Enhanced extrinsic absorption promotes the visible light photocatalytic activity of wide band-gap (BiO)₂CO₃ hierarchical structure

Abstract

(BiO)₂CO₃ hierarchical microspheres and (BiO)₂CO₃ nanoparticles with wide band-gaps were fabricated by a simple hydrothermal method. The former showed higher visible light photocatalytic activity towards the removal of NO than the latter due to the enhanced extrinsic absorption benefiting from the strong light reflecting and scattering effects. The extrinsic absorption over (BiO)₂CO₃ hierarchical microspheres can induce the production of holes and electrons with visible light, subsequently generating active species to participate in photocatalytic reactions.

---

The research of semiconductor materials with different morphologies and dimensions has been an active area over the last few decades because of their shape and size dependent physical, chemical, electronic, and optical properties.¹ Especially, materials with a three dimensional (3D) hierarchical structure have attracted considerable attention due to their widespread potential applications in biology, drugs delivery, photovoltaics, lithium ion storage, catalysis, photocatalysis, sensors and supercapacitors.²

Recently, 3D structured semiconductor photocatalysts have been intensively investigated because of their efficient photocatalytic activity resulting from the large surface areas, special pore structure, efficient light harvesting and charge separation.³ For wide band-gap semiconductors, they are restricted by their deficient visible light absorption. Many efforts have been devoted to overcoming this bottom neck, centring mainly on the improvement of intrinsic absorption (electron transitions from the valance band to conduction band of a semiconductor).^4,5 Most recently, fabrication of 3D structures are reported to be an effective way to boost the visible light absorption of wide band-gap semiconductors due to the reflecting and scattering effects. For example, Que's group synthesized TiO₂ microspheres with enhanced visible light absorption and photocatalytic activity compared with P25 although it is involved with the doping of F atom.⁶ Liu et al. fabricated hierarchical hollow TiO₂ spheres with improved visible light photocatalytic activity towards degradation of RhB.⁷ Likewise, Lu and co-workers prepared BiOCl hierarchical nanostructures with enhanced visible light photocatalytic properties.⁸ Yet, little is known about how the reflecting and scattering effects improve the visible light absorption and the corresponding photocatalytic activities of wide band-gap semiconductors.

(BiO)₂CO₃, a typical wide band-gap semiconductor, displaying promising antibacterial, sensing, super capacitor, and photocatalysis performance, draws increasing research interests.⁹ Herein, (BiO)₂CO₃ hierarchical microspheres (BOC-R3) and (BiO)₂CO₃ nanoparticles (BOC-R6) were fabricated by hydrothermal treatment of bismuth citrate and sodium carbonate via adjusting the dosage of precursors. The as-prepared samples were utilized to photocatalytic removal of NO under both visible and UV light. Except the phenomenon that BOC-R3 showed higher UV light photocatalytic activity than BOC-R6. We found that BOC-R3 displayed enhanced visible light photocatalytic activity compared to BOC-R6. The enhanced visible light photocatalytic activity is supposed to be associated with the improved extrinsic absorption (such as impurity, the excitation of surface states and the photoionization of defects) originating from the strong light reflecting and scattering effects of incident photons in the 3D hierarchical structure. The improved extrinsic absorption can induce free electrons and holes to initiate chemical reactions. This work, in terms of the enhanced visible light absorption and activity of wide band-gap semiconductors with 3D hierarchical structures, provides new insights and sheds light on the dominant role of extrinsic absorption.

The detailed procedure for the experiments was described in ESI. XRD patterns of the BOC-R3 and BOC-R6 samples are shown in Fig. 1a. All diffraction peaks of the two samples could match the orthorhombic (BiO)₂CO₃ crystallites (JCPDS file no. 25-1464). XPS survey spectra in Fig. 1b demonstrate that the two samples only contain Bi, O and C elements, further confirming that the two samples are pure (BiO)₂CO₃. The XPS spectra of Bi, O and C elements shown in Fig. S1 (detailed discussions in ESI†) suggest that each corresponding element in the two samples has similar chemical environment. In addition, four characteristic bands in FT-IR (Fig. S2 (ESI†)) at ν₁ (1067 cm⁻¹), ν₂ (846 and 820 cm⁻¹), ν₃ (1468 and 1391 cm⁻¹), ν₄ (670 cm⁻¹), ν₁ + ν₄ (1756 and 1730 cm⁻¹) corresponding to the internal vibrations of “free” CO₃²⁻ can be observed in the two samples, respectively.¹⁰

Fig. 1 XRD patterns (a) and XPS survey spectra (b) of the BOC-R3 and BOC-R6 samples.

Fig. 2 and 3 present the SEM and TEM images of the as-prepared samples. From low-magnification SEM image of the BOC-R3 sample (Fig. 2a), we can observe many isolated uniform microspheres with a diameter of ∼1.3 μm. These BOC-R3 microspheres are constructed by dozens of nanosheets with a thickness ranging from 20.9 to 44.8 nm. Meantime, many pores are formed among the self-assembled nanosheets. The TEM image (Fig. 2b) shows that the entire microsphere contains a hollow in the center. An HRTEM image of a single nanosheet in the microsphere is displayed in Fig. 2c. The measured lattice spacing is about 0.295 nm, which agrees well with the d-spacing of (013) crystal face of (BiO)₂CO₃. The SAED pattern (Fig. 2d) displays an array of clear and regular diffraction spots of one single nanosheet, indicating that the nanosheet is well-defined single crystalline. Furthermore, the results of EDX analysis indicate the presence of Bi, O and C elements in BOC-R3, confirming that the sample is of high purity (Fig. S3 (ESI†)).

Fig. 2 SEM (a) and TEM (b and c) images of the as-prepared BOC-R3 sample, SAED (d) of a single nanosheet.

Fig. 3 SEM (a) and TEM (b and c) images of the as-prepared BOC-R6 sample, SAED (d) of a single nanoparticle.

In contrast, the BOC-R6 samples with nanoparticles morphology are prepared though increasing the concentration of CO₃²⁻ (Fig. 3a). These nanoparticles with thickness being 71.6–277 nm are not uniform. Further observation in Fig. 3b also shows that these nanoparticles are irregular. The excessive CO₃²⁻ not only induces the preferred growth orientation, generating nanoparticles with increasing thickness, but also accelerates the reaction process, making both nucleation and growth be a kinetic control. Naturally, irregular nanoparticles are achieved. Fig. 3c displays the HRTEM image of an individual nanoparticle. The clear lattice fringes with an interplanar lattice spacing of 0.273 nm correspond to the (110) atomic planes. The corresponding SAED pattern (Fig. 3d) indicates the single-crystalline characteristic of the (BiO)₂CO₃ nanoparticles. The specific surface areas of BOC-R3 and BOC-R6 were determined to be 13.8 and 3.1 m² g⁻¹, respectively. The pore size distribution curve of BOC-R3 displays a pore size distribution centered at ca. 30 nm, which confirmed that the BOC-R3 samples are mesoporous (Fig. S4 (ESI†)). And BOC-R6 exhibits a pore size distribution range <2 nm, indicating that the BOC-R6 samples are microporous (Fig. S4 (ESI†)).

As seen from Fig. 4, BOC-R3 and BOC-R6 exhibit strong absorptions with a steep edge at ca. 365 nm, and BOC-R3 exhibits enhanced UV light absorption compared with BOC-R6. The 3D structure doesn't cause obvious shift in the fundamental absorption edge of (BiO)₂CO₃. Notably, BOC-R3 shows certain light absorption in the visible region while BOC-R6 doesn't. As known to all, light reflecting and scattering effects are obvious in 3D nanostructure. On one hand, the incident light can reach the place though reflecting and scattering where the incident cannot directly achieve, and subsequently increase the absorption probability. On the other hand, it is well-known that the energy of light will be reduced after multiple inelastic scatterings. The semiconductors in reality are extrinsic semiconductors. Except for the intrinsic light absorption, the semiconductor owns extrinsic absorptions including the photoionization of defects absorption, the excitation of surface states absorption, and defects or impurity absorption, which absorb lower energy than the intrinsic absorption.¹¹

Fig. 4 UV-vis DRS of the two samples.

In the UV light region, the absorptions of the two samples are caused by vast intrinsic absorptions and minor extrinsic absorptions. Due to the strong light reflecting and scattering effects, the BOC-R3 sample exhibits high UV light absorption compared to BOC-R6. Nevertheless, in the visible light region, the energy of the light is too low to induce the intrinsic absorption of the two samples due to the large band gaps (the band gaps of the as-synthesized BOC-R3 and BOC-R6 (E_g) samples estimated from the intercept of the tangents to the plots of (Ahν)^1/2 vs. photo energy were 3.34 and 3.41 eV, respectively (inset of Fig. 4)). Thus, the visible light absorption stems from extrinsic absorption. The strong reflecting and scattering effects in BOC-R3 make BOC-R3 possess visible light absorption, but BOC-R6 shows ignorable visible light absorption because of its weak reflecting and scattering effects. The schematic diagram of the reflecting and scattering effects of incident light in (BiO)₂CO₃ hierarchical microspheres structure is shown in Fig. 5. Fig. S5a (ESI†) shows the PL spectra of the two samples with an exciting wavelength of 430 nm. It can be seen that BOC-R3 and BOC-R6 samples exhibit a strong PL signal with a obvious PL peak at about 498 and 503 nm, respectively. Since the intrinsic band transition with 430 nm light is impossible, the PL peak around 500 nm suggests that defects exist in the crystal structure of the two samples (Fig. S5b†).

Fig. 5 Schematic diagram of the reflecting and scattering effects in (BiO)₂CO₃ hierarchical microspheres.

Afterwards, we tested the photoelectrochemical performance of the two samples under visible light as shown in Fig. 6. BOC-R3 sample shows the transient photocurrent responses under visible light irradiation, which implies that the light absorbed by extrinsic absorption could induce the production of holes and electrons, while BOC-R6 displays very low photocurrent responses mainly due to its weak extrinsic absorption (Fig. 6a). The Electrochemical impedance spectroscopy (EIS) is used to investigate the photogenerated charge separation process as shown in Fig. 6b. The diameter of the arc radius on the EIS Nyquist plot of BOC-R3 is smaller than that of BOC-R6 under visible light irradiation, which suggests an effective separation of photogenerated electron–hole pairs and fast interfacial charge transfer due to the special hierarchical nanosheets structure of BOC-R3.

Fig. 6 Photocurrent–time curves (a) and electrochemical impedance spectroscopy spectra (b) of BOC-R3 and BOC-R6 samples under visible light irradiation.

To confirm their photocatalytic activity, the two samples was evaluated by photocatalytic removal of NO in air. Firstly, we investigated the photocatalytic activity of the two samples under UV (360 nm) light irradiation. The 360 nm light photolysis of NO can be negligible without photocatalyst.¹² As displayed in Fig. S6a (ESI†), the removal ratios of NO by BOC-R3 and BOC-R6 products are 42.8% and 25.2%, respectively. In the UV light region, the samples possess intrinsic absorption and extrinsic absorption. The intrinsic light absorption induces electron transitions from the valance band to conduction band to produce free electrons and holes. The extrinsic absorption also can produce free electrons and holes to take part in the photocatalytic reaction as well.¹³ The strong reflecting and scattering of light in BOC-R3 result in high UV light activity. The reaction rate constant k_UV of BOC-R3 is (15.3 × 10⁻² min⁻¹) higher than that of BOC-R6 (10.4 × 10⁻² min⁻¹) as shown in Fig. S6b (ESI†). Also, the normalized rate constants k′_UV was calculated and the result was given in Table 1. Notably, the BOC-R3 shows lower k′_UV than BOC-R6, this phenomenon may be associated with the poor extrinsic absorptions compared with the intrinsic absorptions in the UV region. Anyhow, BOC-R3 showed higher UV light photocatalytic activity than BOC-R6 although surface area plays a major role on the photocatalytic activity in the UV region.

Table 1 Physicochemical properties, reaction rate constants k and normalized rate constants k′ of BOC-R3 and BOC-R6 samples

Samples	SBET (m^2 g^−1)	kUV (×10^−2 min^−1)	k′UV (×10^−3 min^−1)	kvis (×10^−2 min^−1)	k′vis (×10^−3 min^−1)
BOC-R3	13.8	15.3	11.1	7.6	5.5
BOC-R6	3.1	10.4	33.6	0.4	1.3

---

Then, the photocatalytic activities over BOC-R3 and BOC-R6 samples under visible light irradiation with different wavelength were studied. The visible light photolysis of NO can be ignorable without photocatalyst according to the previous research.¹² when a 420 nm cut-off filter was used, from Fig. 7a, we find that BOC-R3 shows a removal ratio of 23%, while BOC-R6 shows negligible visible light activity. This result reveals that the photoexcitation by extrinsic absorption can generate free charge carriers which can take part in chemical reactions, consistent with the finding reported by Serpone et al. who have studied the photoadsorption of H₂, CH₄ (donor) and O₂ (acceptor) molecules on the surface of wide band-gap metal oxides and concluded that photoreaction can be caused by the photoexcitation of extrinsic absorption of a catalyst.¹³ Moreover, the reaction rate constant k_vis of BOC-R3 is 7.6 × 10⁻² min⁻¹, which is about 19 times of BOC-R6 (0.4 × 10⁻² min⁻¹) as shown in Fig. 7b. The normalized rate constants k′_vis was calculated as shown in Table 1. The BOC-R3 (5.5 × 10⁻³ min⁻¹) also shows higher k′_vis than BOC-R6 (1.3 × 10⁻³ min⁻¹). Obviously, BOC-R3 displays high reaction rate. In addition, the reaction intermediate of NO₂ during photocatalytic oxidation of NO under UV light and visible light is monitored on-line as shown in Fig. 8. The fraction of NO₂ generated over BOC-R3 is lower than that with BOC-R6 whether it is under UV light or visible light irradiation. Namely, BOC-R3 promotes the oxidation of intermediate NO₂ to final NO₃⁻ product. We also used a cut-off filter with 450 nm and 510 nm to test the photocatalytic activity, respectively (Fig. S7 (ESI†)). Similarly, BOC-R3 shows higher activity than BOC-R6. Furthermore, the long-term visible (>420 nm) photocatalytic activity of BOC-R3 was performed (Fig. 9). Obviously, there is no obvious deactivation in the removal of NO even after 17 h of visible light irradiation, which indicates the stable and efficient photocatalytic activity of BOC-R3 sample.

Fig. 7 Photocatalytic activity toward removal of NO under visible light irradiation (a) and reaction rate constants k_vis (b) of the BOC-R3 and BOC-R6 samples.

Fig. 8 Monitoring of NO₂ intermediates during UV light irradiation (a) and visible light irradiation (b).

Fig. 9 Long-term photocatalytic activity of BOC-R3 under visible light irradiation (>420 nm).

ESR spectra were carried out under visible light irradiation in aqueous dispersion for DMPO-˙OH (Fig. 10). As displayed in Fig. 10a, the characteristic peaks of DMPO-˙OH can be observed in BOC-R3 under light illumination, indicating the generation of ˙OH radicals. Such result indicates that the photogenerated carriers produced via extrinsic absorption further induce the production of active species. However, nearly no sign of ˙OH radicals can be observed over BOC-R6 (Fig. 10b). The low visible light absorption and then the ignorable production of holes and electrons are responsible for the poor activity. Therefore, we can conclude that the enhanced extrinsic absorptions of BOC-R3 can produce free electrons and holes to take part in the photocatalytic reaction. The high photocatalytic activity of BOC-R3 is directly related to the 3D structure. The special 3D structure allows multiple reflecting and scattering of incident visible light, which will increase the extrinsic absorption. Also, the 3D hierarchical structure is favourable for the diffusion of reaction intermediates and thus accelerating the reaction rate. Namely, the 3D structures enhance the light-harvesting efficiency and increase the quantity of photogenerated carriers to participate in the photocatalytic reaction, consequently dramatically enhancing the visible light photocatalytic activity. Besides, this result also suggests that the visible light photocatalytic activity of wide band-gap semiconductors with 3D hierarchical structure mainly arises from the extrinsic absorption.

Fig. 10 DMPO spin-trapping ESR spectra of BOC-R3 (a) and BOC-R6 (b) in aqueous dispersion for DMPO-˙OH.

In summary, (BiO)₂CO₃ hierarchical microspheres and (BiO)₂CO₃ nanoparticles were fabricated by hydrothermal treatment of bismuth citrate and sodium carbonate by tuning the molar ratio of the precursors. The fabricated samples were applied to photocatalytic removal of NO in air under both UV and visible light. The results indicate that (BiO)₂CO₃ hierarchical microspheres not only showed high UV light photocatalytic activity but also displayed enhanced visible light photocatalytic activity in comparison with (BiO)₂CO₃ nanoparticles. The enhanced visible light photocatalytic activity can be attributed to the increased extrinsic absorption originating from the strong reflecting and scattering effects of incident light in 3D structure. The present work is expected to shed new light on the understanding of enhanced visible light photocatalytic activity of wide band-gap semiconductors mediated by light reflecting and scattering effects in 3D structure. This work could also provide a new perspective for the designed synthesis of visible light photocatalysts by controlling the internal structure of the materials.

Acknowledgements

This research is financially supported by the National Natural Science Foundation of China (51108487, 51478070), the Natural Science Foundation Project of CQ CSTC (cstc2013jcyjA20018), and the Innovative Research Team Development Program in University of Chongqing (KJTD201314).

Notes and references

1. (a) H. Goesmann and C. Feldmann, Angew. Chem., Int. Ed., 2010, 49, 2; (b) M. S. Xu, T. Liang, M. M. Shi and H. Z. Chen, Chem. Rev., 2013, 113, 3766; (c) X. Y. Lai, J. E. Halpert and D. Wang, Energy Environ. Sci., 2012, 5, 5604; (d) Y. Hu, X. H. Gao, L. Yu, Y. L. Wang, J. Q. Ning, S. J. Xu and X. W. Lou, Angew. Chem., Int. Ed., 2013, 52, 5636–5639; (e) L. Zhang and Y. Zhu, Catal. Sci. Technol., 2012, 2, 694–706; (f) K. Zhang and L. Guo, Catal. Sci. Technol., 2013, 3, 1672–1690.
2. (a) M. P. Melancon, M. Zhou and C. Li, Acc. Chem. Res., 2011, 44, 947; (b) M. D'Arienzo, L. Armelao, A. Cacciamani, C. M. Mari, S. Polizzi, R. Ruffo, R. Scotti, A. Testino, L. Wahba and F. Morazzoni, Chem. Mater., 2010, 22, 4083; (c) Y. Ma, G. Ji, B. Ding and J. Y. Lee, J. Mater. Chem., 2012, 22, 24380; (d) L. W. Zhang, T. G. Xu, X. Zhao and Y. F. Zhu, Appl. Catal., B, 2010, 98, 138; (e) M. Zhou, H. B. Wu, J. Bao, L. Liang, X. W. Lou and Y. Xie, Angew. Chem., Int. Ed., 2013, 52, 8679; (f) B. F. Machado and P. Serp, Catal. Sci. Technol., 2012, 2, 54–75; (g) A. Chithambararaj, N. S. Sanjini, A. Chandra Bose and S. Velmathi, Catal. Sci. Technol., 2013, 3, 1405–1414.
3. (a) J. W. Shi, J. H. Ye, Q. Y. Li, Z. H. Zhou, H. Tong, G. C. Xi and L. J. Guo, Chem.–Eur. J., 2012, 18, 3157; (b) Y. S. Jun, J. Park, S. U. Lee, A. Thomas, W. H. Hong and G. D. Stucky, Angew. Chem., Int. Ed., 2013, 52, 1; (c) Z. Sun, J. H. Kim, Y. Zhao, F. Bijarbooneh, V. Malgras, Y. Lee, Y.-M. Kang and S. X. Dou, J. Am. Chem. Soc., 2011, 133, 19314.
4. (a) T. Zhou, Y. Du, A. Borgna, J. Hong, Y. Wang, J. Han, W. Zhang and R. Xu, Energy Environ. Sci., 2013, 6, 3229–3234; (b) W.-J. Ong, L.-L. Tan, S.-P. Chai, S.-T. Yong and A. R. Mohamed, Nano Res., 2014, 7, 1528–1547; (c) J. Ran, J. Zhang, J. Yu, M. Jaroniec and S. Z. Qiao, Chem. Soc. Rev., 2014, 43, 7787–7812; (d) L. Liu, Z. Liu, A. Liu, X. Gu, C. Ge, F. Gao and L. Dong, ChemSusChem, 2014, 7, 618–626.
5. (a) X. B. Chen and C. Burda, J. Am. Chem. Soc., 2008, 130, 5018; (b) H. J. Huang, D. Z. Li, Q. Lin, W. J. Zhang, Y. Shao, Y. B. Chen, M. Sun and X. Z. Fu, Environ. Sci. Technol., 2009, 43, 4164; (c) F. Dong, T. Xiong, R. Wang, Y. J. Sun and Y. K. Jiang, Dalton Trans., 2014, 43, 6631.
6. Z. L. He and W. X. Que, Phys. Chem. Chem. Phys., 2013, 15, 16768.
7. G. L. Li, H. Y. Zhang, J. Lan, J. Li, Q. W. Chen, J. Y. Liu and G. B. Jiang, Dalton Trans., 2013, 42, 8541.
8. J. Y. Xiong, Z. B. Jiao, G. X. Lu, W. Ren, J. H. Ye and Y. P. Bi, Chem.–Eur. J., 2013, 19, 9472.
9. (a) G. Cheng, H. M. Yang, K. F. Rong, Z. Lu, X. L. Yu and R. Chen, J. Solid State Chem., 2010, 183, 1878; (b) F. Qin, G. F. Li, R. M. Wang, J. L. Wu, H. Z. Sun and R. Chen, Chem.–Eur. J., 2012, 18, 16491; (c) F. Dong, W.-K. Ho, S. C. Lee, Z. B. Wu, M. Fu, S. C. Zou and Y. Huang, J. Mater. Chem., 2011, 21, 12428; (d) S. Peng, L. Li, H. Tan, Y. Wu, R. Cai, H. Yu, X. Huang, P. Zhu, S. Ramakrishna, M. Srinivasan and Q. Yan, J. Mater. Chem. A, 2013, 1, 7630; (e) P. Madhusudan, J. Zhang, B. Cheng and G. Liu, CrystEngComm, 2013, 15, 231; (f) X. Zhang, Y. Zheng, D. G. McCulloch, L. Y. Yeo, J. R. Friend and D. R. MacFarlane, J. Mater. Chem. A, 2014, 2, 2275; (g) Y. Zhang, D. Li, Y. Zhang, X. Zhou, S. Guo and L. Yang, J. Mater. Chem. A, 2014, 2, 8273.
10. G. E. Tobon-Zapata, S. B. Etcheverry and E. J. Baran, J. Mater. Sci. Lett., 1997, 16, 656.
11. (a) V. Ryabchuk, Int. J. Photoenergy, 2004, 6, 95–113; (b) M. V. Ganduglia-Pirovano, A. Hofmann and J. Sauer, Surf. Sci. Rep., 2007, 62, 219–270; (c) A. V. Emeline, G. V. Kataeva, V. K. Ryabchuk and N. Serpone, J. Phys. Chem. B, 1999, 103, 9190; (d) X. Pan, M.-Q. Yang, X. Fu, N. Zhang and Y.-J. Xu, Nanoscale, 2013, 5, 3601–3614.
12. F. Dong, A. M. Zheng, Y. J. Sun, M. Fu, B. Q. Jiang, W. K. Ho, S. C. Lee and Z. B. Wu, CrystEngComm, 2012, 14, 3534.
13. (a) A. V. Emeline, K. R. Vladimir and N. Serpone, J. Phys. Chem. B, 1991, 103, 1316; (b) A. V. Emeline, G. N. Kuzmin, D. Purevdorj, V. K. Ryabchuk and N. Serpone, J. Phys. Chem. B, 2000, 104, 2989.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details, characterization, and Fig. S1 to S7. See DOI: 10.1039/c4ra10786a

---