Synergistic effects of Zn and Pd species in TiO₂ towards efficient photo-reduction of CO₂ into CH₄

Abstract

A Zn and Pd co-modified TiO₂ photocatalyst exhibits remarkably enhanced photocatalytic activity upon reduction of CO₂ into CH₄. Owing to the synergistic effect of unique O–Zn–Cl and O–Pd–O species, TiO₂–PdX–ZnY presents enhanced visible absorption and efficient charge carrier separation. It can be deduced from theoretical calculations and experimental results that the energy levels of O–Zn–Cl and O–Pd–O satisfy the redox potential of CO₂ reduction, improving the photocatalytic performance. These results suggest that introducing multi-surface species to produce synergetic effects is a feasible route for efficient photocatalysis.

---

Introduction

Photo-reduction of CO₂ into CH₄via semiconductor photocatalysis has been investigated by many researchers since it uses solar light to convert CO₂ into a valuable carbon fuel, methane.^1–5 The mechanism of photo-reduction of CO₂ into CH₄ is explained as follows:^6–13

Catalyst + hν → e⁻ + h⁺

H₂O + h⁺ → ˙OH + H⁺ (1)

˙OH + H₂O + 3h⁺ → O₂ + 3H⁺

2H₂O + 4h⁺ → O₂ + 4H⁺E⁰_redox = 0.82 V vs. NHE (2)

2CO₂ + 4e⁻ → 2CO + O₂E⁰_redox = −0.12 V vs. NHE

2CO + 4e⁻ → 2˙C + O₂

˙C + 4e⁻ + 4H⁺ → CH₄E⁰_redox = 0.13 V vs. NHE

CO₂ + 8e⁻ + 8H⁺ → CH₄ + 2H₂O E⁰_redox = −0.24 V vs. NHE (3)

Electrons and holes are excited under proper irradiation (eqn (1)). These charge carriers would further react with surface adsorbed CO₂ and H₂O molecules to generate CO and carbon radicals (˙C), which would further react with H⁺ to produce the final product, CH₄ (eqn (2) and (3)). Furthermore, it is considered as necessary for highly active catalysts that the position of the conduction band and the valence band should satisfy the redox potential discussed above.

TiO₂ is one of the most promising photocatalysts for the photoreduction of CO₂ into CH₄, due to its good performance and stability.^{2,6,8,14–18} However, the catalytic application of TiO₂ is still limited because of its large band-gap, its rapid recombination of charge carriers, and especially the mismatch of its redox potential for the photoreduction of CO₂ into CH₄. Its modification with metal or nonmetal elements is considered as an effective method to enhance its visible response and separate the charge carriers as well as design and adjust the energy level of the photocatalyst, leading to full use of the photogenerated charge carriers. It is found that introducing Zn or Mn ions into the TiO₂ system would create new doping energy levels below the conduction band, extending the response into the visible region and separating the charge carriers.^19,20 Moreover, it is also noted that introducing Cu, C and Pd species into the TiO₂ system would enhance the visible response and generate more charge carriers by forming new energy levels above the valence band.^19,21

TiO₂ doped or modified with Pd^22–27 has been investigated to improve the photocatalytic performance recently. Electrons and holes would transfer to the surface to participate in the photo-reduction reaction more easily and rapidly via these doping energy levels. In our previous work, it was found that co-modified TiO₂ with Pd and other elements could show a strong visible response and separate charge carriers efficiently,^28–30 exhibiting remarkable catalytic performance. Moreover, the existence of unique Zn species on the TiO₂ surface is also confirmed.²³

To date, for TiO₂ co-modified with Pd and other elements the band structure, its relationship with the redox potential for the reduction of CO₂ into CH₄, and the behaviour of the photogenerated charge carriers are still not investigated. Furthermore, the synergistic effect of Pd with other elements and their relationships with the visible response, separation of charge carriers and redox ability as well as the photocatalytic activity of the photocatalyst still need further investigation. And the photocatalytic mechanism should be discussed in detail.

In this work, Pd and Zn co-modified TiO₂ exhibited remarkable photocatalytic activity. This paper focuses on the synergistic effect of Pd and Zn species on the surface and their influence on the band structure, visible response, separation of the photogenerated charge carriers and photo-catalytic activity in detail. The photocatalytic mechanism is also investigated for the reduction of CO₂ into CH₄.

Experimental

Synthesis of TiO₂–PdX–ZnY

All chemicals used in this work were of analytical grade. At room temperature, different amounts of PdCl₂ solution, Zn(NO₃)₂ and 1 mL of HCl solution (12 mol L⁻¹) were dissolved in 40 mL of ethanol. Under violent stirring for 30 min, 12 mL of Ti(OC₄H₉)₄ and 1 mL of deionized water were added dropwise (pH = 0.5). The mixed solution was stirred until TiO₂ gel is formed. After 24 hours of aging, the TiO₂ gel was heated at 373 K for 10 h and calcined at 723 K for 2.5 h. These synthesized samples are denoted as TiO₂–PdX–ZnY (X = 1.5–3%; Y = 3–7%), where X and Y stand for the nominal molar ratio of Zn²⁺ and Pd²⁺ to Ti⁴⁺ (Zn/Ti and Pd/Ti), respectively. Pure TiO₂, 5% Zn modified TiO₂ (TiO₂–Zn) and 1.5% Pd modified TiO₂ (TiO₂–Pd) were synthesized by the same process just without PdCl₂ and/or Zn₂(NO₃)₂.

Characterizations

The Raman spectrum was acquired on a Renishaw in Via Raman microscope with the 785 nm line of a Renishaw HPNIR 785 semiconductor laser. The BET surface areas were obtained by measuring nitrogen adsorption–desorption isotherms at 77 K (Micromeritics Automatic Surface Area Analyzer Gemini 2360, Shimadzu), after degassing at 453 K. XPS was performed with an ESCA Lab 220i-XL spectrometer by using an unmonochromated Al Ka X-ray source (148.6 eV). All spectra have been calibrated using the adventitious C1s peak at 284.8 eV. Diffuse reflectance UV-vis absorption spectra (UV-vis DRS) were obtained using a UV-vis spectrometer (U-4100, Hitachi). Photoluminescence (PL) spectra were measured on a spectrometer (Spex 1702), a photomultiplier tube (PMT, Hamamatsu R943), a lock-in amplifier, and a computer for data processing using the 325 nm line of a nano-second Nd:YAG laser (NL303G) as the excitation source. Surface photovoltage spectra (SPS) were acquired on a solid-junction photovoltaic cell of ITO/sample/ITO equipped with a lock-in amplifier (Model SR830, DSP) and synchronized with a light chopper. Monochromatic light was obtained by passing light from a 500 W xenon lamp (CHF-XQ 500W) through a double prism monochromator (WDG30-2).

Photo-reduction of CO₂

The photoreduction of CO₂ into CH₄ is used to estimate the performance of the photocatalyst. A 500 W spherical Xenon arc lamp (Philips, Belgium, 35 mW cm⁻², 290–800 nm) was applied as the light source in this experiment. There was one glass sheet (9.4 cm²) placed at the bottom of the sealed reaction vessel (410 mL). About 150 mg of the as-prepared samples was tiled on the glass. The light beam was vertical to the glass and the lamp was 10 cm away from the surface of the glass. The reaction vessel was filled with CO₂ gas (99.99%) for 45 min before it was sealed. Then the reaction mixture was injected with 2 mL of water. About 0.4 mL of the gas in this vessel was extracted every 120 min. A gas chromatograph (Techcomp GC-7890F, equipped with a 1 m × 3 mm TDX-01 packed column and a flame ionization detector (FID)) was used to detect the concentration of the produced CH₄ and CO. The relative humidity and temperature in the reactor during the photosynthesis process were maintained as 80% and 30 °C.

Results and discussion

The photo-reduction of CO₂ into CH₄ under Xe lamp irradiation is applied to estimate the catalytic performance of the obtained TiO₂-based photocatalysts, as shown in Fig. 1, and Table 1. CO is the intermediate product (Fig. S1 and Table S1, ESI†) and CH₄ is the final product. There is no other gas product detected, such as H₂ and CH₃OH. After 8 h of irradiation, only 0.350 μmol of CH₄ is detected for pure TiO₂. TiO₂–Zn exhibited a finite photocatalytic activity and 0.851 μmol of CH₄ was generated. The photocatalytic performance of TiO₂–Pd was enhanced effectively and 3.59 μmol of CH₄ was generated. After the introduction of Zn and Pd into TiO₂, the TiO₂–PdX–ZnY (X = 1.5–3%; Y = 3–7%) samples exhibited remarkably enhanced photocatalytic activity. Furthermore, the TiO₂–Pd1.5%–Zn5% sample represented the highest catalytic performance of all photocatalysts and about 7.99 μmol of CH₄ was detected, which is almost twice that for TiO₂–Pd and twenty times higher than that for pure TiO₂. The results indicated that the photocatalytic activity for the Zn and Pd co-modified TiO₂ photocatalyst was enhanced remarkably due to the introduction of both Zn and Pd into TiO₂. In addition, it was found from XPS that the amount of C before and after the photocatalytic reaction was 15.34% and 18.63%, respectively, suggesting that the surface absorbed carbon in the photocatalyst did not react with CO₂ to generate CO.

Fig. 1 Amount of CH₄ generation for TiO₂, TiO₂–Pd, TiO₂–Zn and TiO₂–Pd1.5%–Zn5% (a), TiO₂–Pd1.5%–ZnY% (b) and TiO₂–PdX–Zn5% (c) under Xe lamp irradiation after 8 h.

Table 1 Photocatalytic activity of pure TiO₂, TiO₂–Zn, TiO₂–Pd and TiO₂–PdX–ZnY (X = 1.5–3%; Y = 3–7%) samples under Xe lamp irradiation after 8 h

Sample	CH4 (10^−6 mol)	Specific photocatalytic activity (10^−6 mol g^−1 h^−1)^b
a Blank is the photolysis of CO2. b Specific photocatalytic activity of CH4, CH4 generation amount per hour per unit mass of the catalyst. The experimental error in the photocatalytic results is <5% based on triplicate runs.
Blank^a	0.31 ± 0.01	—
TiO2(P25)	0.36 ± 0.02	0.30 ± 0.02
TiO2	0.35 ± 0.03	0.29 ± 0.02
TiO2–Zn	0.85 ± 0.02	0.71 ± 0.02
TiO2–Pd	3.77 ± 0.47	3.14 ± 0.39
TiO2–Pd1.5%–Zn3%	6.48 ± 0.27	5.40 ± 0.23
TiO2–Pd1.5%–Zn5%	7.99 ± 0.28	6.66 ± 0.23
TiO2–Pd1.5%–Zn7%	3.60 ± 0.35	3.00 ± 0.29
TiO2–Pd5%–Zn5%	7.14 ± 0.45	5.95 ± 0.38
TiO2–Pd7%–Zn5%	5.44 ± 0.73	4.53 ± 0.61

---

The structure and the existing states of Pd and Zn were investigated by the Raman, XRD and XPS spectra for TiO₂, TiO₂–Zn, TiO₂–Pd and TiO₂–PdX–ZnY samples. For the Raman spectra (Fig. 2 and Fig. S2, ESI†), all samples showed typical characteristic bands and the vibrational modes of anatase.³¹ In addition, for TiO₂–Zn as well as TiO₂–PdX–ZnY samples (Fig. S2, ESI†), two additional weak Raman peaks around 256 cm⁻¹ and 333 cm⁻¹ were observed, ascribed to the stretching modes of Zn–Cl and Zn–O bonds, respectively.²⁰ XRD results for all samples (Fig. S3 and Table S2, ESI†) also confirm that only anatase exists for these samples, which is in good agreement with the Raman spectrum mentioned above. Furthermore, after the introduction of Pd and Zn, the specific surface areas (BET) increase in the order of TiO₂ < TiO₂–Zn < TiO₂–Pd < TiO₂–Pd–Zn. The crystallite sizes decrease in the order of TiO₂–Pd–Zn < TiO₂–Zn < TiO₂–Pd < TiO₂. The detailed existing states of Zn and Pd are investigated by XPS.

Fig. 2 Raman spectra of TiO₂, TiO₂–Zn, TiO₂–Pd and TiO₂–Pd1.5%–Zn5%.

Fig. 3 shows the XPS spectra of TiO₂, TiO₂–Pd, TiO₂–Zn and TiO₂–Pd1.5%–Zn5%. For the Pd 3d spectra (Fig. 3A), the peaks of Pd 3d_5/2 for TiO₂–Pd and TiO₂–Pd1.5%–Zn5% locate at about 336.7 eV, attributed to the O–Pd–O structure on the TiO₂ surface (one Pd²⁺ ion linked with two unsaturated O²⁻ ions).^19,32 For XPS Cl 2p spectra (Fig. 3B), the peak centered at about 198.1 eV for TiO₂ and TiO₂–Pd is ascribed to Cl 2p of the O–Ti–Cl species on the TiO₂ surface. For the TiO₂–Zn and TiO₂–Pd1.5%–Zn5% samples, the peaks centered at about 198.9 eV and 198.7 eV respectively, are between those of TiCl₄ (198.2 eV) and ZnCl₂ (199.7 eV).⁹ Furthermore, as shown in Fig. 3C, the center of the Zn 2p_3/2 peak for the Zn 2p spectra of TiO₂–Zn (1022.1 eV) and TiO₂–Pd–Zn (1022.0 eV) locates between those of ZnO (1021.5 eV)³³ and ZnCl₂ (1022.5 eV).³⁴ This indicates that one Zn²⁺ ion links with one unsaturated O atom and one Cl⁻ ion to exist as O–Zn–Cl species.^20,34 Moreover, compared with TiO₂ (2.01%) and TiO₂–Pd (2.36%), the atom percentage calculated from the XPS also increases remarkably for TiO₂–Zn (6.49%) and TiO₂–Pd–Zn (5.25%). Hence, it is concluded that unique O–Pd–O and O–Zn–Cl species are formed on the surface of TiO₂.

Fig. 3 (A) XPS Pd 3d spectra of TiO₂–Pd and TiO₂–Pd1.5%–Zn5%; (B) Cl 2p spectra of TiO₂, TiO₂–Pd, TiO₂–Zn and TiO₂–Pd–Zn; (C) Zn 2p of TiO₂–Pd and TiO₂–Pd1.5%–Zn5%.

To determine the band structure of TiO₂–PdX–ZnY, the valence bands of XPS spectra and absorption spectra are shown in Fig. 4. For the XPS valence band spectra (Fig. 4A), the energy levels in all samples are aligned by the Fermi level of the XPS instrument (4.1 eV).³⁵ The BE of the onset edge for O 2p revealed the gap between the valence band top and the Fermi level (E_f). The onset edge of the valence band top for TiO₂ is at about +2.7 eV (+2.3 V, vs. NHE). The onset edge for TiO₂–Zn almost remains unchanged compared with TiO₂, implying that the energy level of O–Zn–Cl species is far from the valence band of TiO₂. It is found from the theoretical calculation (Fig. S4, ESI†) that a new energy level corresponding to O–Zn–Cl species occurred below the conduction band of TiO₂, which mainly consists of Zn 4s and Cl 3p orbitals. Moreover, for TiO₂–Pd and TiO₂–Pd–Zn, a small hump of about +2.1 eV (+1.7 eV, vs. NHE) is caused by the introduced Pd 4d orbital corresponding to the O–Pd–O species, whose energy level can be estimated to be at 0.6 eV above the valence band of TiO₂. The theoretical simulation (Fig. S4, ESI†) also confirmed that new energy levels, composed of the Pd 4d orbital and hybridized with O 2p and Ti 3d states, occurred above the valence band of TiO₂ for TiO₂–Pd.

Fig. 4 (A) XPS valence band spectra, and (B) absorption spectra of TiO₂, TiO₂–Zn, TiO₂–Pd and TiO₂–Pd1.5%–Zn5%.

The absorption spectra of all samples are plotted in Fig. 4B. TiO₂ shows strong absorption in the UV region caused by the band-to-band transition. The band gap for TiO₂ is estimated to be 3.1 eV, as the onset edge is at about 400 nm. Weak absorption centered at about 450 nm from 400 nm to 600 nm for TiO₂–Zn is observable. Electrons in the valence band can be excited to the energy level of O–Zn–Cl species next to the conduction band. The stronger visible response is observed for TiO₂–Pd, as the electrons from O–Pd–O species can be excited to the conduction band of TiO₂. Moreover, as we expect, owing to the synergistic effect of O–Pd–O and O–Zn–Cl species, TiO₂–PdX–ZnY represents the strongest absorption in the visible region (Fig. 4B and Fig. S3, ESI†), and the absorption can be further improved by increasing the amount of Pd and Zn. These results indicate that introducing Zn into the TiO₂–Pd system could lead to the synergetic effect of Pd and Zn, which may enhance the visible response, and match the band structure with the redox potential of photo-reduction of CO₂ into CH₄, resulting in enhanced catalytic activity.

To investigate the behaviour of charge carriers, photoluminescence spectra (PL), time resolved PL and surface photovoltaic spectra (SPS) were acquired, as shown in Fig. 5 and 6 and Table 2. For the photocatalyst, the electrons fell into the oxygen vacancies from the conduction band through a nonirradiative process (τ₁) and recombined with holes via an irradiative process (τ₂), resulting in fluorescence emission. Hence, quenching of the PL intensity and prolonged life-time of charge carriers indicate efficient suppression of the recombination for charge carriers. Fig. 5 shows the PL spectra of TiO₂, TiO₂–Zn, TiO₂–Pd and TiO₂–PdX–ZnY. It is observed from Fig. 5A that the emission intensity of TiO₂–Zn is weakened compared with TiO₂, owing to the surface energy level of O–Zn–Cl species. For TiO₂–Pd (Fig. 5B), the PL intensity is quenched significantly because the holes transfer from the valence band to the O–Pd–O species. And the PL emission for TiO₂–PdX–ZnY is further quenched with the increase of Zn or Pd owing to the contribution from both the Zn and Pd species (Fig. 5A and B). Moreover, the τ₁ and τ₂ values related to the nonirradiative and irradiative processes are shown in Table 2. The τ₂ values for TiO₂–Zn and TiO₂–Pd are greater than that for TiO₂, as the electrons could transfer from the conduction band of TiO₂ to the levels of O–Zn–Cl species or the holes could move from the valence band to the energy levels of O–Pd–O species. The τ₂ values for TiO₂–Pd1.5%–Zn5% samples are greater than those of TiO₂–Zn and TiO₂–Pd due to the synergistic effect of Pd and Zn species. These results suggest that the photo-generated electrons and holes were separated more efficiently for Zn and Pd co-modified TiO₂ samples than TiO₂–Zn and TiO₂–Pd.

Fig. 5 Photoluminescence spectra (PL) of TiO₂, TiO₂–Zn, TiO₂–Pd, and TiO₂–PdX–Zn5% (A) and TiO₂–Pd1.5%–ZnY (B).

Fig. 6 Surface photovoltaic spectra (SPS) of TiO₂, TiO₂–Zn, TiO₂–Pd and TiO₂–PdX–ZnY.

Table 2 Time constant τ₁ and τ₂ through double exponential decay fitting for TiO₂, TiO₂–Zn, TiO₂–Pd, and TiO₂–Pd1.5%–Zn5%

	TiO2	TiO2–Zn	TiO2–Pd	TiO2–Pd1.5%–Zn5%
τ 1 (ns)	0.295	0.298	0.294	0.291
τ 2 (ns)	3.059	3.876	4.038	4.382

---

As shown in Fig. 6, SPS reveals the separation behaviours of photogenerated charge carriers. The higher light response (peak intensity) usually indicates that more charge carriers are separated. For all samples, the response peaks around 350 nm are caused by band–band transition. TiO₂ shows no SPS response in the visible region (400 nm to 800 nm). And TiO₂–Zn and TiO₂–Pd exhibit poor SPS response in the visible region compared with TiO₂. This indicates that the photo-generated charge carriers were not separated efficiently from the surface of the catalyst under visible irradiation, when only Zn or Pd species were introduced into the TiO₂ system. Moreover, the TiO₂–PdX–ZnY samples exhibited a significant light response in the visible region compared to the others. With increasing Pd content, the light response of TiO₂–PdX–Zn5% is further enhanced, which may be attributed to the synergistic effect of O–Zn–Cl and O–Pd–O species. This suggests that more electrons and holes were generated and separated for the TiO₂–PdX–ZnY samples than the TiO₂–Pd or TiO₂–Zn samples, respectively.

According to the SPS results (Fig. 6), the synergistic effect of Pd and Zn species on the surface can be analyzed by the schematic band structure of TiO₂–PdX–ZnY samples (Fig. 7). TiO₂ shows no visible response and the recombination rates of holes and electrons are quite high. For TiO₂–Zn, the weaker SPS response in the visible region is assigned to the limited visible absorption in the visible region. Stronger visible absorption is observed for TiO₂–Pd, caused by the O–Pd–O species. Moreover, the SPS response in the visible region is poor, due to the rapid recombination of charge carriers, as shown in Fig. 7. For TiO₂–PdX–ZnY, the higher SPS light response in the visible region indicates the more efficient generation and separation of photogenerated charge carriers, owing to the synergistic effect of Pd and Zn species on the TiO₂ surface. Electrons were generated from the valence band to the energy level of Zn species. Furthermore, the electrons in the conduction band arising from Pd species can also transfer to the levels of O–Zn–Cl species, instead of recombining with holes in the valence band, separating photogenerated charge carriers efficiently. Therefore, for TiO₂–PdX–ZnY, more charge carriers were excited, separated and transferred to the surface of the photocatalyst than the TiO₂ samples modified with only Pd or Zn on the surface. When the electrons and holes migrate to the surface, the Zn and Pd surface species may act as reactive sites.

Fig. 7 Schematic band structure of TiO₂–PdX–ZnY as well as the photocatalytic mechanism (not drawn to scale).

Because the energy level of O–Zn–Cl (−0.45 V, vs. NHE) locates above the reduction–oxidation potential of reaction (3) (−0.24 V, vs. NHE) and the level of O–Pd–O (1.7 V, vs. NHE) is below the potential of reaction (2) (+0.8V, vs. NHE) respectively, these photo-generated carriers on the surface would participate in the catalytic reaction upon reduction of CO₂ into CH₄. According to the discussion above, owing to the synergistic effect of O–Zn–Cl and O–Pd–O species on the TiO₂ surface, the visible response was improved significantly and charge carriers were separated effectively. Moreover, the band structure for TiO₂–PdX–ZnY photocatalysts match preferably with the redox potential upon the photoreduction of CO₂ into CH₄, because of the energy levels of the O–Pd–O and O–Zn–Cl species. As a result, more photogenerated holes and electrons could be involved in the photocatalytic reduction, leading to a much higher photocatalytic activity for TiO₂–PdX–ZnY than for TiO₂–Pd and TiO₂–Zn.

Conclusions

A series of highly reactive photocatalysts, TiO₂–PdX–ZnY, were synthesized and exhibited significantly improved photo-reduction activity of CO₂ into CH₄. Because of the synergetic effect of surface O–Zn–Cl and O–Pd–O species, visible absorption was enhanced significantly and charge carriers were separated. Because the energy level of surface species is close to the redox potential of photo-reduction, more electrons and holes are enriched in the reactive sites (O–Zn–Cl and O–Pd–O surface species) and further participate in photo-reduction, leading to a remarkably improved photocatalytic activity for TiO₂–PdX–ZnY upon photoreduction of CO₂ into CH₄.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 51372120) and China Postdoctoral Science Foundation (no. 2017M611149). We thank Prof. Yihong Ding from State Key Lab of Theoretical and Computational Chemistry, Jilin University for helping us to carry out the theoretical calculation.

Notes and references

1. S. N. Habisreutinger, L. Schmidt-Mende and J. K. Stolarczyk, Angew. Chem., Int. Ed., 2013, 52, 7372–7408.
2. K. F. Li, X. Q. An, K. H. Park, M. Khraisheh and J. W. Tang, Catal. Today, 2014, 224, 3–12.
3. L. Yuan and Y. J. Xu, Appl. Surf. Sci., 2015, 342, 154–167.
4. O. Ola and M. M. Maroto-Valer, J. Photochem. Photobiol., C, 2015, 24, 16–42.
5. S. Bai, W. Yin, L. Wang, Z. Li and Y. Xiong, RSC Adv., 2016, 6, S7446–S7463.
6. N. M. Dimitrijevic, B. K. Vijayan, O. G. Poluektov, T. Rajh, K. A. Gray, H. He and P. Zapol, J. Am. Chem. Soc., 2011, 133, 3964–3971.
7. Q. Liu, Y. Zhou, Z. Tian, X. Chen, J. Gao and Z. Zou, J. Mater. Chem., 2012, 22, 2033–2038.
8. S. Xie, Y. Wang, Q. Zhang, W. Fan, W. Deng and Y. Wang, Chem. Commun., 2013, 49, 2451–2453.
9. Y. Yan, Y. Yu, D. Wu, Y. Yang and Y. Cao, Nanoscale, 2015, 8, 949–958.
10. Y. Yu, L. Guo, H. Cao, Y. Lv, E. Wang and Y. Cao, Sep. Purif. Technol., 2015, 142, 14–17.
11. E. Karamian and S. Sharifnia, J. CO2 Util., 2016, 16, 194–203.
12. K. Koci, L. Obalova and O. Solcova, Chemical and Process Engineering-Inzynieria Chemiczna I Procesowa, 2010, 31, 395–407.
13. S. T. Seng, L. Zou and E. Hu, Catal. Today, 2008, 131, 125–129.
14. K. Bhattacharyya, A. Danon, B. K. Vijayan, K. A. Gray, P. C. Stair and E. Weitz, J. Phys. Chem. C, 2013, 117, 12661–12678.
15. V. V. Galvita, H. Poelman, C. Detavernier and G. B. Marin, Appl. Catal., B, 2015, 164, 184–191.
16. C. Wang, R. L. Thompson, J. Baltrus and C. Matranga, J. Phys. Lett., 2009, 1, 48–53.
17. J. Low, B. Cheng and J. Yu, Appl. Surf. Sci., 2017, 392, 658–686.
18. L. Liu and Y. Li, Aerosol Air Qual. Res., 2014, 14, 453–469.
19. Y. Yan, Y. Yu, S. Huang, Y. Yang, X. Yang, S. Yin and Y. Cao, J. Phys. Chem. C, 2017, 1089–1098.
20. S. Yan, Y. Yu, Y. Gu, Y. Liu and Y. Cao, Sep. Purif. Technol., 2016, 171, 118–122.
21. Y. Yan, Y. Yu, C. Cao, S. Huang, Y. Yang, X. Yang and Y. Cao, CrystEngComm, 2016, 18, 2956–2964.
22. M. Bowker, D. James, P. Stone, R. Bennett, N. Perkins, L. Millard, J. Greaves and A. Dickinson, J. Catal., 2003, 217, 427–433.
23. Y. Yu, J. Wang, W. Li, W. Zheng and Y. Cao, CrystEngComm, 2015, 17, 5074–5080.
24. F. N. Sayed, O. Jayakumar, R. Sasikala, R. Kadam, S. R. Bharadwaj, L. Kienle, U. Schürmann, S. R. Kaps, R. Adelung and J. Mittal, J. Phys. Chem. C, 2012, 116, 12462–12467.
25. A. T. Kuvarega, R. W. M. Krause and B. B. Mamba, J. Phys. Chem. C, 2011, 115, 22110–22120.
26. Q. Li, Y. W. Li, P. Wu, R. Xie and J. K. Shang, Adv. Mater., 2008, 20, 3717–3723.
27. Q. Li, R. Xie, E. A. Mintz and J. K. Shang, J. Am. Ceram. Soc., 2007, 90, 3863–3868.
28. Y. Yu, E. Wang, J. Yuan and Y. Cao, Appl. Surf. Sci., 2013, 273, 638–644.
29. Y. Cao, T. He, L. Zhao, E. Wang, W. Yang and Y. Cao, J. Phys. Chem. C, 2009, 113, 18121–18124.
30. Y. Cao, Y. Yu, P. Zhang, L. Zhang, T. He and Y. Cao, Sep. Purif. Technol., 2013, 104, 256–262.
31. J. Wang, Y. Yu, S. Li, L. Guo, E. Wang and Y. Cao, J. Phys. Chem. C, 2013, 117, 27120–27126.
32. Y. Yu, T. He, L. Guo, Y. Yang, L. Guo, Y. Tang and Y. Cao, Sci. Rep., 2015, 5(9561), 1–6.
33. W. Li, D. Wu, Y. Yu, P. Zhang, J. Yuan, Y. Cao, Y. Cao and J. Xu, Phys. E, 2014, 58, 118–123.
34. Y. Yu, Y. Gu, W. Zheng, Y. Ding and Y. Cao, J. Phys. Chem. C, 2015, 119, 28190–28193.
35. Y. Cao, T. He, Y. Chen and Y. Cao, J. Phys. Chem. C, 2010, 114, 3627–3633.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7nj03305b

‡ Both authors contributed equally.

---