The O/S heteroatom effects of covalent triazine frameworks for photocatalytic hydrogen evolution

Abstract

S/O heterocyclic covalent triazine frameworks (CTFs i.e., CTF-7 and CTF-8) were synthesized using thiophene and furan as building blocks, respectively. The hydrogen evolution rate of CTF-7 is 7430 μmol g⁻¹ h⁻¹, which is about 5.6 times that of CTF-8. Due to their low electronegativity, sulfur heteroatoms are more favorable for charge separation than oxygen heteroatoms in CTFs. This work provides a guiding principle for the design of high efficiency photocatalyst structures.

---

The adjustment and design of the conjugated structure in porous organic semiconductors (POSCs), including covalent organic frameworks (COFs),¹ conjugated microporous polymers (CMPs)² and covalent triazine frameworks (CTFs),³ have attracted enormous attention in enhancing solar energy conversion. The extended π-conjugation structures in POSCs should be closely associated with the energy levels and charge transfer/separation processes.⁴ The energy levels can be regulated by careful selection of conjugated building blocks.⁵ Construction of donor–acceptor (D–A) conjugated units is a recommended route to efficiently improve charge separation.^6,7 However, the effect of the π-electron distribution on charge separation is still not fully considered in the whole D–A delocalization system.

As a nitrogen-rich subclass of POSCs, CTFs are one of the most likely materials to be used for practical applications due to their good thermal and chemical stability, relatively simple synthesis method, long service life and good economic benefits.^3a In addition, it was found that CTFs have great potential value in many applications including photocatalytic hydrogen evolution owing to the rational tunability of such networks.^3a,b,6a,7 Therefore, the abundant nitrogen content and structural adjustability of CTFs endow them with an amazing heteroatom effect (HAE). The heteroatom doping strategy is adopted to facilitate charge separation/transfer by post-modification⁸ and pre-modification.^6a,9 The sulfur doping in CTFs can enhance the adsorption of visible light, reduce free charge recombination and allow rapid separation and transportation of photogenerated electron–holes.^8a Guo et al.^6a demonstrated that heteroatom engineering can reduce charge recombination by adjusting the push–pull interaction between the donor and acceptor. Recently, Thomas et al.^8b reported that there is a significant red shift in light absorbance and the efficiency of charge separation is greatly improved after the protonation of the D–A type imine-linked COFs.

As an assumption, the introduction of heteroatoms such as oxygen and sulfur in the conjugated structure of CTFs may lead to changes in the chemical environment and π-electron distribution, which may lead to changes in photocatalytic performance. Due to the different electronegativities of oxygen and sulfur, the π-electron distribution on the conjugated structure of CTFs should change, which may affect the photocatalytic performance.

Herein, we designed and synthesized CTF-7 with a sulfur (S) heteroatom and CTF-8 with an oxygen (O) heteroatom using thiophene and furan as building blocks (Fig. 1). Sulfur is less electronegative than oxygen, and has a smaller ability to pull electrons than oxygen. The D–A system formed by thiophene and triazine units in CTF-7 belongs to the π-electron polarization distribution, while the D–A system formed by furan and triazine units in CTF-8 is assigned to the π-electron uniform distribution. It was found that the polarized distribution of the π-electron by the S heteroatom is more favorable for charge separation. The hydrogen evolution rate of CTF-7 can be up to 7430 μmol g⁻¹ h⁻¹ under visible light irradiation, which is about 5.6 times that of CTF-8.

Fig. 1 (a) Synthetic routes of CTF-7 and CTF-8 with thiophene and furan building blocks, respectively. (b) Structural representations of CTF-7 and CTF-8.

The two kinds of CTFs (CTF-7 and CTF-8) were synthesized using an aldehyde amidine condensation reaction under the conditions of feeding rate control.⁹ As shown in the Fourier-transformed infrared (FT-IR) spectra of CTF-7 and CTF-8 (Fig. 2a), the characteristic vibrations at 1517 cm⁻¹ and 1354 cm⁻¹ should be attributed to C=N stretching vibration and C–N stretching vibration of triazine units, respectively.¹⁰ From the solid-state cross-polarization magic angle spinning ¹³C nuclear magnetic resonance (CP-MAS ¹³C-NMR) spectra of CTF-7 (Fig. 2b), the chemical shifts at 169.6 ppm and 146.7 ppm should be ascribed to the sp² carbon signal from triazine and α carbon signal from thiophene rings, respectively.¹¹ Moreover, other peaks at around 138.3 and 128.5 ppm can be assigned to phenyl carbons. The calculated chemical shift of the β carbon from furan is about 127.6 ppm, which may be contained in the main peak at round 128.5 ppm. In the matter of CTF-8 (Fig. 2b), the peaks at 169.6 ppm, 138.3 and 128.5 ppm should be assigned to carbon signals from triazine and benzene rings, respectively.¹² The other peaks at around 153.6 ppm and 118.2 ppm can be attributed to α carbon and β carbon of furan building blocks, respectively. The elemental analysis (EA) results indicate that the carbon, nitrogen, oxygen and sulphur contents are close to the theoretical values (Table S1, ESI†). X-ray photoelectron spectroscopy (XPS) analysis further confirmed the triazine formation of CTF-7 and CTF-8 (Fig. S1, ESI†).

Fig. 2 (a) FT-IR spectra, (b) CP-MAS ¹³C-NMR spectra, (c) CO₂ adsorption curves and (d) corresponding pore size distributions of CTF–7 (black) and CTF-8 (red) at 273 k.

Based on the N₂ adsorption isotherm, the surface areas of CTF-7 and CTF-8 calculated by the Brunauer–Emmett–Teller (BET) method were found to be 449 m² g⁻¹ and 5 m² g⁻¹, respectively (Fig. S2, ESI†). However, according to the CO₂ adsorption isotherm, the BET surface areas of CTF-7 and CTF-8 were 384 m² g⁻¹ and 481 m² g⁻¹, respectively (Fig. 2c). The pore size distributions of CTF-7 and CTF-8 are concentrated at 0.45–0.85 nm and 0.52–0.81 nm, respectively (Fig. 2d). Due to the small pore sizes of CTF-7 and CTF-8, the smaller CO₂ molecules are more suitable for porosity tests than N₂ molecules. From the scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images, it can be seen that CTF-7 is obviously a honeycomb morphology, while CTF-8 is a block morphology (Fig. S3a, b and S4a, b, ESI†). As shown in powder X-ray diffraction (PXRD) (Fig. S5, ESI†), both CTF-7 and CTF-8 are amorphous, despite a series of attempts under various conditions, which should be related to the planarity of monomers.^3d,9,19 The order of aromaticity is benzene > thiophene > furan, which may affect the crystallinity orders of CTF-1, CTF-7 and CTF-8. Thermogravimetric analysis (TGA) reveals that both CTF-7 and CTF-8 can be stable up to about 470 °C under a nitrogen environment (Fig. S6, ESI†).

To analyze the O/S heteroatom effects of CTF-7 and CTF-8, the electron paramagnetic resonance (EPR) was employed to evaluate their unpaired electron intensities in the dark. As shown in Fig. 3a, the EPR signal intensity of CTF-8 was higher than that of CTF-7, indicating that the larger delocalized system of CTF-8 makes the distribution of the π electron in CTF-8 more symmetrical than that of CTF-7.^7,13 The π electron distribution characteristics of CTF-7 and CTF-8 fragments were characterized by the isosurface map and colored map derived from Mulliken analysis.¹⁴ The isosurface map shows that the fragment of thiophene and triazine has a lower π electron density than that of the fragment of furan and triazine (Fig. 3c and d). The colored map can further make it clear that the π electron distribution of the thiophene fragment is not as uniform as that of the furan fragment because of the weaker electronegativity of sulfur than oxygen (Fig. 3e and f). Moreover, the color of CTF-8 is deeper than that of CTF-7 (inset of Fig. 3b). From the UV-visible spectra, it can be unequivocally observed that CTF-8 has higher light absorption than CTF-7 in the visible light region, suggesting that CTF-8 has a larger delocalized system than CTF-7 (Fig. 3b). Therefore, all results revealed that CTF-7 and CTF-8 are the π-electron polarization distribution structure and the π-electron uniform distribution, respectively.

Fig. 3 (a) Radical intensity according to EPR in the dark and (b) UV-visible DRS spectra of CTF–7 and CTF-8 (inset: Photographs of CTFs). The isosurface map of the π electron density of typical fragments in CTF-7 (c) and CTF-8 (d) with an isovalue of 0.4. The colored map of the electron density of typical fragments in CTF-7 (e) and CTF-8 (f) by Mulliken analysis.

The properties of the O/S heteroatom of CTFs may affect the photocatalytic activity. We then evaluated the photocatalytic hydrogen evolution performance of CTFs loaded with 1 wt% Pd under visible light irradiation (λ > 420 nm) according to the literature.^10a No hydrogen signal peaks were found by gas chromatography in the absence of light (Fig. S7, ESI†). The CTFs without Pd loading have low photocatalytic hydrogen evolution activity. The highest hydrogen evolution rate (HER) of CTF-7 without Pd loading is about 1530 μmol g⁻¹ h⁻¹ (Fig. S7, ESI†). After adding Pd as the cocatalyst, the HERs of CTF-7 and CTF-8 reach 7430 μmol g⁻¹ h⁻¹ and 1320 μmol g⁻¹ h⁻¹, respectively (Fig. 4a). Moreover, the HER of CTF-7 is about 5.6 times that of CTF-8, implying that CTF-7 with the S heteroatom has higher photocatalytic activity than CTF-8 with the O heteroatom. It is worth mentioning that the photocatalytic hydrogen evolution activity of CTF-7 with 1 wt% Pd is comparable to that of most of the porous organic polymers and inorganic materials in the literature (Table S2, ESI†). In order to examine the stability of CTF-7, a total of 5 photocatalytic cycles for about 25 h were measured (Fig. 4b). After five photocatalytic cycles, the HER of CTF-7 remains to be about 5320 μmol g⁻¹ h⁻¹, indicating good photocatalytic stability. In order to study the effect of catalyst dosage, a series of different amounts of CTF-7 samples were evaluated (Fig. S8, ESI†). The HERs are 9240 μmol g⁻¹ h⁻¹ for 5.0 mg, 7430 μmol g⁻¹ h⁻¹ for 10.0 mg and 4380 μmol g⁻¹ h⁻¹ for 20.0 mg, respectively. Due to the effect of the dosage of the catalyst on light absorption, the hydrogen evolution performance is not proportional to the dosage.⁷ From the XRD, XPS, SEM and TEM characterization, the structure and morphology of the recycled samples (Re-CTF-7 and Re-CTF-8) do not show much change in comparison to CTF-7 and CTF-8 (Fig. S1, S3, S4 and S9, ESI†).

Fig. 4 (a) Time course of H₂ production of CTFs. (b) Recycle tests for H₂ evolution. (c) TSPV curves of CTF-7 and CTF-8 under pulsed laser excitation at 355 nm. (d) Photocurrent response under visible light irradiation (λ > 420 nm). (e) EPR spectra of CTF-7 and CTF-8 with 5,5-dimethyl-1pyrroline N-oxide (DMPO) as the capture agent. (f) Electron density differences between the first excited state and the ground state in CTF-7 and CTF-8 fragments with an isovalue of 0.003 (note: the green color represents the negative electron density difference; the blue color represents the positive electron density difference).

Transient surface photovoltage (TSPV) measurements are widely used to understand the detailed charge separation dynamics, such as the separation direction, mode and efficiency, by analyzing the shape, time scale and intensity of the TSPV spectra.^15,16 As shown in Fig. 4c, the photovoltage of CTF–7 is much higher than that of CTF–8, indicating that the carrier concentration of CTF–7 is higher than that of CTF-8.¹⁷ Noticeably, the charge separation time of CTF–7 (4.6 × 10⁻⁶ s) is shorter than that of CTF-8 (4.2 × 10⁻⁴ s), suggesting that CTF-7 is better for charge separation in comparison with CTF-8. In the steady-state PL spectra (Fig. S10, ESI†), the relative intensity of CTF-7 is higher than that of CTF-8. In the time-resolved PL spectra (Fig. S11, ESI†), the average lifetime of CTF-7 (1.10 ns) is longer than that of CTF-8 (0.96 ns), indicating that CTF-7 is better for the charge separation than CTF-8.¹⁸ Furthermore, as shown in the T–t curves (Fig. 4d), the photocurrent response of CTF-7 is significantly better than that of CTF-8. In Fig. 4e, the response intensity of DMPO-˙O₂⁻ in CTF-7 was much higher than that in CTF-8 under light irradiation, indicating that more electrons and holes are generated in CTF-7 under the same conditions. These results further confirmed that CTF-7 with the S heteroatom is more favorable for charge separation than CTF-8 with the O heteroatom.

To obtain insights into understanding the O/S heteroatom effects on charge separation in the photocatalytic processes, the electron density differences and the molecular orbitals were calculated using the Gaussian and Mulliken softwares. As shown in Fig. S12 (ESI†), the LUMO mostly localizes on the triazine, and the HOMO mostly localizes on the thiophene and furan units, which implies that the excited electrons will migrate from the thiophene and furan units to the triazine units. Based on the simulated electron density difference distribution (Fig. 4f), both the negative and positive areas of thiophene sulfur are larger than those of furan oxygen, suggesting that the excited electrons of sulfur heteroatoms migrate to triazine units more efficiently than oxygen heteroatoms in CTFs.

In summary, we designed and synthesized CTF-7 with sulfur heteroatoms and CTF-8 with oxygen heteroatoms through thiophene and furan building blocks, respectively. The hydrogen evolution rate of CTF-7 reaches 7430 μmol g⁻¹ h⁻¹ which is about 5.6 times that of CTF-8 (1320 μmol g⁻¹ h⁻¹). It was found that sulfur and oxygen heteroatoms lead to a polarized π electron distribution and a symmetric π electron distribution in CTFs, respectively. The polarized π electron distribution structure is more favourable for charge separation in CTFs.

This study was supported by the National Natural Science Foundation of China (Grant No. 52072327 and 21673200) and the Zhongyuan Thousand Talents (Zhongyuan Scholars) Program of Henan Province (202101510004).

Conflicts of interest

There are no conflicts to declare.

References

1. (a) X. Guan, F. Chen, Q. Fang and S. Qiu, Chem. Soc. Rev., 2020, 49, 1357–1384; (b) Y. Li, W. Chen, G. Xing, D. Jiang and L. Chen, Chem. Soc. Rev., 2020, 49, 2852–2868; (c) C. S. Diercks and O. M. Yaghi, Science, 2017, 355, eaal1585; (d) C. Xia, K. O. Kirlikovali, T. H. C. Nguyen, X. C. Nguyen, Q. B. Tran, M. K. Duong, M. T. Nguyen Dinh, D. L. T. Nguyen, P. Singh, P. Raizada, V.-H. Nguyen, S. Y. Kim, L. Singh, C. C. Nguyen, M. Shokouhimehr and Q. V. Le, Coord. Chem. Rev., 2021, 446, 214117; (e) H. Wang, H. Wang, Z. Wang, L. Tang, G. Zeng, P. Xu, M. Chen, T. Xiong, C. Zhou, X. Li, D. Huang, Y. Zhu, Z. Wang and J. Tang, Chem. Soc. Rev., 2020, 49, 4135–4165; (f) X. Chen, K. Geng, R. Liu, K. T. Tan, Y. Gong, Z. Li, S. Tao, Q. Jiang and D. Jiang, Angew. Chem., Int. Ed., 2019, 59, 5050–5091.
2. (a) J. M. Lee and A. I. Cooper, Chem. Rev., 2020, 120, 2171–2214; (b) J. Kosco, M. Sachs, R. Godin, M. Kirkus, L. Francas, M. Bidwell, M. Qureshi, D. Anjum, J. R. Durrant and I. McCulloch, Adv. Energy Mater., 2018, 8, 1802181; (c) Y. Xu, S. Jin, H. Xu, A. Nagai and D. Jiang, Chem. Soc. Rev., 2013, 42, 8012–8031.
3. (a) M. Liu, L. Guo, S. Jin and B. Tan, J. Mater. Chem. A, 2019, 7, 5153–5172; (b) C. Krishnaraj, H. S. Jena, K. Leus and P. Van Der Voort, Green Chem., 2020, 22, 1038–1071; (c) Z. Yang, H. Chen, S. Wang, W. Guo, T. Wang, X. Suo, D. E. Jiang, X. Zhu, I. Popovs and S. Dai, J. Am. Chem. Soc., 2020, 142, 6856–6860; (d) M. Liu, Q. Huang, S. Wang, Z. Li, B. Li, S. Jin and B. Tan, Angew. Chem., Int. Ed., 2018, 57, 11968–11972.
4. (a) X. Wang, L. Chen, S. Y. Chong, M. A. Little, Y. Wu, W. H. Zhu, R. Clowes, Y. Yan, M. A. Zwijnenburg, R. S. Sprick and A. I. Cooper, Nat. Chem., 2018, 10, 1180–1189; (b) S. Wei, F. Zhang, W. Zhang, P. Qiang, K. Yu, X. Fu, D. Wu, S. Bi and F. Zhang, J. Am. Chem. Soc., 2019, 141, 14272–14279.
5. (a) X. Lan, X. Liu, Y. Zhang, Q. Li, J. Wang, Q. Zhang and G. Bai, ACS Catal., 2021, 11, 7429–7441; (b) L. Chen, L. Wang, Y. Wan, Y. Zhang, Z. Qi, X. Wu and H. Xu, Adv. Mater., 2019, e1904433.
6. (a) L. Guo, Y. Niu, H. Xu, Q. Li, S. Razzaque, Q. Huang, S. Jin and B. Tan, J. Mater. Chem. A, 2018, 6, 19775–19781; (b) Z.-A. Lan, Y. Fang, Y. Zhang and X. Wang, Angew. Chem., 2018, 130, 479–483; (c) Z. A. Lan, M. Wu, Z. Fang, X. Chi, X. Chen, Y. Zhang and X. Wang, Angew. Chem., Int. Ed., 2021, 60, 16355–16359.
7. L. Guo, Y. Niu, S. Razzaque, B. Tan and S. Jin, ACS Catal., 2019, 9, 9438–9445.
8. (a) L. Y. Li, W. Fang, P. Zhang, J. H. Bi, Y. H. He, J. Y. Wang and W. Y. Su, J. Mater. Chem. A, 2016, 4, 12402–12406; (b) Y. J. Yang, A. Acharjya, M. Y. Ye, J. Rabeah, S. Li, Z. Kochovski, S. Youk, J. Roeser, J. Gruneberg, C. Penschke, M. Schwarze, T. Wang, Y. Lu, R. van de Krol, M. Oschatz, R. Schomacker, P. Saalfrank and A. Thomas, Angew. Chem., Int. Ed., 2021, 60, 19797–19803.
9. M. Liu, K. Jiang, X. Ding, S. Wang, C. Zhang, J. Liu, Z. Zhan, G. Cheng, B. Li, H. Chen, S. Jin and B. Tan, Adv. Mater., 2019, 31, e1807865.
10. (a) M. Liu, X. Wang, J. Liu, K. Wang, S. Jin and B. Tan, ACS Appl. Mater. Interfaces, 2020, 12, 12774–12782; (b) K. Wang, L. M. Yang, X. Wang, L. Guo, G. Cheng, C. Zhang, S. Jin, B. Tan and A. Cooper, Angew. Chem., Int. Ed., 2017, 56, 14149–14153; (c) S. Zhang, G. Cheng, L. Guo, N. Wang, B. Tan and S. Jin, Angew. Chem., Int. Ed., 2020, 59, 6007–6014.
11. (a) M. J. Bojdys, J. Jeromenok, A. Thomas and M. Antonietti, Adv. Mater., 2010, 22, 2202–2205; (b) Y. Tang, H. Huang, B. Peng, Y. Chang, Y. Li and C. Zhong, J. Mater. Chem. A, 2020, 8, 16542–16550.
12. S. Kuecken, A. Acharjya, L. Zhi, M. Schwarze, R. Schomacker and A. Thomas, Chem. Commun., 2017, 53(43), 5854–5857.
13. L. Shi, K. Chang, H. Zhang, X. Hai, L. Yang, T. Wang and J. Ye, Small, 2016, 12, 4431–4439.
14. (a) L. U. Tian and C. Fei-Wu, Acta Phys.-Chim. Sin., 2011, 27, 2786–2792; (b) T. Lu and Q. Chen, Theor. Chem. Acc., 2020, 139, 25.
15. F. Chen, T. Ma, T. Zhang, Y. Zhang and H. Huang, Adv. Mater., 2021, 33, e2005256.
16. (a) B. Zhang, Y. Lei, R. Qi, H. Yu, X. Yang, T. Cai and Z. Zheng, Sci. China Mater., 2018, 62, 519–526; (b) Y. Lei, L. Gu, W. He, Z. Jia, X. Yang, H. Jia and Z. Zheng, J. Mater. Chem. A, 2016, 4, 5474–5481; (c) K. Lu, Y. Lei, R. Qi, J. Liu, X. Yang, Z. Jia, R. Liu, Y. Xiang and Z. Zheng, J. Mater. Chem. A, 2017, 5, 25211–25219.
17. H. Wang, J. Jia, L. Wang, K. Butler, R. Song, G. Casillas, L. He, N. P. Kherani, D. D. Perovic, L. Jing, A. Walsh, R. Dittmeyer and G. A. Ozin, Adv. Sci., 2019, 6, 1902170.
18. Z. A. Lan, G. Zhang, X. Chen, Y. Zhang, K. A. I. Zhang and X. Wang, Angew. Chem., Int. Ed., 2019, 58, 10236–10240.
19. X. Chen, M. Addicoat, S. Irle, A. Nagai and D. Jiang, J. Am. Chem. Soc., 2013, 135, 546–549.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1cc05619k

---