Defective TiO2 Nanotube Arrays for Efficient Photoelectrochemical Degradation of Organic Pollutants

Oxygen vacancies (OVs) are one of the most critical factors that enhance the electrical and catalytic characteristics of metal oxide-based photoelectrodes. In this work, a simple procedure was applied to prepare reduced TiO2 nanotube arrays (NTAs) (TiO2–x) via a one-step reduction method using NaBH4. A series of characterization techniques were used to study the structural, optical, and electronic properties of TiO2–x NTAs. X-ray photoelectron spectroscopy confirmed the presence of defects in TiO2–x NTAs. Photoacoustic measurements were used to estimate the electron-trap density in the NTAs. Photoelectrochemical studies show that the photocurrent density of TiO2–x NTAs was nearly 3 times higher than that of pristine TiO2. It was found that increasing OVs in TiO2 affects the surface recombination centers, enhances electrical conductivity, and improves charge transport. For the first time, a TiO2–x photoanode was used in the photoelectrochemical (PEC) degradation of a textile dye (basic blue 41, B41) and ibuprofen (IBF) pharmaceutical using in situ generated reactive chlorine species (RCS). Liquid chromatography coupled with mass spectrometry was used to study the mechanisms for the degradation of B41 and IBF. Phytotoxicity tests of B41 and IBF solutions were performed using Lepidium sativum L. to evaluate the potential acute toxicity before and after the PEC treatment. The present work provides efficient PEC degradation of the B41 dye and IBF in the presence of RCS without generating harmful products.


S2
The XPS measurements were performed in the analysis chamber of the electron spectrometer ESCALAB-Mk II (VG Scientific) with a base pressure of ~ 5 x 10 -8 Pa. The samples in "as prepared" form were mounted on a conductive adhesive tape and C1s, O1s, Ti2p photoelectron and TiLMM Auger spectra were recorded by using AlK α radiation. All spectra were calibrated by using C 1s peak at 285.0 eV as a reference. The surface composition was evaluated from the photoelectron intensities divided by the corresponding photoionization cross sections taken from Scofield [2].

LC-MS analysis
Liquid chromatography coupled with mass spectrometry (LC-MS) measurements were performed with a Water Acquity binary pump based ultra-performance liquid chromatography (UPLC) system (Waters, Milford, USA) binary pump with an Ascentis express C-18 column (2µm, 100 x 2,1 mm i.d). LC column with mobile phases consisting of 0.1 % formic acid in water (A) and acetonitrile (B). The elution gradient was produced by a linear increase from 95% A to 95% B in 7 min and then back to initial conditions at 8 min. The flow rate of mobile phase was 0.3 mL min −1 . Injected volume of 1 to 5 µl samples are introduced through electrospray ionization source (ESI). Compressed nitrogen (99.999%, Messer Slovenia) was used as both the drying and the nebulising gas. The nebulizer gas flow rate was set to 20L h −1 and the de-solvation gas flow rate to 600L h −1 . A cone voltage of 20V and a capillary voltage of 2.5kV in negative ion mode (ESI-) for IBF and 3kV in positive ionization mode (ESI+) for IBF were used. The de-solvation temperature was set to 3000°C and the source temperature to 1200°C. Mass spectra were acquired in centroid mode over an m/z range of 50-1000 in scan time 0.2s and inter scan time 0.02s. For identification and characterization of organic compounds a mass resolution of 10000 was applied for accurate high-resolution mass measurements.
Phytotoxicity test using Lepidium sativum L Lepidium sativum L. was used to assess the acute toxicity of B41 and IBF before and after treatment. B41 and IBF model solutions were prepared and degraded as described in PEC section. In the experiment 15 L. sativum seeds were evenly placed on a filter paper in Petri dish (Ø90 mm) and exposed by adding 3 ml of treated and untreated B41 model solution (sample B41T and B41 before, respectively)), and treated and untreated IBF model solution (sample IBFT and IBF, respectively). Control (C) was also included by adding distilled water to the seeds. Incubation was performed at room temperature and after 72h seed germination and root growth were evaluated. The experiment was done in triplicate for each treatment. Relative S3 germination percentage (RGP), relative radicle growth (RRG) and germination index (GI) were calculated according to the methodology described [3].

Energy-dispersive X-ray spectroscopy (EDS)
Energy-dispersive X-ray spectroscopy (EDS) analysis was used as a rough estimate to determine the atomic ratios of O and Ti elements. In the pristine TiO 2 TNAs the atomic ratio is 1.6 and is higher than in the reduced TiO 2 TNAs which yielded a vlue of 1.3 ( Figure S2e, f).
In our opinion the measured atomic ratio does not reflect the true value since the Ti signal from the foil may also contribute to the total Ti signal.
The oxygen evolution reaction (OER) reaction (eq 1) is inherently slow because it is fourelectron transfer technique, at the same time as the chlorine evolution reaction (CER) (eq 2) includes two electrons and is a far faster reaction. The Cl 2 generated dissolves to HOCl or OCl  depending on the pH of solution. The higher current in NaCl electrolyte is attributed to both reactions: OER and CER. The pathway and kinetics of OER are typically investigated using semi-logarithmically plotted current-potential curve (eq 3), typically called a Tafel plot ( Figure   S8b).
where j is the current density, a is the intercept and b is a coefficient, Tafel slope. Within the case of OER, the values of b are smaller than 500 mV dec −1 and gives facts approximately the mechanism of O 2 evolution. The values of b for efficient water oxidation catalysts are inside the range of 30 -120 mV dec −1 [4]. When the values of b are high then these can be ascribed to various factors like (i) the observed reaction is restricted via mass transfer, (ii) surface diffusion or some chemical steps is the rate determining step, (iii) the accumulation of oxygen bubbles or reactive oxygen species which block the active surface of the electrode, etc [5]. The values of b in this work are 257 mV dec −1 (NaCl), 330 mV dec −1 (NH 4 Cl), and 353 mV dec −1 (Na 2 SO 4 ) and are much more efficient than the Tafel slopes reported in the literature [6].
However, these experimental values are lower than the data reported in the work of Zhu et al [4]. Worth mentioning that in NaCl and NH 4 Cl electrolytes the Tafel slope are not entirely from the OER reaction but involves competitive oxidation reaction of Cl  ions (CER). For efficient catalysts like RuO 2 the Tafel slopes for CER are usually under 100 mV dec −1 [7].
Herein, the Tafel slopes of TiO 2-x photoanode reported in Cl  electrolyte is ascribed both to OER and CER. Since the Tafel slopes in Cl  electrolyte is much lower it can be assumed that CER proceeds much efficiently than OER in the present system.   It is well-known that •OH radicals react with coumarin molecules to form highly fluorescent 7-hydroxycoumarin (7-HC). The PEC experiment was carried out without B41 dye using 30 µM coumarin in 0.01M NaCl. The concentration of 7-HC was recorded via a fluorescence spectrophotometer at an excitation wavelength of 332 nm. Calibration plot at different concentrations of 7-HC is presented in Figure S15a. The lack of emission peak from PEC treated solutions confirm the absence of 7-HC ( Figure S15b).