Aggregation-induced emission luminogen: A new perspective in the photo-degradation of organic pollutants

Guangdong Provincial Key Laboratory of Functional Soft Condensed Matter, School of Material and Energy, Guangdong University of Technology, Guangzhou, China Department of Chemistry, the Hong Kong Branch of Chinese National Engineering Research Center for Tissue Restoration and Reconstruction and Institute for Advanced Study, The Hong Kong University of Science and Technology Clear Water Bay, Hong Kong, China Department of Chemistry and Biochemistry, University of Hull, Hull, UK Institute of Semiconductors Materials and Technology, South China Normal University, Guangzhou, China Beijing Enterprise Water Group Limited, Beijing, China


| INTRODUCTION
Increased energy consumption and environmental issues are becoming increasingly problematic as the global population grows and extensive industrialization continues at an ever-increasing pace. 1 Among them, water pollution is one of the most important environmental problems that mankind faces today. 2 To-date, much effort has been devoted to exploring more efficient technologies for both the prevention and control of water pollution, including the use of chemical and physical methods. Photocatalytic methods are one of the cutting-edge techniques available for water treatment because of mild operating conditions, high efficiency, and relatively low cost. 3 Currently, numerous photocatalytically active materials including inorganic (TiO 2 , ZnO, CdS, etc.), 4 metal-free (g-C 3 N 4 ), 5 organometallic complexes, 6 and metal-nitrides (GaN/InGaN) 7 as well as metal-organic frameworks (MOFs) 8 have been extensively studied for pollutant decomposition due to their good stability, high catalytic activity, and favorable environmental characteristics. In the case of photocatalytically active materials, they can capture photons with suitable light energy, generate electrons and holes in the conduction and valence bands, respectively, which can react with water and oxygen to produce reactive oxygen species (ROS, such as hydroxyl radicals, superoxide, hydrogen peroxide and singlet oxygen), and further degrade organic dyes or produce hydrogen. 9 The ability to generate the ROS is a key point in the use of photoactive materials. 10 Compared to the traditional photoactive materials mentioned above, organic semiconductor materials have attracted considerable attention as photocatalysts, due to their low costs and tunable energy gaps that are responsive to a wide-range of visible light, leading to highly efficient utilization of solar energy. 11 However, examples of photocatalysts are rare due to the aggregation-caused quenching (ACQ) effect, which leads to low efficiency of ROS generation. 12 On the other hand, the lack of catalytic reaction sites in many organic semiconductor materials lowers the photocatalytic efficiency. The heterogeneous structures of alkalinized-C 3 N 4 , TiO 2 -based hybrid materials 13 can accelerate catalytic reactions by improving both the light harvesting ability and ROS generation efficiency via photo-generated electrons and holes.
In 2001, Tang's group developed a series of novel luminogens, defined as "aggregation-induced emission (AIE)," 12a-c which exhibited an abnormal photo-physical phenomenon, viz weak or non-emissive in solution, but emitting bright light with high quantum yield (up to 100%) in the aggregate state. These high-performance AIEgens with their unique optical properties can be used for organic light-emitting diodes, fluorescence probes, as well as in cell imaging etc. 12 On the other hand, AIEgens with appropriate molecular structures in the aggregated state are favorable for the formation of highly ROS for photodynamic therapy (PDT). 14 Thus, if nascent ROS are involved in the photocatalytic reaction of organic pollutants in water, then the AIEgen would be regarded as an excellent candidate for potential clean environmental applications. Previously, Tang with coworker have demonstrated that a tailored cyanostilbene-based molecule with the AIE feature exhibited a controllable regio-and stereo-selective photodimerization or photocyclization reaction under UV irradiation with color-tuneable properties in aqueous medium via a radical process. 15 On the other hand, Zhao et al. reported that cyanopyridiniumbased cationic salts exhibited an enhanced emission with remarkable structure-dependent singlet-oxygen generation ability under white light-illumination for photodynamic anticancer and antibacterial therapy. 16 Both examples indicated that cyanostilbene derivatives possess a typical AIE characteristic with ideal photo-sensitizers for ROS generation. 17 To explore the suitability for employing AIEgens in environmental applications, this article presents a new application of AIEgens in the photodegradation of organic pollution in aqueous solution. As an organic photocatalyst, photoactive AIEgen 17c can play a significant role in the photodegradation of Rhodamine with high efficiency (98.7%); the whole process has been investigated by UV-vis spectroscopy, photoluminescence (PL), photo-electrochemistry, and ultra-performance liquid chromatography/high-definition mass spectrometry (UPLC/ HDMS). More importantly, thermal imaging, electron paramagnetic resonance (EPR) spin-trap, X-ray photoelectron spectroscopic (XPS) techniques, and transmission electron microscopy (TEM) have been used to monitor the entire photocatalytic process and to probe a possible mechanism.

| X-ray crystallography
Crystallographic data for the compound was collected on a Bruker APEX 2 CCD diffractometer with graphite monochromated Mo Kα radiation (λ = 0.71073 Å) in the ω scan mode. The structure was solved by charge flipping or direct methods algorithms and refined by full-matrix least-squares methods on F 2 . All esds (except the esd in the dihedral angle between two l.s. planes) were estimated using the full covariance matrix. The cell esds were considered individually in the estimation of esds in distances, angles, and torsion angles. Correlations between esds in cell parameters were only used when they were defined by crystal symmetry. An approximate (isotropic) treatment of cell esds was used for estimating esds involving l.s. planes. The final cell constants were determined through global refinement of the xyz centroids of the reflections harvested from the entire data set. Structure solution and refinement were carried out using the SHELXTL-PLUS software package. Data (excluding structure factors) on the structures reported here had been deposited with the Cambridge Crystallographic Data Centre with deposition numbers. CCDC 1914539 contain the supplementary crystallographic data for this paper. These data could be obtained free of charge from The Cambridge Crystallographic Data Centre via www. ccdc.cam.ac.uk/data_request/cif.

| X-ray single-crystal diffraction analysis
A single crystal of TPE-Ca (CCDC:1914539) was grown from a mixture of CH 2 Cl 2 and hexane (1:1) at room temperature. The single-crystal X-ray structure of TPE-Ca adopts the monoclinic system with space group P 1 2 1 / n 1. The key parameters are listed in Table 1. The molecule ( Figure 1) adopts a twisted conformation with a large dihedral angle. The torsion angle between the terminal unit of the carbazole group and the phenyl ring is 70.61 . The molecular packing of TPE-Ca can best be described as off-set head-to-tail stacking. Each fragment of the molecule (such as the TPE and carbazole units) are linked to each other by several weak C-HÁ Á Áπ interactions with distances in the range 2.28 to 2.80 Å. Generally, the non-planar conformation of TPE-Ca not only contributes to free intramolecular rotation in solution, but also to the formation of a close-packing matrix, which can inhibit intramolecular rotation and result in enhanced fluorescence efficiency. Indeed, the TPE-Ca molecule exhibits clear AIE characteristics.

| Photophysical properties
The PL spectra of TPE-Ca was measured in THF and in THF with different water fractions (f w ) ( Figure 2). The TPE-Ca exhibited weak emission with a maximum emission peak at 505 nm in THF solution (Φ F = 0.01). On gradual addition of water to the THF from 0 to f w = 99%, the emission intensity of TPE-Ca enhanced ca. 100-fold with limited observed red-shifts (<10 nm) compared to the use of pure THF solution ( Figure 2B) (Φ F = 0.36). Meanwhile, upon photoexcitation, TPE-Ca in the solid state displayed bright green emission at 502 nm with a narrow full width at half maxima (FWHM). The fluorescence quantum yield (Φ solid = 0.6) is higher than that both in solution (0.01) and in aggregated state (0.36). On the other hand, the radiative decay rate (k r ) of TPE-Ca increased from 2 × 10 −9 s −1 in solution to 3.5 × 10 −8 s −1 in the solid state, while the non-radiative decay rate constant (k nr ) decreased from 19.8 × 10 −8 s −1 to 2.4 × 10 −8 s −1 , indicating that the molecular rotation was restricted by several   weak C HÁ Á Áπ interactions, thereby blocking the nonradiative relaxation process 18 (Table 1).

| Radical detection
According to previous reports, many examples have illustrated that AIEgens can be utilized to monitor PDT, cell apoptosis, as well as the imaging of cell organelle by ROS generation under light irradiation, and these processes are closely reliant on the active radical. 19 Liu et al. 20 reported the relationship between the value of the lowest singlet excited state (S1), the lowest triplet excited state (T1) (ΔE S1-T1 , ΔE ST ), and the efficiency to generate ROS by intersystem crossing (ISC); lower ΔE ST values for AIEgens benefit the release of ROS. Thus, in the case of TPE-Ca, the optimized molecular geometry and the electron-density distribution of HOMO and LUMO were calculated using Gaussian 09 software at the B3LYP/6-31G (d, g) basis level (Table S2). 21 The calculated TD-DFT value for ΔE ST is 0.26 eV, indicating that the TPE-Ca would be highly efficient for ROS generation and for the potential photodegradation of organic pollutants.
To investigate the formation of ROS from TPE-Ca, the generation of •OH species was further confirmed using the fluorescent probe, 4-hydroxyphenyl fluorescein (HPF), 22 which showed no emission, but fluoresced strongly after reacting with the •OH radical. As shown in Figure 3A,B, upon illumination of the mixture of TPE-Ca and HPF in water, the fluorescence was greatly enhanced with a positive linear relationship between the irradiation time and fluorescence intensity. Both results were consistent with the origin of the •OH signal being water in the presence of the TPE-Ca under irradiation.
More importantly, EPR spin-trap tests were performed for further confirming the •OH radical in the solid state and on a suspension. As depicted in Figure 3C, the EPR spectra of the solid sample TPE-Ca exhibited a strong EPR single with a g value of 2.002, which corresponds to the value for the free electron. 23 Irradiating the TPE-Ca over 5 minutes using DMPO as the Spin-trapping reagent, a clear EPR •OH signal with four wellresolved lines was detected centered at g = 2.006 in toluene, while two EPR signals were observed with g values of 2.008 and 2.015, respectively in water, due to the free electrons of TPE-Ca and •OH species. Thus, we assume that the origin of the •OH signal is from the H 2 O generated in the presence of a radical of TPE-Ca, which would be further employed for the photocatalytic photodegradation of organic dyes.
Furthermore, to test the separation efficiency of the photo-generated electron-hole pairs, the current density vs time (I-t) characteristic was measured under a repeated ON/OFF AM 1.5 illumination ( Figure 3D). The TPE-Ca exhibited a swift photo response with stable photocurrent density under visible light irradiation. In the absence of light irradiation (OFF), the photocurrent density rapidly dropped to almost zero, and the photocurrent reverted when the light was again ON. In addition, the current density decreased to ca. 10% under light irradiation after 20 cycles, which indicated that the TPE-Ca exhibited a reproducible photo-current with good photo-stability.

| Photocatalytic activity
To evaluate the photocatalytic activity of TPE-Ca, the photo-degradation reaction was performed under visible light irradiation (simulated sunlight irradiation) in aqueous solution using rhodamine B (RhB) as the model organic pollutant. The adsorption equilibria of the mixture of TPE-Ca (5 mg) with RhB (10 mg/L) were obtained by stirring in the dark, and subsequently irradiating with simulated solar light. The photo-degradation behavior was then demonstrated by UV-vis spectra.
Generally, the RhB aqueous solutions remain stable under visible light irradiation over 180 minutes. 24 According to Figure 4A, the UV-vis curves indicate that the RhB was almost completely degraded with the extreme degradation ratio of 98.7% within 120 minutes in the presence of TPE-Ca, and the photodegradation efficiency of TPE-Ca was higher than that with common inorganic photocatalysts, such as C 3 N 4 , 5e TiO 2 , 25 as well as some heterogeneous systems. 26 When extending the irradiation time to 150 minutes, we observed that the suspension color changed from transparent to yellow-green, which may result from the formation of photocatalytic degradation products that accelerate the solubility of TPE-Ca in aqueous solution. Clearly, the TPE-Ca can play a significant role in the photo-degradation of RhB.
Meanwhile, thermal imaging technology was used to follow the entire photo-induced degradation process (Figure 4B). Interestingly, the change in temperature of the mixture TPE-Ca@RhB in water (S1) was remarkably different when compared to the TPE-Ca in water suspension (S2) and in THF (S3) solution when irradiated with sunlight. As Figure 4B shows, the temperature of TPE-Ca in water (S2) increased from 23.6 C to 29.1 C under light The high-resolution XPS spectrum of C1s, N1s and O1s peaks irradiation for 150 minutes. More importantly, the temperature of S1 is higher than S2, which in turn is higher than S3. In addition, the final products of the photocatalytic degradation of RhB were examined by UPLC/HDMS. The UPLC chromatograms retention time at 3.45 minutes was observed, which corresponds to the intermediates A-E ( Figures S8-S13). The HDMS results indicate that the main intermediates consist of N-de-ethylated derivatives (A, B, and C), benzoic acid (D), and phthalic acid (E). Based on the UPLC/HDMS results, we propose a possible degradation pathway for RhB as listed in Figure 5. Firstly, after irradiation of TPE-Ca in water, RhB was degraded to radicals A, B, and C. These unstable intermediates subsequently degrade to the more stable, small molecules (D and E) under visible light irradiation. 27 The HLF and COS7 cells were incubated with final photocatalytic degradation product mixtures 1 (irradiation 120 minutes) and 2 (irradiation 150 minutes) in the medium for 1 hour at 37 C, and then were washed three times with phosphate-buffered saline solution (PBS, pH = 7.4) before imaging under a confocal fluorescence microscope. In Figures S15-S18, the confocal images of HLF and COS7 cells exhibit normal morphology with good health, and both final photo-degradation products 1 and 2 are found to rarely enter the cells. In addition, the cell viability results showed that both cells remain close to 100% (>90%) over 24 hours at a concentration of 0.6-3.0 g/mL, indicating the low cytotoxicity of the degradation compounds. Consistent with previous reports that Rhodamine B has moderate toxicity against normal cells and housefly, [28][29][30] our control experiments also show similar concentration-dependent toxicity ( Figure S18A). [28][29][30] More importantly, after irradiation, the yellow residue was filtered and collected, then characterized by NMR spectroscopy. We observed the proton chemical shift is F I G U R E 6 A, XPS spectrum of TPE-Ca, B, Carbon 1 second, C, Nitrogen 1 second and D, oxygen 1 second XPS spectra of TPE-Ca almost the same as that in the original TPE-Ca, indicating that the photocatalyst has great stability ( Figure S19).

| Photocatalytic mechanism
3.5.1 | XPS spectrum XPS was preformed to analyze the electronic and chemical environment of the radicals and revealed the detailed elements that TPE-Ca is comprised of, viz C1s (283.88 eV), N1s (398.7 eV), and O1s (540 eV) the Auger lines carbon KL1 (1219 eV) and nitrogen KL2 (1139 eV) ( Figures S20 and S21). As Figure 6 shows, the high-resolution XPS spectra of the C1s peaks can be divided into six peaks with binding energies at 283. 35, 283.9, 284.95, 285.95, 288, 290.77 eV, which correspond to the C-C, C-C, C-N, C-N, C-O, and π-plasmon, respectively. 31 The N1s peak at 398. 15, 399.11, 401.71, and 405.65 (Table 2). 22,31 The above analysis indicated that the radical is intact, which is fully consistent with the EPR results in the solid state.

| Possible photocatalytic mechanism
In the absence of UV-irradiation, the mixture of TPE-Ca with RhB in water was stirred overnight and was then submitted for TEM measurements. As shown in Figure S24, the TEM images show that TPE-Ca exists as rounded nanoparticles with diameter of 7 to 10 nm, which disperse well in solution. More importantly, the high-order selected-area electron diffraction pattern suggests the nanoparticles of TPE-Ca still retain high quality crystallization.
Generally, AIEgens emit strong fluorescene in the aggregation state which is attributed to restricted intramolecular rotation (RIR) and restriction of intramolecular vibration (RIV), whereas molecular become more rigidity and block the non-radiative decay channels. 12 The emission intensity enhances in proportion to the morphologies and size of the nanoparticles, thus for more molecules aggregating, more nanoparticle homogeneity there is brighter emissive. 33 In this case, the small size of the AIEgens nanoparticle might annihilate the excitons through non-radiative decay pathways and produce radicals.
Thus, the observations from the TEM, EPR, and XPS results illustrated the possible photocatalytic process: (a) An electron is ejected from TPE-Ca under irradiation resulting in radical TPE-Ca + , (b) the radical can react with water to generate a hydroxyl radical in an exothermic process by cleavage of the oxygen hydrogen bond, and (c) the hydroxyl radical plays a significant role in degrading the RhB to achieve the final products ( Figure 7). On the other hand, for S1, under irradiation, an exothermic reaction occurs when the radical of TPE-Ca reacts with H 2 O to produce •OH. Moreover, the •OH species involved in the oxidation reaction can generate enormous amounts of thermal energy, and S3 exhibited a relatively high temperature in comparison to S2, during the oxidation reactions indicating that the photo-degradation was rapid. 34,35 4 | CONCLUSION In summary, we have explored an efficient photo-degradation method for organic pollutants (RhB) by using a tetraphenylethene-based AIEgen (TPE-Ca) as a photocatalyst. UV-vis spectroscopic monitoring of the TPE-Ca system revealed enhanced photocatalytic ability toward photo-degradation of RhB which occurred with high efficiency (98.7%) under white light irradiation over F I G U R E 7 The possible photocatalytic process of TPE-Ca 120 minutes. Importantly, both chemical and physical interactions generate enormous amounts of thermal energy and enhance the local environmental temperature. The identities of the final degradation products were confirmed by UPLC/HDMS, and these products displayed weak cytotoxicity toward HLF and COS7 cell lines. A photocatalytic mechanism was proposed on the basis of EPR and XPS results, which indicated that the TPE-Ca displayed a free electron signal, which would induce •OH for further photocatalytic processes. We report here the first example of an AIEgen as a high-performance photo-catalyst for environmental applications, and the use of thermal imaging technology for monitoring the photo-induced degradation process. This research opens up new avenues for the application of AIE research.