Thermal catalytic oxidation of octachloronaphthalene over anatase TiO2 nanomaterial and its hypothesized mechanism

As an environmentally-green technology, thermal catalytic oxidation of octachloronaphthalene (CN-75) over anatase TiO2 nanomaterials was investigated at 300 °C. A wide range of oxidation intermediates, which were investigated using various techniques, could be of three types: naphthalene-ring, single-benzene-ring, and completely ring-opened products. Reactive oxygen species on anatase TiO2 surface, such as O2−• and O2−, contributed to oxidative degradation. Based on these findings, a novel oxidation degradation mechanism was proposed. The reaction at (101) surface of anatase TiO2 was used as a model. The naphthalene-ring oxidative products with chloronaphthols and hydroxyl-pentachloronaphthalene-dione, could be formed via attacking the carbon of naphthalene ring at one or more positions by nucleophilic O2−. Lateral cleavage of the naphthalene ring at different C1-C10 and C4-C9, C1-C2 and C4-C9, C1-C2 or and C3-C4 bond positions by electrophilic O2−• could occur. This will lead to the formation of tetrachlorophenol, tetrachloro-benzoic acid, tetrachloro-phthalaldehyde, and tetrachloro-acrolein-benzoic acid, partially with further transformation into tetrachlorobenzene-dihydrodiol and tetrachloro-salicylic acid. Unexpectedly, the symmetric half section of CN-75 could be completely remained with generating the intricate oxidative intermediates characteristically containing tetrachlorobenzene structure. Complete cleavage of naphthalene ring could produce the ring-opened products, such as formic and acetic acids.

based on a risk management evaluation and consideration of the management options, the Committee recommended a Conference of the Parties to consider listing and specifying the relevant control measures for PCNs 8 . The reduction of PCN levels is therefore a matter of public concern in the context of environmental protection.
Catalytic oxidation for the removal of chlorinated aromatic hydrocarbons has attracted much attention as a green technique 9,10 . TiO 2 -based catalysts are generally used for the oxidation of chlorinated aromatic compounds [11][12][13][14] . Lichtenberger et al. 12 examined the oxidation of chlorobenzene, and 1,2-, 1,3-, and 1,4-dichlorobenzene over V 2 O 5 /TiO 2 catalysts. A common reaction mechanism was proposed based on kinetic and in situ fourier transform infrared (FTIR) results. Surface phenolates are formed via nucleophilic attack at the chlorine position in the aromatic ring, followed by electrophilic substitution of the adsorbed partially dechlorinated species in the second step. Krishnamoorthy et al. 11 investigated the catalytic oxidations of 1,2-dichlorobenzene over Cr 2 O 3 , V 2 O 5 , MoO 3 , Fe 2 O 3 , and Co 3 O 4 supported on TiO 2 and Al 2 O 3 . The TiO 2 -supported systems were more active than the corresponding Al 2 O 3 -supported ones, indicating that the support is significant in the catalytic performance of the catalyst in this reaction. Gannoun et al. 15 showed that sulfated TiO 2 nanotubes (HNTs) were a promising support for V 2 O 5 -based materials in the oxidative elimination of chlorobenzene. The formed bridged bidentate Ti and acidic sites on the HNT surface probably govern chlorobenzene oxidation and decrease the reducibility of vanadium, leading to higher reactivity at redox sites and therefore to higher-efficiency catalysts. Thus far, however, the reports to deeply identify the oxidation products and the associated mechanisms of PCNs as new POPs, are particularly scarce. TiO 2 is an important semiconductor material and has been used in a variety of applications such as photosplitting of water 16 , photovoltaic devices 17 , liquid solar cells, surface wettability conversion, and degradation of toxic pollutants 18 . This wide range of applications can be attributed to its nontoxicity, low cost, photostability, redox efficiency, and availability. TiO 2 has three crystal form, i.e., brookite, anatase, and rutile. The crystal form of TiO 2 has a decisive effect on its catalytic performance, because the electronic band gaps (EBGs) of the different forms of TiO 2 are different. It has been reported that the photocatalytic activity of anatase TiO 2 is limited by its small absorption range in the solar spectrum, as a result of its large EBG (Eg = 3.2 eV). However, the larger EBG of anatase TiO 2 has attracted great interest in its better oxidation performance. Therefore, it is of significance that the catalytic oxidation of PCNs is performed by anatase TiO 2 with illustrating the involved deep oxidation mechanism.
In this study, the reactivity of an anatase TiO 2 nanomaterial toward a model compound, i.e., octachloronaphthalene (CN-75), which is fully substituted with chlorine atoms, was evaluated at 300 °C. The degradation products, especially the oxidation products, were comprehensively investigated using gas chromatography-mass spectrometry (GC/MS) combined with silicane derivatization, high-performance liquid chromatography/hybrid quadrupole time-of-flight mass spectrometry (HPLC/Q-TOF-MS/MS), and ion chromatography (IC). Electron spin resonance (ESR) experiments, in combination with X-ray photoelectron spectroscopy (XPS) analysis of the TiO 2 , were used to study the role of reactive oxygen species in the degradation of CN-75. An oxidative degradation mechanism was proposed based on the findings. The results will be useful in developing methods for eliminating PCN-concentrated wastes.

Results
Kinetic study. The time-dependent degradation behavior of CN-75 (990.1 nmol) over anatase TiO 2 at 300 °C was investigated. The black squares in Fig. 1 represent changes in the amount of residual CN-75 with heating time at 300 °C, calculated based on quasi-exponential decay. The amount of CN-75 decreased from 990.1 to 78.28 nmol in 60 min. This suggests that nanosized anatase TiO 2 is an effective catalyst for CN-75 degradation. A linear ln(R CN-75 /I CN-75 ) versus time plot corresponding to pseudo-first-order reaction kinetics with an initial rate constant k obs (min −1 ) of 0.04 was obtained as shown in the inset in Fig. 1( I CN-75 is the initial number of moles of CN-75, and R CN-75 is the number of moles of the remained CN-75 following heating for a given time period). It can be seen from Fig. 1 that only a small amount of 1,2,3,4,5,6,7-heptachloronaphthalene (CN-73) was detected in the hydrodechlorination products from 5 to 60 min. In contrast, in the progress of CN-75 degradation over as-prepared Fe 3 O 4 with the similar dosage for the same reaction phases, a series of hydrodechlorination products from heptachloronaphthalenes to dichloronaphthalenes were detected 19 . The hydrodechlorination reaction of CN-75 was less favored on anatase TiO 2 than on Fe 3 O 4 . This may be because the stability of anatase TiO 2 is higher than that of Fe 3 O 4 , as shown by the higher EBG of anatase TiO 2 (3.2 eV) compared with that of Fe 3 O 4 (0.1 eV). Similarly, the weaker hydrodechlorination of decachlorobiphenyl was also found in its degradation over NiFe 2 O 4 with EBG at 2.19 eV than over Fe 3 O 4 20 . GC/MS analysis of oxidation products after derivatization. Competition between hydrodechlorination and oxidation reactions in the degradation of chlorinated benzenes over metal oxides has often been reported 9,12,20,21 . The reason is that lower chlorinated products and oxidation products, such as phenolate, acetate, and carbon monoxide species, have been detected simultaneously 9,22,23 . This may be explained by different types of active centers on catalysts. One of the reactions will be the main process, depending on the reaction conditions and reactants. A low level of hydrodechlorination suggests that oxidative degradation occurs preferentially. The oxidation intermediate products formed during catalytic degradation of CN-75 were studied to obtain a better understanding of the degradation pathway. Theurich et al. 24 reported that 15 different oxidation intermediates were identified during the photocatalytic degradation of naphthalene in aqueous suspensions of TiO 2 under UV irradiation. To evaluate the existence of oxidative intermediates during the reaction, the dosage of CN-75 increased from 990.1 nmol to 4,950.5 nmol. GC/MS is often used to identify unknown substances. However, the response of the oxidative degradation products often with high polarity was poor in GC/MS. Silylation is one of the derivatization procedures widely used to improve GC behavior of polar compounds containing phenolic and carboxylic groups. In this procedure, the active hydrogens could be replaced by trimethylsilyl groups, producing derivatives which are more volatile and thermally stable. Albero et al. 25 reported that phenolic and carboxylic compounds in soil, such as parabens, bisphenols and triclosan, were determinated by gas chromatography tandem mass spectrometry with in situ derivatization of N,O-bis(trimethylsilyl)trifluoroacetamide with 1% trimethylchlorosilane (BSTFA:TMCS = 99:1, v/v). Saitta et al. 26 also demonstrated 21 phenolic compounds in Italian and Turkish pistachio oil samples by means of the mass spectra of the BSTFA-TMCS derivatives. In present study, the reaction products were derivatized using BSTFA:TMCS (99:1) 27 , and then analyzed using GC/MS in EI full-scan mode. The main derivatization reactions are as follows: Figure 2 shows the GC/MS chromatograms of the chemically derivatized samples after CN-75 degradation over anatase TiO 2 at 300 °C for 5 min. Analysis of the derivatized products showed that tetrachlorophenol, tetrachlorobenzoic acid, tetrachloroacroleinbenzoic acid, tetrachlorophthalaldehyde, tetrachlorosalicylic acid, and hexachloronaphthols were produced. The list of corresponding oxidation products is given in Table 1. Full-scan MS analysis was performed to identify the structures of the detected oxidation derivatives. Qualitative analysis was performed based on the molecular ions, fragment ions, the ratio between 35 Cl and 37 Cl, and comparison with data in the NIST02 standard spectral database 28

HPLC/Q-TOF-MS/MS analysis of oxidation products. LC/MS is a sensitive analytical technique
that is widely used for the separation and quantification of highly polar products 32,33 . During the analytical process, polar oxidation products are efficiently ionized using the ionization techniques associated with LC/MS, enabling their identification 34 . This technique has been often applied together with GC/MS to comprehensively determine the polar species 35 . The oxidation process was therefore further investigated by monitoring the formation of oxidation intermediate products during the catalytic degradation of CN-75 over anatase TiO 2 using HPLC/Q-TOF-MS/MS. Figure 3a shows the HPLC results for the chemical species following reaction between CN-75 (4,950.5 nmol) and anatase TiO 2 (50 mg) at 300 °C for 5 min. Tetra-chlorophenols, tetrachlorobenzenedihydrodiol, hydroxypentachloronaphthalenedione (OH-PeCN-dione), hydroxypentachloronaphthalene (OH-PeCN), and hydroxyhexachloronaphthalene (OH-HxCN) were determined as degradation products (Table 2). However, the isomer patterns of the hydroxyl congeners could not be identified because of limitations associated with the external standards.  Analysis of oxidation products by IC. Literature  . In the current study, ring-cracked products were detected, using IC, in the reaction between CN-75 (990.1 nmol) and anatase TiO 2 (50 mg) at 300 °C. Formic and acetic acids were the main ring-cracked degradation products, as shown in Fig. 3b. The amount of acetic acid rapidly increased to a maximum of 140.6 nmol after heating for about 10 min, and then decreased with heating time. In contrast, the formic acid content increased steadily with heating time, with a maximum content of 90.4 nmol at 60 min. These oxidation products indicate that TiO 2 also facilitates the ring-cracking oxidation pathway of chlorinated aromatics.

Discussion
The presence of active oxygen species on nanosized anatase TiO 2 catalysts is believed to contribute to the occurrence of oxidation reactions during CN-75 degradation 19 . The O 1s XP spectrum of the anatase TiO 2 catalyst is shown in Fig. 4a. The peak at 530.97 eV (denoted by P1) is attributed to surface oxygen and adsorbed oxygen species, and the peak located at 529.15 eV (denoted by P2) is attributed to lattice oxygen 38 . Similar oxygen species were detected on the surface of Ca-doped FeO x hollow microspheres and CaCO 3 /α -Fe 2 O 3 composite catalysts 37 . A high proportion of surface oxygen on the metal oxide catalyst increases the activity in low-temperature oxidative degradation of 1,2-dichlorobenzene.  , and azo dyes over Ag/AgBr/TiO 43 , have been confirmed by ESR spectroscopy. ESR spectroscopy, with DMPO as the spin-trapping agent, was used to obtain information on the active radicals involved, to determine whether O 2 −• and •OH were available products in the decomposition of CN-75 over anatase TiO 2 . A reaction was performed between anatase TiO 2 (50 mg) and CN-75 (990.1 nmol) at 300 °C for 10 min. The reaction products were immediately dissolved in dimethyl sulfoxide (DMSO), and then characterized using an ESR analyzer, as shown in Fig. 4b. Four peaks were observed, and the hyperfine constants, i.e., α N = 12.7429 G, α H = 10.0304 G, and g = 2.0103, coincided with those previously reported for DMPO-O 2 −• (Fig. 4b-I) 9 . The results identify that the superoxide anion may be involved in CN-75 degradation, resulting in the formation of a series of oxidation products and perhaps even into formic acid and acetic acid. The DMPO-•OH species were examined under identical conditions, except water was used as the solvent instead of DMSO. No obvious signal was observed, as shown in Fig. 4b-II. This differs from the photocatalytic degradation of many organic molecules, in which •OH species are often identified [41][42][43] .
An oxidative degradation pathway (Fig. 5) is proposed, based on the available oxygen species and the detected oxidation intermediates. The (101) surface is the most stable and frequent surface of anatase TiO 2 , as shown in Fig. 4c, which was therefore selectively took as a model [44][45][46] . It has the same periodicity as the bulk truncated surface and exposes undercoordinated pentacoordinated Ti cations (Ti 5c ) and dicoordinated oxygen anions (O 2c ), and fully coordinated Ti 6c cations and tricoordinated oxygen anions (O 3c ) 47 . Coordination theory states that unsaturated ions are prone to bond with ligands 23 . It is therefore hypothesized that CN-75 molecules are adsorbed on the anatase TiO 2 surface via coordination interactions between Lewis acid Ti 5c cations and Lewis base Cl 21 . When CN-75 degraded on the surface of the anatase TiO 2 catalyst, firstly, dissociative adsorption of CN-75 on the central Ti 5c cations occurs, followed by the attack of carbon atom potential to accepting the electrons by reactive nucleophilic oxygen O 2− species. This results in C-Cl bond cleavage and subsequent Ti-Cl bond formation. Association of the free chloride ions with Lewis acid Ti ions occurs during CN-75 degradation over anatase TiO 2 . This is confirmed by the Cl 2p core-level XP spectrum of the catalyst after heating for 10 min (Fig. 4d). Three peaks (denoted by P1, P2, and P3) are observed. The peak at 197.8 eV corresponds to Cl bonded to Ti 4+ , with a net charge of − 1, indicating possible formation of TiCl 4 during degradation of CN-75 48 . In this Superoxide O 2 −• species are electrophilic. They have been reported to be formed by transformation of adsorbed O 2 molecules 19,49 . When a subsurface oxygen vacancy is present, it is energetically favorable for O 2 to adsorb at a Ti 5c site close to this defect. On adsorption, the extra charge associated with the defect is transferred to the O 2 molecule, converting it to a superoxide O 2 −• species. The strongly reactive electrophilic O 2 −• species can attack the π -electron cloud of the naphthalene ring, which has a highly dense electron population. This leads to the cracking of the naphthalene ring at different positions. The detection of tetrachlorophenol and the resultant tetrachlorobenzenedihydrodiol indicates that one of the rings in the naphthalene ring of CN-75 is first opened through C 1 -C 10 and C 4 -C 9 bond cleavage. Breakage of the C 1 -C 2 and C 4 -C 9 bonds in one ring could result in the formation of tetrachlorobenzoic acid, which is further oxidized to tetrachlorosalicylic acid. Similarly, the breakage of C 1 -C 2 or and C 3 -C 4 bonds could lead to the formation of tetrachloroacroleinbenzoic acid or and tetrachlorophthalaldehyde, respectively. These results show that lateral cleavage of one naphthalene ring at different C-C bond positions by electrophilic O 2 −• could occur, leading to formation of various single-benzene-ring oxidation products. Unexpectedly, the symmetric half section of CN-75 could be retained along with generation of complex oxidation products containing the tetrachlorobenzene structure.
It is important to note that the reaction pathways via electrophilic and nucleophilic attack by reactive oxygen species such as O 2− and O 2 −• are not independent of each other. The newly formed chlorinated naphthol species can also be attacked by reactive oxygen species such as O 2 −• . Moreover, oxidation products with both naphthalene and single benzene rings can be further attacked by reactive oxygen species, and completely cracked to small molecules such as formic and acetic acids. A wide range of oxidation products such as naphthols, phenols, hydroxy-diones, benzoic acids, acroleinbenzoic acid, phthalaldehyde, salicylic acid, dihydrodiols, and formic and acetic acids, with chlorinated naphthalene or benzene rings, or without aromatic rings, were detected during the degradation of CN-75 over anatase TiO 2 . This is different from the previously reported results for CN-75 degradation over Fe 3 O 4 micro/nanomaterials 19 , in which only chloronaphthol species, and formic acid and acetic acids were detected as the oxidation products under the same experimental conditions. This shows that oxidative degradation of CN-75 on anatase TiO 2 was more extensive that on Fe 3 O 4 micro/nanomaterials. Deep oxidative degradation of CN-75 on anatase TiO 2 occurs possibly because of the electronic structure with an EBG of 3.2 eV and the reactive oxygen species on its surface.

Degradation experiments.
Degradation experiments were performed in sealed glass ampoules.
Prior to the reaction, a hexane solution of CN-75 (990.1 or 4,950.5 nmol) was injected into an ampoule and subsequently evaporated to dryness at room temperature, and then mixed with later added 50 mg of anatase TiO 2 . The samples were heated at 300 °C for an appropriate time. A blank experiment was performed in the absence of TiO 2 under the same conditions. All experiments were performed in triplicate to ensure repeatability of the results.
Degradation product analysis. After the decomposition reaction, the ampoule was cooled to room temperature and crushed, and the sample was extracted. The unreacted CN-75 and newly formed PCNs were analyzed using an Agilent 6890 gas chromatograph equipped with a DB-5 MS capillary column (30 m × 0.25 mm i.d., 0.25 μ m film thickness) and an Agilent 5973 N mass selective detector. Helium (≥ 99.999%) at a flow rate of 1 mL/min was used as the carrier gas, and the injector was set at 260 °C. The column temperature was set at 75 °C for 2 min, gradually increased to 150 °C at 20 °C/min, then increased to 205 °C at 1.5 °C/min, and finally increased to 270 °C at 2.5 °C/min. The diluted sample (1.0 μ L) was injected in split-less mode. An electron ionization system with an ionization energy of 70 eV was used.
For oxidation product analysis, the reaction products obtained after CN-75 degradation over anatase TiO 2 at 300 °C for 5 min were extracted by the mixture solvent of hexane/ methanol/ ethyl acetate (1:1:1, v/v/v). The extract was dehydrated using a column packed with anhydrous sodium sulfate, and then evaporated under stream of nitrogen to dryness. Dry residue was dissolved in 0.2 mL of derivatizing reagent BSTFA:TMCS (99:1) and vortexed. The mixture reacted at room temperature for 60 min, and the derivatization products were analyzed using GC/MS. The column temperature was initially 50 °C, and increased to 180 °C (for 2 min) at 10 °C/min, to 210 °C at 1 °C/min, and to 280 °C at 10 °C/min. The carrier gas was helium at a flow rate of 1 mL/min.
The oxidation products were also analyzed using HPLC/Q-TOF-MS/MS (Micromass Q-TOF micro, Waters, USA). After the degradation of CN-75 (4,950.5 nmol) over anatase TiO 2 , product samples were extracted using HPLC-grade methanol, filtered through a 0.45 μ m mesh membrane, and concentrated to approximately 100 μ L. The oxidation products were detected using a Supelcosiltmlc-18 C18 column (Sigma; 4.6 mm × 250 mm; 5 μ m particle size). The elution flow rate was 0.5 mL/min with a gradient of 0.1% acetic acid in water-acetonitrile [acetonitrile concentrations 0% (isocratic, 5 min), 70% (isocratic, 5 min), 70-90% (linear, 5 min), 90-100% (linear, 5 min), 100% (isocratic, 5 min), and 0% (isocratic, 4 min)]. MS was performed using a Waters Micromass Quattro Premier XE (triple-quadrupole) detector, equipped with an electrospray ionization (EI) source (Micromass, USA). The mass analyzer was operated in negative ionization (EI − ) mode and the optimized parameters were source temperature 120 °C, desolvation temperature 200 °C, capillary voltage 2.50 kV, desolvation gas flow rate 600 L/h, and cone gas flow rate 50 L/h. The organic acid oxidation products such as acetic and formic acid were analyzed using IC. The degradation samples obtained from the reaction of CN-75 (990.1 nmol) and anatase TiO 2 (50 mg) were extracted three times with 15 ml deionized water for 10 min each time under ultrasonication. And then the combined extracts were filtered through a 0.45 μ m mesh membrane for IC measurements. The employed IC was a DIONEX AS 5000 instrument equipped with an AS-AP automated sampler. A Dionex AS11-HC guard column (50 × 4 mm i.d.) and a Dionex AS11-HC analytical column (250 × 4 mm i.d.) were used for the analyses. The analyses were performed at 30 °C with a potassium hydroxide eluent that was generated from a Dionex EG on line and run with a linear gradient at a flow rate of 1.0 mL min −1 .
The radical species formed during degradation were investigated using ESR spectroscopy (ESP 300 E electron paramagnetic resonance spectrometer, Bruker) with 5,5-dimethyl-1-pyrroline N-oxide (DMPO; Sigma Chemical Co.) as the spin-trapping agent. Typically, anatase TiO 2 (50 mg) and CN-75 (990.1 nmol) reacted at 300 °C for 10 min. A reaction using anatase TiO 2 but without CN-75 was also examined under the same conditions for comparison. The settings for the ESR spectrometer were center field, 3,485 G; sweep width, 100.0 G; microwave frequency, 9.8 GHz; and power, 10 mW.