Asymmetric zinc porphyrin-sensitized nanosized TiO2 ...

ChemComm

Published on 26 June 2015. Downloaded by Wuhan University on 07/07/2015 00:51:00.

COMMUNICATION

Cite this: DOI: 10.1039/c5cc03812j Received 7th May 2015, Accepted 26th June 2015

View Article Online View Journal

Asymmetric zinc porphyrin-sensitized nanosized TiO2 for efficient visible-light-driven CO2 photoreduction to CO/CH4† Kan Li, Li Lin, Tianyou Peng,* Yingying Guo, Renjie Li and Jing Zhang

DOI: 10.1039/c5cc03812j www.rsc.org/chemcomm

Asymmetric zinc porphyrin (ZnPy) was synthesized and used to sensitize nanosized TiO2. The visible-light-driven activity of CO2 photoreduction to generate CO/CH4 in the gas phase was observed from the ZnPy-sensitized TiO2 without loading noble metal, and the mechanism was discussed.

Owing to the extensive use of fossil fuels and immoderate deforestation, the CO2 level in the atmosphere has been increasing steadily over the past few centuries. Hence, the conversion of CO2 into C1/C2 energy compounds has become a global issue from the viewpoint of the sustainable development of human society.1 Among various strategies attempted for CO2 conversion,2 CO2 photo-reduction over semiconductors by utilizing the inexhaustible solar energy can efficiently reduce CO2 emission and produce marketable solar fuels, and thus is considered as the most promising way to solve the current energy and environmental problems. Since CO2 does not absorb either visible or UV radiation, CO2 photoreduction is a process requiring a suitable catalyst to absorb UV-vis light and transfer it to CO2.3 However, TiO2 as the most extensively used catalyst only absorbs the UV-light due to its wide bandgap.4 As an efficient route for harvesting visible light, dye sensitization is widely used to extend the spectral response region of wide bandgap semiconductors in the photocatalytic or photovoltaic system.5–7 Similarly, Co or Zn phthalocyanines were used to sensitize TiO2, which exhibited a visible-light-driven activity of CO2 photoreduction to HCOOH in a saturated CO2 solution.8 Recently, a metal complex dyad containing zinc porphyrin as the light-harvesting unit and a rhenium bipyridyl complex as the catalytic moiety for CO2 photoreduction loaded on p-type NiO also showed visible-light-driven activity of CO2 photoreduction.8 Mg porphyrin as the visible light absorption centre can fix CO2 and convert it into carbohydrates in plant photosynthesis, and porphyrins have some intrinsic advantages such as high

College of Chemistry and Molecular Science, Wuhan University, Wuhan 430072, P. R. China. E-mail: [email protected] † Electronic supplementary information (ESI) available: Detailed synthesis procedure and some experimental results and discussion. See DOI: 10.1039/c5cc03812j

This journal is © The Royal Society of Chemistry 2015

extinction coefficient, fluorescence, and quantum yield as well as suitable energy levels for the semiconductors’ photosensitization. Moreover, porphyrins have many reaction sites (i.e. four meso- and eight b-positions) available for dye functionalization,9 and thus it is feasible to construct an oriented electron transfer channel in their sensitized semiconductors by finely tuning the optical, physical, and electrochemical properties through molecular design. Hence, it is reasonable to think that porphyrin-sensitized TiO2 would be an efficient system for CO2 photoreduction. Recent studies suggested that brookite TiO2 or its mixed phase with anatase may be more efficient for CO2 photoreduction among the three TiO2 polymorphs.10 Herein, we present an investigation on a novel asymmetric zinc porphyrin (ZnPy, Fig. S1, ESI†) sensitized TiO2 for visible-light-driven CO2 photoreduction without noble metal loading. Nanosized TiO2 derived from a solvothermal process followed by calcination at 500 1C has a spongy structure composed of aggregated nanoparticles with a mean size of B10 nm (Fig. S2, ESI†), and the lattice fringes with d-spacings of B0.354 nm for those particles in the HRTEM image (Fig. 1) correspond to the (101) planes of anatase TiO2. The XRD pattern (inset in Fig. 1) indicates that the product is anatase/brookite mixed crystal with anatase as the main phase.7a,11 The average crystal size calculated from the anatase (101) peak is 10.6 nm,11 which is consistent with

Fig. 1 HRTEM image and XRD pattern (inset) of the synthesized TiO2 nanoparticles.

Chem. Commun.

View Article Online


Communication

the mean size (B10 nm) observed from the FESEM/TEM images (Fig. S2, ESI†). Raman spectra (Fig. S3, ESI†) indicate that weak peaks at 213, 247, 322, 366, and 460 cm1 ascribable to the B1g, A1g, and B2g vibration modes of the brookite lattice can be observed from the main Raman peaks of anatase, further confirming a small quantity of brookite mixed in anatase.10a,12 Moreover, the product displays a type IV N2 adsorption–desorption isotherm (Fig. S4, ESI†), indicating that there is a mesostructure with a BET specific surface area of 128.5 m2 g1 and a Barret–Joyner–Halenda (BJH) pore size distribution in the range of 2–20 nm centered at B5.5 nm, which is attributed to the inter-crystallite voids among those stacked nanoparticles. Asymmetric ZnPy containing three pyridine and one carboxyl groups synthesized through two steps (Fig. S1, ESI†) has a strong singlet absorption peak (B-band) at 426 nm, and weak triplet light absorption (Q-band) at around 530–630 nm (Fig. 2a).13 Compared with the pristine TiO2 with an absorption edge at B400 nm (DRS spectra, Fig. 2b), ZnPy–TiO2 shows the light absorption characters of ZnPy with significantly enhanced Q-band absorption, which is the outcome of a mixture of metalto-ligand charge transfer (MLCT), ion-to-ligand, and intraligand transitions (ILCTs).14 The increment in Q-band absorption implies the electron transfer from ZnPy to TiO2. In addition, ZnPy–TiO2 shows a strong IR peak at B1600 cm1, which is the superposition of 1603 and 1629 cm1 (Fig. S6a, ESI†), and can be due to the COOH of ZnPy and the Ti–O bond of TiO2, respectively.13,15 This result indicates the connection between ZnPy and TiO2. The optical energy gap (E0–0) of ZnPy is calculated to be 2.06 eV from the intersection of its normalized absorption and emission spectra (Fig. S6b, ESI†),5 and its HOMO (related to Eox, ESI† for details) and LUMO (E* = Eox E0–0) obtained from cyclic voltammograms are 1.16 and 0.90 V (Table S1, ESI†).5

Fig. 2 (a) UV-vis absorption spectrum of ZnPy solution (5 106 mol L1); (b) UV-vis diffuse reflection absorption spectra (DRS) of the synthesized TiO2 and its ZnPy-sensitized products.

Chem. Commun.

ChemComm

Density function theory (DFT) calculation was carried out on ZnPy at the B3LYP/6-31G(d) level to learn more about the frontier molecular orbitals (MOs) and the charge transfer from ZnPy to TiO2.16 The calculated LUMO/HOMO and frontier MO (Table S1 and Fig. S5, ESI†) demonstrate that the electrons of ZnPy’s HOMO2 are mostly dispersed around Zn, while those of HOMO1 are delocalized over Py rings. This suggested ZnPy’s MLCT. Moreover, the HOMO and LUMO of ZnPy are also delocalized over the Py rings, and then the LUMO1 and LUMO2 are respectively delocalized over the Py rings and further move to the carboxyl groups step by step. The calculated energy gap between these orbitals basically agrees with that of Q-band absorption (Fig. 2a), confirming the above MLCT and ILCT assumption of ZnPy.13,14 It indicates that the photoexcited electrons can transfer from the Py skeleton to the carboxyl groups, which is a benefit to the electron injection of the excited ZnPy to TiO2, and then causing the visible-light-driven CO2 photoreduction activity. Control experiments showed that both the photocatalyst and irradiation are necessary for the present gaseous CO2 photoreduction system. The primary results showed that CO2 can be reduced to CO and CH4 in the presence of H2O vapour and ZnPy–TiO2, and no other reduced product such as CH3OH, HCHO or HCOOH is detected in the gas or liquid phase by using the GC-FID method. Moreover, neither CO nor other carboncontaining organic matters can be detected with N2 instead of CO2, demonstrating that CO/CH4 generation stemmed from the CO2 photoreduction process. Fig. 3a shows the effects of the ZnPy-loading level on the CO/CH4 production activities over ZnPy–TiO2 during the initial 2 h irradiation, and the respective data are summarized in Table S2 (ESI†). Pristine TiO2 shows no visible-light-driven photoreduction activity due to the inefficient visible light harvesting. The CO produced rate shows an obvious increasing trend when the ZnPy-loading level is enhanced from 0.1% to 1.0%, and then goes downhill upon further enhancing ZnPy (Fig. 3a). Once 0.8% ZnPy is loaded, CH4 is generated obviously, and 1.0% ZnPy–TiO2 shows a maximum CO/CH4 production activity of 8.07/1.01 mmol g1 h1. The corresponding overall activity for CO/CH4 generation can be estimated by the total consumed electron number (TCEN, ESI,† for details). The highest TECN of 1.0% ZnPy–TiO2 is 24.17 mmol g1 h1, which is 11.2 times higher than that of 0.1% ZnPy–TiO2. Further enhancing ZnPy leads to a significant decrease in the overall activity (TCEN). It can be ascribed to inefficient light harvesting and electron injection due to the p–p stacking of ZnPy on TiO2, which is verified by the wide and weak B-band absorption of 2.0% ZnPy-TiO2 (Fig. 2b).13 Generally, CO2 photoreduction is conducted in the liquid or gas phase system. The catalyst in the liquid phase system can be dispersed uniformly and well attached with the dissolved CO2 species such as CO32, HCO3, CO2 and H2CO3, and thus the photoactivity is usually higher than that of the gaseous system.1,8b However, it is difficult to determine the initial reactants due to the ionization equilibrium among CO32, HCO3, CO2 and H2CO3 in the liquid phase system, and thus the gaseous photoreduction system has increasingly got recognition lately. Although the overall activity (TCEN) is much lower than the similar dye-sensitized


View Article Online


ChemComm

Communication

Fig. 3 (a) Effects of the ZnPy-loading level on the CO/CH4 production activity and the TCEN value for the CO2 photoreduction over the synthesized TiO2 during the initial 2 h light irradiation (l Z 420 nm); (b) typical time course of the amount of CO/CH4 produced over 1.0% ZnPy–TiO2 under l Z 420 nm light irradiation.

catalyst in the liquid phase system, the present TCEN value is relatively high among those gaseous systems reported previously (Table S3, ESI,† for details).10d,17,18 For the photostability test, the recycled 1.0% ZnPy–TiO2 was used in the subsequent photoreduction process (ESI,† for details). As can be seen in Fig. 3b, 1.0% ZnPy–TiO2 has relatively good stability in the first three runs (6 h) with only B5% overall activity loss, and the recycled catalyst kept in the dark for 42 days still exhibits some activity even after eight runs of accumulative 16 h irradiation. Moreover, the recycled catalyst after 16 h irradiation shows a synchronous decrease in the absorbance of both the B-band and Q-band of ZnPy (Fig. S7, ESI†), indicating that some ZnPy molecules are possibly separated from TiO2. In addition to the cycling stability, the typical amount of CO/CH4 produced over 1.0% ZnPy–TiO2 for 1, 2, 4, 6, 8, and 10 h of continuous irradiation are noted (Fig. S8, ESI†). Both the amounts of CO and CH4 produced keep an increasing trend when the irradiation time is no longer than 6 h. Upon further enhancing it to 8 and 10 h, the amount of CH4 produced is reduced although the amount of CO produced shows a limited increasing trend. Since the present system is kept gas-closed without a sacrificial reagent, the resultants would accumulate in the system and the oxidized ZnPy after electron injection cannot be regenerated efficiently, which prevents further shifts of chemical equilibrium for CO2 photoreduction. Therefore, the activity losses after the second run (Fig. 3b) can be related to the resultants’ accumulation, and some ZnPy molecules separated from TiO2 and/or oxidized synergistically. Possibly, CO2/CH4 (0.24 V) has slightly more positive reduction potential than CO2/CO (0.48 V);1,17 thus the oxidized ZnPy molecules tend to be regenerated by the accumulated CH4 rather than CO, thereby causing the amount of CH4 produced to decrease more easily. Although the CO2 photoreduction mechanism is complex and the detailed information is not well understood, it is a consensus that the protons from the water splitting and the adsorption states of carbonates (or its hydrolytes) serve as important intermediates of the reaction.2,17 Hence, in situ DRIFT spectra (Fig. S9, ESI†) are used to detect the adsorption of carbonates. Due to the high CO2 concentration in the present photoreaction system, the pristine TiO2 adsorbs bidentate carbonate (b-CO32) at B1296 cm1 and monodentate carbonate (m-CO32) at B1378 cm1;10,18 and no peak of bicarbonate (HCO3) or


hydroxy is observed. Hence, most TiO2 tend to convert CO2 to CO rather than CH4 under UV irradiation.10 Upon enhancing the ZnPy-loading level from 0.1% to 2.0%, the product shows a slightly decreasing trend in the CO2 adsorption capacity, which can be due to the TiO2 surface sites partly occupied by ZnPy, and thus the carbonates tend to form m-CO32 on TiO2, which causes a new peak of m-CO32 at B1555 cm1 and the peak of b-CO32 shifts from 1296 to 1303 cm1 (Fig. S9, ESI†).10,18 Meanwhile, the IR peak of m-CO32 shifts from 1378 to 1362 cm1 also implying decreased b-CO32 adsorption, and the peak at B1652 and B1430 cm1 can be ascribed to H2O (hydroxy) and HCO3 adsorption,10 respectively. Moreover, the enhanced CH4 produced activity of 0.5–1.0% ZnPy–TiO2 can be ascribed to the additionally increased adsorption of HCO3 (at 1430 cm1) and H2O (at 1652 cm1) as shown in Fig. S9 (ESI†), while the reduced activity of 1.5% and 2.0% ZnPy–TiO2 is due to the more sites occupied by ZnPy, which competes with the HCO3 and H2O adsorption. On the basis of the above experimental results and discussion, a possible mechanism for CO2 photoreduction over ZnPy–TiO2 is proposed in Fig. 4. The asymmetric structure of ZnPy provides a directional electron transfer channel (Fig. S5, ESI†) and a sufficiently negative LUMO as compared with TiO2’s CB (0.53 V vs. NHE),5 thus the electrons of the excited dye (marked as A in Fig. 4) can inject into the TiO2’s CB, oriented from the porphyrin ring to TiO2 connected via a carboxyl group (marked as B). Since the TiO2’s CB (0.53 V) is slightly higher than the reduction potentials of CO2/CO (0.48 V) and CO2/CH4 (0.24 V),1,17 and the CO2 and H2O are adsorbed on TiO2,1,2,17 it is promised to photocatalytically convert CO2 to CO/CH4 on the TiO2 (marked as C). Since the present gaseous system is kept gas-closed without a sacrificial reagent, the resultants would accumulate in the photoreaction system and the oxidized ZnPy cannot be regenerated efficiently. Therefore, the backward electron transfers such as the photogenerated carriers’ self-recombination of ZnPy (marked as D) and the injected electrons in TiO2’s CB recombination with the oxidized ZnPy (marked as E) would occur inevitably, and thus is harmful to the photoactivity of ZnPy–TiO2. Along with the accumulations of the oxidized ZnPy in the system, the CO2 photoreduced products might act as the electron donor to regenerate the oxidized ZnPy (marked as F), which would lead to the reduced activity of ZnPy–TiO2 under longer irradiation as shown in Fig. 3b. In the present system, the protons for CH4 generation are derived

Chem. Commun.

View Article Online

Communication

ChemComm

on the photocatalytic CO2 conversion, and fulfil the deficiency of the research in gaseous dye-sensitized CO2 conversion. This work was supported by the Natural Science Foundation of China (21271146, 21271144, 21171134, 20973128, and 20871096), the Funds for Creative Research Groups of Hubei Province (2014CFA007) of China.


Notes and references Fig. 4 The possible mechanism of CO2 photoreduction to CO/CH4 generation over the ZnPy-sensitized TiO2.

from the limited H2O molecule adsorbed on TiO2, which is unfavourable for the CH4 production reaction involving 8 protons. Hence, the main products in the present system are CO/CH4 rather than other hydrocarbons such as CH3OH, HCHO, and HCOOH.17 Moreover, the reduction potential of CO2/CH4 is slightly more positive than that of CO2/CO, thus the backward electron transfer of CH4 to the oxidized ZnPy is more easy. Hence, the CH4 produced rate (Fig. S8, ESI†) is more easily affected due to the more complex reaction and easier backward electron transfer processes. The effects of ZnPy-loading levels on the charge recombination process can be further validated from the photocurrent responses, the time-resolved PL spectra and EIS spectra. The photocurrent response of 1.0% ZnPy–TiO2 is much stronger than the pristine one (Fig. S10a, ESI†), ascribable to the fast charge transfer from ZnPy to TiO2 and the efficient visible light absorption of ZnPy. Time-resolved PL spectra also support the charge transfer from ZnPy to TiO2 (Fig. S10b, ESI†). ZnPy and TiO2 are both excited under 375 nm laser excitation, and the lifetime (tn 0 ) of ZnPy–TiO2 is 7.78 ns, much longer than that (1.55 ns) of the pristine one (Table S4, ESI†). The prolonged lifetime is due to the ZnPy excitation and the charge transfer from ZnPy to TiO2.5 Moreover, EIS spectra also confirm the frequent charge transfer from ZnPy to TiO2 (Fig. S11 and Table S4, ESI,† for details).7b,16a Hence, it can be concluded that the charge transfer of ZnPy–TiO2 is that the excited ZnPy inject electrons to TiO2 and then convert CO2 over the TiO2 surface. In conclusion, asymmetric zinc porphyrin (ZnPy) with a carboxyl group is successfully synthesized and applied as a novel sensitizer of nanosized TiO2, and the obtained ZnPy–TiO2 exhibits a visible-light-driven activity of CO2 photoreduction to CO/CH4 in a gaseous system without noble metal loading. The optimal ZnPy amount is determined to be 1.0% with the best CO/CH4 production rate. The molecular structural asymmetry and the connection of the carboxyl group produce an oriented electron transfer channel of the excited electrons from ZnPy to TiO2, which make CO2 conversion possible under visible light irradiation. Although the photocatalytic activity and stability of ZnPy–TiO2 needs further improvement, the present results provide an important indication about the effect of dye-sensitization

Chem. Commun.

1 (a) J. Mao, K. Li and T. Y. Peng, Catal. Sci. Technol., 2013, 3, 2481; (b) J. Mao, T. Y. Peng, X. H. Zhang, K. Li, L. Q. Ye and L. Zan, Catal. Sci. Technol., 2013, 3, 1253. 2 (a) W. G. Tu, Y. Zhou and Z. G. Zhou, Adv. Mater., 2014, 26, 4607; (b) K. F. Li, X. Q. An, K. H. Park, M. Khraisheh and J. W. Tang, Catal. Today, 2014, 224, 3; (c) E. V. Kondratenko, G. Mul, J. Baltrusaitis, G. O. Larrazabal and J. Perez-Ramirez, Energy Environ. Sci., 2013, 6, 3112. 3 (a) Y. P. Yuan, S. W. Cao, Y. S. Liao, L. S. Yin and C. Xue, Appl. Catal., B, 2013, 140, 164; (b) Y. H. Fu, D. R. Sun, Y. J. Chen, R. K. Huang, Z. X. Ding, X. Z. Fu and Z. H. Li, Angew. Chem., Int. Ed., 2012, 51, 3364; (c) W. G. Tu, Y. Zhou, Q. Liu, S. C. Yan, S. S. Bao, X. Y. Wang, M. Xiao and Z. G. Zhou, Adv. Funct. Mater., 2013, 23, 1743; (d) F. Sastre, A. V. Puga, L. C. Liu, A. Corma and H. Garcia, J. Am. Chem. Soc., 2014, 136, 6798. 4 L. Z. Wang and T. Sasaki, Chem. Rev., 2014, 114, 9455. 5 (a) X. H. Zhang, L. J. Yu, R. J. Li, T. Y. Peng and X. G. Li, Catal. Sci. Technol., 2014, 4, 3251; (b) X. H. Zhang, U. Veikko, J. Mao, P. Cai and T. Y. Peng, Chem. – Eur. J., 2012, 18, 12103. 6 J. He, J. Q. Wang, Y. J. Chen, J. P. Zhang, D. L. Duan, Y. Wang and Z. Y. Yan, Chem. Commun., 2014, 50, 7063. 7 (a) J. L. Xu, K. Li, W. Y. Shi, R. J. Li and T. Y. Peng, J. Power Sources, 2014, 260, 233; (b) J. N. Clifford, E. Martinez-Ferrero, A. Viterisi and E. Palomares, Chem. Soc. Rev., 2011, 40, 1635. 8 (a) Z. H. Zhao, J. M. Fan, S. H. Liu and Z. Z. Wang, Chem. Eng. J., 2009, 151, 134; (b) Z. H. Zhao, J. M. Fan and Z. Z. Wang, J. Cleaner Prod., 2007, 15, 1894; (c) Y. Kou, S. Nakatani, G. Sunagawa, Y. Tachikawa, D. Masui, T. Shimada, S. Takagi, D. A. Tryk, Y. Nabetani, H. Tachibana and H. Inoue, J. Catal., 2014, 310, 57. 9 (a) C. Hsieh, H. Lu, C. Chiu, C. Lee, S. Chuang, C. Mai, W. Yen, S. Hsu, E. W. Diau and C. Yeh, J. Mater. Chem., 2010, 20, 1127; (b) S. Mathew, A. Yella, P. Gao, R. Humphry-Baker, B. F. E. Curchod, N. Ashari-Astani, I. Tavernelli, U. Rothlisberger, M. K. Nazzeeruddin and M. Gratzell, Nat. Chem., 2014, 6, 242. 10 (a) L. J. Liu, H. L. Zhao, J. M. Andino and Y. Li, ACS Catal., 2012, 2, 1817; (b) L. J. Liu, D. T. Pitts, H. Zhao, C. Y. Zhao and Y. Li, Appl. Catal., A, 2013, 467, 474; (c) H. L. Zhao, L. J. Liu, J. M. Andino and Y. Li, J. Mater. Chem. A, 2013, 1, 8209; (d) J. G. Yu, J. X. Low, W. Xiao, P. Zhou and M. Jaroniec, J. Am. Chem. Soc., 2014, 136, 8839. 11 K. Li, J. L. Xu, X. H. Zhang, T. Y. Peng and X. G. Li, Int. J. Hydrogen Energy, 2013, 38, 15965. 12 S. Zhou, Y. Liu, J. M. Li, Y. J. Wang, G. Y. Jiang, Z. Zhao, D. X. Wang, A. J. Duan, J. Liu and Y. C. Wei, Appl. Catal., B, 2014, 158, 20. 13 (a) J. Akhigbe, M. Luciano, M. Zeller and C. Bruchner, J. Org. Chem., 2015, 80, 499; (b) S. Kim, T. Tachikawa, M. Fujitsuka and T. Majima, J. Am. Chem. Soc., 2014, 136, 11707. 14 W. Sinha, M. G. Sommer, N. Deibel, F. Ehret, B. Sarkar and S. Kar, Chem. – Eur. J., 2014, 20, 15920. 15 Y. S. Li, F. L. Jiang, Q. Xiao, R. Li, K. Li, M. F. Zhang, A. Q. Zhang, S. F. Sun and Y. Liu, Appl. Catal., B, 2010, 101, 118. 16 (a) L. J. Yu, W. Y. Shi, L. Lin, Y. Y. Guo, R. J. Li and T. Y. Peng, Dyes Pigm., 2015, 114, 231; (b) L. J. Yu, W. Y. Shi, L. Lin, Y. W. Liu, R. J. Li, T. Y. Peng and X. G. Li, Dalton Trans., 2014, 43, 8421. 17 H. Zhou, P. Li, J. J. Guo, R. Y. Yan, T. X. Fan, D. Zhang and J. H. Ye, Nanoscale, 2015, 7, 113. 18 (a) J. Mao, L. Q. Ye, K. Li, X. H. Zhang, J. Y. Liu, T. Y. Peng and L. Zan, Appl. Catal., B, 2014, 144, 855; (b) J. Mao, T. Y. Peng, X. H. Zhang, K. Li and L. Zan, Catal. Commun., 2012, 28, 38.
