Kinetics and the mechanism of the photocatalytic degradation of mesotrione in aqueous suspension and toxicity of its degradation mixtures

Article abstract of DOI:10.1016/j.molcata.2014.04.033

The photocatalytic degradation of the herbicide mesotrione (0.05 mM) in aqueous suspensions of TiO₂ Degussa P25 and ZnO was examined as a function of the different operational parameters. The optimum of the catalyst loading was found to be 2.00 mg mL^-1 under UV light at the pH 4.0 (TiO₂ Degussa P25) and pH 9.2 (ZnO). In the first stage of the reaction, the photocatalytic degradation of mesotrione followed the pseudo-first order kinetics, in which the heterogeneous catalysis proceeds mainly via ^?OH radicals, and partly also via holes. The results showed that the disappearance of mesotrione led to the formation of a number of organic intermediates and ionic byproducts, whereas its complete mineralization occurred after about 4 h. Tentative photodegradation pathways were proposed and discussed as a function of the type of photocatalyst. A comparison of the evolution of toxicity that was evaluated in vitro in the rat hepatoma (H-4-II-E) and human fetal lung (MRC-5) cell lines with the degradation kinetics indicates that the cell growth of both cell lines was only mildly affected by the mixture of mesotrione and its photocatalytic degradation intermediates obtained using TiO₂ Degussa P25, and moderately affected by the mixtures obtained using ZnO, whereby at 120 min toxicity decreased.

Full text of DOI:10.1016/j.molcata.2014.04.033

Accepted Manuscript
Title: Kinetics and the mechanism of the photocatalytic
degradation of mesotrione in aqueous suspension and toxicity
of its degradation mixtures
ˇ
´
ˇ ´
Author: Daniela V. Sojic Dejan Z. Orcic Dragana D.
ˇ
´
´
´
Cetojevic-Simin Vesna N. Despotovic Biljana F. Abramovic
PII:
S1381-1169(14)00171-X
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcata.2014.04.033
MOLCAA 9095
To appear in:
Journal of Molecular Catalysis A: Chemical
Received date:
Revised date:
Accepted date:
18-9-2013
10-4-2014
25-4-2014
ˇ
ˇ
´
ˇ ´
´
Please cite this article as: D.V. Sojic, D.Z. Orcic, D.D. Cetojevic-Simin, V.N.
´
´
Despotovic, B.F. Abramovic, Kinetics and the mechanism of the photocatalytic
degradation of mesotrione in aqueous suspension and toxicity of its
degradation mixtures, Journal of Molecular Catalysis A: Chemical (2014),
http://dx.doi.org/10.1016/j.molcata.2014.04.033
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Kinetics and the mechanism of the photocatalytic degradation of mesotrione in aqueous
suspension and toxicity of its degradation mixtures
Daniela V. Šojića, Dejan Z. Orčića, Dragana D. Četojević-Siminb, Vesna N. Despotovića, Biljana F.
Abramovića*
a University of Novi Sad, Faculty of Sciences, Department of Chemistry, Biochemistry and
Environmental Protection, Trg D. Obradovića 3, 21000 Novi Sad, Serbia
b University of Novi Sad, Faculty of Medicine, Oncology Institute of Vojvodina, Dr Goldmana 4,
21204 Sremska Kamenica, Serbia
*Professor Biljana Abramović
Department of Chemistry, Biochemistry and Environmental Protection
Faculty of Sciences
Phone: +381 21 4852753
Fax: +381 21 454065
e-mail: biljana.abramovic@dh.uns.ac.rs
Trg D. Obradovića 3
21000 Novi Sad
Serbia
E-mail addresses: daniela.sojic@dh.uns.ac.rs (Daniela V. Šojić)
dejan.orcic@dh.uns.ac.rs (Dejan Z. Orčić)
ddaaggeerr@gmail.com (Dragana D. Četojević-Simin)
vesna.despotovic@dh.uns.ac.rs (Vesna N. Despotović)
1
Page 1 of 49

Abstract
The photocatalytic degradation of the herbicide mesotrione (0.05 mM) in aqueous suspensions of
TiO2 Degussa P25 and ZnO was examined as a function of the different operational parameters. The
optimum of the catalyst loading was found to be 2.00 mg mL−1 under UV light at the pH 4.0 (TiO2
Degussa P25) and pH 9.2 (ZnO). In the first stage of the reaction, the photocatalytic degradation of
mesotrione followed the pseudo-first order kinetics, in which the heterogeneous catalysis proceeds
mainly via OH radicals, and partly also via holes. The results showed that the disappearance of
mesotrione led to the formation of a number of organic intermediates and ionic byproducts, whereas
its complete mineralization occurred after about 4 h. Tentative photodegradation pathways were
proposed and discussed as a function of the type of photocatalyst. A comparison of the evolution of
toxicity that was evaluated in vitro in the rat hepatoma (H-4-II-E) and human fetal lung (MRC-5)
cell lines with the degradation kinetics indicates that the cell growth of both cell lines was only
mildly affected by the mixture of mesotrione and its photocatalytic degradation intermediates
obtained using TiO2 Degussa P25, and moderately affected by the mixtures obtained using ZnO,
whereby at 120 min toxicity decreased.
Key words: mesotrione, TiO2 Degussa P25, ZnO, photodegradation mechanism, mammalian cell
growth.
1. Introduction
Not so long ago, atrazine (6-chloro-N2-ethyl-N4-isopropyl-1,3,5-triazine-2,4-diamine), one of
the most popular herbicides among the triazine group, was introduced as an herbicide to control
weeds in corn fields. Numerous studies showed that atrazine is toxic to aquatic organisms [1]. Since
atrazine has been banned in the European Union countries since 2007, mesotrione {2-[(4-
methylsulfonyl)-2-nitrobenzoyl]cyclohexane-1,3-dione} appeared as a new selective herbicide for
use in corn, which has been substituted for atrazine [2]. Mesotrione is an herbicide from the family
2
Page 2 of 49

of benzoyl-ciclohexanedione, to control wide spectrum of broad leaf weeds, especially in corn
farming. It was developed by the company Syngeta Crop Protection and it was registered in Europe
in 2000, and in the United States in 2001 [3]. The herbicide is marketed under the name Callisto®.
Mesotrione is a hot research topic since it is a relatively new substance on the market, still
insufficiently investigated. Also, little is known about its intermediates and metabolites. Further,
there is still insufficient data about the pathway of its degradation in nature and about toxicity of its
intermediates. As far as the persistence of mesotrione in soil is concerned, Rouchaud et al. [4]
reported results of their investigation based on a three-month study, carried out at four different
places in Belgium with different soil composition. This study showed that in the soil surface layer
of 0−10 cm there was 90% of the applied dose of mesotrione, whereas in the layers at the depths of
15−20 cm it could not be detected. According to the report of the European Commission from 2003,
mesotrione is stable in water in the presence of light (DT50) for about 90 days [5]. A survey of the
literature shows that there are only few studies concerning the decomposition of this herbicide and
the herbicides from the same family, whereas the photocatalytic degradation of mesotrione has not
been reported yet. Recently, removal of mesotrione in aqueous medium by photolysis with
simulated natural sunlight irradiation [6,7] and UV-B light [6], Fenton process [8], and electro-
Fenton [9] has been reported. However, photocatalysis with UV and TiO2/ZnO, one of the most
promising advanced oxidation processes for the destruction of aquatic pollutants, as well as the
toxicity of mesotrione and its photocatalytic degradation byproducts in mammalian cell lines has
not been reported. One of the most important aspects of environmental photocatalysis is the
selection of semiconductor materials such as TiO2 and ZnO, two ideal photocatalysts in several
respects. For example, they are relatively inexpensive and, due to their wide band gap energy,
provide photogenerated holes with high oxidizing power. Since ZnO has nearly the same band gap
energy (3.2 eV) as TiO2, its photocatalytic capacity is anticipated to be similar to that of TiO2.
The objective of this work was to monitor the kinetics of photocatalytic degradation of
mesotrione in the presence of TiO2 Degussa P25 and ZnO as photocatalysts, by irradiation of
3
Page 3 of 49

aqueous suspensions using mercury and halogen lamps, as well as natural sunlight, under different
conditions of the catalyst load and pH of the solution. An attempt was also made to identify the
reaction intermediates formed during the photo-oxidation process, using the LC–ESI–MS/MS
technique. Also, the effect of mesotrione and its photocatalytic degradation intermediates on cell
growth activity was evaluated in vitro in two mammalian cell lines.
2. Experimental
2.1. Chemicals, catalysts and solutions
All chemicals were of reagent grade and were used without further purification. Mesotrione
(C14H13NO7S, 99.9%, Mr = 339.3, CAS No 104206-82-8), 98.9%, pestanal quality, was
manufactured by Fluka; 85% H3PO4 was obtained from Lachema (Neratovice, Czech Republic),
Na2SO3 and NaOH from ZorkaPharm (Šabac, Serbia). Of the other chemicals, 96% ethanol was
obtained from Centrohem (Stara Pazova, Serbia), and 98–100% formic acid (HCOOH) and 60%
HClO4 from Merck, while 99.8% acetonitrile (ACN) and HPLC gradient grade methanol (MeOH)
were products of J.T. Baker. All solutions were made using doubly distilled water (DDW). The pH
of the reaction mixture was adjusted using a dilute aqueous solution of HClO4 or NaOH.
Dulbecco’s modified Eagle’s medium (DMEM), and fetal calf serum were purchased from PAA
Laboratories GmbH (Pashing, Austria), penicillin and streptomycin from Galenika (Belgrade,
Serbia), trypsin from Serva (Heidelberg, Germany), EDTA, trichloroacetic acid (TCA) from
Laphoma (Skopje, FYR Macedonia), and tris(hydroxymethyl)amino methane from Sigma Aldrich.
TiO2 Degussa P25 (75% anatase and 25% rutile form, 50 m2 g1, about 20 nm, non-porous),
TiO2 Hombikat, anatase (Sigma-Aldrich), TiO2, rutile [10] and ZnO (Kemika, 99.96%) were used
as photocatalysts.
2.2. Photodegradation procedure
4
Page 4 of 49

Photocatalytic degradation was carried out in a cell made of Pyrex glass (total volume of ca. 40
mL, liquid layer thickness 35 mm), with a plain window on which the light beam was focused. The
cell was equipped with a magnetic stirring bar and a water circulating jacket. A 125 W high-
pressure mercury lamp (Philips, HPL-N, emission bands in the UV region at 304, 314, 335 and 366
nm, with maximum emission at 366 nm), together with an appropriate concave mirror, was used as
the radiation source. Irradiation in the visible spectral range was performed using a 50 W halogen
lamp (Philips). The outputs for the mercury and halogen lamps were calculated to be ca. 8.8 × 10−9
Einstein mL−1 min−1 and 1.7 × 10−9 Einstein mL−1 min−1 (potassium ferrioxalate actinometry),
respectively. With the aim of examining the use of natural sunlight, the experiments were also
carried out at daylight during the whole day (23±1 oC) in the presence/absence of the catalyst, from
February 1st to September 15th 2011.
The experiments were carried out using 20 mL of 0.05 mM solution of mesotrione containing
40 mg of TiO2 Degussa P25 (except for the study of the effect of catalyst load and direct
photolysis). After that, the aqueous suspension of TiO2 Degussa P25 was sonicated (50 Hz) in the
dark for 15 min before illumination, to make distribution of the photocatalyst particles uniform and
attain adsorption equilibrium. The suspension thus obtained was thermostated at 25±0.5 oC, and
then irradiated at a constant stream of O2 (3.0 mL min1). During the irradiation, the mixture was
stirred at a constant speed. The control experiments, carried out under O2 flow but by stopping the
irradiation, showed that there were no losses of volatile compounds during the degradation. All
experiments were performed at the natural pH (~ 4.0), except when studying the influence of the
pH. Where applicable, ethanol (so that the its concentration after the addition was 0.34 M) was
added as a hydroxyl radical scavenger and Na2SO3 as hole scavenger (so that the concentration of
sulfite after the addition was 3 mM).
Photonic efficiencies ζ have been calculated using Eqs. (1) and (2).

V c
I0 At
degradation rate [mol/s]
(1)
 
100%
100%
photon flux [molh/s]
5
Page 5 of 49


I 
NA  h c
(2)
I0 
where V is the volume of the suspension in reactor (20.0 cm3), Δc/Δt the degradation rate, A the
illuminated area (9.60 cm2), I0 the photon flux, I is the light intensity (53.04 W m2), λ is 365 nm,
NA is Avogadro’s constant, h is Planck’s constant and c is the velocity of light.
2.3. Analytical procedures
For the kinetic studies of the mesotrione photodegradation, samples of the reaction mixture
were taken at the beginning of the experiment and at regular time intervals (volume variation ca.
10%). The obtained suspensions were filtered through Millipore (Millex-GV, 0.22 µm) membrane
filters. The preliminary check confirmed the absence of mesotrione adsorption on the filters. After
that, a 20-µl sample was injected and analyzed on an Agilent Technologies 1100 Series liquid
chromatograph, with a UV/VIS DAD set at 225 nm (maximum of absorption for mesotrione), as
well as at 210, 260, 270 and 280 nm (for intermediates) and an Eclypse XDB-C18 (150 mm × 4.6
mm i.d., particle size 5 µm, 25 oC) column. The mobile phase (flow rate 1 mL min–1, pH 2.6) was a
mixture of ACN and DDW (4:6, v/v), the water being acidified with 0.1% H3PO4.
For the spectrophotometric monitoring, samples were prepared in the same way as for LC
measurements, and the corresponding aliquots were acidified with cc HClO4 to the pH 2.6, and UV
spectra were recorded on a T80 +UV-vis Spectrophotometer, PG Instruments Ltd (UK), at a fixed
slit width (2 nm), using 1 cm quartz cell and computer-loaded UV Win 5 data software. Kinetics of
the aromatic ring degradation was monitored at 225 nm.
Ion chromatographic (IC) determinations were performed on a Dionex ICS 3000 Reagent-Free
IC system with conductometric detector. For the anion determinations use was made of an IonPac
AS18 analytical column (250 mm × 4 mm i.d., bead diameter 8 µm). The mobile phase was a
solution of KOH (2040 mM) at a flow rate of 1 mL min1. Cations were determined using an
6
Page 6 of 49

IonPac CS12A analytical column (250 mm × 4 mm i.d., bead diameter 7.5 µm), and the mobile
phase was a solution of 40 mM of methane sulfonic acid at a flow rate of 1 mL min1.
For the LC–ESI–MS/MS evaluation of intermediates, a solution of 0.05 mM of mesotrione was
prepared and 20 µL sample was analyzed on an Agilent Technologies 1200 series LC with Agilent
Technologies 6410A series electrospray ionization triple-quadrupole MS/MS and Agilent
MassHunter B.03.03 software, using Agilent Technologies Zorbax XDB-C18 column (50 mm × 4.6
mm i.d., particle size 1.8 µm) held at 50 oC. The mobile phase (flow rate 1.0 mL min1) consisted of
0.05% aqueous HCOOH and MeOH (gradient: 0 min 30% MeOH, 1012 min 100% MeOH, post-
time 3 min). Analytes were ionized using the electrospray ion source, with nitrogen as drying gas
(temperature 350 oC, flow 10 L min1) and nebulizer gas (40 psi), and a capillary voltage of 4.0 kV.
High-purity nitrogen was used as the collision gas. Full scan mode (m/z range 100–800, scan time
200 ms, fragmentor voltage 100 V) in both polarities, was used to select precursor ion for
mesotrione and each degradation product, as well as to examine isotopic peaks distribution (Table
1). Then, the product ion scan MS2 mode (fragmentor voltage 100 V, scan time 200 ms, collision
energy 0–30 V in increments of 10 V), was used for structure elucidation of each degradation
product. Finally, the multiple reaction monitoring (MRM) mode (parameters are given in Table 1)
was used for obtaining peak areas of the analytes, in order to track the reaction kinetics.
Table 1
The total organic carbon (TOC) analysis was performed on an Elementar Liqui TOC II
according to Standard US EPA Method 9060A. In studying the influence of the initial pH on the
photocatalytic degradation use was made of a combined glass electrode (pH-Electrode SenTix 20,
WTW) connected to the pH-meter (pH/Cond 340i, WTW).
In repeated runs, the results agreed within 3−10%.
7
Page 7 of 49

The UV energy flux was measured using a Delta Ohm HD 2102.2 (Padova, Italy) radiometer
that was fitted with the LP 471 UVA sensor (spectral range 315−400 nm).
The cell lines of rat hepatoma (H-4-II-E) and human fetal lung (MRC-5) were grown in DMEM
medium as described previously [11].
The concentrations of mesotrione were in the range from 0.31 to 5 μM. The photodegradation
mixtures obtained after regular time intervals of irradiation were used for the preparation of four
additional two-fold serial dilutions in DDW by taking 1000 μL of stock solution and diluting it with
DDW to 2000 μL (dilution 2). Of this solution, three more serial dilutions (4, 8 and 16) were made
in the same way. A volume of 20 μL of the original mixture and of its four dilutions was added to
180 μL of the cell culture medium, so that the final dilutions were in the range from 10 to 160. The
same volume of DDW (20 μL) was added to the control wells.
The blank test was performed using the aqueous suspension of TiO2 Degussa P25/ZnO (2.00
mg mL1 without mesotrione), also sonicated in the dark for 15 min and filtered through Millipore
membrane filter. All samples were filtered through 0.22 µm microfilters (Sartorius), to obtain
sterility.
The cell growth was evaluated by the colorimetric sulphorhodamine B (SRB) assay of Skehan
et al. [12], modified by Četojević-Simin et al. [13].
Results were expressed as mean of two independent experiments, each performed in
quadruplicate (n = 8).
3. Results and discussion
3.1. Effect of the type of catalyst and light source
The photocatalytic degradation of mesotrione in the presence of different crystal structures of
TiO2 catalysts was studied (Fig. 1a). The obtained results demonstrate that the efficiency of TiO2
Degussa P25 and TiO2 Hombikat was almost the same, whereas the efficiency of TiO2, rutile was
lower compared to that observed in the presence of both above catalysts. The higher photocatalytic
8
Page 8 of 49

activity of TiO2 Degussa P25 has been ascribed to its mixture crystallites with small rutile
dispersing into an anatase matrix. Mixed-phase TiO2 photocatalysts have been reported to have
significantly higher activity than single-phase TiO2, which has been interpreted as due to electronic
interactions between the anatase and rutile phases [14]. Since the TiO2 Degussa P25 was showed
the same or better efficiency compared to other investigated TiO2 catalysts, as well as the
heterogeneous photodegradation in presence TiO2 Degussa P25 appeared as one of the most
efficient method for eliminating a number of pesticides from water, this catalyst was used in the
following experiments the photodegradation of mesotrione. Besides, by comparing the efficiency of
TiO2 Degussa P25 and ZnO in the photocatalytic degradation of mesotrione, it can be noticed that
the degradation rates are very similar (Fig. 1b). Also, the photonic efficiencies calculated from data
obtained for 10 min of irradiation TiO2 Degussa P25 suspension (4.12%) is very similar to the one
obtained using ZnO (3.99%). These results are in agreement with the literature data. Namely,
Dindar and Içli [15] in their study of the photolysis of phenol in the presence of ZnO, TiO2 and
Fe2O3 as catalysts, found that ZnO is a potential substitution for TiO2 because of their very similar
activities. To investigate the stability of mesotrione solution, the experiments were performed under
conditions of natural sunlight irradiation, as well as in the dark, in the presence/absence of TiO2
Degussa P25/ZnO. It was found that under irradiation by natural sunlight, mesotrione degraded
completely in two days in the presence of both catalysts, while in the absence of the catalyst, only
about 60% degraded during 6 months. In the dark, no mesotrione degradation occurred in the period
of 6 months. This indicates an even greater importance of studying the efficiency of the process of
photocatalytic degradation of mesotrione. Also, practically no degradation was observed under the
UV and visible light (halogen lamp) irradiation in the absence of catalysts (Fig. 1b). Bearing in
mind that mesotrione absorbs UV irradiation with the maximum absorption band at 225 nm, as well
as that the maximum emission of high-pressure mercury lamp is at 366 nm, no direct photolysis of
mesotrione could be expected in this case. In the case of halogen lamp, only a small part of the
spectrum belongs to the wavelengths below 400 nm, and also no direct photolysis of mesotrione
9
Page 9 of 49

could be expected. Thus, it appears that mesotrione can be degraded in the visible region, but at a
significantly lower rate compared to the UV region in the case of both catalysts (Fig. 1b). These
results indicate that, as was expected, the UV/catalyst treatment is most efficient for the removal of
the substrate.
Fig. 1
3.2. Effect of catalyst loading
The effect of the loading of TiO2 Degussa P25 on the efficiency of mesotrione
photodegradation was examined in the range from 0.50 to 4.00 mg mL1. Under the relevant
experimental conditions, the reaction followed pseudo-first order kinetics. On the basis of the
kinetic curves lnc (substrate concentration) vs. t, the values of the pseudo-first order rate constant,
k′, were calculated. It was found that the optimum of the catalyst loading is 2.00 mg mL1 (Table 2).
An increase in the load of the catalyst has two opposed contributions to the degradation process. On
the one hand, it makes available a higher number of active sites for adsorption on the surface of the
particles, and on the other, the amount of light dispersed by the catalyst particles is higher. These
particles are therefore not available for the generation of electronhole (eh+) pairs [16], and part
of the originally activated TiO2 may also be deactivated through collision [17]. Based on the data
presented in Table 2, the amount of TiO2 Degussa P25 was kept constant at the optimal load of 2.00
mg mL1 in all the subsequent photocatalytic degradation experiments.
As can be seen from Table 2, the presence of the photocatalyst caused a tremendous
enhancement of the degradation rate of mesotrione in comparison with the photolysis.
Table 2
3.3. Effect of pH
10
Page 10 of 49

The effect of pH is very important in the heterogeneous photocatalytic removal of organic
molecules since it influences both the surface charge of TiO2 and the ionic form of the reactant,
influencing thus the electrostatic interactions between the reactant species and the catalyst surface.
Moreover, the pH influences the sizes of TiO2 aggregates, interaction of the solvent molecules with
the catalyst, and the type of radicals or intermediates formed during the photocatalytic reaction [18].
The effect of the initial pH on the efficiency of mesotrione photodegradation was investigated in the
range from 2.011.0 (UV/TiO2 Degussa P25 process, Table 3). It was found that the efficiency of
photodegradation at pH 4.0 reaches its maximum, which can be explained by the possibility of
adsorption of mesotrione on the surface of the photocatalyst, bearing in mind that the point of zero
charge (pHpzc) of TiO2 is about 6 [19], and that the value of pKa of mesotrione is 3.12 [5]. The lower
degradation rate at lower pH may be probably attributed to the fact that mesotrione at the pH 2.0 is
present in its molecular form, and the TiO2 Degussa P25 surface is positively charged, so that their
interactions are not significant. On the other hand, at about pH 6, the degradation rate shows a
decrease. This can be explained by the fact that TiO2 Degussa P25 surface at about pH 6 is neutral
and mesotrione is in its anionic form, so that their interactions are not significant.
Table 3
The effect of pH on the degradation efficiency of mesotrione in aqueous suspension of ZnO
was studied in the range from 6.7 to 10.3 (Table 4). In the pH range from 6.7 to 9.2, the rate
constant shows a slight increase, to decrease at pH 10.3. This could be expected since the optimal
pH value for the interaction of mesotrione and ZnO is between the pKa of substrate and pHpzc of
photocatalyst [20]. Namely, the isoelectric point of ZnO is ~ 9.3 [21], and, as was already
mentioned, at the pH > 3.12 mesotrione is present in the anionic form, which leads to strong
attractive electrostatic interactions between the substrate and the catalyst. On the other hand, at the
pH > 9.3, repulsive electrostatic forces come into play, which results in a decreased rate of
11
Page 11 of 49

degradation of the substrate. Besides, the reduced removal efficiency in basic medium may be due
to the photocorrosion induced by the self-oxidation, as well as to the tendency of ZnO to dissolve
[21]. Similar findings in the studies of degradation of various substrates in the presence of ZnO
have been also reported by other authors [15,22−24]. Keeping in mind the similar behavior of ZnO
in acidic media, experiments at the pH < 6.7 were not carried out.
Table 4
3.4. Mechanism of photodegradation
In order to identify the type of intermediates formed during the degradation, UV spectra were
recorded in the range of 200–400 nm for mesotrione in the presence TiO2 Degussa P25/ZnO. It was
found that the degradation rate of the substrate measured by spectrophotometry was lower
compared to their transformation rates obtained by LC–DAD in the presence of both of catalysts.
Namely, the respective first-order rate constants for the reaction in the presence of TiO2 Degussa
P25 and ZnO measured by the spectrophotometric method and calculated for the first 10 min of
irradiation are: k = 7.37×102 min−1; r = 0.993, for TiO2 Degussa P25, and: k = 10.44×102 min−1; r
= 0.995, for ZnO, while the corresponding values measured by LC–DAD are: k = 38.44×102
min−1, r = 0.997, and k = 37.22×102 min−1; r = 0.993. This indicates the presence of the
intermediates which absorb at the same wavelength (225 nm) as mesotrione.
3.4.1. Evaluation of mineralization
The process of mineralization of mesotrione was studied by the IC (Fig. 2) and TOC
measurements. Since mesotrione contains nitrogen and sulfur, one can expect the formation of
ammonium and/or nitrate/nitrite and sulfate ions. The total transformation of nitrogen in the
presence of ZnO is significantly higher, and the amount of ammonium formed is the same in both
cases, while no nitrate was formed in the presence of TiO2 Degussa P25 (Fig. 2). Besides, the
12
Page 12 of 49

presence of sulfur in the mesotrione molecule results in the generation of sulfate. During 4 h of
degradation, 20% of sulfur was transformed to sulfate in the presence of TiO2 Degussa P25 and
even about 60% in the presence of ZnO, for the same irradiation time. Under the same experimental
conditions, acetate and formate were also liberated, which then transformed to CO2 and H2O. In the
case of formate, there are differences in the degree of its mineralization in dependence of the type of
catalyst. All these results lead to the conclusion that, irrespective of the fact that the kinetics of
degradation of mesotrione by applying the two catalysts is very similar, there are still differences in
the mechanisms of the degradation process.
Fig. 2
The TOC technique was employed to get a deeper insight into the degree of mineralization of
mesotrione in the presence of TiO2 Degussa P25 and ZnO. It was found that in the case of TiO2
Degussa P25, 90% of the substrate was mineralized during 4 h of irradiation, although only small
amounts of mineralized nitrogen and sulfur species were observed (Fig. 2a). It can be explained by
the possibility that small polar molecules. Also, sulfate ion can be adsorbed on catalyst surface
which is positively charged [25]. However, the same irradiation time in the presence of ZnO
resulted in the complete mineralization.
3.4.2. Effect of the scavengers of OH radicals and holes
In order to examine whether the heterogeneous catalysis occurs via OH radicals or holes,
ethanol or Na2SO3 was added to the reaction mixture containing mesotrione and the catalyst.
Namely, it is known that alcohols, such as ethanol and methanol, act as scavengers of OH radicals
[23,26]. On the other hand, Na2SO3 acts as a hole scavenger [27]. The obtained results indicate that
the reaction is slower compared to that in the absence of the scavengers (Fig. 3). Namely, 10 min of
irradiation of the suspension, regardless of the type of catalyst, resulted in a practically complete
13
Page 13 of 49

decomposition of mesotrione, whereas only about 30% decomposed under the same experimental
conditions, but with the presence of ethanol. The presence of Na2SO3 decreased also the degradation
rate, but to a smaller extent compared to ethanol – the amounts of decomposed mesotrione were
about 80% and 90% in the presence of TiO2 Degussa P25 and ZnO, respectively. The difference in
the degradation rate of mesotrione in the first few minutes depending on the type of the catalyst suggests
that in the case of ZnO decomposition takes place at a somewhat greater extent via holes than via OH
radicals. On the contrary, in the case of TiO2 Degussa P25 the degradation rate of substrate is
somewhat lower using ethanol which indicates that OH radicals have a slightly greater role in relation to
the process where ZnO is applied. This phenomenon can be explained by the difference in the mechanism of
mesotrione degradation, i.e. formation of intermediate P1 only in the case of ZnO. Finally, these results
prove that the reaction of the photocatalytic degradation of mesotrione proceeds mainly via OH
radicals, and to a smaller extent via holes. This is understandable if it is borne in mind that the
reactions involving OH radicals are prevailing in the processes of photocatalytic degradation in
aqueous media, although those involving holes cannot be neglected.
Fig. 3
3.4.3. Identification of intermediates
In the negative ionization (NI) mode, a number of peaks of photodegradation intermediates
were detected, indicating the presence of acidic groups (carboxyl, phenolic hydroxyl, or enolato-
structure). On the other hand, only one peak (P1) was prominent in the positive ionization (PI)
chromatograms. The kinetic curves of the identified intermediates are shown in Fig. 4.
Peak P1, eluting at 0.49 min, was found only among ZnO-catalyzed photodegradation products.
The only observable ion in the PI MS1 spectrum was at m/z 134. Assuming it represents an [M+H]+
ion, odd molecular weight would indicate the presence of nitrogen. However, no odd molecular
weight losses were observed in the PI MS2 spectra. Instead, only a number of fragments common to
14
Page 14 of 49

oxygenated compounds were detected, corresponding to losses of ketene (134→92), CO2 (116→72)
and water (134→116, 92→74, 74→56). No signal was present in the NI MS1 chromatogram,
suggesting the absence of acidic groups. The very short retention time and low molecular weight
indicate a very polar product of extensive degradation. At the moment, it is not possible to identify
any structural features of P1.
Peak P2, eluting at 0.68 min, represents a dominant peak in the NI MS1 chromatograms of both
TiO2 Degussa P25 and ZnO-catalyzed photodegradation reaction mixture. The odd monoisotopic
molecular weight of 245 Da indicates an odd number of nitrogen atoms. In the NI MS2 spectra, a

series of neutral losses were observed: CO (244→200), NO (200→170), CO (170→142),
CH3
2
(142→127) and CH2SO2 (142→64). The decarboxylation reaction (Δm/z = 44), as well as an
intense signal in NI, indicate the presence of aromatic carboxyl group. The consecutive losses of
NO and CO are consistent with the mechanism of nitroaromatics fragmentation that includes
nitro→nitrosooxy isomerization, loss of NO, and subsequent loss of CO accompanied by ring

contraction [28]. Finally, the loss of CH SO and the presence of methanesulfinate ion (
,
CH3SO2
2
2
m/z 79), formed by alternative fragmentation with the charge remaining on the sulfur-containing
group, confirm the presence of methylsulfonyl group. Thus, P2 represents 4-(methylsulfonyl)-2-
nitrobenzoic acid, also known as MNBA. This compound was already identified as mesotrione
photodegradation product and metabolite [7].
Peak P3, eluting at 1.20 min, corresponds to a compound with Mmi = 215, i.e. containing an
odd number of nitrogen atoms. The compound was detected in both TiO2 Degussa P25 and ZnO-
catalyzed photodegradation reaction mixture, and was detectable in both NI and PI modes. The
observed loss of CO2 (214→170) in NI, loss of H2O (216→198) in PI, as well as a strong signal in

NI, suggest the presence of carboxyl group. Subsequent losses of
(170→155) and SO2
CH3
(155→91), and the presence of methanesulfinate ion (m/z 79), confirm the presence of
methylsulfonyl group. This is supported by the observed loss of methanesulfinic acid (CH3SO2H) in
PI mode (216→136). At the same time, the absence of NO and HNO2 losses (Δm/z of 30 and 47
15
Page 15 of 49

mass units, respectively) points out to the absence of nitro group. Based on the molecular weight
and increased basicity (reflected in the detectability in PI mode), it was concluded that P3 likely
represents 2-amino-4-(methylsulfonyl)benzoic acid, also known as AMBA. This compound was
also well known mesotrione photodegradation product and metabolite [7].
Peak P4, with the retention time of 1.45 min, was detected in both TiO2 Degussa P25 and ZnO-
catalyzed photodegradation reaction mixture. The odd monoisotopic molecular weight of 217 Da
indicates the presence of an odd number of nitrogen atoms; the loss of NO (137→107, as well as
weak 153→123) identifies nitrogen atom as a part of nitro moiety. The absence of CO2 loss
indicates that no carboxyl groups are present, and the ionizability in NI is due to some other acidic

group, possibly phenolic hydroxyl. The observed losses of
(216→201) and SO2 (201→137)
CH3
confirm the presence of CH3SO2-group. Based on that, P4 is identified as hydroxy-1-
(methylsulfonyl)-3-nitrobenzene.
Peak P5, eluting at 1.57 min, was detected in both TiO2 Degussa P25 and ZnO-catalyzed
photodegradation reaction mixture. The odd molecular weight of 371 Da, 32 Da higher than that of
mesotrione, indicates the presence of one nitrogen atom and possible introduction of two oxygen
atoms. In the NI MS2 mode, the compound underwent extensive fragmentation; unfortunately, it
was not possible to identify any structural features.
Peak P6, eluting at 1.62 min, was detected in both TiO2 Degussa P25 and ZnO-catalyzed
photodegradation reaction mixture. Unlike P5, its signal in PI was enough to allow acquisition of PI
MS2 spectra. A monoisotopic molecular weight of 355 Da, 16 Da higher than that of mesotrione,
suggests the introduction of one oxygen atom. The observed losses of 47 (HNO2) and 79 mass units
(
 ) in the NI MS2 spectra confirm the presence of nitro and methylsulfonyl groups,
CH3SO2
respectively. The loss of a water molecule, observed early both in the PI (356→338) and NI
(307→289) fragmentation pathways may indicate the presence of alcoholic hydroxyl. The
prominent ion at m/z 228, also detected in the PI MS2 spectra of mesotrione, probably represents
{[4-(methylsulfonyl)-2-nitrophenyl]methylidyne} oxonium, thus suggesting that the phenyl part of
16
Page 16 of 49

the molecule is intact, and hydroxyl is introduced into the cyclohexyl ring. In that case, peak P6
represents either 4- or 5-hydroxy-2-[4-(methylsulfonyl)-2-nitrobenzoyl]cyclohexane-1,3-dione.
The different kinetics of some intermediates with the two photocatalysts and the detection of
intermediate P1 only in the case of ZnO suggest that the degradation pathways with TiO2 Degussa
P25 and ZnO might not be identical. On the basis of the identified intermediates and kinetic data,
we propose a tentative scheme of mesotrione photocatalytic degradation (Fig. 5).
Fig. 4
Fig. 5
3.5. Cell growth activity
In order to evaluate the environmental hazardousness of the water containing mesotrione and
its intermediates, toxicity assessments in mammalian cell lines were performed during the
photodegradation. The effect of the examined samples on the growth of selected cell lines was
dependent on the type of catalyst, time of irradiation, concentration, and the cell line (Fig. 6).
The mixture of mesotrione and its photocatalytic degradation intermediates obtained using TiO2
Degussa P25 catalyst produced mild effects on the growth of both cell lines – the most pronounced
inhibition being observed at 30 and 120 min of irradiation at the lowest dilution (highest
concentration), but in all cases it was below 10% (Fig. 6a and c). Similar effects were obtained in
the H-4-II-E cell line using ZnO catalyst, but at 30, 60 and 120 min of irradiation (Fig. 6b), while
the effects in the MRC-5 cell line derived from healthy tissue were more pronounced, and were
observed at 30 and 60 min of irradiation at the lowest dilutions, reaching 28% of the cell growth
inhibition (Fig. 6d). This implies that the irradiation contributed to the increase in the toxicity of the
mixture that was no longer dominated by the parent compound. In the examined concentration
range, none of the treatments produced cell growth inhibition higher than 50%, either by the parent
17
Page 17 of 49

compound or by the samples obtained after different irradiation times. Hence, their toxicity was
substantially lower compared to the cytotoxic drugs and the known toxicant, HgCl2 [29]. In the
blank tests, the effects in both cell lines were below 10% (Fig. 7).
Generally, a higher toxicity of the photodegradation mixtures was observed when using ZnO
catalyst especially in the case of MRC-5 (Fig. 5). This is probably a consequence of the different
degradation mechanisms operating in the presence of the two catalysts, which means that, in
addition to the intermediates that are common for both catalysts, some catalyst-specific product is
generated (P1). Apart from the difference in the mechanism, there exists also a difference in the
kinetics of formation of the same intermediates (Fig. 4), which can entail the difference in toxic
properties of the investigated systems, as well as in their intensities. Based on the results, it appears
that the intermediate P5 is present in the system with ZnO in a higher concentration compared to
that with TiO2 Degussa P25 and a consequence of this could be the observed more pronounced
effect on the cell activity. On the other hand, the intermediates P2 and P4, which appeared at higher
concentrations in the system with TiO2 Degussa P25, showed no significant effect on the selected
cell lines, which indicates their lower toxicity.
Fig. 6
Fig. 7
4. Conclusions
The results of this study clearly indicate that the photocatalytic treatment using TiO2 Degussa
P25 and ZnO with UV irradiation can efficiently eliminate mesotrione from water. The reaction
followed the pseudo-first order kinetics. The optimum of the catalyst loading was 2.0 mg mL1 at
the pH 4.0 (for TiO2 Degussa P25) and 9.2 (for ZnO). It was found that the presence of ethanol, and
to a smaller extent of Na2SO3, inhibited the mesotrione photodecomposition, suggesting that the
18
Page 18 of 49

reaction proceeds mainly via free OH radicals, and partly via holes. The mesotrione disappearance
is accompanied by the release of ionic intermediates and byproducts,
 , NO3 , SO24 , acetate
NH 4
and formate. While mesotrione is removed from water in about 15 min, its mineralization is
completed after about 4 h of irradiation in the case of both catalysts. By using the LC–ESI–MS/MS
method, it was possible to identify several intermediates and conclude that the degradation
pathways with TiO2 Degussa P25 and ZnO might not be identical. The analysis of the intermediate
products formed during the photocatalytic degradation served as the basis to postulate the
degradation pathways. The cell growth in both cell lines was mildly affected by the mixture of
mesotrione and its photocatalytic degradation intermediates obtained using TiO2 Degussa P25
catalyst, and moderately affected using mixture obtained using ZnO catalyst. A comparison of the
evolution of the toxicity and degradation kinetics indicates that the toxicity toward the MRC-5 cell
line increased after irradiation using ZnO catalyst, whereby at 120 min decreased. This implies that
the irradiation contributed to the increase of the toxicity of the mixtures that were no longer
dominated by the parent compound. In the examined concentration range, none of the treatments
produced cell growth inhibition higher than 50%, either by the parent compound itself or by the
samples obtained after different irradiation times.
Acknowledgment
The authors greatly appreciate the financial support from the Ministry of Education, Science and
Technological Development of the Republic of Serbia (Project No. 172042). Besides, they thank
MSc Nada Jurišić for the technical assistance.
19
Page 19 of 49

References
[1] D.J. Beale, N.A. Porter, F.A. Roddick, Talanta 78 (2009) 342–347.
[2] H. Barchanska, M. Rusek, A. Szatkowska, Environ. Monit. Assess. 184 (2012) 321–334.
[3] P. Alferness, L.Wiebe, J. Agric. Food Chem. 50 (2002) 3926–3934.
[4] J. Rouchaud, O. Neus, K. Cools, R. Bulcke, Toxicol. Environ. Chem. 77 (2000) 31–40.
[5] European Commission Report (2003). Commission working document. Review Report for the
Active Substance Mesotrione. Health and Consumer Protection Directorate General. Directorate
E–Food Safety: Plant health, animal health and welfare, international questions, E1. Plant health.
p. 22.
[6] H. Chaabane, E. Vulliet, F. Joux, F. Lantoine, P. Conan, J.-F. Cooper, C.-M. Coste, Water Res.
41 (2007) 1781–1789.
[7] A. ter Halle, C. Richard, Environ. Sci. Technol. 40 (2006) 3842–3847.
[8] N. Bensalah, A. Khodary, A. Abdel-Wahab, J. Hazard. Mater. 189 (2011) 479–485.
[9] M. Murati, N. Oturan, J.-J. Aaron, A. Dirany, B. Tassin, Z. Zdravkovski, M.A. Oturan,
Environ. Sci. Pollut. Res. 19 (2012) 1563–1573.
[10]N.D. Abazović, L. Mirenghi, I.A. Janković, N. Bibić, D.V. Šojić, B.F. Abramović, M.I. Čomor,
Nanoscale Res. Lett. 4 (2009) 518–525.
[11]B.F. Abramović, V.N. Despotović, D.V. Šojić, D.Z. Orčić, J.J. Csanadi, D.D. Četojević-Simin,
Chemosphere 93 (2013b) 166–171.
[12] P. Skehan, R. Storeng, D. Scudiero, A. Monks, J. McMahon, D. Vistica, J.T. Warren, H.
Bokesch, S. Kenney, M.R. Boyd, J. Natl. Cancer Inst. 82 (1990) 1107–1112.
[13] D.D. Četojević-Simin, S.A. Velićanski, D.D. Cvetković, L.S. Markov, Z.J. Mrđanović, V.V.
Bogdanović, S.V. Šolajić, Food Bioprocess Tech. 5 (2012) 1756–1765.
[14] D.C. Hurum, A.G. Agrios, K.A. Gray, J. Phys. Chem. B 107 (2003) 4545–4549.
[15] B. Dindar, S. Içli, J. Photochem. Photobiol. A: Chem. 140 (2001) 263–268.
[16] F. Méndez-Arriaga, S. Esplugas, J. Giménez, Water Res. 42 (2008) 585–594.
20
Page 20 of 49

[17] B. Neppolian, H.C. Choi, S. Sakthivel, B. Arabindoo, V. Murugesan, J. Hazard. Mater. 89
(2002) 303–317.
[18]C. Tizaoui, K. Mezughi, R. Bickley, Desalination 273 (2011) 197–204.
[19] M. Kosmulski, Adv. Colloid Interface Sci. 99 (2002) 255–264.
[20] N. Daneshvar, S. Aber, M.S. Seyed Dorraji, A.R. Khataee, M.H. Rasoulifard, Sep. Purif.
Technol. 58 (2007) 91–98.
[21]R. Comparelli, E. Fanizza, M.L. Curri, P.D. Cozzoli, G. Mascolo, A. Agostiano, Appl. Catal.
B: Environ. 60 (2005) 1–11.
[22]B.F. Abramović, N.D. Banić, J.B. Krstić, Ind. Eng. Chem. Res. 52 (2013a) 5040–5047.
[23] N. Daneshvar, D. Salari, A.R. Khataee, J. Photochem. Photobiol. A: Chem. 162 (2004) 317–
322.
[24] A. Shafaei, M. Nikazar, M. Arami, Desalination 252 (2010) 8–16.
[25] M. Muruganandham, M. Swaminathan, Dyes Pigments 68 (2006) 133–142.
[26] I. Ilisz, A. Dombi, Appl. Catal. A: Gen. 180 (1999) 35–45.
[27] J. Han, W. Qiu, W. Gao, J. Hazard. Mater. 178 (2010) 115–122.
[28]A.C. Schmidt, R. Herzschuh, F.M. Matysik, W. Engewald, Rapid Commun. Mass Spectrom.
20 (2006) 2293–2302.
[29] D.D. Četojević-Simin, S.J. Armaković, D.V. Šojić, B.F. Abramović, Sci. Total Environ. 463–
464 (2013) 968–974.
21
Page 21 of 49

Captions for figures
Fig. 1. Kinetics of the photolytic and photocatalytic degradation of mesotrione (c0 = 0.05 mM)
under (a) different crystal structures of TiO2 and (b) UV and visible light irradiation. When
present, the catalyst loading was 2.00 mg mL1.
Fig. 2. Evolution of different ions in mesotrione (c0 = 0.05 mM) photocatalytic degradation under
UV irradiation by (a) TiO2 Degussa P25 and (b) ZnO (2.00 mg mL1).
Fig. 3. Effect of ethanol and Na2SO3 on the kinetics of mesotrione photodegradation (c0 = 0.05
mM) in the presence of different catalysts (2.00 mg mL1) and UV irradiation.
Fig. 4. Kinetics of appearance/disappearance of mesotrione (c0 = 0.05 mM) and its
photodegradation intermediates with (a) TiO2 Degussa P25 and (b) ZnO (2.00 mg mL1) and UV
irradiation using LCESIMS/MS (ESI).
Fig. 5. Tentative pathways for the photocatalytic degradation of mesotrione in the presence of TiO2
Degussa P25 and ZnO.
Fig. 6. Effect of mesotrione and its photodegradation mixtures obtained after different irradiation
times using TiO2 Degussa P25 (a, c) and ZnO catalysts (b, d) on the growth of the H-4-II-E (a, b)
and MRC-5 (c, d) cell line.
Fig. 7. Growth activity of the (a) H-4-II-E and (b) MRC-5 cells in the blank tests.
22
Page 22 of 49

Table 1. MS/MS fragmentation data for detected compounds at three collision
energies. The collision energies and product ions were selected for MRM
MS/MS method.
Mmi
Peak label
tR (min)
Mode
PI
Precursor ion m/z
Vcol (V)
Product ions m/z, rel. abundance
(g mol−1)
339
Mesotrione
3.09
340
0
10
20
30
0
10
20
30
340 (100), 228 (14)
340 (20), 228 (100)
228 (100), 104 (17)
228 (69), 166 (12), 104(100), 79(19), 77(33), 76(20)
338 (11), 291 (100)
291 (100)
291 (100), 249 (10), 212 (38)
291 (33), 249 (19), 212 (100), 170 (5)
NI
PI
338
134
244
P1
P2
P3
0.49
0.68
1.20
133*
245
0
10
20
30
134 (100)
134 (100), 116 (16), 92 (25), 74 (35), 72 (21), 56 (18)
134 (10), 92 (10), 74 (41), 72 (23), 56 (100)
56 (100), 56 (100)
NI
0
10
20
30
244 (7), 200 (100), 170 (19), 142 (6)
200 (21), 170 (66), 142 (100)
171 (9), 142 (100), 79 (7)
142 (100), 127 (14), 79 (48), 64 (11)
215
PI
216
214
0
10
20
30
0
216 (66), 198 (100)
198 (100), 136 (18)
198 (44), 136 (100), 119 (11), 108 (12), 107 (17)
136 (100), 119 (59), 108 (48), 107 (52), 91 (39)
214 (100), 170 (16)
NI
10
20
30
214 (13), 170 (100), 155 (17)
170 (70), 155 (100), 106 (10), 91 (79), 79 (34)
155 (14), 106 (5), 91 (100), 79 (49), 64 (13)
P4
1.45
217
NI
216
0
10
20
30
216 (100)
216 (100), 201 (17), 153 (6), 137 (18)
216 (30), 201 (12), 153 (22), 137 (100)
137 (100), 107 (14)
P5
P6
1.57
1.62
371
355
NI
PI
370
356
0
10
370 (63), 305 (100), 277 (9), 261 (14)
305 (100), 277 (31), 261 (48), 226 (32), 182 (40), 169 (28)
0
10
20
30
0
10
20
30
356 (100), 338 (18)
356 (32), 338 (94), 310 (13), 296 (39), 228 (100)
250 (9), 228 (100), 104 (25)
228 (100), 104 (19)
354 (29), 307 (100), 289 (10)
307 (100), 289 (45)
NI
354
307 (37), 289 (100), 210 (25)
289 (74), 210 (100)
* Since only one ion was observed in MS1, its assignation and thus Mmi determination is uncertain
23
Page 23 of 49

Table 2. Values of the reaction rate constant (k) for the degradation of mesotrione as a
function of the catalyst loading.
TiO2 (mg mL−1)
r2†
k 102 (min−1)§
0.38
0
0.989
0.999
0.972
0.997
0.980
0.965
0.50
1.00
2.00
3.00
4.00
21.41
22.73
38.44
30.12
30.98
§ k determined after 10 min of irradiation;
† correlation coefficient.
24
Page 24 of 49

Table 3. Values of the reaction rate constant (k) for the mesotrione degradation in the presence of
TiO2 as a function of pH.
pH
2.0
r2†
k 102 (min−1)§
16.25
0.996
0.997
0.998
0.999
0.999
4.0
6.0
38.44
14.07
9.0
12.27
11.0
8.63
§ k determined after 10 min of irradiation;
† correlation coefficient.
25
Page 25 of 49

Table 4. Values of the reaction rate constant (k) for the mesotrione degradation in the presence of
ZnO as a function of pH.
pH
6.7
r2†
k 102 (min−1)§
29.85
0.997
0.998
0.998
0.999
8.3
9.2
33.99
37.22
10.3
25.40
§ k determined after 10 min of irradiation;
† correlation coefficient.
26
Page 26 of 49

Highlights
 UV/TiO2 and UV/ZnO treatments eliminate efficiently mesotrione from water.
 Photocatalytic degradation follows the pseudo-first order kinetics.
 Mineralization of mesotrione is completed after about 4 h of irradiation.
 Four intermediates were identified by LC–ESI–MS/MS.
 Toxicity toward the H-4-II-E and MRC-5 cell lines increased by irradiation with ZnO.
27
Page 27 of 49

*Graphical Abstract (for review)
Page 28 of 49

Figure 1a black white
Page 29 of 49