Highly photoactive heterojunction based on g-C₃N₄ nanosheets decorated with dendritic zinc(II) phthalocyanine through axial coordination and its ultrasensitive enzyme-free sensing of choline

Abstract

A heterojunction with excellent photocatalytic performance based on graphene-like carbon nitride (g-C₃N₄) nanosheets and dendritic zinc(II) phthalocyanine was proposed. Herein, the g-C₃N₄ with excellent photo-activity and high nitrogen content was readily available as a functional material. The g-C₃N₄ acted as an electron pair donor for dendritic zinc(II) phthalocyanine through axial coordination, forming the p–n heterojunction. Then by taking advantage of the distortion of dendritic zinc(II) phthalocyanine, the spatial charge separation of photo-generated charge carriers in this metal macrocycle achieved high efficiency, resulting in the enhanced photo-to-electric conversion efficiency. Therefore, the optoelectronic sensing device based on the heterojunction led to an enhanced photocurrent, and made it a promising candidate for establishing photoelectrochemical biosensors. Moreover, the p–n heterojunction was successfully applied to the detection of choline with a wide linear range from 10 nM to 5 μM, which could be oxidized by the photo-generated holes. Along with these attractive features, the as-proposed biosensor also displayed a remarkable specificity against other interferents and could be successfully used for detecting choline in real samples. The heterojunction with enhanced photoelectronic properties provides a promising format for the future development of photoelectrochemical biosensors.

---

Photoelectrochemistry is a newly emerging yet dynamically developing analytical method which displays promising analytical applications and has attracted considerable interest.^1,2 Because of its different forms of energy for excitation and detection, the photoelectrochemical (PEC) analytical methods not only possess good temporal and spatial control, but also have high sensitivity with low background signals.^3–7 In recent years, fascinating inorganic semiconductor nanostructures and ruthenium bipyridine have been widely regarded as attractive candidate photosensitizers for photoelectric applications resulting from their optical and electronic properties.^4,8–10 Recently, g-C₃N₄ nanosheets, an appealing class of two-dimensional semi-conductive layered nanomaterials, have been explored for various applications, especially in the fields of visible light photocatalysis and photodegradation due to their large specific surface area, short diffusion length, high chemical stability and appealing electronic structure with a band gap of 2.7 eV.^11–13 In addition, the high nitrogen content in the g-C₃N₄ frameworks plays an essential role in the high photoelectric conversion efficiency, where the nitrogen atoms affect the spin density and the charge distribution of carbon atoms, including an activation region on the layer surface.¹⁴ On one hand, the band gap of g-C₃N₄ between the conduction band (CB) and the valence band (VB) could be opened, because the nitrogen causes the Fermi level to shift above the Dirac point, and the density of the states near the Fermi level is suppressed.^15,16 However, the photo-to-current conversion efficiency of bare g-C₃N₄ is still cursed by the high recombination rate of photo-generated electron–hole pairs and the quantum efficiency for the pristine semiconductor.^13,17

To improve the photo-to-current conversion efficiency of g-C₃N₄, dendritic zinc(II) phthalocyanine (DZP) bearing poly(aryl benzyl ether) dendritic substituents was employed in this text. Due to the excellent thermal and photochemical stability, phthalocyanines are obviously versatile and it is possible to change the central atom and to introduce substituents in the peripheral and axial positions, leading to the possibility to tune their applications as photosensitizers for photodynamic therapy.^18,19 And yet, the traditional phthalocyanines always suffer from molecular aggregation attributed to their intrinsic large π-conjugation, resulting in shortening the triplet state lifetime and reducing the quantum yield.²⁰ To resolve this problem, in this paper, DZP was synthesized with hydrophilic stability to avoid aggregation. The dendritic architectures of DZP employed in the photoelectrochemical system could achieve an antenna effect, which could conduce a larger light-harvesting system and facilitate photon absorption in the visible region.²¹ Even though dendrimeric phthalocyanine has such attracted much attention, till now, few groups have utilized dendrimeric phthalocyanine to establish a photoelectrochemical sensor. Herein, it was found DZP could couple with g-C₃N₄ nanosheets through axial coordination making this compound distorted, where g-C₃N₄ as a Lewis base exhibited strong adsorption capacity towards the zinc ion of DZP. The electrostatic repulsive force also existed between the electronegative g-C₃N₄ and DZP, which then induced distortion in the phthalocyanine ring, and the Lewis acidity of the Zn(II) to be stronger, bringing about the stronger axial coordination.¹⁸ Additionally, the formation of p–n heterojunctions between n-type g-C₃N₄ and the organic p-type charge transport material DZP could realize a low recombination rate for the electron–hole pairs and lead to an efficient charge separation process. Consequently, by using the stronger Lewis acidity of Zn(II) in the non-planar Zn(II) complex of distorted DZP and g-C₃N₄, a supermolecular electron donor–acceptor assembly could be constructed.^22,23 And the photovoltaic devices based on the p–n heterojunctions formed by g-C₃N₄ and DZP could not only facilitate the charge transfer, but also improve the performance of the optoelectronic systems, resulting in the enhanced photocurrent and making it a promising candidate for establishing photoelectrochemical biosensors. To evaluate the application of the excellent photoactive nanocomposites in analysis detection, choline (Ch), a dietary essential nutrient involved in several bodily functions, as a research sample was introduced into the determination.

In this text, our group developed a non-enzymatic PEC biosensor for direct, rapid and sensitive detection of Ch based on Ch oxidation by photo-generated holes, thus resulting in an obviously enhanced photocurrent. The detection result indicated that the g-C₃N₄/DZP axial coordinated composite provided a promising method for the future development of other PEC biosensors and a versatile tool in determining a low abundance analyte in bioanalysis and clinical biomedicine. The electrode decorated with the high charge separation efficiency g-C₃N₄/DZP compound for non-enzymatic PEC determination of Ch is depicted in Scheme 1. DZP was coupled with g-C₃N₄ nanosheets through axial coordination, where the nitrogen in g-C₃N₄ acted as an electron donor and the zinc ion of DZP was regarded as the electron acceptor. The energy levels of the LUMO and HOMO of DZP could match with that of the g-C₃N₄ nanosheets.²⁴ Hence, when visible light irradiated the axial coordinative compound, the holes emanated from the VB of g-C₃N₄, then they injected into the HOMO of the DZP, resulting in the high efficiency of the charge separation. Afterward, since Ch could be oxidized by the holes, it could act as an effective electron donor for scavenging of holes, leading to the inhibition of the electron–hole recombination, then the photocurrent intensity enhanced dramatically.

Scheme 1 Schematic illustration of the axial coordination between g-C₃N₄ and DZP and schematic model for the PEC process of g-C₃N₄/DZP composites under visible light.

As shown in Fig. 1, open-circuit voltage-decay measurements were conducted to investigate the recombination kinetics of the structure. The open-circuit voltage after stopping the illumination was measured by the following the technique as reported previously. And it could be analyzed using the approximation derived by Bisquert. The photovoltage decay rate directly related to the electron lifetime by the following equation:^25,26

Here, τ is the potential dependent lifetime, k_B is Boltzmann’s constant, T is the temperature in K, e is the charge of a single electron, and V_oc is the open-circuit voltage at time t. Fig. 1B is the plot of the response time which was obtained by applying eqn (1) to the data in Fig. 1A. By a comparison, the response time data for the g-C₃N₄/DZP nanocomposite was significantly longer, which resulted from that the energy levels of the LUMO and HOMO of DZP could match with those of the g-C₃N₄ nanosheet. The result exhibited long life-time and efficient charge separation was achieved in the axial coordination composites.²⁷ The effective charge separation led to improved IPCE in g-C₃N₄/DZP, which is shown in Fig. S4.† Therefore, the g-C₃N₄/DZP compound possessed a promising application for PEC sensors.

Fig. 1 (A) V_oc time profiles of g-C₃N₄ (a), g-C₃N₄/DZP (b); (B) electron life-time measurements determined from the V_oc decay in the dark by applying eqn (1), g-C₃N₄ (a), g-C₃N₄/DZP (b).

The dendritic zinc phthalocyanine which was an artificial light harvesting system, possessed the special dendritic architecture. Due to the periphery of dendritic macromolecules acting as the energy absorbing and electron transferring antenna, the light harvesting property of DZP was improved dramatically. The photo-physical properties of DZP were characterized by fluorescence spectroscopy in Fig. S5A.† Upon excitation at 290 nm, the fluorescence emission spectra of water-soluble DZP in water was at 678 nm, and a strong fluorescent intensity was observed obviously without the aid of surfactants, which was unprecedented. Fig. S5B† shows the optical absorption spectra of g-C₃N₄ (a) and g-C₃N₄/DZP (b) in aqueous solution, normalized to their most intense absorption bands. The g-C₃N₄ had relatively obvious UV absorption peaks at 231 nm and 328 nm. After the g-C₃N₄ was functionalized by DZP, changes were observed in the spectra. Compared with the pure g-C₃N₄, the absorption bands of the composites were characterized at 223 and 321 nm with 7–8 nm blue shifts, proving that the axial coordination between DZP and g-C₃N₄ successfully formed. Then the axial coordination bond between DZP and g-C₃N₄ underwent charge separation and made the composites to construct supermolecular assemblies of light harvesting and charge separation units, which provided potential perspective for the development of efficient light energy conversion systems. The result in accord with that of Fig. S6.†

The TEM image of g-C₃N₄ (Fig. 2A) showed a distinctly layered structure, revealing that g-C₃N₄ consisted of graphitic planes stacking. The planes were constructed from tri-s-triazine units displayed in the inset of Fig. 2A. The SEM image of g-C₃N₄ shown in Fig. 2B suggested the nanosheets were loose and soft, and the basic unit was a sheet. The large 2D aromatic surface of g-C₃N₄ makes it an ideal substrate in potential uses including biosensing, photoelectrochemistry, and electroanalysis, etc. To gain insight into the morphology and thickness of the nanosheets, the atomic force microscopy (AFM) image was recorded, Fig. 2C. It described the topographic heights of these sheets with the range between 0.3 nm and 8.1 nm, suggesting that g-C₃N₄ consists of only several layers of nanosheets. Additionally, the high water-dispersity is shown in Fig. 2D–F. Especially, the obvious Tyndall phenomenon suggested that the particles of the g-C₃N₄ in solution were stable and distributed quite uniformly in water. The analytical results of XPS of g-C₃N₄ nanosheets displayed that the nanosheets were composed of carbon and nitrogen elements. And it also proved that the atomic ratio of carbon and nitrogen was 3 to 4, verifying that the g-C₃N₄ was successfully synthesized in this work, which was prepared for the further use.

Fig. 2 (A) TEM image of g-C₃N₄, (B) SEM image of g-C₃N₄, (C) AFM image of g-C₃N₄; the photos of diluted g-C₃N₄ nanosheets solution under white light (D) and 390 nm visible light (E), (F) the photo of the Tyndall phenomenon of the g-C₃N₄. Table 1: the results of XPS of g-C₃N₄ nanosheets.

Fig. 3 displays the PEC response of the sensor in the absence and presence of Ch (1 μM). As we can see, the photocurrent density of g-C₃N₄/GCE was higher than that of DZP/GCE which was so low that it can be ignored. The excellent photocurrent benefited from the appealing electronic structure of g-C₃N₄ and the highly efficient conversion of light energy to electricity. However, after the hybrid of DZP and g-C₃N₄ was dropped onto the electrode surface, the photoelectrochemical performance enhanced dramatically, as demonstrated in Fig. 4A–C. It strongly proved that the presence of DZP could overcome the large activation barrier during a short period, allowing the light-initiated reactivity by the conversion of the light energy into chemical energy. The finding results from two aspects, on one hand, the g-C₃N₄ was functionalized spontaneously by dendritic phthalocyanine with the carboxylic groups, which acted as an energy absorbing and electron transferring antenna to enhance the PEC response to visible light. DZP was coupled with g-C₃N₄ nanosheets through axial coordination, where the nitrogen in g-C₃N₄ acting as a Lewis base showed a high adsorption capacity for the zinc ion of DZP. Then the spatial structure of DZP was distorted, leading to the spatial charge separation of photo-generated charge carriers. In addition, there was an electrostatic repulsive force between the electronegative g-C₃N₄ and DZP. The interaction led to the supermolecular ring being distorted further, the Lewis acidity of the Zn ion in DZP became stronger, leading to the stronger axial coordination between g-C₃N₄ and DZP. Therefore an electron donor–acceptor assembly was constructed, resulting in achieving high efficiency and a long lifetime for the charge separation of photo-generated carriers. On the other hand, the energy levels of the LUMO and HOMO of DZP could match with that of the g-C₃N₄ nanosheets, which prevented the recombination of the electron–hole pairs. Thus it could be employed as a potential sensitizer for the photoelectrochemical application of the g-C₃N₄.

Fig. 3 (A) The photocurrent response of various modified electrodes (a) DZP/GCE, (b) g-C₃N₄/GCE, (c) g-C₃N₄/DZP/GCE in 0.1 M PBS (pH 7.0), (B) the photocurrent response of various modified electrodes in 0.1 M PBS (pH 7.0) containing 1 μM Ch, (C) the photocurrent density contrasting pattern of different electrodes in 0.1 M PBS (pH 7.0) and 1 μM Ch solution, (D) CV curves of the g-C₃N₄/DZP/GCE in the absence (a), (a′) and (b), (b′) presence of 1 μM Ch, and (a′), (b′) with irradiation, (a), (b) without irradiation.

Fig. 4 (A) The effect of amount of g-C₃N₄ in g-C₃N₄/GCE (a), the effect of amount of DZP in DZP/GCE, (B) the effect of amount of g-C₃N₄ in g-C₃N₄/DZP/GCE (a), the effect of amount of DZP in g-C₃N₄/DZP/GCE, (C) excitation wavelength vs. photocurrent responses of g-C₃N₄/DZP/GCE in 0.1 M PBS (pH 7.0) containing 1 μM choline.

The contrasting pattern in Fig. 3C showed the photocurrent density at various modified electrodes in the solution containing Ch was higher than that in the blank PBS. Since Ch acted as the effective electron donor for scavenging holes, and the recombination of electron–hole was inhibited, then the photocurrent intensity enhanced dramatically. The cyclic voltammogram of the g-C₃N₄/DZP electrode, as Fig. 3D demonstrates, did not display obvious oxidation peaks in the potential range from 0 to 1.0 V in the presence and absence of 1 μM Ch in 0.1 M PBS (pH = 7) solution, which suggested Ch could not be oxidized under these conditions. However, the CV profile increased in the presence of Ch under irradiation due to the Ch oxidation caused by the holes on the g-C₃N₄. The phenomenon was consistent with that above. Therefore, g-C₃N₄/DZP axial coordinated composites have potential applications in PEC sensors.

In order to optimize the experimental conditions for Ch detection, the effects of the amount of g-C₃N₄ and DZP and the excitation wavelength were investigated. As depicted in Fig. 4A, the photocurrent enhanced obviously with the increasing amount of g-C₃N₄ on the g-C₃N₄ modified electrode. When the g-C₃N₄ amount was greater than 3 mg mL⁻¹, the photocurrent decreased, resulting from the molecular aggregation and large π-conjugation blocked the photo-generated electron transfer. Fig. 4A illustrates the effect of DZP amount on the photocurrent of the modified electrode, depicting the photocurrent almost did not change with the amount of DZP, which strongly proved the photocurrent was generated by g-C₃N₄, and DZP just acted as a sensitizer for g-C₃N₄. Additionally, the amount of g-C₃N₄ and DZP in the g-C₃N₄/DZP/GCE was optimized for choline measurement. As described in Fig. 4B, the photocurrent on the g-C₃N₄/DZP electrode substantially increased with the increased amount of g-C₃N₄ from 0 to 2 mg mL⁻¹ (curve a), then it decreased quickly afterward. Curve b in Fig. 4B depicts that the photoelectrochemical response enhanced with the amount of DZP increasing from 0 to 8 μL, beyond that it decreased dramatically. Therefore, 2 mg mL⁻¹ of g-C₃N₄ and 8 μL DZP solutions were chosen for the further study. The irradiation wavelength was one of the significant factors quite relevant to the photocurrent response. As displayed in Fig. 4C, in the choline solution, the photocurrent density increased as the exciting wavelength was increased from 355 nm to 375 nm, subsequently, the photocurrent density was decreased only a little from 375 nm to 390 nm, then it dramatically decreased beyond 390 nm. The exciting wavelength of 390 nm was in the range of visible light, so it was chosen for the photoelectrochemical determination of choline.

Under the optimum experimental conditions (Fig. 4), the operational stability of the sensor is displayed in Fig. 5A. The relative standard deviation for ten parallel measurements with the g-C₃N₄/DZP modified electrode was 4.3%, indicating an excellent precision. Moreover, the reproducibility of the sensor was estimated by detection of 1 μM Ch with ten biosensors prepared at the different electrodes. A relative standard deviation of 4.58% was attained from ten measurements, giving an acceptable fabrication reproducibility of the biosensor. Moreover, no evident decrease in the photocurrent was observed after 15 days of storage at 4 °C, implying that the fabricated sensor possessed good storage stability.

Fig. 5 (A) The stability of g-C₃N₄/DZP/GCE in 0.1 M PBS (pH 7.0) containing 1 μM Ch, (B) selectivity of PEC biosensor for 0.1 M PBS, 100 μM L-glutamate, fructose, trisodium citrate, glucose, L-aspartate and 1 μM Ch, (C) effect of the concentration of Ch 0.01 μM, 0.05 μM, 0.1 μM, 0.5 μM, 1 μM, 5 μM. (D) The plot of photocurrent density against the logarithm of concentration of Ch.

As Fig. 5B reveals, the present biosensor showed an excellent selectivity for Ch over other agents. To further demonstrate the analytical reliability and real application of the PEC biosensor, the quantitative determination of Ch was developed, as Fig. 5C displays. Hence, by tracking the photocurrent density related to the logarithm of concentration of Ch (Fig. 5D) the curve had a linear range from 10 nM to 5 μM and the detection limit was 3 nM, which was lower than other Ch sensors.^28–32 The prepared biosensor was used for detecting trace amounts of Ch in practical analytical applications as shown in Table S2.†

In summary, an enzyme-free PEC sensing scaffold based on the axial coordinated g-C₃N₄/DZP composites was presented. In this scaffold, by coupling the unique textural and electronic features of g-C₃N₄ with dendritic phthalocyanine through axial coordination, a high charge separation efficiency could be obtained owing to the blocking of recombination of the electron–hole pairs. Ch was introduced in this analytical system which acted as an effective electron donor, further generating an amplified photocurrent. The fabrication of the biosensor which possesses high sensitivity, good reproducibility and long-term stability marks a starting point for further research in this field based on g-C₃N₄ nanosheets, offering the possibility of coupling g-C₃N₄ nanosheets with dendritic phthalocyanine in potential applications including photoelectroanalysis, photocatalytic degradation etc.

Acknowledgements

This project was financially supported by NSFC (21205016, 21275031, 21274021), National Science Foundation of Fujian Province (2011J05020), Education Department of Fujian Province (JA14071, JB14036, JA13068), Foundation of Fuzhou Science and Technology Bureau (2013-S-113) and Fujian Normal University outstanding young teacher research fund projects (fjsdjk2012068).

Notes and references

1. D. Chen, H. Zhang, X. Li and J. Li, Anal. Chem., 2010, 82, 2253.
2. R. Gill, M. Zayats and I. Willner, Angew. Chem., Int. Ed., 2008, 47, 7602.
3. X. M. Zhao, S. W. Zhou, Q. M. Shen, L. P. Jiang and J. J. Zhu, Analyst, 2012, 137, 3697.
4. W. W. Zhao, Z. Y. Ma, P. P. Yu, X. Y. Dong, J. J. Xu and H. Y. Chen, Anal. Chem., 2012, 84, 917.
5. W. J. Wang, J. P. Lei, W. W. Tu and H. X. Ju, Anal. Chim. Acta, 2012, 744, 33.
6. M. Liang, S. Jia, A. Zhu and L. Guo, Environ. Sci. Technol., 2008, 42, 635.
7. Y. J. Li, M. J. Ma and J. J. Zhu, Anal. Chem., 2012, 84, 10492.
8. P. P. Wang, W. J. Dai, L. Ge, M. Yan, S. G. Ge and J. H. Yu, Analyst, 2013, 138, 939.
9. Q. M. Shen, X. M. Zhao, S. W. Zhou, W. H. Hou and J. J. Zhu, J. Phys. Chem. C, 2011, 115, 17958.
10. W. J. Yao, L. G. Alan, S. Nicolas, H. Michael, W. D. Guo, S. Dan, D. Eric and C. Serge, Biosens. Bioelectron., 2013, 42, 556.
11. C. M. Cheng, Y. Huang, X. Q. Tian, B. Z. Zheng, Y. Li, H. Y. Yuan, D. Xiao, S. P. Xie and M. M. Choi, Anal. Chem., 2012, 84, 4754.
12. X. S. Zhou, B. Jin, R. Q. Chen, F. Peng and Y. P. Fang, Mater. Res. Bull., 2013, 48, 1447.
13. B. Chai, T. Y. Peng, J. Mao, K. Li and L. Zan, Phys. Chem. Chem. Phys., 2012, 14, 16745.
14. L. P. Zhang and Z. H. Xia, J. Phys. Chem. C, 2011, 115, 11170.
15. M. Wu, C. Cao and J. Z. Jiang, Nanotechnology, 2010, 21, 505202.
16. H. B. Wang, T. Maiyalagan and X. Wang, ACS Catal., 2012, 2, 781–794.
17. J. Q. Tian, Q. Liu, C. J. Ge, Z. C. Xing, A. M. Asiri, A. O. Al-Youbi and X. P. Sun, Nanoscale, 2013, 5, 8921.
18. N. Karousis, J. Ortiz, K. Ohkubo, T. Hasobe, S. Fukuzumi, A. S. Santos and N. Tagmatarchis, J. Phys. Chem. C, 2012, 116, 20564.
19. C. Ingrosso, A. Petrella, P. Cosma, M. L. Curri, M. Striccoli and A. Agostiano, J. Phys. Chem. B, 2006, 110, 24424.
20. Y. R. Peng, H. Zhang, H. L. Wu, B. Q. Huang, L. Gan and Z. Chen, Dyes Pigm., 2010, 87, 10.
21. H. Q. Yang, Y. R. Peng, L. S. Huang, H. Zhang, Y. H. Wang and S. S. Xie, J. Lumin., 2013, 135, 26.
22. S. Fukuzumi, T. Honda, K. Ohkubo and T. Kojima, Dalton Trans., 2009, 3880.
23. B. K. Chandra, K. Ohkubo, P. A. Karr, S. Fukuzumi and F. D. Souza, Chem. Commun., 2013, 49, 7614.
24. M. M. Liu, R. Liu and W. Chen, Biosens. Bioelectron., 2013, 45, 206.
25. X. J. Zhang, L. J. Mao, D. Zhang and L. Zhang, J. Mol. Struct., 2012, 1022, 153.
26. B. H. Meekins and P. V. Kama, ACS Nano, 2009, 11, 3437.
27. N. Karousis, J. Ortiz, K. Ohkubo, T. Hasobe, S. Fukuzumi, A. S. Santos and N. Tagmatarchis, J. Phys. Chem. C, 2012, 116, 20564.
28. C. Z. Zhao, J. Yu, G. S. Zhao and K. Jiao, Anal. Chem., 2011, 39, 886.
29. Z. Arie, G. Miri and B. Juan, ChemPhysChem, 2003, 4, 859.
30. S. Çevik, S. Timur and Ü. Anik, Microchim. Acta, 2012, 179, 299.
31. H. Dai, X. P. Wu, H. F. Xu, M. D. Wei, Y. M. Wan and G. N. Chen, Electrochem. Commun., 2009, 11, 1599.
32. Y. N. Li, H. Huang, F. P. Shi, Y. Li and X. G. Su, Microchim. Acta, 2013, 180, 1135.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental details. See DOI: 10.1039/c4ra09841b

---