Journal of Environmental Chemical Engineering 10 (2022) 108311

Contents lists available at ScienceDirect

Journal of Environmental Chemical Engineering

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

Theoretical insight into the degradation of diclofenac by hydroxyl and

sulfate radicals in aqueous-phase: Mechanisms, kinetics and eco-toxicity
Jiaoxue Yang a, *, Guochun Lv b, Tingting Li a, Shuchen Sun a, Xiaomin Sun c, *
a
School of Geography and Environment, Liaocheng University, Liaocheng 252000, China
b
College of Environmental Sciences, Sichuan Agricultural University, Chengdu 611130, China
c
Environment Research Institute, Shandong University, Qingdao 266237, China

A R T I C L E I N F O A B S T R A C T

Editor: Xin Yang Diclofenac (DCF) is a widely used non-steroidal anti-inflammatory drug, which has attracted more and more
attention due to its biotoxicity and refractory degradation. The objective of this study was to systematically
Keywords: analysis the transformation mechanisms, kinetics and toxicity of DCF initiated by •OH and SO4 •– in persulfate
Diclofenac oxidation process. In this study, the degradation mechanism of DCF has been performed by density functional
Radicals
theory (DFT) methods, and the toxicity of the by-products was studied by computational toxicology. The results
Degradation mechanism
show that the reactivity of benzylic position of DCF is higher than that of chlorinated aromatic ring, indicating
Rate constant
Toxicity assessment •OH and SO4 •– are more likely to react in benzylic position. By comparing the initiation reactions of •OH and
SO4 •–, it was found that •OH-initiations were the predominant pathway in the degradation process. The kinetic
calculations show that the degradation rate constants were positively correlated with temperature. Furthermore,
the calculated toxicological results indicate that the toxicity level of most products is reduced, but the hydro­
lysates remain toxic and deserve more attention for the degradation by-products of DCF. The comprehensive
theoretical study is helpful to reveal the microscopic mechanism of DCF transformation and AOP-wastewater
remediation.

1. Introduction DCF has got much worldwide attention [11]. The zebra fish showed
some acute toxic symptoms like unusual swimming, staying at bottom
Pharmaceuticals and personal care products (PPCPs) have been and rapid tumbling in the DCF solution [12]. Thus, it is a critically
recognized as one of the most pressing environmental concern due to its important issue for researching the DCF removal in aquatic
frequent occurrence and persistence in the aquatic environment [1–3]. environment.
There is a rising concern over the risk it may pose to potential adverse Conventional sewage treatment technologies (such as coagulation
effect on ecosystems and human health [4]. Diclofenac (DCF, (2-[2′ , and biotreatment) have limited degradation efficiency for DCF [13]. In
6′ -dichlorophenyl)amino] phenylacetic acid), as representative PPCPs, recent years, advanced oxidation processes (AOPs) have become alter­
is considered the most popular painkiller and the most commonly used natives to traditional treatment and have been proven to be promising
non-steroidal anti-inflammatory drug for humans and domestic animals technologies for PPCPs removal [14,15]. Sulfate radical (SO4 •–) based
[5]. The market share of DCF is close to that of the other three advanced oxidation processes (SR-AOPs) have attracted more and more
concomitant drugs (ibuprofen, methanic acid and naproxen) [6,7]. DCF attention because of its high efficiency of mineralization of organic
is continually monitored in waste water from sewage disposal plants and pollutants [16,17]. The SO4 •– can be generated by activation of per­
surface water sources because of the low biodegradability and easy sulfate (S2O2–
8 , PS) and peroxymonosulfate (HSO5, PMS) by UV, heat,
–

excretion with urine [8,9]. Previous studies shown that the average transition metals and by catalysts [18–20]. SO4 •– can be generated by
concentration of DCF in surface water is 1.2 μg L− 1[10]. The sudden radical chain reaction (S2O2– 8 →2SO4 ) in the UV/PS system [21].
•–

extinction of vultures by eating carcasses containing traces of DCF was Meanwhile, it was reported that SO•– 4 can react with OH to form hy­
–

the first widely publicized case of drug-induced ecological damage. droxyl radical (•OH) with a rate constant of 6.5 × 10 M− 1 s− 1 under
7
2–
After a series of ecological pollution events in the early 20th century, alkaline condition (SO•–4 + OH →•OH + SO4 ) [22]. The research of
–

* Corresponding authors.
E-mail addresses: [email protected] (J. Yang), [email protected] (X. Sun).

https://doi.org/10.1016/j.jece.2022.108311
Received 11 May 2022; Received in revised form 8 July 2022; Accepted 20 July 2022
Available online 26 July 2022
2213-3437/© 2022 Elsevier Ltd. All rights reserved.
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

Fig. 1. Molecular surface electrostatic potential distribution diagram of DCF. The blue color represents an electron- deficient region; and the red area represents an
electron-rich region.

PCBs indicated that the SO•–4 and •OH coexisted as 9 <pH< 11 [22]. The Meanwhile, vibrational frequencies were calculated under the same
•OH is non-selective and its redox potential ranges from 1.8 to 2.8 V methodology and basis to characterized the local minima with all real
[23]. SO•–4 is a strong oxidant with high oxidation (E0 ≈2.5–3.1 V) ef­ frequencies, or transition states with single imaginary frequency [38].
ficiency to organic pollutants [24]. SO•–
4 is more selective than •OH and Intrinsic reaction coordinate (IRC) calculations of the transition state
prefers to react with electron-rich groups by electron-transfer reactions were carried out to demonstrate the correctness of the TS configuration
[25,26]. (TS connects the designed reactant to the product) [39]. In addition, a
There are some researches elucidated that SO•– 4 -based degradation high-precision basis set, 6–311 + +G(3df,3pd), was used to further
has a high effective removal of PPCPs [27–29]. It was reported that the refine the single point energies of all optimized geometries to obtain the
reaction rate constants of SO•– 4 towards aminopyralid and picloram accurate energies. The influence of the aquatic environment was
herbicides (containing amino group on pyridine ring) were determined considered using the polarizable continuum model (PCM) of a
at 1.56 × 109 M− 1s− 1 and 1.21 × 109 M− 1s− 1 [30]. The experimental self-consistent reaction field (SCRF) theory [40–42]. The PCM method
study shown that the biomolecular rate constant of DCF for •OH was has been verified to be efficient and reliable in calculating the solvation
(9.29 ± 0.11) × 109 M− 1s− 1 [14]. However, there are less detailed energies for molecular properties of chemical system [43].
mechanism as well as their transformation pathways study of DCF in Moreover, Marcus theory were used to calculated the activation
SO•–4 -based degradation. Theoretical calculation method is widely used barrier (ΔG∕
SET ) in the single-electron transfer (SET) process [44,45]. The
=

to investigate the reaction mechanism and kinetic parameter [31,32].

calculation formula for activation barrier (ΔG∕
SET ) of SET process is
=
Understanding the degradation mechanism of DCF will help predict its
shown below [46]:
environmental fate in wastewater.
The purpose of this paper was to establish the degradation mecha­ (λ + ΔGSET )2
ΔG∕
=
SET = (1)
nism, reaction rate constants as well as the degradation products for 4λ
reaction of SO•–4 and •OH with DCF via quantum chemistry method.
Furthermore, the computational toxicology method was adopted to Where ΔGSET is the free energy of reaction, λpresent the nuclear reor­
predict the ecotoxicity of DCF and by-products in aqueous environment. ganization energy of SET.
The addition reaction, abstract reaction and single electron transfer of
transformation mechanisms, the reaction efficiency of DCF during SO•– 4 2.2. Kinetic calculation
and •OH-initiated in aquatic environments were systematically eluci­
dated [33]. Meanwhile, ECOSAR and TEST software were chosen to The analysis of dynamic results was performed through the
evaluate the relative toxicity of the by-products compared to their transition-state theory (TST), a general method to calculate the dynamic
parent compounds (DCF) in order to estimate the potential environ­ of infrequent event [47,48]. The algorithm, which is practical for re­
mental risks. The obtained theoretical results would help us to better actants containing dozens of atoms, predicts both microcanonical and
understand the degradation of a wide variety of PPCPs in SO•– 4 -AOPs canonical rate constants. In this work, the rate constants were calculated
system and make up for the limitations of experimental methods. over the temperature ranging of 273–343 K at 1 M standard state. The
rate constants were calculated using Eq. (2):
2. Computational method ( )
kB T ΔG∕=
ka = σ κ exp − (2)
h RT
2.1. Quantum chemistry calculation
Where σ is reaction path degeneracy; κ is wigner tunneling correction; T
In this study, all configuration optimizations in the calculation of
stands for temperature, kB, h and R represent Boltzmann’s constant,
DCF degradation mechanisms were carried out by Gaussian 09 program
Planck’s constant and the gas constant, respectively; ΔG∕ =
is the Gibbs
software [34,35]. The molecular geometries of stationary points (mini­
free energy of activation of the radical reaction. In this study, Col­
mum value of reactants, intermediates and products), as well as transi­
lins–Kimball theory was used to correct the values of the thermal rate
tion states (TS) involved in the degradation of DCF were calculated using
constant (ka ) obtained by TST, considering the effect of diffusion limit
M06–2X functional with a level 6–311 +G(d, p) basis set [36,37].
[49]. The corrected constant was expressed as Eq. (3):

2
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

Fig. 2. All the reaction pathways in the initial reaction of DCF with •OH and SO4•‾.

kD ka 3. Results and discussion

k= (3)
kD + ka
3.1. DCF structure analysis
kD is the steady-state rate constant associated with irreversible
bimolecular diffusion control reactions, defined by Smoluchowski
In this study, the average local ionization energy of DCF is analyzed
equation [50]:
via Multiwfn program to understand its electronic property [57,58]. In
kD = 4πRDAB NA (4) the Fig. 1, the bluer the color is, the more negative the electrostatic
potential is; the redder the area is, the more positive the electrostatic
Where R is reaction distantce; NA is the Avogadro number; DAB express potential is, and the white area is near 0. A small value of the molecular
the mutual diffusion coefficient of reactants A and B [51], which was surface electrostatic potential indicates that electrons are more reactive
calculated by Stokes Einstein method (Eq. (5)): and more prone to electrophilic and free radical reactions. As shown in
kB T Fig. 1, the electrostatic potential diagram of DCF confirms that the ar­
D= (5) omatic region is a relatively electron-deficient active region, and is
6πηa
easily attacked by free radicals, and thus easily occurs nucleophilic ar­
Where kB is Boltzmann constant; η is the viscosity of solvent, and the omatic substitution reaction. Based on the electronic and structural
viscosity of water in this chapter is 8.91 × 10− 4 Pa s; a is the solute optimizations, the DCF degradation mechanisms, including radical
radius. addition reaction, H-abstraction reaction and single electron transfer
reaction, were calculated.
2.3. Ecotoxicity evaluation

The ecological structure-activity relationships (ECOSAR) software 3.2. Degradation mechanisms of DCF during initial step
[52] based on QSAR models was used to study the toxicity of DCF and its
oxidation by-products to aquatic organism (green algae, daphnia, and The •OH and SO•– 4 radicals have a strong redox potential and can
fish). The acute toxicity of DCF and degradation by-products were participate in the degradation process of DCF in aqueous-phase UV-
measured in terms of 50 % lethal concentration (LC50) of fish and activated persulfate process. Fig. 2 expressed all possible pathways of
daphnia after 96 and 48 h exposures, or 50 % effective concentration initial reaction. The structure of DCF consists of two aromatic rings
(EC50) of green algae after 96 h [53]. Chronic toxicity is estimated in connected by -NH group. In this study, three different reaction types
terms of ChV [54]. In addition, the developmental toxicity and muta­ including radical adds to the benzene ring (Radd), H transfer reaction
genicity of DCF and products were performed using the toxicity esti­ (Rabs) and single electron transfer process (Rset) were taken into account.
mation software tool (TEST) program [55]. Based on the Globally The mechanism of reaction initiated by •OH and SO•– 4 radicals is dis­
Harmonized System of Classification and Labelling of Chemicals (GHS), cussed in detail below.
the estimated ecotoxicity values were classified into four levels [56].
3.2.1. •OH-initiated reaction of DCF
As shown in Fig. 3 (c), Gibbs free energy barrier of DCF reaction with
•OH of Rset pathway is 22.91 kcal mol− 1, which is calculated by Marcus

3
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

Fig. 3. Schematic diagram of free energy for all the possible •OH-initiated reactions of DCF at the M06–2X/6–311 +G (d, p) level.

4
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

Fig. 4. The free energy of activation (ΔG∕

=
) and Gibbs free energy (.) changes for the overall •OH and SO4•‾ initial at all reaction sites.

theory based the free energy of reaction (ΔGSET ). It can be found that the H-abstraction reaction mechanisms.
activation barrier of SET pathway is high than that of •OH-addition For addition (DCF + •OH ̅̅̅̅→ •DCF…OH) and abstraction (DCF
Radd

(7.54–16.77 kcal mol− 1) and abstraction (11.22–15.77 kcal mol− 1) + •OH →•Rabs + H2O) mechanism of DCF reaction initiated by •OH, the
process, which probably caused by the fact that DCF contains weak stability of the radicals is calculated through Eq. (6) and Eq. (7) and the
electron-withdrawing group (-COOH). In addition, SET pathway is an Gibbs free energy of all the possible •OH-initiated pathways are shown
endothermic reaction with a reaction energy of 21.62 kcal mol− 1. While in Fig. S1.
all the addition and abstraction reactions are exothermic processes. The
above analysis demonstrates that SET pathway is less likely to occur ΔGr-add=ΔG (•DCF…OH) - ΔG (DCF) - ΔG (•OH) (6)
comparing with Radd and Rabs channel from the viewpoint of kinetics and
ΔGr-abs=ΔG (•Rabs) + ΔG (H2O) - ΔG (DCF) - ΔG (•OH) (7)
thermodynamics. Thus, the SET process can be excluded in the •OH
initiation degradation process of DCF in the aqueous phase. In the OH-initiated reactions can occur at different sites of the DCF mole­
following analysis, we will pay more attention on the •OH-addition and cule as shown in Fig. 3. •OH can attack the C atom in the aromatic rings

5
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

Fig. 5. Schematic diagram of free energy for SO4•‾-initiated reactions of DCF at the M06–2X/6–311 +G (d, p) level.

Table 1
1
Calculated overall rate constants (M− s− 1) between 273 and 343 K in the reaction of DCF with •OH, SO4•‾.
T (K) 273 288 298 313 328 343

kOH
abs
4.46 × 105 1.41 × 106 2.85 × 106 7.61 × 106 1.87 × 107 4.26 × 107
kOH
add
2.94 × 108 6.75 × 108 1.12 × 109 2.22 × 109 4.09 × 109 6.95 × 1010
5 4 3 2 1 1
kOH
set
5.85 × 10− 5.88 × 10− 2.41 × 10− 1.70 × 10− 1.01 × 10− 5.12 × 10−
kOH
total
2.95 × 108 6.76 × 108 1.12 × 109 2.2 × 109 4.11 × 109 6.99 × 109
kSO 4 57.9 2.86 × 102 7.60 × 102 2.94 × 103 1.02 × 104 3.17 × 104
abs
kSO 4 2.43 × 107 6.46 × 107 1.18 × 108 2.70 × 108 5.74 × 108 1.14 × 109
add
kSO
set
4 2.35 × 106 6.63 × 106 1.25 × 107 3.02 × 107 6.74 × 107 1.41 × 108
kSO 4 2.67 × 107 7.13 × 107 1.30 × 108 3.00 × 108 6.42 × 108 1.29 × 109
total

Fig. 6. Schematic diagram of free energy for subsequent reactions of IM11 in the •OH-addition reaction of DCF in the degradation process.

6
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

There are two reaction sites on DCF for SO4•‾ addition, including
benzylic position and chlorinated aromatic ring addition. Fig. 5 shows
the potential energy profiles of the lowest barrier reactions at these two
different sites. For SO4•‾ addition reactions, the most likely site to occur
is C9, where the barrier is 8.96 kcal mol− 1, and 0.25 kcal mol− 1of en­
ergy is released. Furthermore, H atoms linked to C atom at site 10
(Rabs10a) from the benzylic position are more likely to be extracted to
form HSO–4. The Gibbs free energy barrier for the abstraction reaction
(Rabs10a) is 1.8 times higher than that for the most reactive addition
process (Radd9). As shown in Fig. 4 (a), it can be seen that the Gibbs free
energy barriers of SO4•‾- addition reactions are lower than that of
extraction reactions. By comparing the energy barriers of H-abstraction
reactions at different sites, it can be found that the barriers on benzylic
position are lower than those on aromatic rings containing chlorine
substituents. The results indicate that H-extraction reaction that occurs
at benzylic position is easier to react from a thermodynamic and kinetic
standpoint. The results are consistent with previous reports that
hydrogen is preferentially extracted from phenyl carbon than aromatic
carbon. Among all the possible pathways of SO•– 4 -initiated reaction of
Fig. 7. The developmental toxicity and mutagenicity changes for DCF and DCF, it can be concluded that addition of SO4•‾ at C9 (Radd9) and single
degradation by-products. electron transfer reaction are determined as primary pathways during
the UV/PS-AOP degradation process of DCF.
and the formed intermediate is not stable. Fig. 3 (a) expressed the po­ Therefore, •OH-addition reaction plays an important role in initia­
tential energy surface (PES) and optimized structures of •OH-addition tion. Also, single electron transfer cannot be ignored in SO•– 4 -initiated
reaction. The Gibbs free energy barriers for •OH addition on aromatic reaction. The calculated results show that the reactivity of benzylic
ring containing chlorine (Radd1–6) ranges from 10.42 to position is higher than that of chlorinated aromatic ring according to
16.77 kcal mol− 1. Also, for another aromatic ring, the barriers are thermodynamics and kinetic values. We evaluate the contribution of
7.54–11.13 kcal mol− 1. It is clear that the addition barriers on carbox­ each reaction path through kinetic calculations of rate constants in the
ylated benzene rings are lower than that on chlorinated aromatic rings. following sections.
In the case of H-abstraction reaction, the Gibbs free energy barriers are
determined as 11.22–15.77 kcal mol− 1. The energy barriers are lower 3.3. Kinetic calculation
on the aromatic ring containing carboxyl, which is consistent with the
•OH addition reaction. It can be seen that the lowest barrier As shown in Fig. 4 (b), all paths (except SET process) of •OH -
(11.22 kcal mol− 1) of Rabs pathway occurs at C12 site. The result in­ initiated reactions are exothermic, indicating that •OH reactions are
dicates that Rabs12a pathway were more likely to occur among these more likely to occur thermodynamically. In addition, Fig. 4 (a) shows
abstraction process. Fig. 4 was drawn to intuitively display the energy that the barrier of •OH at each reaction site is lower than that of SO•– 4
level of each point of •OH-initiated degradation step with green. As except for Radd 4, 6 and Rset.
shown in Fig. 4 (b), all the Radd and Rabs path of •OH are exothermic Table 1 shows the overall rate constants of •OH / SO•– 4 -initiated re­
processes with reaction energies in the range of − 6.19 to action between 273 K and 343 K, including addition, abstraction and
− 13.33 kcal mol− 1, which indicates that they potentially contribute to electron transfer reaction. Reaction rate constants including addition
the degradation of DCF based on thermodynamics principle. Among all and abstraction reaction of all possible sites initiated by •OH and SO•– 4
possible reactions, Radd11 pathway is kinetically more likely to occur are shown in Table S1 and Table S2, respectively. The theoretical overall
with the lowest barrier (7.54 kcal mol− 1), this process would release rate constant of •OH-initiated reaction is 1.12 × 109 M− 1 s− 1 in the
reaction energies of 9.31 kcal mol− 1. These results indicated that Radd11 degradation of DCF at 298 K. Yu et al. [14] reported that the bimolecular
of •OH-addition reaction is dominant in the degradation of DCF, form­ reaction rate constants for DCF for •OH-AOP was (9.29 ± 0.11) × 109
ing the addition complex intermediate. The intermediates containing M− 1 s− 1, which is in the same order of magnitude as the theoretical value
free radicals are unstable and are prone to subsequent reactions to calculated in this paper. This result also proves the reliability of the
produce relatively stable products. theoretical kinetic calculations. Furthermore, the rate constants of the
8
SO•–
4 -addition and SET pathways are determined as 2.70 × 10 and
7 − 1 − 1
3.2.2. SO•–4 -initiated reaction of DCF
3.02 × 10 M s at 298 K, respectively, indicating that the single
The mechanisms of SO•– 4 -initiated reactions are similar to the •OH
electron transfer cannot be ignored in the SO•–4 -initiated reaction. It can
reactions of DCF, including SO•– 4 addition, H-transfer and electron
be concluded that the reaction rate constant of •OH is one order of
8 − 1 − 1
transfer reactions. The Gibbs free energy and geometries of all the magnitude higher than that of SO•– 4 (1.30 × 10 M s ). The reaction
possible SO4•‾-initiated pathways are drawn in Fig. S2. Previous studies rate constant is positively correlated with temperature, which indicates
have shown that the single electron transfer reaction of SO•– 4 plays an
that high temperature is beneficial to degradation of DCF in aquatic
important role in the degradation of organic pollutants. The radical environments. The •OH-Radd, SO•– 4 -Radd and SO4 -Rset are faster by
•–

cation (DCF+•) and sulfate anion (SO24‾) were formed in the single comparing all the reaction rate constants, which is consistent with the
electron transfer process. The Gibbs energy barrier of the Rset pathway is result of thermodynamics.
calculated to be 9.66 kcal mol− 1, which is 13.25 kcal mol− 1 lower than
that of •OH-SET process. This result shows that the electron transfer of 3.4. Formation of the degradation by-products
SO•–
4 is more reactive than •OH from a kinetic perspective. Meanwhile,
the process needs to absorb a small amount of energy of 1.93 kcal mol− 1, The addition complexes (•DCF…OH) produced from the •OH-initi­
which may be caused by the weak electron-withdrawing effect of ated degradation of DCF are highly reactive intermediated containing
-COOH. The calculation demonstrates that the single electron transfer free radicals. Thus, the subsequent transformations of these in­
process is very important for SO4•‾-initiated degradation of DCF. termediates are further studied in this study.
Structurally, when •OH radical attacks C atoms connected to

7
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

Fig. 8. Acute toxicity (A) and chronic toxicity (B) (unit: mg L− 1) of DCF and their degradation by-products.

chlorine substituents in the aromatic ring, the carbon-chlorine bond is leading to the formation of IM-O2 intermediate. This process requires
broken and •OH is directly added to form phenolic products (P1 and P2). overcoming the Gibbs energy barrier of 14.27 kcal mol− 1. Subsequently,
Furthermore, Cl- is a potential ion when chlorinated organic pollutants the •OO radical on IM-O2 can extract the H atom on adjacent C to form
are oxidized by SO•– 4 . Chlorine derived radicals (•ClO) can also react the •OOH and the rest part changes into the P5 product. The energy
with DCF to form corresponding intermediates. Fig. S3 shows the reac­ barrier of the pathways is determined as 14.23 kcal mol− 1. The by-
tion pathways of DCF with •ClO. product of P5 can generate dichlorobenzene P6 and phenol product P7
Among all the •OH and SO•– 4 -initiation reactions, the •OH-addition through hydrolysis reaction by overcoming certain reaction energy
reaction at C11 site is the optimal reaction path, because this site has the barrier. The Gibbs energy barrier of this process is relatively high, but
lowest barrier and the highest reaction rate constant. In the following the hydrolysis reaction is easy to occur in the presence of catalyst in
study, intermediate at C11 site (IM11) was selected as research object to aquatic environment. Some studies have detected the transformation
study the transformation of by-products in the degradation process of products of DCF using an ultra performance liquid chromatograph
DCF. In the presence of O2 in aquatic environment, IM11 can readily coupled with a quadrupole-time of fight-mass spectrometry [59,60].
react with O2 exothermically, with energy of 0.56 kcal mol− 1 (Fig. 6), Notably, the product P5 (m/z 312, C14H11NO3Cl2) was proposed by

8
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

hydroxylation [59,61]. Also, the reaction pathway of C-N bond cleavage processes in aquatic environments require more consideration, and the
is speculated experimentally [62]. potential environmental risks arising from the degradation by-products
To sum up, •OH radicals have high degradation efficiency on DCF. need to be carefully assessed.
•OH attacks the C on the phenyl position to form an intermediate con­
taining radical instability. Then, the addition intermediates can undergo 4. Conclusion
subsequent reactions to form stable products (P5, dichlorobenzene P6
and phenol product P7). It is critical to estimate the biological reactivity There are concerns about the potential eco-toxicity and human
of these products in the aquatic environment. health effects risk of DCF on organisms in aquatic environments. In
summary, the quantum chemistry and computational toxicology
3.5. Ecotoxicity assessment methods were used to research the degradation mechanism and toxicity
of DCF induced by •OH and SO•– 4 radicals in aqueous-phase UV-activated
DCF and its degradation by-products exit widely in water bodies. persulfate process. We found that Radd11 pathway of •OH-initiated re­
Thus, it is necessary to evaluate their impact on ecosystem. In this study, action is more likely to occur with the lowest Gibbs barrier
ECOSAR and TEST programs, computational toxicity software, were (7.54 kcal mol− 1), this process would release reaction energies of
used to estimate the ecotoxicity of DCF and its by-products. Although the 9.31 kcal mol− 1. For the single electron transfer reaction of SO•–4 , the
toxicity experiment is irreplaceable, the computational toxicity program Gibbs energy barrier to be overcome is 9.66 kcal mol− 1, which is
can provide a lot of toxicity property values conveniently at the 13.25 kcal mol− 1 lower than that of •OH-SET process. This result shows
screening level [63,64]. that the electron transfer of SO•–
4 and addition reaction of •OH played a
The acute and chronic toxicities of DCF and its degradation by- major role in the degradation process of DCF from a kinetic perspective.
products for target organism (green alga, daphnid and fish) are predi­ Furthermore, the reactivity of benzylic position of DCF is higher than
cated using ECOSAR software. Fig. 8 shows the toxicities of DCF and that of chlorinated aromatic ring according to thermodynamics and ki­
products. The structure configurations of products are given in Fig. S4. netic values. In addition, the degradation rates of •OH and SO•– 4 with
According to the Globally Harmonized System of Classification and DCF were 1.12 × 109 and 1.30 × 108 M− 1 s− 1 at 298 K, respectively.
Labelling of Chemicals (GHS) (Table S3), toxicities are divided into four Although the toxicity level of most products is reduced, the hydrolysates
levels (not harmful, harmful, toxicity and very toxicity). The octanol- remain toxic and deserve more attention for the degradation by-
water partitioning coefficient (log Kow) of DCF was 4.02 by computa­ products of DCF.
tional toxicology method, which was close to the experimental value log
Kow 4.5 [65]. The values of acute, chronic and developmental toxicity, CRediT authorship contribution statement
and mutagenicity for DCF and degradation by-products are expressed in
Table S4. As shown in Fig. 8 (a), the acute toxicity of DCF is classified as Jiaoxue Yang: Conceptualization, Investigation, Writing – original
harmful level for target organism based on the values of LC50 and EC50. draft, Writing – review & editing, Software. Guochun Lv: Visualization,
For fish and daphnia, acute toxicity of P1–4 was reduced, especially P3 Software, Data curation. Tingting Li: Visualization. Shuchen Sun:
and P4 are completely reduced to not harmful products. It is obvious Writing – review & editing. Xiaomin Sun: Resources, Project adminis­
that the toxicity is enhanced for aromatic compounds (P6 and P7) pro­ tration, Funding acquisition, Supervision.
duced by hydrolysis. The change trend of chronic toxicity is consistent
with acute toxicity, the toxicity of other products almost degraded to not
Declaration of Competing Interest
harmful grade except for P6 and P7. In addition, daphnia is the most
sensitive organism to chronic toxic changes of DCF and its degradation
The authors declare that they have no known competing financial
products. The acute and chronic toxicity changes showed that the hy­
interests or personal relationships that could have appeared to influence
drolysates remained at toxic or very toxic levels compared to the parent
the work reported in this paper.
compound, although some products are not harmful. The known LD50
and the mass concentration of organic compounds can be used to
Data availability
calculate the theoretical toxicity of the products over time during the
SO•–
4 -AOPs process. The Eq. 8 is the calculation formula of theoretical
Data will be made available on request.
toxicity [66]. The change of the time-dependent total theoretical toxicity
during the entire oxidation process can be assessed in the experiment
[67]. Therefore, more attention should be paid to the toxicity changes of Acknowledgements
by-products during DCF degradation.
This work was supported by National Natural Science Foundation of
Theoretival toxicity =
themassconcentrationoforganiccompounds
(8) China (21976109), Natural Science Foundation of Shandong Province
LD50 (ZR2021QB195 and ZR2020QE234), and The Research Foundation of
Furthermore, the developmental toxicity and mutagenicity of DCF Liaocheng University (318052114, 318051943 and 318012114).
and its by-products are predicated using TEST software. The results are
shown in Fig. 7. Developmental toxicity is caused by environmental Appendix A. Supporting information
factors (such as drugs, environmental toxic chemicals), which can
interfere with homeostasis, normal growth, differentiation and devel­ Supplementary data associated with this article can be found in the
opment of organisms [68]. The developmental toxicity of DCF is 1.03, online version at doi:10.1016/j.jece.2022.108311.
which is higher than the toxicity threshold of 0.5. The result clearly
indicates that DCF is a developmentally toxic compound. This indicates References
that DCF pollution may have an adverse impact on biological develop­
[1] A.J. Ebele, M. Abou-Elwafa Abdallah, S. Harrad, Pharmaceuticals and personal
ment in the prenatal development period. Although the developmental
care products (PPCPs) in the freshwater aquatic environment, Emerg. Contam. 3
toxicity of DCF degradation by-products decreased, it remains at (2017) 1–16.
developmental toxicity levels (except for P6). In addition, DCF has a [2] J. Wang, S. Wang, Removal of pharmaceuticals and personal care products (PPCPs)
positive correlation with mutagenic according to the mutagenic results. from wastewater: a review, J. Environ. Manag. 182 (2016) 620–640.
[3] Y. Yang, Y.S. Ok, K.-H. Kim, E.E. Kwon, Y.F. Tsang, Occurrences and removal of
The by-products are transformed into non-mutagenic toxic compounds pharmaceuticals and personal care products (PPCPs) in drinking water and water/
with the degradation of DCF. Therefore, these DCF degradation sewage treatment plants: a review, Sci. Total Environ. 596-597 (2017) 303–320.

9
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

[4] B. Hayes Tyrone, V. Khoury, A. Narayan, M. Nazir, A. Park, T. Brown, L. Adame, [30] X. Yang, X. Cao, L. Zhang, Y. Wu, L. Zhou, G. Xiu, C. Ferronato, J.-M. Chovelon,
E. Chan, D. Buchholz, T. Stueve, S. Gallipeau, Atrazine induces complete Sulfate radical-based oxidation of the aminopyralid and picloram herbicides: The
feminization and chemical castration in male African clawed frogs (Xenopus role of amino group on pyridine ring, J. Hazard. Mater. 405 (2021), 124181.
laevis), Proc. Natl. Acad. Sci. 107 (2010) 4612–4617. [31] W. Liu, Y. Li, Y. Wang, Y. Zhao, Y. Xu, X. Liu, DFT insights into the degradation
[5] T.J. Gan, Diclofenac: an update on its mechanism of action and safety profile, Curr. mechanism of carbendazim by hydroxyl radicals in aqueous solution, J. Hazard.
Med. Res. Opin. 26 (2010) 1715–1731. Mater. 431 (2022), 128577.
[6] P. McGettigan, D. Henry, Use of non-steroidal anti-inflammatory drugs that elevate [32] Y. Gao, G. Li, Y. Qin, Y. Ji, B. Mai, T. An, New theoretical insight into indirect
cardiovascular risk: an examination of sales and essential medicines lists in low-, photochemical transformation of fragrance nitro-musks: mechanisms, eco-toxicity
middle-, and high-income countries, PLoS Med. 10 (2013) 1001388. and health effects, Environ. Int. 129 (2019) 68–75.
[7] L. Lonappan, R. Pulicharla, T. Rouissi, S.K. Brar, M. Verma, R.Y. Surampalli, J. [33] S. Luo, Z. Wei, D.D. Dionysiou, R. Spinney, W.-P. Hu, L. Chai, Z. Yang, T. Ye,
R. Valero, Diclofenac in municipal wastewater treatment plant: quantification R. Xiao, Mechanistic insight into reactivity of sulfate radical with aromatic
using laser diode thermal desorption—atmospheric pressure chemical contaminants through single-electron transfer pathway, Chem. Eng. J. 327 (2017)
ionization—tandem mass spectrometry approach in comparison with an 1056–1065.
established liquid chromatography-electrospray ionization–tandem mass [34] M. Frisch, G. Trucks, H. Schlegel, G. Scuseria, M. Robb, J. Cheeseman, G. Scalmani,
spectrometry method, J. Chromatogr. A 1433 (2016) 106–113. V. Barone, B. Mennucci, G. Petersson, Gaussian 09; Gaussian, Inc, Wallingford, CT,
[8] A. Ziylan, N.H. Ince, The occurrence and fate of anti-inflammatory and analgesic 32 (2009) 5648–5652.
pharmaceuticals in sewage and fresh water: Treatability by conventional and non- [35] M. Frisch, gaussian, http://www. gaussian. com.
conventional processes, J. Hazard. Mater. 187 (2011) 24–36. [36] M. Walker, A.J.A. Harvey, A. Sen, C.E.H. Dessent, Performance of M06, M06-2X,
[9] I. Martínez-Alcalá, J.M. Guillén-Navarro, C. Fernández-López, Pharmaceutical and M06-HF density functionals for conformationally flexible anionic clusters: M06
biological degradation, sorption and mass balance determination in a conventional functionals perform better than B3LYP for a model system with dispersion and
activated-sludge wastewater treatment plant from Murcia, Spain, Chem. Eng. J. ionic hydrogen-bonding interactions, J. Phys. Chem. A 117 (2013) 12590–12600.
316 (2017) 332–340. [37] K.B. Wiberg, Basis set effects on calculated geometries: 6-311++G** vs. aug-cc-
[10] L. Lonappan, S.K. Brar, R.K. Das, M. Verma, R.Y. Surampalli, Diclofenac and its pVDZ, J. Comput. Chem. 25 (2004) 1342–1346.
transformation products: environmental occurrence and toxicity - a review, [38] C. Sosa, J. Andzelm, B.C. Elkin, E. Wimmer, K.D. Dobbs, D.A. Dixon, A local density
Environ. Int. 96 (2016) 127–138. functional study of the structure and vibrational frequencies of molecular
[11] L. Lonappan, S.K. Brar, R.K. Das, M. Verma, R.Y. Surampalli, Diclofenac and its transition-metal compounds, J. Phys. Chem. 96 (1992) 6630–6636.
transformation products: environmental occurrence and toxicity-a review, Environ. [39] L. Deng, T. Ziegler, The determination of intrinsic reaction coordinates by density
Int. 96 (2016) 127–138. functional theory, Int. J. Quantum Chem. 52 (1994) 731–765.
[12] V. Vishnu Priyan, T. Shahnaz, E. Suganya, S. Sivaprakasam, S. Narayanasamy, [40] B. Mennucci, Polarizable continuum model, WIREs Comput. Mol. Sci. 2 (2012)
Ecotoxicological assessment of micropollutant Diclofenac biosorption on magnetic 386–404.
sawdust: phyto, microbial and fish toxicity studies, J. Hazard. Mater. 403 (2021), [41] J. Tomasi, R. Cammi, B. Mennucci, Medium effects on the properties of chemical
123532. systems: an overview of recent formulations in the polarizable continuum model
[13] S. Suárez, M. Carballa, F. Omil, J.M. Lema, How are pharmaceutical and personal (PCM), Int. J. Quantum Chem. 75 (1999) 783–803.
care products (PPCPs) removed from urban wastewaters? Rev. Environ. Sci. Bio/ [42] O. Tapia, O. Goscinski, Self-consistent reaction field theory of solvent effects, Mol.
Technol. 7 (2008) 125–138. Phys. 29 (1975) 1653–1661.
[14] H. Yu, E. Nie, J. Xu, S. Yan, W.J. Cooper, W. Song, Degradation of Diclofenac by [43] M. Cossi, G. Scalmani, N. Rega, V. Barone, New developments in the polarizable
advanced oxidation and reduction processes: kinetic studies, degradation pathways continuum model for quantum mechanical and classical calculations on molecules
and toxicity assessments, Water Res. 47 (2013) 1909–1918. in solution, J. Chem. Phys. 117 (2002) 43–54.
[15] L. Varanasi, E. Coscarelli, M. Khaksari, L.R. Mazzoleni, D. Minakata, [44] R.A. Marcus, Electron transfer reactions in chemistry. Theory and experiment, Rev.
Transformations of dissolved organic matter induced by UV photolysis, hydroxyl Mod. Phys. 65 (1993) 599.
radicals, chlorine radicals, and sulfate radicals in aqueous-phase UV-based [45] R. Marcus, Transfer reactions in chemistry. Theory and experiment, Pure Appl.
advanced oxidation processes, Water Res. 135 (2018) 22–30. Chem. 69 (1997) 13–30.
[16] Y. Feng, Q. Song, W. Lv, G. Liu, Degradation of ketoprofen by sulfate radical-based [46] M. Chou, C. Creutz, N. Sutin, Rate constants and activation parameters for outer-
advanced oxidation processes: kinetics, mechanisms, and effects of natural water sphere electron-transfer reactions and comparisons with the predictions of Marcus
matrices, Chemosphere 189 (2017) 643–651. theory, J. Am. Chem. Soc. 99 (1977) 5615–5623.
[17] M.Y. Kilic, W.H. Abdelraheem, X. He, K. Kestioglu, D.D. Dionysiou, Photochemical [47] P. Pechukas, Transition state theory, Annu. Rev. Phys. Chem. 32 (1981) 159–177.
treatment of tyrosol, a model phenolic compound present in olive mill wastewater, [48] E. Vanden-Eijnden, F.A. Tal, Transition state theory: variational formulation,
by hydroxyl and sulfate radical-based advanced oxidation processes (AOPs), dynamical corrections, and error estimates, J. Chem. Phys. 123 (2005), 184103.
J. Hazard. Mater. 367 (2019) 734–742. [49] F.C. Collins, G.E. Kimball, Diffusion-controlled reaction rates, J. Colloid Sci. 4
[18] R.L. Johnson, P.G. Tratnyek, R.O.B. Johnson, Persulfate persistence under thermal (1949) 425–437.
activation conditions, Environ. Sci. Technol. 42 (2008) 9350–9356. [50] J. Dzubiella, J. McCammon, Substrate concentration dependence of the diffusion-
[19] A. Ghauch, A. Baalbaki, M. Amasha, R. El Asmar, O. Tantawi, Contribution of controlled steady-state rate constant, J. Chem. Phys. 122 (2005), 184902.
persulfate in UV-254nm activated systems for complete degradation of [51] D.G. Truhlar, Nearly encounter-controlled reactions: the equivalence of the steady-
chloramphenicol antibiotic in water, Chem. Eng. J. 317 (2017) 1012–1025. state and diffusional viewpoints, J. Chem. Educ. 62 (1985) 104.
[20] G. Ayoub, A. Ghauch, Assessment of bimetallic and trimetallic iron-based systems [52] ECOASR, in, 2014. 〈https://www.epa.gov/chemicals-under-tsca〉.
for persulfate activation: application to sulfamethoxazole degradation, Chem. Eng. [53] M. El-Harbawi, Toxicity measurement of imidazolium ionic liquids using acute
J. 256 (2014) 280–292. toxicity test, Procedia Chem. 9 (2014) 40–52.
[21] C. Liang, Z.-S. Wang, N. Mohanty, Influences of carbonate and chloride ions on [54] N. Dom, D. Knapen, R. Blust, Assessment of aquatic experimental versus predicted
persulfate oxidation of trichloroethylene at 20◦ C, Sci. Total Environ. 370 (2006) and extrapolated chronic toxicity data of four structural analogues, Chemosphere
271–277. 86 (2012) 56–64.
[22] G.-D. Fang, D.D. Dionysiou, Y. Wang, S.R. Al-Abed, D.-M. Zhou, Sulfate radical- [55] TEST, in, 2016. 〈https://www.epa.gov/chemical-research/toxicity-estimation-so
based degradation of polychlorinated biphenyls: effects of chloride ion and ftware-tool-test〉.
reaction kinetics, J. Hazard. Mater. 227-228 (2012) 394–401. [56] T. Jefferson, The globally harmonized system of classification and labelling of
[23] K. Li, M.I. Stefan, J.C. Crittenden, Trichloroethene degradation by UV/H2O2 chemicals (GHS), Laboratory Safety for Chemistry Students, (2010).
advanced oxidation process: product study and kinetic modeling, Environ. Sci. [57] T. Lu, F. Chen, Multiwfn: a multifunctional wavefunction analyzer, J. Comput.
Technol. 41 (2007) 1696–1703. Chem. 33 (2012) 580–592.
[24] R.H. Waldemer, P.G. Tratnyek, R.L. Johnson, J.T. Nurmi, Oxidation of chlorinated [58] S. Manzetti, T. Lu, The geometry and electronic structure of Aristolochic acid:
ethenes by heat-activated persulfate: kinetics and products, Environ. Sci. Technol. possible implications for a frozen resonance, J. Phys. Org. Chem. 26 (2013)
41 (2007) 1010–1015. 473–483.
[25] P. Neta, V. Madhavan, H. Zemel, R.W. Fessenden, Rate constants and mechanism of [59] H. Shi, G. Zhou, Y. Liu, Y. Fu, H. Wang, P. Wu, Kinetics and pathways of diclofenac
reaction of sulfate radical anion with aromatic compounds, J. Am. Chem. Soc. 99 degradation by heat-activated persulfate, RSC Adv. 9 (2019) 31370–31377.
(1977) 163–164. [60] R. Salgado, V.J. Pereira, G. Carvalho, R. Soeiro, V. Gaffney, C. Almeida, V.
[26] A. Rastogi, S.R. Al-Abed, D.D. Dionysiou, Sulfate radical-based V. Cardoso, E. Ferreira, M.J. Benoliel, T.A. Ternes, A. Oehmen, M.A.M. Reis, J.
ferrous–peroxymonosulfate oxidative system for PCBs degradation in aqueous and P. Noronha, Photodegradation kinetics and transformation products of ketoprofen,
sediment systems, Appl. Catal. B: Environ. 85 (2009) 171–179. diclofenac and atenolol in pure water and treated wastewater, J. Hazard. Mater.
[27] G.P. Anipsitakis, D.D. Dionysiou, Degradation of organic contaminants in water 244-245 (2013) 516–527.
with sulfate radicals generated by the conjunction of peroxymonosulfate with [61] J.M. Monteagudo, H. El-taliawy, A. Durán, G. Caro, K. Bester, Sono-activated
cobalt, Environ. Sci. Technol. 37 (2003) 4790–4797. persulfate oxidation of diclofenac: degradation, kinetics, pathway and contribution
[28] C. Zhou, J. Wu, L. Dong, B. Liu, D. Xing, S. Yang, X. Wu, Q. Wang, J. Fan, L. Feng, of the different radicals involved, J. Hazard. Mater. 357 (2018) 457–465.
G. Cao, Removal of antibiotic resistant bacteria and antibiotic resistance genes in [62] X. Lu, Y. Shao, N. Gao, J. Chen, Y. Zhang, H. Xiang, Y. Guo, Degradation of
wastewater effluent by UV-activated persulfate, J. Hazard. Mater. 388 (2020), diclofenac by UV-activated persulfate process: kinetic studies, degradation
122070. pathways and toxicity assessments, Ecotoxicol. Environ. Saf. 141 (2017) 139–147.
[29] S. Khan, M. Sohail, C. Han, J.A. Khan, H.M. Khan, D.D. Dionysiou, Degradation of [63] P. Reuschenbach, M. Silvani, M. Dammann, D. Warnecke, T. Knacker, ECOSAR
highly chlorinated pesticide, lindane, in water using UV/persulfate: kinetics and model performance with a large test set of industrial chemicals, Chemosphere 71
mechanism, toxicity evaluation, and synergism by H2O2, J. Hazard. Mater. 402 (2008) 1986–1995.
(2021), 123558.

10
J. Yang et al. Journal of Environmental Chemical Engineering 10 (2022) 108311

[64] H. Sanderson, D.J. Johnson, C.J. Wilson, R.A. Brain, K.R. Solomon, Probabilistic [66] A. Zhang, F. Wang, W. Chu, X. Yang, Y. Pan, H. Zhu, Integrated control of CX3R-
hazard assessment of environmentally occurring pharmaceuticals toxicity to fish, type DBP formation by coupling thermally activated persulfate pre-oxidation and
daphnids and algae by ECOSAR screening, Toxicol. Lett. 144 (2003) 383–395. chloramination, Water Res. 160 (2019) 304–312.
[65] M.F. Morissette, S. Vo Duy, H.P.H. Arp, S. Sauvé, Sorption and desorption of [67] H. Xiang, N. Gao, X. Lu, Y. Xiang, W. Chu, Degradation of sulfamethazine in UV/
diverse contaminants of varying polarity in wastewater sludge with and without monochloramine process: kinetics, by-products formation, and toxicity evaluation,
alum, Environ. Sci.: Process. Impacts 17 (2015) 674–682. Chem. Eng. J. 430 (2022), 133008.
[68] P. Grandjean, P.J. Landrigan, Neurobehavioural effects of developmental toxicity,
Lancet Neurol. 13 (2014) 330–338.

11