3.1 Ti/TiO2 nanotubes electrodes
The Ti/TiO2 or Ti/TiO2-(metallo)porphyrin electrodes (H2P, MnP1, or MnP2) were prepared by the anodic growth process (Albu and Schmuki 2013; Berger et al 2010; Brammer et al 2012; Tan et al 2012). This is an efficient technique for the controlled growth of films on metal surfaces. For titanium, the majority of TiO2 growth occurs on the surface of the metal plate.
The dissolution reaction of the titanium cations generates negatively charged vacancies in the oxide, which migrate to the metal/oxide interface because of the potential gradient. The presence of vacancies at the metal/oxide interface facilitates the oxidation of Ti to Ti4+, since the cations can migrate easily to the vacancy and, consequently, an increase in current density occurs. During this step, the nucleation of nanopores occurs on the oxide surface and the growth of the nanotubular oxide layer occurs when the current density reaches a constant value over time (Kim et al 2018).
3.2 Characterization of the electrodes
The morphology of the film formed on the surface of the titanium plate is directly related to the photoelectrochemical response of the material. Some factors such as anodizing time, the solvent used as an electrolyte, and potential range influence the morphology of the film. The use of ammonium fluoride (NH4F) at low concentrations increases the titanium oxide layer and, consequently, increases the reactive area of the electrode. The anodic growth process in the presence of fluoride anions occurs as follows: anodic oxidation of the plate (where an amorphous TiO2 film grows), followed by formation of [TiF6]2- (water-soluble) complexes, and oxide dissolution due to the migration of fluoride anions through the oxide layer (Roy et al 2011).
The length of the nanotubes and the pore size are directly related to the anodizing time and the voltage used. The higher the voltage and time, the longer the tube lengths and diameters, until a limit is reached, from which the nanotubes stop growing (Bauer et al 2006). Another factor that affects the length of the nanotubes is the distance between the working and auxiliary electrodes in the anodizing process; the closer the electrodes are, the longer the tubes will be (Sun et al 2009).
The degree of electrode coating with the (metallo)porphyrins were calculated through spectroscopic analyzes in the visible ultraviolet region (UV-Vis), evaluating the initial and final (metallo)porphyrin solution absorbance used during the coating process. The (metallo)porphyrin coating degrees were 15, 12, and 13% for the H2P, MnP1, or MnP2, respectively.
Surface studies of Ti/TiO2 and Ti/TiO2-H2P electrodes were performed (porphyrin system that obtained better degradation results) by scanning electron microscopy (SEM). The images are shown in Figure 3.
The SEM analyses were obtained to verify the formation of the TiO2 nanotubes. The electrodes underwent treatment before the deposition of an Au/Pd film, which was added to become the conductive material and facilitate the construction of the microscopic image. The most intense white spots observed for the two electrodes (Ti/TiO2 and Ti/TiO2-H2P) are related to the treatment of the electrode with an Au/Pd film. By comparing the surfaces of the two electrodes, it is possible to observe that the deposition of porphyrin occurred mainly among the TiO2 nanotubes (Figure 3).
It is observed in Figure 3 that the formation of a homogeneous film of TiO2 nanotubes, uniformly organized under the titanium plate, occurs. The nanotubes formed have an average diameter of 150 nm (± 10 nm), as previously reported for other similar films (Macak et al 2007; Osugi et al 2008). As already described by one of the authors in a previous study (Brugnera et al 2010), the XRD patterns of this prepared TiO2 nanotubes allowed us to characterize the TiO2 on the electrode as the anatase phase. This is considered the most active phase for the use of TiO2 as a semiconductor (Li et al 2009; Roy et al 2011) and had been described after calcination of titanium dioxide nanotube arrays at 450 oC (Aeimbhu 2018; Kirkey et al 2019).
The UV–vis diffuse reflectance spectra of all materials are illustrated in Figure S1 (Supplementary material). It is possible to observe that the pure TiO2 nanotubes exhibited an absorption edge at 380 nm as already described by Han and coworkers (Han et al 2016). For the porphyrin-modified TiO2 nanotubes, it can be also noticed the presence of broad absorption peaks in the visible light region. The diffuse reflectance measurements could be converted into an equivalent absorption coefficient using the equation (Ahν)2 = K (hν - Eg), in which A corresponds to the absorbance, K is a proportionality constant, hν is the photo energy, and Eg is the band gap energy (Han et al 2016). The band gap calculated for the TiO2 nanotube (3.32 eV) was close to the already described in the literature (3.2 eV) for the anatase phase (Cai et al 2013), while the band gap values of the Ti/TiO2-porphyrin were slightly changed to 3.19; 3.23; 3.25 eV for Ti/TiO2-H2P, Ti/TiO2-MnP1 and Ti/TiO2-MnP2, respectively. As already expected (Brito et al 2020; Han et al 2016; Wei et al 2015), the porphyrins could enlarge the wavelength response range and contract the band gap, which is beneficial for the photocatalytic activity of the electrodes.
The photoactivity of the electrodes were evaluated by photocurrent measures, Figure S2 (Supplementary material). The photocurrent in the dark was negligible. The highest photocurrent was obtained for Ti/TiO2-H2P electrode (28 mA at 1.5 V). The photocurrent voltammograms show the generation of electron-hole pairs. So, the electrode Ti/TiO2-H2P is able to generate more hole and consequently is more efficient for the degradation (Zanoni et al 2003). The photocurrent for the other electrodes at 1.5 V potential were 9.8 mA, 3.6 mV and 5.3 mA, for Ti/TiO2, Ti/TiO2-MnP1 and Ti/TiO2-MnP2, respectively.
3.3 Atrazine degradation reactions
The effect of heterogenized water-soluble manganese porphyrins in Ti/TiO2 nanotubes electrodes was evaluated in a preliminary atrazine degradation light-mediated through photocatalysis (PC) and photoelectrocatalysis (PEC) and the performance was compared with that on Ti/TiO2 and photolysis system (atrazine degradation degree through the photolysis, photocatalysis (PC), and photoelectrocatalysis (PEC) reactions mediated by porphyrin-modified Ti-TiO2 electrodes are presented at Table S1 in Supplementary Material). Manganese porphyrins were the first chosen since they have presented better results compared to iron and cobalt porphyrin on photocatalytic degradation dyes by TiO2 modified by some metalloporphyrins (Zhou et al 2012). It is important to note that ATZ did not undergo degradation in the absence of light and/or catalyst, as can be seen in Figure 4 (control system). In the presence of Ti/TiO2 or Ti/TiO2-porphyrins electrodes (MnP1 or MnP2), a high degree of ATZ degradation (up to 92%) was observed in the PC or PEC reaction systems, when using UV light (Hg lamp). Unfortunately, these results did not overcome those observed for the photolysis system (without any electrode) which also provided 92% of ATZ degradation at 60 minutes.
In the photo(electro)catalysis systems, despite the different reaction orders, the manganese porphyrins had no positive contribution in the degradation of the substrate. As already proposed by Li and coworkers (Li et al 2011), the manganese porphyrin (a similar one; their manganese porphyrin macrocycle has no -COOH groups) could not be able to transfer its photo-induced electrons to the conduction band of the titanium nanotubes. The ATZ degradation observed in the PC had an exclusive contribution to the light since the presence of the TiO2 did not provide any change concerning the photolysis system. By evaluating the ATZ degradation profile over time in photo(electro)catalysis systems, a different behavior was observed for the electrodes coated with different manganese porphyrins, although the degree of final degradation of the pesticide was not very different. In this way, it was decided to evaluate the kinetic constant for atrazine degradation (Table 1) to understand the influence of MnP in Ti/TiO2 nanotubes electrodes. The kinetic coefficients were determined by fitting the data by the first- or second-order kinetic model (Table S2 in Supplementary Material).
Table 1 Kinetic data of the photolysis, and photo(electro)catalysis reactions
System
|
Order
|
R2
|
kobs (s-1)
|
t½ (min)
|
kobs (L mol-1 s-1)
|
Photolysis
|
First
|
0.9893
|
0.0006
|
19
|
-
|
Ti/TiO2 (PC)
|
Second
|
0.9941
|
-
|
-
|
54.357
|
Ti/TiO2-MnP1 (PC)
|
First
|
0.9939
|
0.0006
|
19
|
-
|
Ti/TiO2-MnP2 (PC)
|
First
|
0.9905
|
0.0005
|
23
|
-
|
Ti/TiO2 (PEC)
|
Second
|
0.9613
|
-
|
-
|
90.832
|
Ti/TiO2-MnP1 (PEC)
|
Second
|
0.9718
|
-
|
-
|
58.383
|
Ti/TiO2-MnP2 (PEC)
|
First
|
0.9973
|
0.0004
|
29
|
-
|
The Langmuir–Hinshelwood (L–H) kinetic model was used. The first-order linear kinetic model equation is shown below and by plotting Ln (Ct/C0) versus time, the correlation coefficient(kobs) was obtained.
in which C0 is the original concentration of atrazine and Ct is the concentration after irradiated for time t. kobs is the overall degradation rate constant, which demonstrated the activity of the photocatalytic system. The half-life values were calculated from the kobs values using the following formula: ![](data:image/png;base64,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)
The coefficients (R2) of the first-order kinetic model were ca. 0.99, which were much better than that of any other kinetic model for the photolysis, Ti/TiO2-MnP1 (PC), Ti/TiO2-MnP2 (PC), and Ti/TiO2-MnP2 (PEC) systems. In this way, according to the experimental data, the first-order kinetic model was more suitable for these systems. Thus, the kinetics of these atrazine removal reactions follow the L–H model. Otherwise, second-order kinetics prevailed over Ti/TiO2 (PC), Ti/TiO2 (PEC), and Ti/TiO2-MnP1 (PEC) systems.
Once the Mn-porphyrin had an inhibitory effect on the ATZ degradation activity by Ti/TiO2 nanotubes electrodes, we decided to heterogenize a water-soluble free-base porphyrin (with the same carboxylic group) in Ti/TiO2 nanotubes electrodes and to evaluate the atrazine degradation light-mediated through photocatalysis (PC) and photoelectrocatalysis (PEC) (Figure 5 and Table 2).
Table 2 Kinetic data of the photolysis, photocatalysis, and photoelectrocatalysis reactions
System
|
Order
|
R2
|
k
(s-1)
|
t½
(min)
|
k
(L mol-1 s-1)
|
Photolysis
|
First
|
0.9893
|
0.0006
|
19
|
-
|
Ti/TiO2(PC)
|
Second
|
0.9941
|
-
|
-
|
54.357
|
Ti/TiO2-H2P (PC)
|
Second
|
0.9716
|
-
|
-
|
32.881
|
Ti/TiO2(PEC)
|
Second
|
0.9613
|
-
|
-
|
90.832
|
Ti/TiO2-H2P (PEC)
|
Second
|
0.9169
|
-
|
-
|
94.077
|
Ti/TiO2-H2P (PEC) (in surface water)
|
First
|
0.9761
|
0.0006
|
19
|
-
|
We have replicated the PEC reactions using Ti/TiO2-H2P undersurface water collected from a river in Cuiabá, MT, Brazil. Comparing the photoelectrocatalytic systems with the Ti/TiO2-H2P electrode in distilled water and surface water (removal of the Cuiabá River), we observed a change in reaction orders (from second to first order, respectively, Table 2). This change may be occurring due to the presence of organic matter which, consequently, may lead to competition between atrazine and organic matter in the degradation process. The presence of ions and/or other species in surface water could be responsible for changing the electrostatic potentials of the system.
To the best of our knowledge, only three studies have described atrazine degradation presenting kinetics data (see the Supplementary Material to verify the studies which used other types of materials to catalyze this degradation, Table S3). We could check that, comparing our data with the already published studies, our systems could lead to the highest atrazine degree (92%) in only 60 minutes. Other studies have shown this degree of ATZ degradation after 300 minutes (5 hours). In most cases, the photoelectrocatalytic system had a better performance compared to the photocatalytic and electrochemical systems (Table S3 at Supplementary Material). In our study, the performance depends on the porphyrin attached to the TiO2 nanotubes. The non-modified Ti/TiO2 electrode and that one modified by the free-base porphyrin (H2P) favor a second-order rate independently of the catalyst system, but the photoelectrocatalytic system presented a higher kinetic constant. Surprisingly, the Ti/TiO2-H2P material described the first-order rate when it was studied in surface water. Probably, the presence of other compounds (salts, organic matter) could interfere, reducing the rate at which the atrazine degradation occurs. The manganese porphyrins presented different behavior when modifying the Ti/TiO2 electrode; MnP2 practically did not change the constant rate, while MnP1 changed from a first-order reaction at the photocatalytic system to a second-order reaction at photoelectrocatalytic one.
In previous work, we have shown that in a system using ATZ contaminated river water, the homogeneous catalysts could promote the complete degradation of ATZ, differently from the model systems in ACN. Besides, the proposed products have not shown any toxicity (Lage et al 2019). Although the Ti/TiO2-H2P (PEC) experiment (in surface water) did not lead to the best degradation system, in the present work, we have decided to perform the analysis of the reaction products by mass spectrometry and evaluate their cytotoxicity.
In Figure 6 the proposed products of atrazine degradation are shown by the reaction with the Ti/TiO2-H2P electrode. The product of m/z 202 was observed at the time of 5 minutes; in the time of 10 minutes, the products of m/z 140 and 111 were observed; at 15 and 30 minutes, the m/z 111 product was still detected, in addition to the product of m/z 128; in 60 minutes, a product of m/z 102 was observed. It is important to highlight that these products have not been described for porphyrin-mediated systems (Martins et al 2022).
The optimized geometry of atrazine is shown in Figure 7. Frequency calculations revealed that the obtained structure corresponds to a local minimum due to the absence of imaginary frequencies. From the optimized geometry, the electronic spectrum was then simulated at the PBE-D3(BJ)/aug-cc-pVTZ(SMD) level of theory.
Table 3 shows the two principal electronic transitions and the contributions of Kohn-Sham orbitals involved in each excited state and their respective contributions to the state (c2). It is possible to observe that both states are principally constituted by two main contributions, to know: (i) HOMO ® LUMO and (ii) HOMO – 1 ® LUMO + 1 (Figure 8). The principal contribution, approximately 80%, corresponds to the HOMO ® LUMO transition. This transition can be characterized as a n®π* with respect to the non-ligand orbitals located on the nitrogen atoms and the C–N anti-bonding orbitals located on the ring. Besides, the first excited state also contains a filled non-ligand orbital on the chlorine atom. Table 3 also shows that the calculations are in good agreement with the experimental UV/Vis spectrum.
From these results, the degradation products proposed above might be corroborated. As the atrazine degradation takes place, the experimental response is the quenching of the maximum absorption band at 221 nm. As the simulated electronic spectrum shows, the excited electronic state of this transition has a filled non-bonded orbital on chlorine. Since the intensity of an electronic band is proportional to the product of the wave function of the ground state (GS) times the transition dipole moment operator is applied to the wave function of the excited state
– the quenching of the maximum absorption band with the degradation of atrazine might be explained by the dehalogenation proposed in Figure 6. Since the absence of a chlorine atom will not allow the population of the non-bonded orbital, the coupling between the product of the transition dipole moment with the excited-state wave function will then vanish as the dehalogenation occurs. This supports the observed quenching of the maximum absorption band in the UV/Vis spectrum as dehalogenation occurs.
Table 3 Experimental and calculated electronic transitions – at PBE-D3(BJ)-aug-cc-pVTZ(SMD) level of theory – for atrazine and their respective ∆E, fosc, major orbital contribution, and c2
Wavelength (nm)
|
∆E (eV)
|
fosc
|
Major contribution
|
c2(%)
|
Experimental
|
TD-DFT
|
221
|
219.71
|
5.643
|
0.291
|
HOMO®LUMO
|
0.75
|
HOMO-1®LUMO+1
|
0.18
|
219.09
|
5.659
|
0.125
|
HOMO®LUMO
|
0.83
|
HOMO-1®LUMO+1
|
0.12
|
Besides, it is essential to assess the cytotoxicity of the degradation products. For this heterogeneous system, the cytotoxicity of atrazine and the formed products when the Ti/TiO2-H2P electrode was evaluated (Figure 9). We have utilized the human peripheral blood mononuclear cell (PBMC) model, as described by Lage and coworkers (Lage et al 2019). Immunotoxicity is a seriously deleterious effect of chemicals against a human being. The use of PBMCs to predict toxicity has the advantage to predict the hazards of compounds over the immunological function of human being, once the assay employs fresh human peripheral blood, where the defense cells are viable and functional. So, it is possible to identify damage from different substances that will compromise human health (Villani et al 2018). The viability of human PBMCs after treatment with different concentrations of atrazine or its degradation products is displayed in Figure 9. The results for ATZ and the reaction system were compared to the growing of the cells in PBS, considered the control system. It is observed that the degradation products at the highest concentration of 3 μM, significantly reduced the cell viability of PBMC. No differences in cell viability were observed with the lowest concentrations of these products, neither with atrazine at 3 μM. By the preliminary results, we could verify that the products formed during photoelectrocatalytic processes by Ti/TiO2-H2P electrode are equal or more toxic than the original atrazine.