A nine-fold enhancement of visible-light photocatalytic hydrogen production of g-C₃N₄ with TCNQ by forming a conjugated structure

Abstract

Photocatalytic hydrogen evolution by water splitting has become a very effective way to solve the energy crisis. For use in that process, graphitic carbon nitride (g-C₃N₄) has drawn much attention for its response in the visible region. However, its insufficient sunlight absorption efficiency and easy recombination of photoinduced carriers restrict its photocatalytic activity. Herein, we demonstrate a two-step liquid ultrasonic method in water to synthesize a series of tetracyanoquinodimethane (TCNQ)–C₃N₄ photocatalysts aiming to form a conjugated structure by 7,7,8,8-TCNQ. g-C₃N₄ was treated with APTES firstly on its surface in order to give a better interface contact with TCNQ. Benefiting from the conjugation effect between TCNQ and g-C₃N₄, the separation and transport efficiency of photogenerated carriers were significantly improved. Besides, introducing TCNQ also broadened the absorption region. Both of these points lead to the enhancement of photocatalytic H₂ production rate, with the optimized 5% TCNQ–C₃N₄ giving a rate nearly 9.48 times that of pure g-C₃N₄. Also, 5% TCNQ–C₃N₄ (U) was prepared with unmodified g-C₃N₄, which exhibited a rate only 6.87 times that of pure g-C₃N₄, thus validating the necessity of surface modification. Our work reveals that the rational conjugated structure could modulate the electrical and optical properties of g-C₃N₄, yielding an improvement of photocatalytic activities.

---

1. Introduction

The energy crisis has become a serious problem which threatens human development.¹ Photocatalytic technology is known as an effective solution for producing hydrogen, reducing carbon nitride and degrading organic dyes, and photocatalytic hydrogen evolution has attracted much attention.^2–5 Among various photocatalysts, graphitic carbon nitride (g-C₃N₄), which is considered as a kind of typical metal-free organic semiconductor, has been widely investigated as a promising photocatalyst in H₂ production by water splitting because of its favorable bandgap (2.7 eV) corresponding well with visible-light response, excellent thermal stability, low cost and nontoxicity characteristics. Nonetheless, the fast recombination of photogenerated electron–hole pairs and narrow absorption in the visible region (up to 460 nm) limit its application.^3,6–8 Therefore, various modification methods of g-C₃N₄ have been developed aiming to promote its photocatalytic performance, such as chemical heteroatom doping (B, S, and P),^9–12 morphology control,^13,14 dye sensitization,^15–18 defect engineering, and g-C₃N₄ heterojunctions.²

Herein, we introduce a highly conjugated polymer, TCNQ (7,7,8,8-tetracyanoquinodimethane), with plenty of cyano groups.¹⁹ Many works have confirmed that TCNQ has strong π–π stacking interactions with carbon nanotubes and graphene. The conjugative π-electron structures of polymer hybridized photocatalysts exhibit elevated photoactivity because of rapid photogenerated charger transfer.^19–21 Although conjugated polymers have been demonstrated to form polymer/C₃N₄ heterojunctions,^22–30 there are few researches which have applied TCNQ on pure g-C₃N₄ for hydrogen generation utilizing water splitting with visible light.

To reinforce the interfacial contact between g-C₃N₄ and TCNQ, the cationic surfactant 3-aminopropyltriethoxysilane (C₉H₂₃NO₃Si, APTES) was utilized to change the negative charge of the g-C₃N₄ surface into positive charge,³¹ which intensifies the integration of negative TCNQ with g-C₃N₄ using an electrostatic self-assembly approach.

In this study, we report TCNQ–C₃N₄ composites with a good interfacial contact achieved by surface modification. The g-C₃N₄ synthesized by condensation at high temperature was surface-charge-modified by APTES to get positive charges. Then, the positively charged g-C₃N₄ was interacted with TCNQ which has negative charges. After that, the photocatalytic H₂ production rates from water of as-synthesized (2%, 5%, 10%, 15%) TCNQ–C₃N₄ composites with different doses of TCNQ were investigated under visible light irradiation. All the TCNQ–C₃N₄ photocatalysts show a higher H₂ evolution rate than pure C₃N₄, which is ascribed to the broader response in the visible region and faster transfer of photogenerated electron–hole pairs. The hydrogen production rate of 5% TCNQ–C₃N₄ reaches 425.64 μmol g⁻¹ h⁻¹, which is 9.48 times higher than that of pure g-C₃N₄. Besides, the photocatalytic performance of 5% TCNQ–C₃N₄ (U) composed of TCNQ and unmodified g-C₃N₄ was also studied for the same experimental conditions, which exhibits a poorer H₂ production rate than the 5% TCNQ–C₃N₄ composite, thus confirming the necessity of surface charge modification for electrostatic interaction between TCNQ and g-C₃N₄. This work might offer a new insight into improving the photoactivity of g-C₃N₄ via surface modification and constructing conjugated structures.³²

2. Experimental

2.1 Materials

Urea (H₂NCONH₂), barium sulfate (BaSO₄), anhydrous sodium sulfate (Na₂SO₄), N,N-dimethylformamide (DMF), and ethanol (C₂H₆O) were purchased from Sinopharm. APTES (C₉H₂₃NO₃Si) and TCNQ were purchased from Aladdin. All the materials were used without any further purification.

2.2 Preparation of g-C₃N₄ (bulk)

Pure g-C₃N₄ (bulk) was obtained by thermal polycondensation of carbamide in air atmosphere. The typical preparation method of g-C₃N₄ (bulk) was as follows: 5 g of carbamide was put in a tube furnace, followed by heating to 550 °C for 4 hours to complete the reaction. The heating rate was 2.3 °C min⁻¹.

2.3 Synthesis of APTES-modified g-C₃N₄ (bulk)

The as-prepared g-C₃N₄ (bulk) was modified with APTES as follows. Firstly, 0.5 g of g-C₃N₄ (bulk) was dispersed in 200 mL of ethanol followed by ultrasonic agitation for 20 min. After that, 0.5 mL of APTES was added into the solution, and the obtained mixture was subjected to a reflux operation for 4 h at 65 °C. Lastly, the solution was centrifuged and washed with ethanol, followed by drying at 60 °C for 12 hours to obtain the APTES-modified g-C₃N₄ (bulk).

2.4 Synthesis of TCNQ–C₃N₄ composite photocatalysts

The TCNQ–C₃N₄ composites were synthesized by a liquid ultrasonic agitation method in deionized water as shown in Fig. 1. Firstly, the appropriate amount of g-C₃N₄ (bulk) was added into deionized water and subsequently treated with ultrasonic agitation for 3 hours to get a homogeneous dispersed solution of g-C₃N₄. A pre-prepared TCNQ–DMF solution with a concentration of 5 mg mL⁻¹ was added into the mixture above. Then, the mixture was treated by ultrasonication for an hour and stirring for 24 hours, followed by drying at 80 °C after volatilizing the solvent. Finally, photocatalysts with various mass ratios of TCNQ–C₃N₄ from 2% to 15% were synthesized, and named as (2%, 5%, 10%, 15%) TCNQ–C₃N₄. Besides, a control sample was prepared with unmodified g-C₃N₄ and TCNQ in the same way, aiming to investigate the effect of modification with APTES, which was named as 5% TCNQ–C₃N₄ (U). In all the analyses described below, we abbreviated the samples as bulk, 2%, 5%, 10%, 15% and 5% (U), respectively.

Fig. 1 Schematic illustration of the synthesis of TCNQ–C₃N₄ composite photocatalysts.

2.5 Characterization of the photocatalysts

Zeta potentials of pure g-C₃N₄ (bulk), g-C₃N₄ (APTES), and TCNQ dispersed in aqueous solutions at pH = 6.5 were tested using a Zetasizer3000HSA analyzer (Malvern Instruments). Scanning electron microscopy (SEM, Hitachi S-4800) and high-resolution transmission electron microscopy (HRTEM, JEM-2100) were carried out to explore the morphologies of the samples. Powder X-ray diffraction (XRD) patterns of all samples were measured using an X'Pert powder diffractometer with Cu Kα radiation (λ = 0.15418 nm). Fourier transform infrared (FT-IR) spectroscopy was performed with a Tensor 27 FT-IR spectrometer at a resolution of 4 cm⁻¹, which was utilized to confirm the characteristic functional groups and the conjugated structures of the composites. Electron paramagnetic resonance (EPR) spectra were obtained with a Bruker model ESRA-300 spectrometer at ambient temperature. X-ray photoelectron spectroscopy (XPS) was performed with a monochromatic Al Kα radiation source (Kratos AXIS Supra). UV-visible diffuse reflectance spectroscopy (DRS) was conducted with a Shimadzu UV-3600 spectrophotometer.

Photoelectrochemical measurements were carried out using a CHI760B workstation (Chenhua Instrument) with a three-electrode cell (Pt plate as the counter electrode and Ag/AgCl as the reference electrodes, and 0.5 M Na₂SO₄ solution as the supporting electrolyte). A 300 W Xe lamp (Perfectlight Co., Beijing, China) with a cutoff channel (AM1.5G) served as the simulated solar light source. The working electrodes were fabricated as follows: 20 mg of each photocatalysts was separately dispersed in 20 μL Nafion (5%) aqueous solution and 500 μL ethanol to make a slurry by ultrasonication, which was dip-coated on conductive fluorine tin oxide glass substrates to produce electrodes with ca. 0.8 cm² of active area. These electrolytes were dried completely overnight. The photocurrent intensity measurement was taken at 0 V versus Ag/AgCl with the light on and off. Electrochemical impedance spectroscopy (EIS) was conducted with an AC voltage of 10 mV amplitude in a frequency range of 10⁻²–10⁶ Hz under an open circuit potential.

2.6 Photocatalytic H₂ production performance

The photocatalytic hydrogen evolution activities of the different photocatalysts were determined in a quartz reaction vessel connected with a glass closed gas circulation system. Normally, 10 mg of catalyst powder was dispersed in 30 mL (5 mL methyl alcohol as hole scavenger, 1 wt% Pt as cocatalyst) aqueous solution. The aqueous reactant mixture was irradiated under visible light using a 300 W xenon lamp (Perfectlight Co., Beijing, China) with a 400 nm cutoff filter. During the photocatalytic process, the suspension was kept at room temperature by flowing cooling water. The produced H₂ gas was extracted every half an hour and measured by an online gas chromatograph (FULI 9790) using a thermal conductivity detector and Ar gas carrier. In addition, apparent quantum yield (AQY) measurements were carried out by using a multichannel photochemical reaction system (PCX50B Discover, Perfectlight, Beijing) with monochromatic light (λ = 365, 420, 450, 485, 535, 595, 630 nm). The AQY value was calculated by eqn (1):

3. Results and discussion

3.1 Synthesis and characterization of C₃N₄–TCNQ

Zeta potential analysis was carried out to investigate the superiority of surface modification in deionized water at pH value of 6.5 for TCNQ, g-C₃N₄ (bulk) and g-C₃N₄ (APTES), as shown in Fig. 2. The zeta potential of TCNQ is −35.2 mV which is ascribed to the abundant cyano groups on its surface, while the zeta potential of g-C₃N₄ (bulk) is −19.2 mV. Apparently, it goes without saying that it is somewhat difficult to make a favorable surface interaction by mechanical agitation when the g-C₃N₄ (bulk) and TCNQ have the same electrical properties. After modification with APTES, g-C₃N₄ (APTES) shows a zeta potential of +3.14 mV, which could help form an electrostatic attractive interaction by electrostatic assembly when TCNQ and modified g-C₃N₄ (bulk) are stirred in aqueous solution.

Fig. 2 Zeta potential of TCNQ, g-C₃N₄ (bulk) and g-C₃N₄ (APTES).

The morphologies of the synthesized photocatalysts were scrutinized by field emission SEM and HRTEM. As shown in the HRTEM images of bulk g-C₃N₄ and 5% TCNQ–C₃N₄ photocatalyst in Fig. 3, there are lamellar nanosheets without obvious crystal structure due to the intrinsic amorphous phase of graphitic carbon nitride. The bulk g-C₃N₄ displayed big and smooth layer sheets, while 5% TCNQ–C₃N₄ composite presented thin and unfolded flake sheets with many wrinkles. As shown in Fig. S1,† the series of g-C₃N₄ have very similar stacking lamellar structures, and TCNQ was larger than any other samples. After modification with APTES, g-C₃N₄ (APTES) shows similar structures as pure g-C₃N₄ (bulk), while the same thing happens with (2%, 5%, 10%, 15%) TCNQ–C₃N₄, as shown in Fig. S1(d–g),† which provides plenty of active sites that originated from more g-C₃N₄ hybridizing with TCNQ on the surface.

Fig. 3 TEM images: (a) g-C₃N₄ (bulk); (b) 5% TCNQ–C₃N₄.

The XRD patterns of g-C₃N₄ (bulk), g-C₃N₄ (APTES) and (2%, 5%, 10%, 15%) TCNQ–C₃N₄ are shown in Fig. 4, indicating that the crystal phase of pure graphitic carbon nitride does not change on modification with APTES or loading with TCNQ. Notably, the obvious diffraction peaks of g-C₃N₄ (bulk) in Fig. 4(a) are observed at 13.1° and 27.7° corresponding to the (100) and (002) facets (PDF no. 87-1526, JCPDS), respectively. In addition, the peak at 13.1° is attributed to the repeated facet of tri-s-triazine units, while that at 27.7° arises from the interplanar stacking of π-conjugated system. After surface modification, g-C₃N₄ (APTES) which is abbreviated as (APTES) in Fig. 4(a) exhibits the characteristic peaks at the same positions, thus indicating that surface treatment does not alter the crystalline structure. From Fig. 4(b), only the intensities of the 27.7° peak are improved with increasing content of TCNQ, thus suggesting that TCNQ–C₃N₄ composites are successfully synthesized having consistent crystal structure without lattice expansion.

Fig. 4 XRD patterns: (a) g-C₃N₄ (bulk) and g-C₃N₄ (APTES); (b) g-C₃N₄ (bulk) and (2%, 5%, 10%, 15%) TCNQ–C₃N₄.

The chemical structures of g-C₃N₄ (bulk) and (2%, 5%, 10%, 15%) TCNQ–C₃N₄ were investigated by FT-IR spectroscopy (Fig. 5). For pristine g-C₃N₄, a series of peaks at 1200 cm⁻¹ to 1650 cm⁻¹ belonged to the characteristic stretching vibration of C–N heterocycles, and the sharp peak at 811 cm⁻¹ was attributed to the out-of-plane breathing mode of s-triazine units. A vibration peak of TCNQ was clearly observed at 2223 cm⁻¹, which originated from the stretching of C≡N, while similar peaks also appeared in the spectra of the TCNQ–C₃N₄ composites. Obviously, the stretching of C–N heterocycles and the breathing mode of triazine units in TCNQ–C₃N₄ became weaker as shown in Fig. 5(a), suggesting the formation of intermolecular interactions between carbon nitride and TCNQ to construct rigid conjugated structures.²⁷ Besides, the vibration band at 1635 cm⁻¹ of pure g-C₃N₄ shifted to 1640 cm⁻¹ in the spectra of TCNQ–C₃N₄ composites as shown in Fig. 5(b), which indicated that charges may transfer from g-C₃N₄ to TCNQ.²⁸ And, a stronger EPR intensity of 5% TCNQ–C₃N₄ compared with bulk g-C₃N₄ is observed in Fig. S2,† revealing a higher concentration of unpaired electrons, which was attributed to the cyano groups in TCNQ.

Fig. 5 FT-IR spectra in the regions (a) 2500 to 700 cm⁻¹ and (b) 1700 to 700 cm⁻¹ of g-C₃N₄ (bulk) and (2%, 5%, 10%, 15%) TCNQ–C₃N₄.

To confirm the surface interfacial interactions and to elucidate the corresponding chemical states of the photocatalysts, XPS analysis was performed for g-C₃N₄ (bulk) and 5% TCNQ–C₃N₄, as shown in Fig. 6. As seen in the C 1s XPS spectra of Fig. 6(a), three photoelectron peaks are present at binding energies of 284.8 eV, 285.6 eV and 288.4 eV, which are assigned to residual C, sp³-coordinated carbon nitride (C–N) and sp²-bonded C in the C–N heterocycle, respectively.³² Moreover, the C 1s XPS peaks of 5% TCNQ–C₃N₄ are fitted to three positions at 284.8 eV, 286.1 eV and 288.5 eV, with the 286.1 eV peak being associated with –C≡N.³³ The binding energy shifts in the binary hybrids are ascribed to an intense interaction between the two components.³⁴ Besides, the peak area of the 286.1 eV peak exhibited a decrease which may be attributed to the sp³ hybridization transforming to sp² hybridization. For the N 1s XPS spectrum of g-C₃N₄ (bulk) in Fig. 6(b), four peaks are observed at 398.9 eV, 400.2 eV, 401.5 eV and 404.4 eV which can be attributed to the sp²-bonded N (C–N=C) in the C–N heterocycle, tertiary N bonded to carbon atoms in the form of N–C₃, amino groups (C–N–H) and π excitations respectively.³⁵ Similarly, the binding energies of N 1s for 5% TCNQ–C₃N₄ shift to higher values by ca. 0.2 eV, which is attributed to the strong interactions between TCNQ and carbon nitride.³²

Fig. 6 XPS spectra of (a) C 1s and (b) N 1s for g-C₃N₄ (bulk) and 5% TCNQ–C₃N₄.

The photo-absorption properties of g-C₃N₄ (bulk) and (2%, 5%, 10%, 15%) TCNQ–C₃N₄ were investigated by UV-visible absorption spectra, as shown in Fig. 7. The g-C₃N₄ (bulk) exhibited negligible absorption above 400 nm, while (2%, 5%, 10%, 15%) TCNQ–C₃N₄ photocatalysts displayed obvious absorption in the visible light region. Remarkably, two charge-transfer bands of the TCNQ–C₃N₄ composites were observed at 497 nm and 659 nm, which were enhanced with increasing content of TCNQ. This may be attributed to the intense conjugation effect of cyano groups derived from TCNQ in the TCNQ–C₃N₄ composites, thereby producing more photoinduced carriers and improving the photosensitivity under visible light.

Fig. 7 UV-visible DRS of g-C₃N₄ (bulk) and (2%, 5%, 10%, 15%) TCNQ–C₃N₄.

3.2 Photocatalytic performance

The photocatalytic hydrogen production rate of the pure graphitic carbon nitride and TCNQ–C₃N₄ composites was measured using methanol as sacrificial agent to quench generated holes during the photocatalytic process under visible light illumination (λ ≥ 400 nm). As displayed in Fig. 8, the TCNQ–C₃N₄ composites gave higher hydrogen production rates than pure g-C₃N₄ (bulk), with pure g-C₃N₄ (bulk) exhibiting a low H₂ production rate of 44.92 μmol g⁻¹ h⁻¹. The enhanced hydrogen production rates could be attributed to the abundant cyano groups which were introduced into the TCNQ–C₃N₄ conjugated structures, thus broadening the visible-light absorption region and improving the separation efficiency of photogenerated carriers. Apparently, 5% TCNQ–C₃N₄ presented the highest H₂ evolution rate of 425.64 μmol g⁻¹ h⁻¹, which was 9.48 times higher than that of the pure g-C₃N₄. Further increasing the content of TCNQ results in a decrease in photoactivity of the TCNQ–C₃N₄ composites, meaning that the yield of H₂ does not increase linearly with the content of TCNQ. It was found that further increasing the mass ratios of the TCNQ beyond the optimized ratio at 5% led to a decrease in the photoactivity. This could be attributed to excess TCNQ not only leading to fewer reaction sites but also causing less light to enter into the suspension system. Besides, we also carried out the hydrogen evolution experiment with 5% TCNQ–C₃N₄ (U) aiming to confirm the necessity of surface modification of g-C₃N₄. Not surprisingly, 5% TCNQ–C₃N₄ (U) showed a lower H₂ evolution rate of 308.64 μmol g⁻¹ h⁻¹, which was lower than that of 5% TCNQ–C₃N₄, being only 6.87 times higher than that of pure g-C₃N₄.

Fig. 8 H₂ production of g-C₃N₄ (bulk), (2%, 5%, 10%, 15%) TCNQ–C₃N₄ and 5% TCNQ–C₃N₄ (U).

Moreover, we compared the DRS and wavelength-dependent AQY for the 5% TCNQ–C₃N₄ photocatalyst under various monochromatic light illuminations, which exhibited the best hydrogen production performance among as-synthesized samples. As shown in Fig. 9, the AQY values were in accordance with the changing curve of the DRS line. The highest value among the different monochromatic light wavelengths was at 595 nm with the AQY reaching 1.79%. Besides, AQY values of 0.34, 0.50, 0.48, 0.30, 0.50, and 0.16% were obtained under λ = 365, 420, 450, 485, 535, and 630 nm monochromatic illumination for 5% TCNQ–C₃N₄, respectively.

Fig. 9 Wavelength-dependent AQY and DRS of 5% TCNQ–C₃N₄ (λ = 365, 420, 450, 485, 535, 595 and 630 nm).

3.3 Photoelectrochemical properties

To further investigate the charge-transfer properties of the TCNQ–C₃N₄ composites, transient photocurrent measurements were carried out. As seen from Fig. 10, all TCNQ–C₃N₄ electrodes showed significantly higher photocurrent density compared with the pure g-C₃N₄ (bulk) during on/off irradiation cycles, which agrees with their photocatalytic H₂ production performance. As expected, 5% TCNQ–C₃N₄ showed the highest photocurrent response of 45.85 μA cm⁻², nearly 17.24 times as high as that of pristine g-C₃N₄ (bulk) of 2.66 μA cm⁻², reflecting a better light response and more efficient separation of photoinduced electrons and holes. The higher transfer and separation efficiency of photoinduced carriers and the broader visible-light responses are caused by the interaction between TCNQ and g-C₃N₄, as a result of the strongly conjugated structure.

Fig. 10 Photocurrent density of g-C₃N₄ (bulk) and (2%, 5%, 10%, 15%) TCNQ–C₃N₄.

Beyond that, the transfer efficiency of photogenerated electrons was further confirmed by the EIS results as shown in Fig. 11. The Nyquist plots for 5% TCNQ–C₃N₄ showed the minimal semicircle arc at high frequencies compared to the other samples, validating the reduced charge-transfer resistance and fast interfacial electron transfer of the 5% TCNQ–C₃N₄ system.

Fig. 11 EIS Nyquist plots of g-C₃N₄ (bulk) and (2%, 5%, 10%, 15%) TCNQ–C₃N₄ at open circuit voltage of 10 mV.

3.4 Possible photocatalytic mechanism

On the basis of the above analyses, a possible mechanism for the enhanced photocatalytic performance of the TCNQ–C₃N₄ composites could be proposed as shown in Fig. 12. In fact, both g-C₃N₄ and TCNQ can generate electrons and holes under visible light. It is regrettable that generated photocarriers are easily recombined in g-C₃N₄, which limits its photoactivity. In the present work, the TCNQ–C₃N₄ composites were successfully prepared via the intermolecular interaction between TCNQ and g-C₃N₄. Since both materials possess plenty of delocalized conjugated π-electrons, this allows the construction of strongly conjugated structures in order to transfer electrons, which has been confirmed by FT-IR analysis (Fig. 5). Therefore, the improvement of the photocatalytic performance is attributed to the interactions of π-conjugated electrons in the Z-scheme.

Fig. 12 Schematic band structure and photoinduced charge transfer of TCNQ–C₃N₄ under visible light irradiation.

In parallel, new absorption peaks appear in the visible region at 497 nm and 659 nm, whose intensities become stronger with increasing content of TCNQ, contributing to improve the utilization of visible light and enhance the photocatalytic activity of the TCNQ–C₃N₄ composites. This result reveals that inducing TCNQ could improve the optical absorption intensity in the visible-light region of g-C₃N₄.

We speculate that the electrons would be excited to the conduction band (CB) of g-C₃N₄ from its valence band (VB), and then transferred to the LUMO of TCNQ, leaving the holes in the VB of g-C₃N₄, as shown in Fig. 12. Meanwhile, the holes in the VB of g-C₃N₄ are consumed by methyl alcohol, thus resulting in efficient separation of photogenerated carriers in TCNQ–C₃N₄ and a much enhanced photocatalytic performance.^26–28 However, excess TCNQ would reduce the photoactivities of the composites due to fewer reaction sites and less incident light acting on the surface of g-C₃N₄.

In general, the remarkably improved photocatalytic H₂ production performance of TCNQ–C₃N₄ is mainly ascribed to the introduction of TCNQ, which not only enhances the photo-absorption but also promotes the separation of photoinduced holes and electrons effectively.

4. Conclusions

In conclusion, a series of TCNQ–C₃N₄ photocatalysts were synthesized via a two-step liquid ultrasonic method in water by modifying the surface of graphitic carbon nitride with APTES, which leads to excellent photocatalytic H₂ evolution performance. The optimum photocatalytic H₂ production rate of 5% TCNQ–C₃N₄ (425.64 μmol g⁻¹ h⁻¹) is nearly 9.48 times higher than that of pure g-C₃N₄ (bulk) (44.92 μmol g⁻¹ h⁻¹), while that of 5% TCNQ–C₃N₄ (U) (308.64 μmol g⁻¹ h⁻¹) is about 6.87 times higher than that of pure carbon nitride, thus confirming the importance of surface modification. In that way, the main reason for the improvement of H₂ production is ascribed to the construction of a strongly conjugated structure, which efficiently promotes the separation of photoinduced carriers and widely broadens the absorption in the visible-light region, thus boosting photocatalytic activity. All in all, this work demonstrates the construction of molecular conjugated structures between TCNQ and graphitic carbon nitride, putting forward a new insight into promising conjugation effects in photocatalytic performance.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by National Natural Science Foundation of China under grant no. 51972283 and 91833301, and Zhejiang Provincial Natural Science Foundation of China under grant no. LY17E020005.

References

1. R. Asahi, T. Morikawa, H. Irie and T. Ohwaki, Chem. Rev., 2014, 114, 9824–9852.
2. J. Chen, C. Dong, D. Zhao, Y. Huang, X. Wang, L. Samad, L. Dang, M. Shearer, S. Shen and L. Guo, Adv. Mater., 2017, 29, 1606198.
3. S. Chu, Y. Wang, Y. Guo, J. Feng, C. Wang, W. Luo, X. Fan and Z. Zou, ACS Catal., 2013, 3, 912–919.
4. Z. Lou, Y. Li, L. Zhu, W. Xie, W. Niu, H. Song, Z. Ye and S. Zhang, J. Mater. Chem. A, 2017, 5(6), 2732–2738.
5. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38.
6. X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2009, 8, 76–80.
7. W. J. Ong, L. L. Tan, Y. H. Ng, S. T. Yong and S. P. Chai, Chem. Rev., 2016, 116, 7159–7329.
8. H. Wang, L. Zhang, Z. Chen, J. Hu, S. Li, Z. Wang, J. Liu and X. Wang, Chem. Soc. Rev., 2014, 43, 5234–5244.
9. X. Han, C. Yao, A. Yuan, F. Xi, X. Dong and J. Liu, Mater. Res. Bull., 2018, 107, 477–483.
10. S. Yan, Z. Li and Z. Zou, Langmuir, 2010, 26, 3894–3901.
11. K. Wang, J. Fu and Y. Zheng, Appl. Catal., B, 2019, 254, 270–282.
12. J. Ran, T. Ma, G. Gao, X. Du and S. Qiao, Energy Environ. Sci., 2015, 8, 3708–3717.
13. Z. Huang, F. Li, B. Chen, T. Lu, Y. Yuan and G. Yuan, Appl. Catal., B, 2013, 136–137, 269–277.
14. M. Shalom, S. Inal, C. Fettkenhauer, D. Neher and M. Antonietti, J. Am. Chem. Soc., 2013, 135, 7118–7121.
15. S. Kushwaha and L. Bahadur, J. Lumin., 2015, 161, 426–430.
16. Z. Li, Y. Wu and G. Lu, Appl. Catal., B, 2016, 188, 56–64.
17. K. Takanabe, K. Kamata, X. Wang, M. Antonietti, J. Kubota and K. Domen, Phys. Chem. Chem. Phys., 2010, 12, 13020–13025.
18. Y. Zheng, J. Liu, J. Liang, M. Jaroniec and S. Qiao, Energy Environ. Sci., 2012, 5, 6717.
19. X. Guegano, A. Kanibolotsky, C. Blum, S. Mertens, S. Liu, A. Neels, H. Hagemann, P. Skabara, S. Leutwyler, T. Wandlowski, A. Hauser and S. Decurtins, Chemistry, 2009, 15, 63–66.
20. J. Lu, X. Qu, G. Peleckis, J. Boas, A. Bond and L. Martin, J. Org. Chem., 2011, 76, 10078–10082.
21. X. Mei and J. Ouyang, Langmuir, 2011, 27, 10953–10961.
22. J. Chen, X. Tao, L. Tao, H. Li, C. Li, X. Wang, C. Li, R. Li and Q. Yang, Appl. Catal., B, 2019, 241, 461–470.
23. X. Han, D. Xu, L. An, C. Hou, Y. Li, Q. Zhang and H. Wang, Appl. Catal., B, 2019, 243, 136–144.
24. S. Patra, S. Rahut and J. Basu, New J. Chem., 2018, 42, 18598–18607.
25. R. Sprick, J. Jiang, B. Bonillo, S. Ren, T. Ratvijitvech, P. Guiglion, M. Zwijnenburg, D. Adams and A. Cooper, J. Am. Chem. Soc., 2015, 137, 3265–3270.
26. F. Yang, F. Xia, J. Hu, C. Zheng, J. Sun and H. Yi, RSC Adv., 2018, 8, 1899–1904.
27. F. Yu, Z. Wang, S. Zhang, H. Ye, K. Kong, X. Gong, J. Hua and H. Tian, Adv. Funct. Mater., 2018, 28, 1804502.
28. M. Zhang, W. Yao, Y. Lv, X. Bai, Y. Liu, W. Jiang and Y. Zhu, J. Mater. Chem. A, 2014, 2, 11432–11438.
29. Z. Zhang, L. Zhang, X. Yan, H. Wang, Y. Liu, C. Yu, X. Cao, L. van Eijck and B. Wen, J. Power Sources, 2019, 410–411, 162–170.
30. W. Zhou, T. Jia, H. Shi, D. Yu, W. Hong and X. Chen, J. Mater. Chem. A, 2019, 7, 303–311.
31. F. Wang, Y. Zhou, X. Pan, B. Lu, J. Huang and Z. Ye, Phys. Chem. Chem. Phys., 2018, 20, 6959–6969.
32. M. Zhu, S. Kim, L. Mao, M. Fujitsuka, J. Zhang, X. Wang and T. Majima, J. Am. Chem. Soc., 2017, 139, 13234–13242.
33. W. Wang, H. Zhang, S. Zhang, Y. Liu, G. Wang, C. Sun and H. Zhao, Angew. Chem. Int. Ed., 2019, 58, 16644–16650.
34. Z. Zhang, K. Liu, Z. Feng, Y. Bao and B. Dong, Sci. Rep., 2016, 6, 19221.
35. Q. Xu, B. Zhu, B. Cheng, J. Yu, M. Zhou and W. Ho, Appl. Catal., B, 2019, 255, 117770.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9ra10819j

---