Capturing photogenerated electrons and holes at the B/Cl co-modified rutile TiO₂ nanorods during organic pollutant degradation

Abstract

Degrees for capturing photogenerated holes by residual chemically-adsorbed Cl as donators and electrons by increased amount of adsorbed O₂ as receptors after H₃BO₃ modification in rutile nanorods are quantitatively controlled based on the surface photovoltage measurements, suggesting that capturing electrons is dominant compared to capture holes in colorless pollutant degradation.

---

Semiconductor photocatalysis has attracted significant attention as an inexpensive and environmental-friendly technique to chemically transform organic pollutants into non-hazardous compounds in the past several decades.¹ Among numerous photocatalysts, TiO₂ is the most popular due to its high stability, low cost and safety toward both humans and the environment.² Compared with anatase, rutile is often neglected in photocatalysis. However, it possesses several features, involved with high chemical stability, narrow band gap, high refractive index, high hardness, and high Young's modulus value.³ Interestingly, it was clearly demonstrated that the nanosized rutile TiO₂ synthesized by hydrothermal processes could exhibit higher photocatalytic activities for decomposing dye pollutants than anatase.⁴ Thus, it is much possible to develop good-performance rutile-based photocatalysts, with great significance.

In general, the photocatalytic performance mainly depends on the separation of photogenerated electron–hole pairs.⁵ To enhance the charge separation, it would be highly desirable to effectively capture photogenerated charges on the surfaces of solid photocatalysts. In this case, which is much dominant to capture photogenerated electrons and photogenerated holes in the photocatalytic degradation of colorless pollutants? Obviously, it is much meaningful for environmental photocatalysis to clarify the above issues from the views of scientific and engineering points.

In our previous work,⁶ it was found that the residual chloride on the surfaces of rutile nanorods could capture photogenerated holes, and the phosphate modification could promote photogenerated electrons captured by the adsorbed O₂. In that case, the residual chloride as donator to capture photogenerated holes is still much effective compared with the adsorbed O₂ as acceptors to capture photogenerated electrons. To clarify the above assumptions, it is necessary to make photogenerated electrons captured much effectively compared with photogenerated holes. It has been well demonstrated that the modification with acidic substances could enhance the amount of O₂ adsorbed on TiO₂ so as to be favorable to capture photogenerated electrons.⁷ Among common inorganic acids, boric acid (H₃BO₃) has seldom been employed to modify semiconductor photocatalysts to date. This is mainly because it usually displays low solubility, so that it is difficult to obtain highly dispersed H₃BO₃ molecules modified on the photocatalyst. However, there are vacancy atomic orbits in the B of H₃BO₃, whereas lone-pair electrons exist in the Cl of residual chloride. Thus, it is expected that the boric acid molecules could be modified on the Cl-residual rutile nanorods so as to greatly change the capturing processes for photogenerated charges. To the best of our knowledge, related works have not been reported up to now.

In this work, we first prepare TiO₂ by a hydrochloric acid-modified hydrothermal process⁶ and then modify TiO₂ in different contents of boric acid solution by a hydrothermal process of 160 °C for 6 h (ESI† experimental). As seen from Fig. 1A and S1,† the obtained TiO₂ is in the rutile phase with nanorod morphology, and the modification does not change the phase composition, nanorod morphology and optical absorption. In addition, no related-B species is observed, indicating that the modified borate is uniformly dispersed on the surfaces. For the resulting rutile TiO₂ nanorod (R), it is confirmed from Fig. 1B and Table S1† that the residual Cl element displays two kinds of chemical states, including the doped form incorporated into the crystal lattice at about 199.3 eV and the chemically-adsorbed one (–Ti–Cl) on the surfaces at about 197.0 eV.^6,8

Fig. 1 XRD patterns with TEM images as insets (A) and XPS spectra (B) of R and XB–R samples. R means the obtained rutile TiO₂, and XB–R is the H₃BO₃ modified TiO₂, in which X is the mass ratio of H₃BO₃ and TiO₂.

Noticeably, a new XPS peak at 191.4 eV appears in the modified rutile (XB–R). As the B amount gradually increases, its XPS peak slightly shifts to the high binding energy. This is related to the aggregation of H₃BO₃ molecules with dehydration since the binding energy of B1s in H₃BO₃ is low compared with that in B₂O₃.⁹ Meanwhile, the amount of chemically-adsorbed Cl decreases a little with the tiny shift to the low binding energy. As for the doped Cl, its binding energy does almost not change. Thus, it is suggested that the H₃BO₃ molecules be modified on the surfaces of rutile nanorods via the connections (–Ti–Cl: B–OH) between chemically-adsorbed Cl and B in H₃BO₃. This is further supported by the XPS result that the binding energy of B in 2B–R is slightly higher compared to that in the H₃BO₃-modified rutile without chloride. According to the atomic number ratios of adsorbed Cl to B, it is confirmed that each Cl corresponds to multiple B, meaning the aggregation of the boric acid molecules.

To investigate the capturing processes for photogenerated charges, we have completed the atmosphere-controlled surface photovoltage (SPV) measurements, as described in Fig. 2A. It is widely accepted that the SPV signal of a semiconductor nanoparticle mainly results from the photogenerated charge separation via the carrier diffusion process.¹⁰ For the nanosized pure TiO₂ without any donator or acceptor, no SPV response is observed in N₂ because the surface charge amount is not changed (Fig. 2B). However, an obvious SPV response is seen in air, which is ascribed to the acceptor role of adsorbed O₂ to capture photogenerated electrons.¹¹ Differently, the as-prepared R sample could exhibit a strong SPV response in N₂ (Fig. S2†), attributed to the residual chloride as donators to capture photogenerated holes so as to make photogenerated electrons preferentially diffuse to the electrode surfaces (Fig. 2C).⁶ If the SPV measurement is controlled in air, it is deduced that there would simultaneously exist donators and acceptors. Based on the different O₂-content of SPV responses (Fig. S2†), it is confirmed that the photogenerated holes are captured much more effectively compared with photogenerated electrons in the chloride-residual R because the SPV intensity is gradually decreased with the increasing O₂ content. Interestingly, the H₃BO₃-modified rutile TiO₂ samples not only exhibit obvious SPV responses in N₂, but also have corresponding increased SPV intensities as the O₂ content rises (Fig. S2†). The result demonstrates that the modification with H₃BO₃ changes the advantage of photogenerated holes to be captured, and the photogenerated electrons to be captured becomes dominant after modification in air (Fig. 2D). This would provide the necessary conditions to clarify the above issues, which is very crucial to capture photogenerated electrons and photogenerated holes in the photocatalysis for degrading colorless pollutants.

Fig. 2 (A) Schematic of a photovoltage cell for SPV measurement; (B–D) schematic of the photogenerated charge transfer in different atmospheres.

To quantitatively reflect the degrees for capturing photogenerated charges by donators and receptors in air, we tried to evaluate the charge amounts (CA) resulting from the residual chloride as donators and the adsorbed O₂ as acceptors based on the SPV responses in different atmospheres. Because there is only the residual chloride as donators, the observed SPV response [CA(h⁺)] in N₂ could directly reflect the degree for capturing photogenerated holes (Fig. 3). One can notice that the degree for capturing photogenerated holes is gradually decreased with increasing the amount of modified H₃BO₃. In combination with the above XPS analyses, it is deduced that the chemically-adsorbed Cl could capture photogenerated holes other than the doped ones, and it would not capture photogenerated holes after connecting with boric acid molecules.

Fig. 3 The calculated degree for capturing photogenerated holes [CA(h⁺)] and photogenerated electrons [CA(e⁻)] based on the SPV responses.

In air atmosphere, besides the chemically-adsorbed Cl as donators, there is the adsorbed O₂ as acceptors for all resulting rutile samples. Thus, the SPV response mainly depends on the synergistic effects of photogenerated electrons and photogenerated holes. Based on the SPV attribute in air, the degree [CA(e⁻)] for capturing photogenerated electrons (Fig. 3) could be calculated. As to the unmodified R sample, CA(e⁻) = CA(in N₂) − CA(in air), while for the H₃BO₃-modified R, CA(e⁻) = CA(in air) + CA(in N₂). It is seen from Fig. 3 that the degree [CA(e⁻)] for capturing photogenerated electrons is enhanced after modification with a proper amount of H₃BO₃, and the 2B–R exhibits the largest CA(e⁻) among the modified ones. If the amount of used H₃BO₃ continuously increases, the CA(e⁻) begins to decrease. Since only the adsorbed O₂ is taken as the receptor, it is expected that CA(e⁻) mainly depends on the amount of adsorbed O₂. This is well supported by the O₂ temperature-programmed desorption (TPD) curves (Fig. S3†) and the electrochemical reduction curves of O₂ on the as-obtained rutile nanorod film electrodes (Fig. S4†). It is confirmed that the amount of adsorbed O₂ is increased after modification with an appropriate amount of H₃BO₃, especially at the high temperature, meaning the chemically-adsorbed O₂. It is much obvious in the 2B–R sample, corresponding to the high current of O₂ electrochemical reduction. However, the desorbed O₂ amount and the corresponding O₂ reduction current would become small if the amount of modified borate is excess, attributed to the aggregation of H₃BO₃ molecules. This is in good agreement with the above CA(e⁻) results.

To investigate the photogenerated charge separation, we have carried out the measurements of formed hydroxyl radical (˙OH) amounts on different rutile TiO₂ after irradiation for 1 h by the widely used coumarin fluorescent method (Fig. S5†), since the ˙OH group is generally known as a key active species in photocatalytic processes.¹² Generally, the larger is the formed ˙OH amount, the higher is the photogenerated charge separation. Also, we take the charge amount maximum at about 370 nm to clearly reflect the degrees for capturing photogenerated charges, as shown in Fig. 4. One can notice that the formed ˙OH amount of H₃BO₃-modified TiO₂ is gradually enhanced with increasing the amount of used H₃BO₃, and the 2B–R one displays the largest amount. If the amount of used boric acid continuously increases, the formed ˙OH amount begins to decrease. By comparison, it is found that the amount of formed ˙OH agrees well with the degree for capturing photogenerated electrons, and it seems to have no relationship with that for capturing photogenerated holes. Hence, it is expected that the step where the adsorbed O₂ captures photogenerated electrons is rather crucial in the photocatalytic process.

Fig. 4 The formed hydroxyl radical amount-related fluorescence peak intensities at 456 nm after irradiation for 1 h as column diagram and their corresponding charge amounts maximum at about 370 nm for capturing photogenerated holes [CA(h⁺)] (shown as green square) and photogenerated electrons [CA(e⁻)] (shown as red circle).

As expected, it is confirmed based on the photocatalytic data for degrading colorless gas-phase acetaldehyde and liquid-phase phenol (Fig. 5) that the activity mainly depends on the degree for capturing photogenerated electrons, with certain stability since the activity almost remains unchanged after the 5-cycle photocatalytic degradation experiments (Fig. S6†). This demonstrates that capturing photogenerated electrons is important compared to capturing photogenerated holes, that is the step that the adsorbed O₂ captures photogenerated electrons is rate determining. In addition, to prove the role of residual chloride to connect with B in H₃BO₃, we prepared Cl-free rutile TiO₂ by directly calcining nanocrystalline anatase at 800 °C and tried to modify it with H₃BO₃ by the similar hydrothermal process (Fig. S7†). As a result, modification with boric acid is unfavorable for the photogenerated charge separation and the photocatalytic degradation reactions. Therefore, it is suggested that the residual Cl could link B via lone-pair electrons in Cl and vacancy atomic orbit in B, similar to the suggested points of N and B,¹³ so as to make H₃BO₃ molecules disperse.

Fig. 5 Photocatalytic degradation rate of gas-phase acetaldehyde and liquid-phase phenol in the inset.

In conclusion, we have prepared chloride-residual rutile TiO₂ nanorods, and then modified them further with H₃BO₃ molecules. The H₃BO₃ molecules are modified on the surfaces by the –Ti–Cl: B–OH connections. It is confirmed that the residual chemically-adsorbed Cl as donators could capture photogenerated holes, and the modified H₃BO₃ is favorable for the increased amount of adsorbed O₂ as receptors to capture photogenerated electrons. Thus, the degrees for capturing photogenerated charges by donators and receptors in rutile nanorods are controlled. Based on the quantitative evaluation, it is suggested for the first time that to capture photogenerated electrons is dominant compared to capture photogenerated holes in the photocatalytic degradation of colorless pollutants. This work would help us well understand the main factors influencing the photocatalytic activities for degrading pollutants and provide with new guides to design high-activity photocatalysts.

Acknowledgements

We are grateful for financial support from the National Scientific Fund for Distinguished Young Scholars (51125033), Funds for Creative Research Group of China (51121062), NSFC (21071048), the Program for Innovative Research Team in Chinese Universities (IRT1237), the Project of Chinese Ministry of Education (213011A), the Specialized Research Fund for the Doctoral Program of Higher Education (20122301110002), the Science Foundation for Excellent Youth of Harbin City of China (2014), and State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology (Grant no. 2013DX08).

Notes and references

1. (a) C. Chen, W. Ma and J. Zhao, Chem. Soc. Rev., 2010, 39, 4206; (b) L. G. Devi and R. Kavitha, Appl. Catal., B, 2013, 140–141, 559.
2. (a) X. Chen, L. Liu, P. Y. Yu and S. S. Mao, Science, 2011, 331, 746; (b) T. Froschl, U. Hormann, P. Kubiak, G. Kucerova, M. Pfanzelt, C. K. Weiss, R. J. Behm, N. Husing, U. Kaiser, K. Landfester and M. Wohlfahrt-Mehrens, Chem. Soc. Rev., 2012, 41, 5313.
3. (a) B. Liu and E. S. Aydil, J. Am. Chem. Soc., 2009, 131, 3985; (b) E. Hosono, S. Fujihara, K. Kakiuchi and H. Imai, J. Am. Chem. Soc., 2004, 126, 7790.
4. (a) C. Xiong, X. Deng and J. Li, Appl. Catal., B, 2010, 94, 234; (b) Y. Liu, H. Wang, Y. Wang, H. Xu, M. Li and H. Shen, Chem. Commun., 2011, 47, 3790.
5. (a) D. P. Colombo and R. M. Bowman, J. Phys. Chem., 1995, 99, 11752; (b) D. P. Colombo and R. M. Bowman, J. Phys. Chem., 1996, 100, 18445.
6. Y. Luan, L. Jing, Q. Meng, H. Nan, P. Luan, M. Xie and Y. Feng, J. Phys. Chem. C, 2012, 116, 17094.
7. (a) Y. Luan, L. Jing, Y. Xie, X. Sun, Y. Feng and H. Fu, ACS Catal., 2013, 3, 1378; (b) H. Cui, Y. Cao, L. Jing, Y. Luan and N. Li, ChemPlusChem, 2014, 79, 318; (c) L. Jing, Y. Cao, H. Cui, J. R. Durrant, J. Tang, D. Liu and H. Fu, Chem. Commun., 2012, 48, 10775.
8. (a) H. Xu and L. Zhang, J. Phys. Chem. C, 2010, 114, 11534; (b) H. Xu, Z. Zheng, L. Zhang, H. Zhang and F. Deng, J. Solid State Chem., 2008, 181, 2516.
9. W. Zhao, W. Ma, C. Chen, J. Zhao and Z. Shuai, J. Am. Chem. Soc., 2004, 126, 4782.
10. (a) L. Jing, W. Zhou, G. Tian and H. Fu, Chem. Soc. Rev., 2013, 42, 9509; (b) L. Jing, X. Sun, J. Shang, W. Cai, Z. Xu, Y. Du and H. Fu, Sol. Energy Mater. Sol. Cells, 2003, 79, 133.
11. Y. Cao, L. Jing, X. Shi, Y. Luan, J. R. Durrant, J. Tang and H. Fu, Phys. Chem. Chem. Phys., 2012, 14, 8530.
12. (a) J. G. Yu, L. F. Qi and M. Jaroniec, J. Phys. Chem. C, 2010, 114, 13118; (b) B. Hauchecorne and S. Lenaerts, J. Photochem. Photobiol., C, 2013, 14, 72.
13. G. Liu, C. Sun, L. Cheng, Y. Jin, H. Lu, L. Wang, S. C. Smith, G. Q. Lu and H. M. Cheng, J. Phys. Chem. C, 2009, 113, 12317.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedures, DRS spectra (Fig. S1), XPS data (Table 1), SPV responses (Fig. S2), O₂-TPD curves (Fig. S3), electrochemical reduction of O₂ (Fig. S4), amounts of hydroxyl radicals (Fig. S5), cycling runs of photocatalytic degradation (Fig. S6), XRD patterns, SPV responses, and photocatalytic activity (Fig. S7). See DOI: 10.1039/c4ra03619k

---