Self-assembly of aligned rutile@anatase TiO₂ nanorod@CdS quantum dots ternary core–shell heterostructure: cascade electron transfer by interfacial design

Abstract

A novel self-assembly approach based on electrostatic interactions has been developed for the synthesis of a rutile@anatase TiO₂ nanorod (NR)@CdS quantum dots (QDs) ternary core–shell heterostructure, in which an in situ formed monodisperse anatase TiO₂ layer was intimately sandwiched between rutile TiO₂ NRs and CdS QDs. It has been demonstrated that the well-defined bilayer interface significantly improves the photocatalytic performance of the ternary heterostructure (i.e. rutile@anatase TiO₂ NR@CdS QDs), owing predominantly to the appropriate band alignment of the constituent semiconductors, thus facilitating photogenerated electron–hole separation and charge collection under simulated solar light irradiation.

---

Introduction

One-dimensional (1D) nanostructured metal oxide arrays, such as nanotubes, nanorods and nanowires, have received growing interest in a wide range of application fields because of their unique geometrical structure providing large surface-to-volume ratios, direct pathways for charge transport, and low reflectance induced by light scattering and trapping.^1–5 Particularly, TiO₂ nanorod arrays (TiO₂ NRs) grown vertically on transparent conductive substrates have received the most attention for the development of novel nanostructured electrodes as they have shown significant enhancements in charge separation, transport, and light absorption.^6–8 In spite of these great efforts, TiO₂, as a quintessential wide bandgap semiconductor, fails to absorb visible photons, which limits its potential in photocatalysis remarkably.

To improve visible light absorption, a number of research efforts have been devoted to reducing the bandgap of TiO₂ by doping with transition metal cations or non-metal anions,^9,10 depositing noble metal nanoparticles,¹¹ and coupling with secondary semiconductors.¹² Coupling TiO₂ with narrow bandgap semiconductors such as CdX (X = S, Se, Te),^13–15 PbS,¹⁶ and InP,¹⁷ substantially extends the visible absorption of TiO₂ and improves the quantum efficiency.^18–20 Although many TiO₂ based binary nanocomposites have been extensively investigated for this purpose by targeted modifications, nevertheless, higher excitonic efficiency of nanocomposites can be further achieved if a ternary hybrid system is well-designed via judicious integration of band alignment at the interface of different semiconductor components. For example, it has been well-established that an intermediate ZnS or ZnSe layer integrated between two different semiconductors contributes to enhanced charge separation by forming a built-in potential gradient band structure,^21,22 thus resulting in remarkably enhanced photoelectrochemical performances.^23–25 On the other hand, a number of works have previously demonstrated the advantage of the core–shell nanocable architecture, in which carrier separation takes place in the radial direction, with a carrier collection distance smaller or comparable to the minority of carrier diffusion lengths.^26–30

Inspired by these ideas, herein we have designed a rutile@anatase TiO₂ NRs (TN)@CdS QDs ternary core–shell heterostructure via a facile self-assembly approach, in which an intermediate anatase TiO₂ layer was sandwiched in the interface between rutile TiO₂ NRs and CdS QDs, and intimate integration between the nano-building blocks was achieved. It was found that the core–shell bilayer system plays an important role in light absorption and charge separation, therefore contributing to significantly improved photocatalytic performances.

Results and discussion

Scheme 1 depicts a flowchart for the self-assembly of the rutile@anatase TN@CdS QDs ternary core–shell heterostructure. Single-crystalline rutile TiO₂ NRs were grown vertically on the fluorine-doped tin oxide (FTO) substrate via a hydrothermal approach. In order to obtain a positively charged surface, cationic poly(diallyldimethylammonium chloride) (PDDA) was used to modify the TiO₂ NRs surface. In this way, negatively charged titanium(IV) bis(ammonium lactato)dihydroxide (TALH), a typical anatase TiO₂ precursor,^31–33 could be spontaneously adsorbed on the surface of the TiO₂ NRs via pronounced electrostatic interactions, leading to a rutile@anatase TN core–shell heterostructure after facile refluxing.³³ It should be emphasized that the as-prepared rutile@anatase TN nanostructure was subjected to calcination (450 °C, 1 h, in air) in order to improve its photoelectrochemical performance and simultaneously remove the PDDA layer. Finally, tailor-made negatively charged CdS QDs were self-assembled onto the 3-aminopropyltrimethoxysilane (APS) modified positively charged rutile@anatase TN scaffold, yielding a hierarchical rutile@anatase TN@CdS QDs ternary core–shell heterostructure, in which intermediate anatase TiO₂ layer was sandwiched closely at the interface.

Scheme 1 Schematic illustration of the self-assembly of rutile@anatase TN@CdS QDs ternary core–shell heterostructure.

As revealed by the field-emission scanning electron microscopy (FESEM) images shown in Fig. 1a and b and Fig. S2,† self-oriented TiO₂ NRs with an average diameter of ca. 120 nm are uniformly distributed over the entire surface of the FTO and are nearly perpendicular to the substrate. The top surface of some large nanorods demonstrates a tetragonal shape, which contains many rough step edges (Fig. S2d†), while the side surface is very smooth (Fig. S2c†). Intriguingly, as displayed in Fig. 1c and S3,† the surface of the TiO₂ NRs was evenly wrapped with a rough layer consisting of many monodisperse anatase TiO₂ nanoparticles (NPs), resulting from the hydrolysis and condensation of TALH after refluxing; this is in stark contrast to the original morphology of the pure TiO₂ NRs (Fig. S2†). Fig. 1d demonstrates a typical FESEM image of the rutile@anatase TN@CdS QDs ternary core–shell heterostructure, the morphology of which is distinct from that of pure TiO₂ NRs and rutile@anatase TN. Note that the TiO₂ NRs were covered by compact and monodisperse CdS QDs over the entire FTO surface (Fig. S4†), and no agglomeration of the CdS QDs was observed.

Fig. 1 Panoramic FESEM images of (a and b) rutile TNs, (c) rutile@anatase TN and (d) rutile@anatase TN@CdS QDs ternary core–shell heterostructure.

The morphology and crystal structure of TiO₂ NRs were characterized by transmission electron microscopy (TEM) and high-resolution TEM (HRTEM). The NRs consist of single crystalline rutile phase TiO₂ with a [001] growth direction (Fig. 2a and b), which is consistent with TiO₂ NRs reported previously.^6,34 The image in Fig. 2c clearly shows that a thin anatase TiO₂ layer was sandwiched at the interface between rutile TiO₂ NRs and CdS QDs, and CdS QDs closely cover the surface with increased surface roughness. The HRTEM image in Fig. 2d demonstrates well-resolved lattice fringes of about 3.52 Å, which corresponds to the (101) crystal plane of anatase TiO₂, suggesting the good crystallinity of the intermediate anatase TiO₂ layer. The HRTEM image of the part shown in Fig. 2c reveals a distinct lattice fringe of ca. 3.36 Å, corresponding to the (220) crystal plane of CdS QDs.

Fig. 2 TEM images of (a) a single TiO₂ NR and (c) the rutile@anatase TN@CdS QDs core–shell heterostructure. HRTEM images of (b)TiO₂ NR and (d) the rutile@anatase TN@CdS QDs core–shell heterostructure.

XRD patterns of the samples (Fig. S5†) reveal that the TiO₂ NRs deposited on the FTO substrate are single-crystalline rutile TiO₂, which is in good agreement with our previous results.⁶ Note that only weak diffraction peaks belonging to anatase TiO₂ and CdS were observed (Fig. S5c†), which may be attributed to the relatively low deposition percentage of anatase TiO₂ and CdS QDs ingredients. Moreover, the uniform deposition of the anatase TiO₂ layer between the rutile TiO₂ NRs and CdS QDs may also shield the characteristic diffraction peaks of the intermediate anatase TiO₂ component.

The composition and chemical valence states of the elements in the ternary core–shell heterostructure were determined by X-ray photoelectron spectroscopy (XPS), which is displayed in Fig. S6.† As shown in Fig. S6f,† the Ti 2p_3/2 and Ti 2p_1/2 peaks located at ca. 458.35 and 464.10 eV, respectively, indicate that the Ti element is in the +4 oxidation state.³⁵ In the high-resolution XPS spectrum of Cd 3d (Fig. S6e†), the peaks at ca. 405.03 and 411.76 eV for Cd 3d_5/2 and Cd 3d_3/2 are ascribed to the +2 oxidation state of Cd.³⁶ Moreover, combined with high-resolution spectrum of S 2p (Fig. S6d†) which corresponds to chemical species of S²⁻,³⁷ it can be concluded that CdS QDs were successfully deposited onto the rutile@anatase TN substrate, chemical valence states of which were retained during the self-assembly process. On the other hand, predominant peaks in the high-resolution spectrum of O 1s (Fig. S6c†) positioned at ca. 529.61 and 530.90 eV are assigned to lattice oxygen in TiO₂ and Ti–OH groups on the TiO₂ surface, respectively, suggesting that the synthesized TiO₂ NRs are well crystallized. The detailed chemical bond species versus binding energy is tabulated in Table S1.†

UV-vis diffuse reflectance spectra (DRS) were used to determine the optical absorption properties of the samples. It can be seen from Fig. 3a that enhanced absorbance in the visible region ranging from ca. 400 to 800 nm is observed with the introduction of CdS QDs into the core–shell heterostructure. A plot obtained via transformation based on the Kubelka–Munk function versus the energy of light (Fig. 3b) reveals the estimated bandgaps of the samples to be ca. 3.02, 3.00, and 2.76 eV, corresponding to blank rutile TN, rutile@anatase TN, and the rutile@anatase TN@CdS QDs heterostructure, respectively. This suggests a narrowing of the bandgap of TiO₂ NRs due to the integration of CdS QDs into the heterostructure. Meanwhile, we note that the light absorbance of rutile@anatase TN is significantly enhanced in comparison with that of pure rutile TN in the UV region, which can be ascribed primarily to the homogeneous passivation of rutile TiO₂ NRs by the anatase TiO₂ layer, thus contributing to the enhanced harvesting of UV light.

Fig. 3 (a) UV-vis diffuse reflectance spectra (DRS) of rutile TiO₂ NRs, rutile@anatase TN and the rutile@anatase TN@CdS QDs ternary core–shell heterostructure, and (b) plot of transformed Kubelka–Munk function versus energy of light.

As displayed in Fig. S7,† Fourier transformed infrared (FTIR) spectra of the samples further corroborate the deposition of CdS QDs on the TiO₂ NRs framework via the self-assembly strategy, since a substantial stretching vibration peak positioned at ca. 1578 cm⁻¹, arising from various deprotonated carboxyl groups (–COO⁻) on the CdS QDs surface, was clearly observed. Besides, characteristic peaks corresponding to Ti–OH and Ti–O bonds of TiO₂ could also be clearly identified.

Fig. 4a shows the transient photocurrent responses of the samples under intermittent simulated solar light irradiation. The introduction of CdS QDs significantly enhances the photocurrent of the nanocomposites, which rapidly decreases to zero once the light is switched off. It has been well-established that photocurrent is formed due to the diffusion of photogenerated electrons to the back contact and the simultaneous take up of the photogenerated holes by the hole acceptor in the electrolyte.³⁸ As such, the superior photocurrent of the ternary core–shell heterostructure over its rutile TN@CdS QDs counterpart prepared via the direct self-assembly of CdS QDs onto APS-modified rutile TiO₂ NRs using the same method indicates the more efficient separation and longer lifetime of the photoexcited electron–hole pairs. This result highlights the advantage of the anatase TiO₂ layer between CdS QDs and rutile TiO₂ NRs for light absorption and charge separation.

Fig. 4 (a) Transient photocurrent responses and (b) current density versus potential curves of rutile TN, rutile@anatase TN, rutile@CdS QDs and the rutile@anatase TN@CdS QDs ternary core–shell heterostructure, (c) electrochemical impedance spectroscopy (EIS) Nyquist plots of samples in 0.1 M Na₂S aqueous solution with zero bias under simulated solar light irradiation (100 mW cm⁻²), the amplitude of the sinusoidal wave was set at 10 mV and the frequency varied from 100 kHZ to 0.01 HZ.

Fig. 4b demonstrates the photocurrent density versus the applied voltage plot recorded under irradiation with simulated solar light. It is found that the photocurrent density of the samples increases with the forward bias voltage, indicative of a typical n-type semiconductor.³⁹ Among the samples, the ternary core–shell heterostructure demonstrates remarkably enhanced photocurrent compared to the other structures over the whole potential profile, which agrees well with the result in Fig. 4a. On the other hand, the electrochemical impedance spectroscopy (EIS) Nyquist plots in Fig. 4c show the dramatic decrease of the high-frequency arc for the ternary core–shell heterostructure as compared to that of pure rutile TN, suggesting that interfacial charge transfer over ternary core–shell heterostructure is quite efficient. Consequently, based on the above analysis, the imperative role of the intermediate anatase TiO₂ layer in the ternary core–shell heterostructure was ascertained for prolonging the lifetime of the photogenerated charge carriers.

Photocatalytic performances of the samples was examined by the photodegradation of an organic pollutant under ambient conditions and simulated solar light irradiation. As shown in Fig. S8,† blank experiments (without light or photocatalyst) show negligible degradation, confirming that the reaction was driven by a photocatalytic process. It can be seen from Fig. 5a that the pure rutile TN exhibited almost no photoactivity, and in distinct contrast, the rutile@anatase TN binary nanostructure demonstrates improved photoactivity, which can be ascribed to the uniform deposition of a photoactive anatase TiO₂ layer on the rutile TN surface. Moreover, when CdS QDs were additionally self-assembled on the rutile@anatase TN substrate, the photocatalytic performance of the ternary core–shell heterostructure was remarkably enhanced, along with favorable photostability (Fig. S9†). It should be noted that the photocatalytic performance of rutile TN@CdS QDs is inferior to that of the ternary heterostructure under the same experimental conditions, thus once again highlighting that the intermediate anatase TiO₂ layer is of great importance for suppressing the recombination of photoexcited electron–hole charge carriers due to the intimate interfacial interaction between rutile TiO₂ NRs and CdS QDs. On the other hand, the photocatalytic reduction of 4-nitrophenol (4-NP) to 4-aminophenol (4-AP) over the different samples was also probed under ambient conditions (Fig. 5b and S10†). During the photocatalytic reduction process, photogenerated holes were completely quenched by a hole scavenger (HCO₂NH₄) in the presence of saturated argon flow, and thus the majority of the surviving photoelectrons could selectively reduce 4-NP to 4-AP. It was found that the photoreduction performances of the samples exhibited an analogous trend, among which the ternary core–shell heterostructure demonstrates the best activity among the samples.

Fig. 5 Photocatalytic performances of rutile TN, rutile@anatase TN, rutile TN@CdS QDs, and rutile@anatase TN@CdS QDs ternary core–shell heterostructure toward (a) the photooxidation of MO and (b) the selective photocatalytic reduction of 4-nitrophenol (4-NP) under simulated sunlight irradiation (AM 1.5, 100 mW cm⁻²) at ambient conditions.

Accordingly, a possible photocatalytic mechanism for the rutile@anatase TN@CdS QDs ternary core–shell heterostructure is proposed, as illustrated in Fig. S11.† Under simulated solar light irradiation, the photogenerated electrons are excited from the valence band (VB) of the CdS QDs to the conduction band (CB) under visible light irradiation, along with a small portion of the photoelectrons originating from the CB of the anatase and rutile TiO₂ phases in the near-UV photon region, thereby leaving holes in the VB. Notably, the CB of anatase TiO₂ is located between the CBs of CdS QDs and rutile TiO₂ (Fig. S11†), thus forming a unique band arrangement which efficiently facilitates the smooth and gradual transfer of photogenerated electrons from CdS QDs via the intermediate anatase TiO₂ layer to rutile TiO₂, and retards their back diffusion as well as recombination within the CdS QDs.⁷ Moreover, it is worth noting that the framework of rutile TiO₂ NRs is a single-crystalline electron collector self-aligned on the FTO substrate with strong adhesion, which further promotes the shuttling of injected electrons from the CdS QDs to TiO₂ and collection by the FTO substrate.⁷ In the meantime, photogenerated holes from the intermediate anatase TiO₂ layer and rutile TiO₂ NRs resulting from UV light excitation may also be effectively and gradually transferred to the VB of the CdS QDs, contributing to a more efficient separation and longer lifetime of the electron–hole pairs in the ternary core–shell heterostructure.

Conclusions

In summary, a facile and efficient self-assembly strategy for the preparation of a rutile@anatase TN@CdS QDs ternary core–shell heterostructure was developed, in which an intermediate anatase TiO₂ layer was judiciously integrated at the interface between rutile TiO₂ NRs and CdS QDs. The ternary core–shell heterostructure demonstrated remarkably enhanced photoelectrochemical and photocatalytic performances as a result of the monodisperse deposition of an anatase TiO₂ layer and CdS QDs in an intimate fashion onto a framework of rutile TiO₂ NRs, which provides the gradual transfer of photogenerated electrons for charge separation and visible light absorption. It is expected that our work could open the way for the design of various 1D semiconductor based core–shell heterostructures for promising photocatalytic applications.

Acknowledgements

This work was supported by the Nanyang Technological University startup grant: M4080977.120, and the Singapore Ministry of Education Academic Research Fund (AcRF) Tier 1: M4011021.120. We thank Mr. Song Xiaohui, Mr. Fang Zhanxi, Prof. Zhang Hua and Prof. Chen Hongyu for their kind help for TEM measurements.

References

1. A. H. Hochbaum and P. D. Yang, Chem. Rev., 2010, 110, 527–546.
2. M. Law, L. E. Greene, J. C. Johnson, R. Saykally and P. Yang, Nat. Mater., 2005, 4, 455–459.
3. O. L. Muskens, J. G. Rivas, R. E. Algra, E. P. A. M. Bakkers and A. Lagendijk, Nano Lett., 2008, 8, 2638–2642.
4. F. Zhang, S. M. Niu, W. X. Guo, G. Zhu, Y. Liu, X. L. Zhang and Z. L. Wang, ACS Nano, 2013, 7, 4537–4544.
5. K. S. Leschkies, R. Divakar, J. Basu, E. Enache-Pommer, J. E. Boercker, C. B. Carter, U. R. Kortshagen, D. J. Norris and E. S. Aydil, Nano Lett., 2007, 7, 1793–1798.
6. B. Liu and E. S. Aydil, J. Am. Chem. Soc., 2009, 131, 3985–3990.
7. J. T Li, M. W. G. Hoffmann, H. Shen, C. Fabrega, J. D. Prades, T. Andreu, F. Hernandez-Ramirez and S. Mathur, J. Mater. Chem., 2012, 22, 20472–20476.
8. J. Bang and P. V. Kamat, Adv. Funct. Mater., 2010, 20, 1970–1976.
9. W. Choi, A. Termin and M. R. Hoffmann, J. Phys. Chem., 1994, 98, 13669–13679.
10. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269–271.
11. A. Dawson and P. V. Kamat, J. Phys. Chem. B, 2001, 105, 960–966.
12. I. Robel, V. Subramanian, M. Kuno and P. V. Kamat, J. Am. Chem. Soc., 2006, 128, 2385–2393.
13. W. T. Sun, Y. Yu, H. Y. Pan, X. F. Gao, Q. Chen and L. M. Peng, J. Am. Chem. Soc., 2008, 130, 1124–1125.
14. S. Banerjee, S. K. Mohapatra, P. P. Das and M. Misra, Chem. Mater., 2008, 20, 6784–6791.
15. A. Kongkanand, K. Tvrdy, K. Takechi, M. Kuno and P. V. Kamat, J. Am. Chem. Soc., 2008, 130, 4007–4015.
16. R. Plass, S. Pelet, J. Krueger and M. Gratzel, J. Phys. Chem. B, 2002, 106, 7578–7580.
17. A. Zaban, O. I. Micic, B. A. Gregg and A. J. Nozik, Langmuir, 1998, 14, 3153–3156.
18. A. L. Linsebigler, G. Q. Lu and J. T. Yates, Chem. Rev., 1995, 95, 735–758.
19. L. M. Peter, D. J. Riley, E. J. Tull and K. G. Wijayantha, Chem. Commun., 2002, 1030–1031.
20. Y. Tak, H. Kim, D. Lee and K. Yong, Chem. Commun., 2008, 4585–4587.
21. I. Mora-Sero and J. Bisquert, J. Phys. Chem. Lett., 2010, 1, 3046–3052.
22. K. Y. Yan, L. X. Zhang, J. H. Qiu, Y. C. Qiu, Z. L. Zhu, J. N. Wang and S. H. Yang, J. Am. Chem. Soc., 2013, 135, 9531–9539.
23. M. Shalom, J. Albero, Z. Tachan, E. Martinez-Ferrero, A. Zaban and E. Palomares, J. Phys. Chem. Lett., 2010, 1, 1134–1138.
24. Y. Myung, D. M. Jang, T. K. Sung, Y. Sohn, G. Jung, Y. Cho, H. Kim and J. Park, ACS Nano, 2010, 4, 3789–3800.
25. H. Kim, J. Kim, W. Kim and W. Choi, J. Phys. Chem. C, 2011, 115, 9797–9805.
26. N. S. Lewis, Science, 2007, 315, 798–801.
27. Y. Zhang and L.-W. Wang, Nano Lett., 2007, 7, 1264–1269.
28. B. Tian, X. Zheng, T. J. Kempa, Y. Fang, N. Yu, G. Yu, J. Huang and C. M. Lieber, Nature, 2007, 449, 885–889.
29. B. Tian, T. J. Kempa and C. M. Lieber, Chem. Soc. Rev., 2009, 38, 16–24.
30. Y. J. Hwang, A. Boukai and P. Yang, Nano Lett., 2009, 9, 410–415.
31. F. Caruso, X. Y. Shi, R. A. Caruso and A. Susha, Adv. Mater., 2001, 13, 740–744.
32. A. Hanprasopwattana, T. Rieker, A. G. Sault and A. K. Dayte, Catal. Lett., 1997, 45, 165–175.
33. Y.-G. Guo, J.-S. Hu, H.-P. Liang, L.-J. Wan and C.-L. Bai, Adv. Funct. Mater., 2005, 15, 196–202.
34. I. S. Cho, Z. B. Chen, A. J. Forman, D. R. Kim, P. M. Rao, T. F. Jaramillo and X. L. Zheng, Nano Lett., 2011, 11, 4978–4984.
35. O. I. Micic, M. T. Nenadovic, M. W. Peterson and A. J. Nozik, J. Phys. Chem., 1987, 91, 1295–1297.
36. J. Zhang, J. G. Yu, M. Jaroniec and J. R. Gong, Nano Lett., 2012, 12, 4584–4589.
37. T. Nakanishi, B. Ohtani, K. Shimazu and K. Uosaki, Chem. Phys. Lett., 1997, 278, 233–237.
38. N. Zhang, Y. H. Zhang, X. Y. Pan, X. Z. Fu, S. Q. Liu and Y.-J. Xu, J. Phys. Chem. C, 2011, 115, 23501–23511.
39. M. J. Natan, J. W. Thackeray and M. S. Wrighton, J. Phys. Chem., 1986, 90, 4089–4098.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c3mh00097d

---