Yonggang
Xiang
a,
Xuepeng
Wang
a,
Xiaohu
Zhang
a,
Huijie
Hou
a,
Ke
Dai
b,
Qiaoyun
Huang
b and
Hao
Chen
*a
aCollege of Science, Huazhong Agricultural University, Wuhan 430070, P. R. China. E-mail: hchenhao@mail.hzau.edu.cn
bState Key Laboratory of Agricultural Microbiology, Huazhong Agricultural University, Wuhan 430070, China
First published on 23rd November 2017
The search for high-efficient TiO2-based heterojunction photocatalyst for photocatalytic H2 evolution and pollutant removal remains a great challenge. In this study, linear conjugated polymer B-BT-1,4-E consisting of alternating electron donor and electron acceptor was incorporated on the surface of TiO2 to form a binary composite in the facile in situ strategy. B-BT-1,4-E/TiO2 exhibited superior photocatalytic activity under visible light irradiation (λ ≥ 420 nm). The hydrogen evolution rate (HER) of 16.7% B-BT-1,4-E/TiO2 reached 220.4 μmol h−1 without additional noble metal (Pt), and the optimized 13.3% B-BT-1,4-E/TiO2 also displayed outstanding CIP degradation efficiency with a kinetic constant of 0.232. Based on extensive characterization results from UV-Vis diffused reflectance spectra (DRS), electron spin resonance (ESR), photocurrent, electrochemical impedance spectroscopy (EIS) and photoluminescence (PL), the enhanced photocatalytic activity of B-BT-1,4-E/TiO2 heterojunction can be attributed to broader visible light absorption range, faster charge separation and transfer. Moreover, a reasonable photocatalytic mechanism is also proposed. In comparison to reported CMP(BBT)/TiO2, HER and kinetic constants of CIP degradation with B-BT-1,4-E/TiO2 as the photocatalyst were increased by 7.3 times and 3.1 times, respectively. This study demonstrated that intrinsic merits of conjugated polymers made them promising candidates for exploring more efficient organic semiconductor–inorganic semiconductor heterojunction photocatalysts.
Besides the state-of-the-art graphitic carbon nitride (g-C3N4) polymer,6,7 most recently, conjugated polymers consisting of alternating electron donors (D) and electron acceptors (A) are emerging as potential photocatalysts; also HOMO and LUMO positions of the delocalized π-system in the conjugated polymers can be fine-tuned by facile molecular backbone adjustment (linear to three-dimensional network) or by varying donor/acceptor moieties of repeating sub-entities.8–11 Cooper et al. synthesized a series of pyrene-based conjugated microporous polymers (CMP1-15) by introducing different comonomers; the bandgap of CMP varied from 2.95 to 1.94 eV, indicating the strong visible light absorption ability, and the CP-CMP10 exhibited an average hydrogen evolution rate (HER) of 17.4 μmol h−1.12 Since then, structures of CMP have been optimized to improve the HER under visible light irradiation.13–16 Moreover, CMP also exhibited promising overall water splitting.17 With the same strategy, several linear or planarized conjugated polymers were also obtained, displaying superior photocatalytic activities than their CMP counterparts despite intrinsically high Brunauer–Emmett–Teller (BET) surface areas.18–22 However, only a few conjugated polymers have been used for photodegradation of organic pollutants. The Remita et al. recently fabricated one-dimensional poly(diphenylbutadiyne) (PDPB) nanostructures by photopolymerization.23 In comparison to traditional conjugated polymers such as PPy, PANI, and PTh, PDPB exhibited superior photocatalytic removal ability of methyl orange and phenol with high stability.
Based on the strategy of constructing semiconductor–semiconductor heterojunction, polymer–TiO2 composites have been designed with enhanced photocatalytic ability for H2 evolution and pollutant removal.24–30 However, most of the polymers were focused on the general PANI and P3HT, and the photocatalytic performance was lower than some inorganic semiconductor–TiO2. The merits of D–A conjugated polymers inspired us to fabricate D–A conjugated polymer–TiO2 composites. We recently designed a CMP(BBT)/TiO2 heterojunction through in situ polycondensation of 4,7-dibromobenzo[c][1,2,5]thiadiazole and 1,3,5-triethynylbenzene in the presence of commercial TiO2 (P25).31 Under visible light irradiation (λ ≥ 420 nm), 6.7 wt% BBT/TiO2/Pt exhibited HER at 178 μmol h−1 with a Pt content of 0.25%, and HER reduced to ca. 30 μmol h−1 if Pt was not doped. In addition, the kinetic rate constant of 13.3% BBT/TiO2 is 0.076 for photodegradation of CIP, and the photocatalytic performance was reduced slightly after three repeated cycles.
To further enhance the photocatalytic performance of D–A conjugated polymer–TiO2 composites, in the present study, linear conjugated poly(benzothiadiazole) (B-BT-1,4-E) (Scheme 1) was selected as the compatible conjugated polymer to afford the desired B-BT-1,4-E/TiO2 heterojunction in a similar in situ polycondensation procedure. In comparison to CMP(BBT)/TiO2, B-BT-1,4-E/TiO2 exhibited more efficient charge mobility and broader visible light absorption. HER of 16.7% B-BT-1,4/TiO2 was 220.4 μmol h−1 without addition of Pt, which is about 7.3 times higher than that of the CMP-TiO2 counterpart, and the kinetic rate constant of 13.3% B-BT-1,4/TiO2 was improved to 0.232 with high stability. To gain insight into the mechanism of superior photocatalytic activity, photochemical and other characterizations have been utilized.
For ciprofloxacin photodegradation, 10 mg of the photocatalyst was dispersed in an aqueous solution of ciprofloxacin (CIP, 50 mL, 20 ppm). Prior to irradiation, the suspension was continuously stirred in dark for 1 h to establish adsorption–desorption equilibrium. During the illumination, 5 mL aliquots from each sample were taken at an interval of 5 min for 20 min, followed by filtration to remove the photocatalyst. The concentration of CIP was determined by HPLC with a UV detector at 275 nm. In the recycling experiments, the photocatalyst was collected by centrifugation, washed with methanol and dried after each photocatalytic experiment.
Fig. 1 (a) XRD patterns of TiO2, 13.3% B-BT-1,4-E/TiO2 and pure B-B-T-1,4-E. (b) HRTEM of 10.0% B-BT-1,4-E/TiO2. |
The successful synthesis of B-BT-1,4-E/TiO2 heterojunction was first confirmed by solid-state 13C CP/MAS NMR (Fig. 2a), and the corresponding peaks of pure B-BT-1,4-E were also detected in the 13C CP/MAS NMR spectra of 13.3% B-BT-1,4-E/TiO2, suggesting that B-BT-1,4-E was introduced into the composite. Fig. 2b shows Raman spectra of the as-prepared B-BT-1,4-E/TiO2 heterojunction, pure B-BT-1,4-E and pure TiO2. For pure conjugated polymer, band wavenumbers appeared at 1362 cm−1, 1537 cm−1 and 2204 cm−1, which were identified as ring stretch, CC stretch, and CC stretch, respectively. P25 exhibits characteristic peaks at 146 cm−1 (Eg(1)) 401 cm−1 (B1g), 520 cm−1 (A1g) and 642 cm−1 (Eg(2)). In comparison with pure P25, the corresponding P25 Raman peak intensity of the heterojunction became weaker due to coating of B-BT-1,4-E on the surface. However, the Raman peak signals of B-BT-1,4-E in the heterojunction showed stronger intensity without any visible shifts, supporting that more ordered and longer conjugated segments formed in the B-BT-1,4-E/TiO2 composites (Fig. S1†).
Fig. 2 (a) Solid-state 13C-CP/MAS NMR of 13.3% B-BT-1,4-E/TiO2 and pure B-BT-1,4-E. (b) Raman spectra of TiO2, 13.3% B-BT-1,4-E/TiO2 and pure B-BT-1,4-E. |
UV-Vis diffuse reflectance spectra (DRS) of the as-prepared 10.0% B-BT-1,4-E/TiO2, pure B-BT-1,4-E and pure TiO2 were also measured. As shown in Fig. 3, the optical onsets of B-BT-1,4-E/TiO2 heterojunction extended to around 800 nm in the visible light region. In comparison to pure B-BT-1,4-E, the composite displayed similar absorption range and intensity, and it can be clearly seen that B-BT-1,4-E acted as the visible light sensitizer, indicating that the B-BT-1,4-E/TiO2 could possess visible light photocatalytic activity. The BET surface area of the composite was then investigated (Fig. S2†), and the surface area of the as-prepared 10% B-BT-1,4-E/TiO2 was 61 m2/g−1, which is close to that of pure TiO2, while the BET surface area of pure B-BT-1,4-E was 17 m2 g−1.
XPS measurement was applied to gain insight into chemical states and composition of the heterojunction. As shown in Fig. 4a, the surface chemical composition of the as-prepared 13.3% B-BT-1,4-E/TiO2 was first analyzed by the XPS survey spectrum; the chemical compositions of Ti, O, C, S, and N elements in the samples were identified with the binding energy at 458, 531, 285, 167 and 400 eV, respectively. Fig. S3(a)† shows the high-resolution Ti 2p spectra of the composite in comparison with pure TiO2, and it indicated that both the samples had two peaks located at 458.4 eV and 464.2 eV, which ascribed to Ti 2p3/2 and Ti 2p1/2, respectively. Next, the high-resolution O 1s XPS spectra of TiO2 were detected at 530.88 eV and 529.41 eV, assigning to OH, and Ti–O–Ti on the surface of TiO2, respectively, while O 1s XPS spectra of the heterojunction displayed peaks at 530.51 eV and 529.41 eV, and it indicated that there exists a slight energy shift for OH (Fig. 4b). Moreover, the peaks in the C 1s spectrum at 284.8 eV, 285.7 eV and 289.0 eV were detected, and the signal around 284.8 eV was assigned to C–C bonds, while higher binding energy (285.7 eV) was attributed to CN bonds. The weaker signal at 289.0 eV arises from CO bonds and no binding energy change was observed after the heterojunction was formed (Fig. S3(b)†).32
Fig. 4 XPS spectra of TiO2, 13.3% B-BT-1,4-E/TiO2 and B-BT-1,4-E. (a) Survey spectra. (b) O 1s. (c) N 1s. (d) S 2p. |
Next, we investigated the nitrogen atoms in pure B-BT-1,4-E polymer and the composite. N 1s peaks of the polymers appeared at 399.1 eV in comparison to the corresponding 398.9 eV of 10% B-BT-1,4-E/TiO2− (Fig. 4c). As shown in Fig. 4d, the pure B-BT-1,4-E exhibits typical S 2p peaks that are deconvoluted into S 2p3/2 and S 2p1/2 with binding energies of 165.0 V and 166.2 eV, respectively, which were consistent with the reported S 2p XPS 33; however, S 2p peaks of the heterojunction appeared at 164.7 and 166.8 eV, respectively. Accordingly, S 2p and N 1s XPS shifts for the polymer after heterojunction formation indicated that there could exist an interfacial interaction between B-BT-1,4-E and TiO2, which would facilitate the charge transfer.
Fig. 5 (a) Compared photocatalytic HER of TiO2, x% B-BT-1,4-E/TiO2 and B-BT-1,4-E under visible light irradiation. (b) Long-term stability of 16.7% B-BT-1,4-E/TiO2. |
In terms of the practical application, the stability of the as-prepared photocatalyst is also a very important factor along with extraordinary photocatalytic activity. Accordingly, the recycling experiments were conducted, and stability was evaluated. As shown in Fig. 6d, there was no substantial decrease observed in terms of the photocatalytic activity after four repeated photodegradation cycles, revealing high stability of B-BT-1,4-E/TiO2. The kinetic constant and stability of B-BT-1,4-E/TiO2 for photocatalytic CIP degradation were compared with the reported CMP(BBT)/TiO2,31 and it demonstrated that the degradation rate was increased by about 3.1 times along with superior stability.
To identify the types of active species involved in the photocatalytic degradation process of CIP, trapping experiments of the as-prepared 10.0% B-BT-1,4-E/TiO2 were carried out. The procedure was similar to that of the photocatalytic degradation measurement except the addition of various scavengers. In detail, isopropanol (IPA), K2Cr2O7, superoxide dismutase (SOD), triethanolamine (TEOA) or β-carotene was added to the CIP aqueous solution to capture hydroxyl radicals (˙OH), electron (e−), superoxide radical anions (˙O2−), hole (h+) or singlet oxygen (1O2), respectively. As shown in Fig. S6,† the photocatalytic ability of the heterojunction was almost not affected in the presence of IPA, K2Cr2O7 or β-carotene, indicating that the active species ˙OH, e−, and 1O2 did not contribute to the oxidation of CIP. When superoxide dismutase (SOD) was added, the photocatalytic rate slightly reduced. However, the photodegradation of the organic pollutant was completely inhibited if the scavenger TEOA was added. Accordingly, superoxide radical anions acted as the main active species in the process of the CIP degradation.
ESR spin-trap with DMPO technique was subsequently used to investigate active species ˙O2− and ˙OH generated over pure B-BT-1,4-E and 13.3% B-BT-1,4-E/TiO2 under visible light. As shown in Fig. 7, signals of both ˙OH and ˙O2− for 13.3% B-BT-1,4-E/TiO2 were stronger than those of B-BT-1,4-E. Although ˙OH and ˙O2− were not the main active species of photocatalytic H2 or CIP degradation, ˙O2− and ˙OH were generally considered as the product of O2 reduction by e− and water oxidation by h+. Accordingly, an increase in both ˙O2− and ˙OH radical generation can also explain the superior photocatalytic activity of 13.3% B-BT-1,4-E/TiO2 compared with B-BT-1,4-E.
Photocurrent and electrochemical impedance spectroscopy analyses were used to investigate the interface charge transfer and charge separation efficiency of photogenerated e−/h+. Fig. 8a illustrates the steady and reversible photocurrent responses under four on/off visible light irradiation cycles. It can be observed that photocurrent intensity of 13.3% B-BT-1,4-E/TiO2 was almost 60 times higher than that of B-BT-1,4-E, indicating faster charge transfer as a result of a synergetic effect between B-BT-1,4-E and TiO2. Under the same conditions, the photocurrent intensity of 13.3% CMP(BBT)/TiO2 was much weaker than 13.3% B-BT-1,4-E/TiO2. Moreover, 13.3% B-BT-1,4-E/TiO2 heterojunction showed much smaller radius in the EIS Nyquist plot than pure B-BT-1,4-E and pure TiO2, indicating that charge transfer was slightly inhibited. In addition, the PL intensity of 13.3% B-BT-1,4-E/TiO2 at ca. 420 nm was much lower than that of pure TiO2, suggesting a faster charge separation (Fig. S7†). Together with the photocurrents and EIS measurements, it was inferred that strong interfacial interaction between B-BT-1,4-E and TiO2 in the heterojunction accelerated the electron–hole separation, which was beneficial for the enhancement of photocatalytic activities of the as-prepared composites.
Fig. 8 Photocurrent intensity (a) and EIS spectra (b) comparison of TiO2, 13.3% B-BT-1,4-E/TiO2 and B-BT-1,4-E. |
The bandgap of pure B-BT-1,4-E was determined to be 1.64 eV by DRS, and the corresponding LUMO potential was determined to be −0.69 eV vs. NHE measured via the cyclic voltammetry method. According to the bandgap and LUMO potential, the HOMO potential was calculated to be 0.95 eV. In comparison to the corresponding VB and CB of commercial TiO2 at 2.7 eV and −0.5 eV, respectively, B-BT-1,4-E had a more negative LUMO value and less positive HOMO value. Based on the above results and discussions, an interfacial electron transfer heterojunction mechanism for photocatalytic H2 evolution and CIP degradation of the as-prepared B-BT-1,4-E/TiO2 is proposed (Scheme 2). Under visible light irradiation, photogenerated electron–hole pairs were only generated for the conjugated polymer B-BT-1,4-E since TiO2 cannot absorb visible light due to large bandgap. The excited electrons on the HOMO of B-BT-1,4-E was injected into the CB of TiO2; then the electrons were further transferred to the surface of TiO2 for photocatalyst H2 evolution. In the process of photocatalytic CIP degradation, photoexcited electrons could react with O2 to form ˙O2−. Moreover, photogenerated holes in the HOMO of B-BT-1,4-E were also collected due to well-aligned straddling band structures of B-BT-1,4-E/TiO2 heterojunction at the interface. As a result, the photogenerated electron/hole could be effectively separated, and then the active species hole and ˙O2− participated in the photocatalytic degradation of CIP.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7ta09374h |
This journal is © The Royal Society of Chemistry 2018 |