contents lists available at ScienceDirect

Ecotoxicology and Environmental Safety

journal homepage: www.elsevier.com/locate/ecoenv

Indirect photodegradation of fludioxonil by hydroxyl radical and singlet T

oxygen in aquatic environment: Mechanism, photoproducts formation and
eco-toxicity assessment
Jiaoxue Yanga, Zehua Wanga, Guochun Lva, Wen Liua, Yan Wanga, Xiaomin Suna,∗, Jian Gaob,∗∗
a
Environment Research Institute, Shandong University, Qingdao, 266237, China
b
State Key Laboratory of Environmental Criteria and Risk Assessment, Chinese Research Academy of Environmental Sciences, Beijing, 100012, China

A R T I C LE I N FO A B S T R A C T

Keywords: Fludioxonil has been proven valuable as a broad-spectrum fungicide. However, there are concerns about its risk
Fludioxonil posed to non-target organisms in aquatic environments. In this paper, the mechanism, photoproducts trans-
Hydroxyl radical formation and eco-toxicity of fludioxonil during •OH/1O2-initiated process were systematically studied using
Singlet oxygen quantum chemistry and computational toxicology. The results indicate that the two favorable pathways of
Transformation mechanism
•OH/1O2-initiated reactions are both occurred in pyrrole ring. It can conclude that the rate constants of •OH and
Photoproducts
Eco-toxicity
1
O2 are 1.23 × 1010 and 3.69 × 107 M−1 s−1 at 298K, respectively, which results in half-lives of < 2 days in
surface waters under sunlit near-surface conditions. Based on toxicity assessments, these photoproducts showed
a decreased aquatic toxicity but the majority products are still toxic. This study gives more insight into the
chemical transformation mechanism of fludioxonil in aquatic environments.

1. Introduction cytosol enzymes. Now, based on the research of fludioxonil's pre-

decessor (fenpiclonil), fludioxonil is reported to inhibit transport-asso-
Fludioxonil is commonly used as fungicides in agriculture to protect ciated phosphorylation of glucose(Jespers and De Waard, 1995; Avenot
fruits and vegetable crops(Pose-Juan et al., 2006). It can protect crops and Michailides, 2015). The chemical behaviors of fludioxonil deserve
from Sclerotinia spp., Botrytis spp., Alternaria spp., and Fusarium spp. further consideration in aquatic environments owning to the concerns
(Kanetis et al., 2007; Schirra et al., 2005; Fenoll et al., 2011). Excessive about the persistence and the risks to aquatic animals.
amounts of fludioxonil dissolved in water or adsorbed on soil particles Direct photodegradation of fludioxonil in environmental systems is
may be easily transported by run-off surface water to large aquatic possible since fludioxonil absorbs light at wavelengths longer than
reservoirs(Dewez et al., 2005; Arias et al., 2005). It was found that 290 nm(Nicol et al., 2016). The direct photodegradation half-life was
fludioxonil have a long persistence in aquatic environment since re- reported to be 9.9 and 8.7 days during summer at 30 and 40°N, re-
sidues of this fungicide were present up to 5 years in vineyards region spectively(Duan et al., 2013). Together with direct photolysis, indirect
(Cabras et al., 1997a, 1997b). Based on the concentration and persis- photochemical transformation of fludioxonil in aquatic environments
tence of fludioxonil in the aquatic environment, it may have additional could also likely occur due to the presence of various reactive species
effects on non-target aquatic plant species(Brandhorst and Klein, 2019; including hydroxyl radicals (•OH) and singlet oxygen (1O2)(Wilkinson
Verdisson et al., 2001). et al., 1995; Costa et al., 2008; Yang et al., 2017). In fact, high con-
Fludioxonil is one of two existing commercial phenylpyrrole fun- centration of •OH was reported in various advanced oxidation processes
gicides (the other being fenpiclonil) derived from the antibiotic pyr- (AOPs) (e.g. UV/H2O2 system)(Zhang et al., 2016; Luo et al., 2018; Xiao
rolnitrin(Hu et al., 2019). The fungicides can inhibit growth of two et al., 2018). •OH have been used for the destruction of recalcitrant
aquatic plants (aquatic vascular plants and algae), thereby showing a organic contaminants from drinking water and wastewater, because it
potential toxicity against non-target species(Dewez et al., 2005; Lassalle can react with many organic chemicals at near diffusion-controlled
et al., 2015) (Verdisson et al., 2001). Dewez et al. (2005) proposed that rates(Li et al., 2014; Nayebzadeh et al., 2020). The pyrrole ring in
fludioxonil can induce strong antioxidative activities associated with fludioxonil also is a likely reaction site for indirect photodegradation

∗
Corresponding author.;
∗∗
Corresponding author.
E-mail addresses: [email protected] (X. Sun), [email protected] (J. Gao).

https://doi.org/10.1016/j.ecoenv.2020.110644
Received 20 February 2020; Received in revised form 12 April 2020; Accepted 13 April 2020
Available online 20 April 2020
0147-6513/ © 2020 Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).
J. Yang, et al. Ecotoxicology and Environmental Safety 197 (2020) 110644

(Scott et al., 1983; Cermola et al., 2006; Razavi et al., 2011). In pre- et al., 2009; Bhoorasingh et al., 2017). In this paper, the KiSThelP
vious study, the results have shown that the rate constant for the re- program (Canneaux et al., 2014) is used to calculate the rate constants
action of fludioxonil with •OH was 1.11( ± 0.02) × 1010 M−1 s−1 using which based on TST with wigner tunneling correction. For the aqueous
competition kinetics and hydrogen peroxide or nitrite as a source of bimolecular reaction, we have transferred the calculated rate constants
•OH. And the rate constant of 1O2 + fludioxonil was to the corresponding aqueously values because the KiSThelP use the
4.90( ± 0.02) × 107 M−1 s−1(Apell et al., 2019). However, the de- gas-phase standard-state as the reaction quotient.
tailed degradation mechanisms have not been discussed, and thus the
performance of fludioxonil transformation processes is not clear. 2.4. Eco-toxicity assessment
To understand the aquatic degradation behavior of fludioxonil, it is
critical to investigated their transformation mechanism. Some previous The eco-toxicity of fludioxonil and its transformation products were
studies have shown that quantum chemical calculation has been suc- evaluated using the ecological structure-activity relationships
cessfully applied to the study on the chemistry degradation mechanism (ECOSAR) model(Burden et al., 2016; Takata et al., 2020). In this work,
of pesticides(Shi et al., 2011; Zhou et al., 2011; Gao et al., 2014). In this three aquatic organisms including green algae, daphnia and fish were
work, both quantum chemical calculation and computational tox- chosen as target objects for biotoxicity exposure assessments. The
icology are used to explore the mechanisms and environmental impacts toxicity during transformation of fludioxonil was expressed by the acute
of fludioxonil during indirect photochemical transformation. The toxicity and chronic toxicity. Acute toxicity was shown by the median
transformation mechanism, kinetics of •OH/1O2-initiated indirect pho- lethal concentration (LC50) and media effect concentration (EC50) after
tolysis, the aquatic environmental persistence and the potential eco- 96 h exposure.
toxicity of fludioxonil and its products are elucidated in detail.
3. Results and discussion
2. Computational methods
3.1. Initial reaction of fludioxonil with •OH and 1O2
2.1. Mechanism computation
In this study, we discussed the reaction of fludioxonil towards •OH
All quantum chemical calculations were carried out using the and 1O2 in aquatic environments. •OH-initiated oxidation reaction of
Gaussian 09 package. In this study, the geometries of the reactants, fludioxonil were discussed in terms of three different reaction me-
transition states (TSs), intermediates, and products were optimized chanisms: (i) •OH-addition reaction (Radd), •OH addition to the neutral
using M06–2X functional(Walker et al., 2013) with a standard 6- reactant leading to a complex; (ii) H-abstraction reaction(Rabs), •OH
311+G(d,p) basis set. M06–2X has been proven to be a reliable cal- could abstract a H atom from the neutral reactant leading to H2O; (iii)
culation for transformation mechanism, kinetics and environmental the single electron transfer mechanism from the fludioxonil molecular
fates of organic pollutants(Zhao and Truhlar, 2008; Gao et al., 2019). (Rset). All possible pathways are depicted in Fig. 1.
The corresponding vibrational frequency calculations were kept at the As shown in Fig. 1, addition reaction of fludioxonil with •OH can
same level and the results are used to confirm the structure of local occur at different carbon atoms, which can be classified to three major
minimal point and transition-states. Each transition state was verified categories: benzene ring/pyrrole ring/branched carbon addition. Fig. 2
to connect the designed reactants and products by performing an in- shows reaction processes of the most favorable addition pathways
trinsic reaction coordinate (IRC) analysis(Maeda et al., 2015). More- (Radd4, Radd6 and Radd11) of the three major categories. Other pathways
over, the single point energy of all species was optimized at the and optimized geometries are given in Fig. S1. The Gibbs free energy
M06–2X/6–311++G(3df,3pd) level in order to obtain more accurate barriers of the addition reaction in C4 (Radd4), C6 (Radd6), and C11
energy. In addition, the polarized continuum model (PCM) within a (Radd11) are 1.95, 5.73 and 12.81 kcal mol−1 (the difference between
self-consistent reaction field (SCRF) theory was employed to determine the transition state and the pre-reactive complex), respectively. And the
the effect of solvent water(Zheng et al., 2013; Xia et al., 2009; Fudickar rate constants of the three reaction at 298K (Table 1) are 1.23 × 1010,
and Linker, 2019). The geometries were visualized using the CYLview 8.31 × 105, and 8.15 M−1 s−1, respectively. These results indicate that
software package. the reaction occurred at C4 of the pyrrole ring is most favorable due to
For the mechanisms involving single-electron transfers (SET), the the lowest Gibbs free energy barrier and the highest rate constants.
Marcus theory was used(Mayer, 2011; Chou et al., 1977). Within this Considering all the pathways of the addition reaction, it is clear that the
≠
transition-state formalism, the SET activation barrier ( ΔGSET ) is defined reaction energies are negative (−5.11~-19.05 kcal mol−1) except a
in terms of the free energy of reaction ( ΔGSET ) and the nuclear re- little 1.61 kcal mol−1 at C2 atom. The results imply that these processes
organization energy (λ ): are exothermic and potentially contribute to the transformation of
fludioxonil from standpoint thermodynamics.
≠ (λ + ΔGSET ) 2
ΔGSET = In addition, the free energy profile and the optimized geometries for
4λ (1)
the H-abstraction reactions of fludioxonil by •OH are shown in Fig. S2.
As shown in Fig. S2, the free energy barriers of the Rabs pathways are
2.2. The reorganization energy (λ ) has been calculated as within a range of 12.84–17.81 kcal mol−1 (the difference between the
transition state and the reactant), which is higher than those of •OH-
addition reaction. These results mean that the •OH-addition reaction
λ = ΔESET − ΔGSET (2)
could more likely to occur than the abstraction pathways in aquatic
Where ΔESET has been calculated as the non-adiabatic energy difference environments.
between reactants and vertical products. For the electron-transfer reaction (Dadashi-Silab et al., 2016), flu-
dioxonil reacts with •OH via one electron transfer with the formation of
2.3. Kinetics computation radical cation species and hydroxyl ion. The energy barrier of the Rset
pathway was calculated to be as high as 40.79 kcal mol−1 by Marcus
The transition state theory (TST) (Miller, 1993) is a method to theory. And this reaction is an endothermic process with a reaction
calculate the dynamic parameters of the elementary reaction by using energy of 33.11 kcal mol−1. The results confirm that the electron
molecular structure and thermodynamic properties(Truhlar and transfer pathway is unlikely to occur during •OH-initiated transforma-
Kuppermann, 1971). It has been widely applied to calculate the rate tion of fludioxonil. This finding could probably be ascribed to the
processes in previous studies(Georgievskii and Klippenstein, 2003; Ju structural characteristic of fludioxonil.

2
J. Yang, et al. Ecotoxicology and Environmental Safety 197 (2020) 110644

Fig. 1. All the reaction pathways in the initial reaction of fludioxonil with •OH.

Additionally, previous studies have shown that pyrrole with alkyl and C1-1 (or C1) → PC1 (or IMI). The Gibbs free energy barriers of the
and phenyl substituents can undergo reactions with singlet oxygen addition reactions in C1–C2 and C1–C4 are 21.85 and 4.51 kcal mol−1,
(1O2) in organic solvents and water(Cote et al., 2018; Zeinali et al., respectively. This means the Radd1-4 pathway is the main routes for
2019; Kaur and Anastasio, 2017). Therefore, the mechanism of the 1
O2-initiated photochemical transformation of fludioxonil in aquatic
reaction between fludioxonil and 1O2 also are considered. The free environments. Among the two pathways, 1O2 cycloaddition reaction
energy profile and the optimized geometries for the reactions between would release reaction energy of 8.8 and 15.72 kcal mol−1.
fludioxonil and 1O2 are depicted in Fig. S3 and Fig. 3. The 1O2-initiated
reaction of fludioxonil is cycloaddition, which is different from the •OH-
3.2. Kinetic calculation and environmental persistence
addition reaction. Depending on the experimental products, the initia-
tion reaction of 1O2 with fludioxonil can react on C1–C2 (Radd1-2) or
In order to evaluate the significance of the •OH and 1O2 in aquatic
C1–C4 (Radd1-4) site. The initiation processes of 1O2 include two steps,
environments, the initiation oxidation reaction rate of fludioxonil by
which correspond to the fludioxonil + 1O2 → the complex C1-1 (or C1),
•OH and 1O2 at different temperature (273K–343K) were calculated

Fig. 2. Schematic diagram of free energy for the addition reactions of fludioxonil with •OH at different sites at the M06–2X/6–311+G (d, p) level.

3
J. Yang, et al. Ecotoxicology and Environmental Safety 197 (2020) 110644

Table 1
Calculated rate constants (M−1 s−1) between 273 and 343 K in the reaction of fludioxonil and •OH.
T(K) 273 288 298 313 328 343

7 7 7 7 7
Radd1 5.47 × 10 5.29 × 10 5.18 × 10 5.06 × 10 4.96 × 10 4.90 × 107
Radd2 1.57 × 102 2.44 × 102 3.20 × 102 4.64 × 102 6.53 × 102 8.94 × 102
Radd3 1.97 × 104 2.36 × 104 2.64 × 104 3.10 × 104 3.59 × 104 4.10 × 104
Radd4 1.81 × 1010 1.42 × 1010 1.23 × 1010 1.01 × 1010 8.38 × 109 7.13 × 109
Radd5 1.20 × 103 1.66 × 103 2.02 × 103 2.66 × 103 3.42 × 103 4.30 × 103
Radd6 7.52 × 105 8.00 × 105 8.31 × 105 8.77 × 105 9.27 × 105 9.74 × 105
Radd7 4.55 × 104 5.58 × 104 6.32 × 104 7.53 × 104 8.84 × 104 1.03 × 105
Radd8 1.16 × 105 1.34 × 105 1.47 × 105 1.67 × 105 1.88 × 105 2.09 × 105
Radd9 2.19 × 105 2.28 × 105 2.34 × 105 2.43 × 105 2.51 × 105 2.60 × 105
Radd10 2.62 × 105 2.60 × 105 2.59 × 105 2.57 × 105 2.58 × 105 2.58 × 105
Radd11 2.92 5.51 8.15 14.00 22.7 35.7
Rabs1a 0.44 1.08 1.87 4.00 8.02 1.52
Raba4a 0.74 1.71 2.86 5.82 11.1 20.1
Rabs6a 1.13 × 103 1.70 × 103 2.20 × 103 3.12 × 103 4.31 × 103 5.80 × 103
Rabs7a 3.49 × 103 5.33 × 103 6.90 × 103 9.89 × 103 1.37 × 104 1.86 × 104
Rabs8a 7.29 132 191 317 503 768
OH
ktotal 1.82 × 1010 1.43 × 1010 1.23 × 1010 1.01 × 1010 8.44 × 109 7.18 × 109
10
experimental 1.11( ± 0.02) × 10

Fig. 3. Schematic diagram of free energy for the reactions of fludioxonil with 1O2 at C1–C4 site at the M06–2X/6–311+G (d, p) level.

Table 2
Calculated rate constants (M−1 s−1) between 273 and 343 K in the reaction of fludioxonil and 1O2.
T(K) 273 288 298 313 328 343

−6 −6 −5 −5 −5
Radd1-2 1.33 × 10 5.02 × 10 1.13 × 10 3.50 × 10 9.76 × 10 2.49 × 10−4
Radd1-4 6.48 × 107 4.56 × 107 3.69 × 107 2.74 × 107 2.11 × 107 1.66 × 107
O1 6.48 × 107 4.56 × 107 3.69 × 107 2.74 × 107 2.11 × 107 1.66 × 107
t1/22
experimental 4.90( ± 0.02) × 107

with the TST method based on acquired thermodynamic data. The rate pre-reactive complex cannot be found, leading to that the bimolecular
constants of each pathway, including •OH-addition and H-abstraction reaction strategy is used to calculated the rate constant of H-abstraction
and 1O2-cycloaddition, as well as the overall rate constants and ex- reactions. For the addition reactions, as the pre-reactive complexes
perimental rate constants were displayed in Tables 1 and 2. The reac- exist, the initial reaction of fludioxonil with the •OH and 1O2 can be
tion rate constants of H-abstraction were calculated by the kinetic of described as follows:
bimolecular reactions due to none of pre-reactive complex. For the •OH- k1
addition and 1O2-cycloaddition reaction, Shiroudi and Deleuze (2014) fludioxonil + X ⇄ fludioxonil⋅⋅⋅X
k−1
proposed a method to calculated reaction rate constant involving pre-
reactive complex. As shown before, for the H-abstraction reactions, the k2
fludioxonil⋅⋅⋅X → post reactive complex

4
J. Yang, et al. Ecotoxicology and Environmental Safety 197 (2020) 110644

which X are •OH and 1O2. was confirmed by several experimental approaches. For the 1O2, the
1 1 1
The formula for the total rate constant is shown as follows: O2 O2
half-life (t1/2 ) was expressed by the formula t1/2 O2
= ln2/(ktotal × [1O2]),
where [1O2] is the concentration of 1O2. The value of [1O2] = 10−13 M
ktotal = Keq k 2 (3)
was used because it is typical for reported laboratory measurements on
The calculated overall rate constant OH
(ktotal the •OH with flu-
)for whole water sample. In theoretical calculation, it is noteworthy that the
1
dioxonil is 1.23 × 1010 M−1 s−1 at 298K, which is near the experi- O2
calculated t1/2 is 2.17 days using [1O2] = 10−13 M at 298K. It is near
mental data (1.11( ± 0.02) × 1010 M−1 s−1). And for the 1O2, the total the experimental data (1.6 days). The results could explain the de-
rate constant is 3.69 × 107 M−1 s−1 at 298K, and the experimental gradation rates observed.
value is 4.90( ± 0.02) × 107 M−1 s−1. Comparing the rate constants of Although the rate constants of •OH is higher than 1O2, the role of
theoretical calculation and experimental data, we find that the values 1
O2 cannot be ignored due to its high concentration in aquatic en-
are on the same order of magnitude, which proves that our result is vironments. The degradation of fludioxonil can occur simultaneously
reliable. Furthermore, with the temperature increased, the overall rate according to the •OH and 1O2 initial reaction pathways when both of
constants during fludioxonil transformation decreased. For instance, •OH and 1O2 exist in the system. Considering the concentration of the
the total rate constants for the 1O2-cycloaddition is 6.48 × 107 M−1 s−1 •OH and 1O2 in aquatic environments, the degradation rate can be ex-
X
at 273K, which is 3.9 times higher than that at 343K pressed as the formula r = ktotal [X] [fludioxonil], where [X] represents
(1.66 × 107 M−1 s−1). Therefore, it could be inferred that a relatively the concentration of •OH or O2. Furthermore, the concentration of 1O2
1

lower temperature is more efficient for the transformation of fludiox- is high than OH in aquatic environments. Taken into account, the de-
onil. gradation pathways of fludioxonil was mainly caused by 1O2-initiated
According to the rate constants of the reactions of fludioxonil in- reaction. In the near-surface sunlit waters of ambient environments
itiated by •OH, the half-life (t1/2
OH
) can be calculated and analyzed by the O2
(298K), the t1/2
1
OH
(2.17 days) is shorter than the t1/2 (20.6 days). The
OH OH
formula t1/2 = ln2/(ktotal × [•OH]), where [•OH] is the concentration of results also indicate that 1O2 plays a dominant role in the degradation
•OH in aquatic environment(Sinkkonen and Paasivirta, 2000). And process of fludioxonil. Thus, the 1O2 is indispensable to assess the
lifetime can express the persistence of fludioxonil during the •OH-in- subsequence reaction of fludioxonil and the transformation of photo-
itiated reaction transformation. In this study, the half-lifetime was products.
calculated within the range of 273–343 K, which is listed in Fig. 4. In
surface water, the [•OH] is 10−15–10−18 M approximately(Brezonik
and Fulkerson-Brekken, 1998; Haag and Hoigné, 1985). And in lake 3.3. Photoproducts formation of 1O2-initiation
water and special aquatic environments, such as acid mine wastewater,
the concentration of •OH can be increased to a range of 10−12–10−14 M For the purpose of identifying the environmental fate of fludioxonil
(Brezonik and Fulkerson-Brekken, 1998). Fig. 4 shown that the t1/2 OH
of during 1O2 initiated photochemical transformation in environmental
fludioxonil initiated by •OH decreased from 3.06 year to 9.6 h with the water, the subsequent reactions of the intermediates were also con-
increase of [•OH] from 10−18–10−15 M at the temperature 273–343K. sidered. According to the measured products in experiment and reac-
OH tion mechanisms proposed by Apell et al. (2019), we studied the re-
Furthermore, in ambient environments(298K), the t1/2 decreased from
1.79 year to 15.6 h as the [•OH] increase from 10−18-10−15 M, and action pathways of fludioxonil with 1O2, which is shown in Fig. S4.
when [•OH] up to 10−12 M, the t1/2 OH
(56.3 s) is short-lived. This study Also, Fig. S3 shows the free energy profile for the initiation reaction of
indicates that the fate and persistence of fludioxonil in aquatic en- fludioxonil by 1O2 (Radd1-2) and the subsequent reactions. As can be
vironmental deserve more attention. seen in Fig. S3, in the initial reaction, the Gibbs free energy of pre-
As mentioned above, the experiment gives a rate constants of flu- reactive complexes (C1-1) is higher than that of the reactants due to the
dioxonil with •OH, which is 1.11( ± 0.02) × 1010 M−1 s−1. The result larger contribution of entropy. The complex C1-1 can add to the pyrrole
would lead to a half-life (t1/2) on the order of 23 days considering ring and the process need to overcome the barrier of 21.85 kcal mol−1
[•OH] = 10−16.5 M in near-surface sunlit waters. Corresponding to the to form the dioxetanes product (PC1), which was detected by laser flash
theoretical calculation result, the t1/2 is 20.6 days. The theoretical value photolysis experiments. Furthermore, the PC1 can undergo rearrange-
is close to the experimental data. The reaction of fludioxonil with 1O2 ment and ring-opening to the intermediate IM2-1. In this reaction, the
Gibbs free energy barrier of transition state TS2-1 is 32.6 kcal mol−1
and this is an exothermic process with reaction energy 79.6 kcal mol−1.
Then, the intermediate IM2-1 can undergo hydrolysis reaction with a
barrier of 54.2 kcal mol−1, thereby resulting in the formation of IM3-1.
Furthermore, the IM3-1 could react with H2O or CH3•OH through an
exothermic process with a reaction energy of 81.7, or 80.9 kcal mol−1,
leading to the formation of epoxy acid product (PC3) or epoxy ester
product (PC2).
In addition, 1O2 has also been proposed to react with pyrroles by
1,4-cycloaddition to form endo-peroxide intermediates. The initial and
subsequent reactions of C1–C4 are shown in Fig. 3. Obviously, the free
energy barrier of Radd1-4 initial pathway (4.51 kcal mol−1) is
17.34 kcal mol−1 lower than that of Radd1-2 initial pathway. The IM1 is
formed by the cycloaddition process with releasing energy of
15.72 kcal mol−1. The IM1 can successively undergo C–O bond broken
reaction at C1 site (or C4 site) with the assistance of water molecular. In
this process, the intermediate IM2-1 (or IM2) is produced by passing
through the transition state TS2-1 (or TS2) with the free energy barrier
24.16 kcal mol−1 (or 26.03 kcal mol−1). Then, the IM2-1 can hydrolyze
with water molecular to form the hydroperoxide intermediate IM3-1.
The reaction needs to overcome the free energy barrier
Fig. 4. Calculated half-life (t1/2) of fludioxonil transformation as a function of 3.76 kcal mol−1, and the process is a high exothermic process with
[•OH] (unit: M) in temperature range of 273–343 K in natural waters. 29.16 kcal mol−1. The IM3-1 can undergo isomer transformation

5
J. Yang, et al. Ecotoxicology and Environmental Safety 197 (2020) 110644

reaction with a free energy barrier of 37.61 kcal mol−1. After the IM4-1
formation, the dehydration reaction proceeds via a transition state TS5-
1 (Gibbs free energy barriers of 60.19 kcal mol−1), into PC4 simulta-
neously with releasing large energies of 150.74 kcal mol−1. Subsequent
reaction of PC4 by losing of a molecular H2 is proposed to yield PC6. As
shown in Fig. 3, the intermediate IM2, generated by C–O broken at C4
site can undergo similar process to form the product PC5 and PC6.
Considering the previous research, the products PC1, PC4, PC5 and PC6
were identified in the indirect photodegradation experiment of flu-
dioxonil with 1O2.

3.4. Eco-toxicity assessment

ECOSAR program is an important tool for computational toxicology,

which is widely applied to predict environmental and health risks
(Sanderson et al., 2003). Thus, this study used the ECOSAR model to
evaluated the acute toxicity (LC50 and EC50) and chronic toxicity (Chv)
of fludioxonil and its transformation products. Moreover, the results of
eco-toxicity assessment of fludioxonil and its degradation products are
preliminary predictions based on the computational toxicology pro-
gram. And experimental evidence is needed for more accurate toxicity
assessment. The relationship between ecotoxicity (LC50, EC50, and ChV)
and photoproducts was illustrated in Fig. 5. The obtained LC50 of flu-
dioxonil for fish was lower than 1.0 mg L−1, and the range
(LC50 < 1.0 mg L−1) is considered as a very toxic compound in ac-
cordance with the aquatic toxicity criteria of the European Union
(Table S1). Thus, fludioxonil can be regard as a high acute toxicity
compound. The chronic toxicity of fludioxonil is evaluated as ChV for
fish, daphnia and green algae, and these values are 0.03, 0.02, and
0.09 mg L−1, respectively. The results suggest that fludioxonil is a very
chronically toxic compound. Therefore, the aquatic toxicity of its pro-
ducts needs to pay more attention.
The PC1, PC2 and PC3 are formed in the Radd12 cycloaddition re-
action pathway. The LC50 of PC1 is 0.92 mg L−1, which is higher than
the fludioxonil, but it still be regard as very toxic level. Furthermore,
the acute toxicity of product PC2 (or PC3) is classified as harmful (or
not harmful) level due to its LC50 of 14.5 mg L−1 (or 336.2 mg L−1) for
fish. For the chronic toxicity, the ChV values are 0.5 (PC1), 0.9 (PC2)
and 5.5 (PC3) mg L−1 for fish, respectively. The results indicate that the
toxicity is decreased by the degradation processes comparing with
fludioxonil. As shown in Fig. 5, the LC50 values for fish are 55.1, 2.9,
and 1.2 mg L−1 for products PC4, PC5, and PC6, which are higher than Fig. 5. Acute toxicity (a) and chronic toxicity (b) (unit: mg L−1) of fludioxonil
that of fludioxonil (0.2 mg L−1). This indicates that their acute toxicity and its transformation products of 1O2 pathways.
is lower than the original contaminant. Although the toxicities are de-
creased with the degradation of fludioxonil, some products still keep at epoxy ester products. In order to understand the eco-toxicity of the
toxic level. Thus, the eco-toxicity from these products and original fludioxonil and the transformation products, the ECOSAR program, as a
contaminants needs to be seriously considered in the indirect photo- reliable method was used to predict aquatic toxicities. Although these
degradation process. photoproducts showed a decreased aquatic toxicity, much attention
should be paid to the transformation process.
4. Conclusion

This study demonstrated that the degradation of fludioxonil can CRediT authorship contribution statement
occur through the indirect photodegradation with •OH and 1O2. The
Gibbs free energy barrier of •OH addition reaction at C4 of fludioxonil is Jiaoxue Yang: Investigation, Writing - original draft. Zehua Wang:
1.95 kcal mol−1, which is lower than other pathways, and the Gibbs Software. Guochun Lv: Writing - review & editing. Wen Liu: Software.
free energy barrier of addition reaction by 1O2 is 4.51 kcal mol−1 Yan Wang: Validation. Xiaomin Sun: Resources, Supervision. Jian
through Radd14 process. The two favorable pathways are both occurred Gao: Supervision.
in pyrrole ring. The rate constants of initiation reactions through the
two oxidants are calculated to evaluate their environmental fate. We
found that the rate constants of the degradation reaction initiated by Declaration of competing interest
•OH and 1O2 are 1.23 × 1010 and 3.69 × 107 M−1s−1 at 298K, re-
spectively, which results in half-lives of < 2 days in surface waters The authors declare that they have no known competing financial
under sunlit near-surface conditions. The results of theoretical calcu- interests or personal relationships that could have appeared to influ-
lation are in agreement with experimental data. The main transfor- ence the work reported in this paper.
mation mechanism was proposed as an 1O2 cycloaddition reaction of
fludioxonil, leading to the formation of endo-peroxide, epoxy acid and

6
J. Yang, et al. Ecotoxicology and Environmental Safety 197 (2020) 110644

Acknowledgements risk assessment for fludioxonil in Sclerotinia homoeocarpa in China. Pestic. Biochem.
Physiol. 156, 123–128.
Jespers, A.B.K., De Waard, M.A., 1995. Effect of fenpiclonil on phosphorylation of glucose
This work was supported by National Natural Science Foundation of in Fusarium sulphureum. Pestic. Sci. 44, 167–175.
China (21976109), Natural Science Foundation of Shandong Province Ju, L.-P., Han, K.-L., Zhang, J.Z.H., 2009. Global dynamics and transition state theories:
(ZR2018MB043), The Fundamental Research Funds of Shandong comparative study of reaction rate constants for gas-phase chemical reactions. J.
Comput. Chem. 30, 305–316.
University (2018JC027), Shandong Province Key Research and Kanetis, L., Förster, H., Adaskaveg, J.E., 2007. Comparative efficacy of the new post-
Development Program (2019GSF109037). harvest fungicides azoxystrobin, fludioxonil, and pyrimethanil for managing citrus
green mold. Plant Dis. 91, 1502–1511.
Kaur, R., Anastasio, C., 2017. Light absorption and the photoformation of hydroxyl ra-
Appendix A. Supplementary data dical and singlet oxygen in fog waters. Atmos. Environ. 164, 387–397.
Lassalle, Y., Nicol, É., Genty, C., Bourcier, S., Bouchonnet, S., 2015. Structural elucidation
Supplementary data to this article can be found online at https:// and estimation of the acute toxicity of the major UV–visible photoproduct of flu-
dioxonil – detection in both skin and flesh samples of grape. J. Mass Spectrom. 50,
doi.org/10.1016/j.ecoenv.2020.110644.
864–869.
Li, C., Xie, H.-B., Chen, J., Yang, X., Zhang, Y., Qiao, X., 2014. Predicting gaseous reaction
References rates of short chain chlorinated paraffins with ·OH: overcoming the difficulty in ex-
perimental determination. Environ. Sci. Technol. 48, 13808–13816.
Luo, S., Gao, L., Wei, Z., Spinney, R., Dionysiou, D.D., Hu, W.-P., Chai, L., Xiao, R., 2018.
Apell, J.N., Pflug, N.C., McNeill, K., 2019. Photodegradation of fludioxonil and other Kinetic and mechanistic aspects of hydroxyl radical‒mediated degradation of na-
pyrroles: the importance of indirect photodegradation for understanding environ- proxen and reaction intermediates. Water Res. 137, 233–241.
mental fate and photoproduct formation. Environ. Sci. Technol. 53, 11240–11250. Maeda, S., Harabuchi, Y., Ono, Y., Taketsugu, T., Morokuma, K., 2015. Intrinsic reaction
Arias, M., Torrente, A.C., López, E., Soto, B., Simal-Gándara, J., 2005. coordinate: calculation, bifurcation, and automated search. Int. J. Quant. Chem. 115,
Adsorption−Desorption dynamics of cyprodinil and fludioxonil in vineyard soils. J. 258–269.
Agric. Food Chem. 53, 5675–5681. Mayer, J.M., 2011. Understanding hydrogen atom transfer: from bond strengths to
Avenot, H.F., Michailides, T.J., 2015. Detection of isolates of Alternaria alternata with Marcus theory. Acco. Chem. Res. 44, 36–46.
multiple-resistance to fludioxonil, cyprodinil, boscalid and pyraclostrobin in Miller, W.H., 1993. Beyond transition-state theory: a rigorous quantum theory of che-
California pistachio orchards. Crop Protect. 78, 214–221. mical reaction rates. Acco. Chem. Res. 26, 174–181.
Bhoorasingh, P.L., Slakman, B.L., Seyedzadeh Khanshan, F., Cain, J.Y., West, R.H., 2017. Nayebzadeh, M., Vahedpour, M., Shiroudi, A., Rius-Bartra, J.M., 2020. Kinetics and
Automated transition state theory calculations for high-throughput kinetics. J. Phys. oxidation mechanism of pyrene initiated by hydroxyl radical. A theoretical in-
Chem. 121, 6896–6904. vestigation. Chem. Phys. 528, 110522.
Brandhorst, T.T., Klein, B.S., 2019. Uncertainty surrounding the mechanism and safety of Nicol, E., Chayata, H., Genty, C., Bouchonnet, S., Bourcier, S., 2016. Photodegradation of
the post-harvest fungicide fludioxonil. Food Chem. Toxicol. 123, 561–565. cyprodinil under UV–visible irradiation – chemical and toxicological approaches.
Brezonik, P.L., Fulkerson-Brekken, J., 1998. Nitrate-induced photolysis in natural Waters: Rapid Commun. Mass Spectrom. 30, 2201–2211.
controls on concentrations of hydroxyl radical photo-intermediates by natural Pose-Juan, E., Cancho-Grande, B., Rial-Otero, R., Simal-Gándara, J., 2006. The dissipa-
scavenging agents. Environ. Sci. Technol. 32, 3004–3010. tion rates of cyprodinil, fludioxonil, procymidone and vinclozoline during storage of
Burden, N., Maynard, S.K., Weltje, L., Wheeler, J.R., 2016. The utility of QSARs in pre- grape juice. Food Contr. 17, 1012–1017.
dicting acute fish toxicity of pesticide metabolites: a retrospective validation ap- Razavi, B., Ben Abdelmelek, S., Song, W., O'Shea, K.E., Cooper, W.J., 2011.
proach. Regul. Toxicol. Pharmacol. 80, 241–246. Photochemical fate of atorvastatin (lipitor) in simulated natural waters. Water Res.
Cabras, P., Angioni, A., Garau, V.L., Melis, M., Pirisi, F.M., Minelli, E.V., Cabitza, F., 45, 625–631.
Cubeddu, M., 1997a. Fate of some new fungicides (cyprodinil, fludioxonil, pyr- Sanderson, H., Johnson, D.J., Wilson, C.J., Brain, R.A., Solomon, K.R., 2003. Probabilistic
imethanil, and tebuconazole) from vine to wine. J. Agric. Food Chem. 45, hazard assessment of environmentally occurring pharmaceuticals toxicity to fish,
2708–2710. daphnids and algae by ECOSAR screening. Toxicol. Lett. 144, 383–395.
Cabras, P., Angioni, A., Garau, V.L., Melis, M., Pirisi, F.M., Farris, G.A., Sotgiu, C., Minelli, Schirra, M., D'Aquino, S., Palma, A., Marceddu, S., Angioni, A., Cabras, P., Scherm, B.,
E.V., 1997b. Persistence and metabolism of folpet in grapes and wine. J. Agric. Food Migheli, Q., 2005. Residue level, persistence, and storage performance of citrus fruit
Chem. 45, 476–479. treated with fludioxonil. J. Agric. Food Chem. 53, 6718–6724.
Canneaux, S., Bohr, F., Henon, E., 2014. KiSThelP: a program to predict thermodynamic Scott, J.C., Pfluger, P., Krounbi, M.T., Street, G.B., 1983. Electron-spin-resonance studies
properties and rate constants from quantum chemistry results†. J. Comput. Chem. 35, of pyrrole polymers: evidence for bipolarons. Phys. Rev. B 28, 2140–2145.
82–93. Shi, J., Liu, H., Sun, L., Hou, H., Xu, Y., Wang, Z., 2011. Theoretical study on hydro-
Cermola, F., DellaGreca, M., Iesce, M.R., Montanaro, S., Previtera, L., Temussi, F., 2006. philicity and thermodynamic properties of polyfluorinated dibenzofurans.
Photochemical behavior of the drug atorvastatin in water. Tetrahedron 62, Chemosphere 84, 296–304.
7390–7395. Shiroudi, A., Deleuze, M.S., 2014. Theoretical study of the oxidation mechanisms of
Chou, M., Creutz, C., Sutin, N., 1977. Rate constants and activation parameters for outer- naphthalene initiated by hydroxyl radicals: the H abstraction pathway. J. Phys.
sphere electron-transfer reactions and comparisons with the predictions of Marcus Chem. 118, 3625–3636.
theory. J. Am. Chem. Soc. 99, 5615–5623. Sinkkonen, S., Paasivirta, J., 2000. Degradation half-life times of PCDDs, PCDFs and PCBs
Costa, D., Gomes, A., Lima, J.L.F.C., Fernandes, E., 2008. Singlet oxygen scavenging ac- for environmental fate modeling. Chemosphere 40, 943–949.
tivity of non-steroidal anti-inflammatory drugs. Redox Rep. 13, 153–160. Takata, M., Lin, B.-L., Xue, M., Zushi, Y., Terada, A., Hosomi, M., 2020. Predicting the
Cote, C.D., Schneider, S.R., Lyu, M., Gao, S., Gan, L., Holod, A.J., Chou, T.H.H., Styler, acute ecotoxicity of chemical substances by machine learning using graph theory.
S.A., 2018. Photochemical production of singlet oxygen by urban road dust. Environ. Chemosphere 238, 124604.
Sci. Technol. Lett. 5, 92–97. Truhlar, D.G., Kuppermann, A., 1971. A test of transition state theory against exact
Dadashi-Silab, S., Doran, S., Yagci, Y., 2016. Photoinduced electron transfer reactions for quantum mechanical calculations11This work was supported in part by the United
macromolecular syntheses. Chem. Rev. 116, 10212–10275. States Atomic Energy Commission, Report Code No. CALT-767P4-74. Chem. Phys.
Dewez, D., Geoffroy, L., Vernet, G., Popovic, R., 2005. Determination of photosynthetic Lett. 9, 269–272.
and enzymatic biomarkers sensitivity used to evaluate toxic effects of copper and Verdisson, S., Couderchet, M., Vernet, G., 2001. Effects of procymidone, fludioxonil and
fludioxonil in alga Scenedesmus obliquus. Aquat. Toxicol. 74, 150–159. pyrimethanil on two non-target aquatic plants. Chemosphere 44, 467–474.
Duan, Y., Ge, C., Liu, S., Chen, C., Zhou, M., 2013. Effect of phenylpyrrole fungicide Walker, M., Harvey, A.J.A., Sen, A., Dessent, C.E.H., 2013. Performance of M06, M06-2X,
fludioxonil on morphological and physiological characteristics of Sclerotinia scler- and M06-HF density functionals for conformationally flexible anionic clusters: M06
otiorum. Pestic. Biochem. Physiol. 106, 61–67. functionals perform better than B3LYP for a model system with dispersion and ionic
Fenoll, J., Ruiz, E., Hellín, P., Flores, P., Navarro, S., 2011. Heterogeneous photocatalytic hydrogen-bonding interactions. J. Phys. Chem. 117, 12590–12600.
oxidation of cyprodinil and fludioxonil in leaching water under solar irradiation. Wilkinson, F., Helman, W.P., Ross, A.B., 1995. Rate constants for the decay and reactions
Chemosphere 85, 1262–1268. of the lowest electronically excited singlet state of molecular oxygen in solution. An
Fudickar, W., Linker, T., 2019. Theoretical insights into the effect of solvents on the [4 + expanded and revised compilation. J. Phys. Chem. Ref. Data 24, 663–677.
2] cycloaddition of singlet oxygen to substituted anthracenes: a change from a Xia, F.-F., Yi, H.-B., Zeng, D., 2009. Hydrates of copper dichloride in aqueous solution: a
stepwise process to a concerted process. J. Phys. Org. Chem. 32, e3951. density functional theory and polarized continuum model investigation. J. Phys.
Gao, R., Sun, X., Yu, W., Zhang, Q., Wang, W., 2014. Mechanism and rate constants for Chem. 113, 14029–14038.
complete series reactions of 19 fluorophenols with atomic H. J. Environ. Sci. 26, Xiao, R., Luo, Z., Wei, Z., Luo, S., Spinney, R., Yang, W., Dionysiou, D.D., 2018. Activation
154–159. of peroxymonosulfate/persulfate by nanomaterials for sulfate radical-based advanced
Gao, Y., Li, G., Qin, Y., Ji, Y., Mai, B., An, T., 2019. New theoretical insight into indirect oxidation technologies. Current Opinion in Chemical Engineering 19, 51–58.
photochemical transformation of fragrance nitro-musks: mechanisms, eco-toxicity Yang, Y., et al., 2017. Degradation of sulfamethoxazole by UV, UV/H2O2 and UV/per-
and health effects. Environ. Int. 129, 68–75. sulfate (PDS): formation of oxidation products and effect of bicarbonate. Water Res.
Georgievskii, Y., Klippenstein, S.J., 2003. Transition state theory for multichannel addi- 118, 196–207.
tion Reactions: multifaceted dividing surfaces. J. Phys. Chem. 107, 9776–9781. Zeinali, N., Oluwoye, I., Altarawneh, M., Dlugogorski, B.Z., 2019. The mechanism of
Haag, W.R., Hoigné, J., 1985. Photo-sensitized oxidation in natural water via .OH radi- electrophilic addition of singlet oxygen to pyrrolic ring. Theoretical Chemistry
cals. Chemosphere 14, 1659–1671. Accounts 138, 90.
Hu, J., Zhou, Y., Gao, T., Geng, J., Dai, Y., Ren, H., Lamour, K., Liu, X., 2019. Resistance Zhang, R., Yang, Y., Huang, C.-H., Li, N., Liu, H., Zhao, L., Sun, P., 2016. UV/H2O2 and

7
J. Yang, et al. Ecotoxicology and Environmental Safety 197 (2020) 110644

UV/PDS treatment of trimethoprim and sulfamethoxazole in synthetic human urine: anthracene and tetracyanoethylene dimers: a computational study based on a range
transformation products and toxicity. Environ. Sci. Technol. 50, 2573–2583. separated hybrid functional and charge constrained self-consistent field with
Zhao, Y., Truhlar, D.G., 2008. The M06 suite of density functionals for main group switching Gaussian polarized continuum models. J. Chem. Theor. Comput. 9,
thermochemistry, thermochemical kinetics, noncovalent interactions, excited states, 1125–1131.
and transition elements: two new functionals and systematic testing of four M06-class Zhou, Q., Sun, X., Gao, R., Hu, J., 2011. Mechanism and kinetic properties for OH-in-
functionals and 12 other functionals. Theoretical Chemistry Accounts 120, 215–241. itiated atmospheric degradation of the organophosphorus pesticide diazinon. Atmos.
Zheng, S., Geva, E., Dunietz, B.D., 2013. Solvated charge transfer states of functionalized Environ. 45, 3141–3148.