English

• With the exhaustion of fossil resources and the related serious environmental pollution problems, it is highly demanded to exploit renewable energy sources. Since 1991, dye-sensitized solar cells (DSSCs) have been developed very fast owing to their relatively high energy-conversion efficiency (PCE) and low cost availability [1-3]. To date, lots of excellent dyes have been successfully developed, which can be classified as ruthenium complex sensitizers [4], metal-free organic dyes [5-7] and porphyrin dyes [8, 9]. Among them, porphyrins are particularly promising due to their structural similarity to chlorophylls, strong absorption, excellent redox behavior and high efficiencies [10-21]. However, despite the strong light absorption in the Soret (400-450 nm) and Q (550-650) bands, obvious absorption drawbacks of porphyrins are observed in the near-infrared (NIR) region, which limits the light-harvesting in this range [22-25]. To address this issue, rational design of porphyrin dyes has been demonstrated to be effective. For example, the insertion of a triple bond between the donor group and the porphyrin framework was applied in a number of dyes [10, 18, 26]. Recently, incorporating an auxiliary electron accepting benzothiadiazole (BTD) unit has been demonstrated to be an effective approach for extending the π-conjugation and shifting the absorption to longer wavelengths [27-29]. However, one adverse effect of these approaches is the severe dye aggregation and charge recombination induced by the extended π-conjugated framework, which may result in decreased photovoltages (V_oc)[30-32]. To overcome this disadvantage, introduction of multiple bulky groups to the donor may be effective for improving the V_oc by suppressing the dye aggregation [33-37]. In addition, electrolyte is also an important factor for achieving efficient DSSCs [38]. In this regard, the Co³⁺/Co²⁺ based electrolyte has been widely used in recent years owing to their weak visible light absorption, high photovoltages induced by its higher redox potential and weak corrosiveness towards metallic conductors [39-41].

  Herein, we report a series of novel porphyrin dyes by modifying XW4 [8]. Initially, XW57 (Fig. 1) was designed by incorporating a BTD unit as the auxiliary acceptor to improve the light harvesting ability in the NIR region. As expected, it exhibits a higher photocurrent (J_sc) value (13.72 mA/cm²) than that of XW4 (12.82 mA/cm²). On the other hand, the carbazole donor group was further modulated by introducing bulky dihexyloxyphenyl groups to afford XW58, which is finally wrapped with ten alkoxy chains. Consequently, it exhibits a relatively higher V_oc value of 844 mV relative to those of XW4 and XW57, which mainly result from the efficient suppression of dye aggregation, accompanied with preventing the Co³⁺/Co²⁺ ions in the electrolyte from approaching the TiO₂ surface. The respective strategy of long wavelength absorption (higher J_sc) or anti-aggregation (higher V_oc) results in enhanced efficiencies of 7.15% (XW57) and 6.54% (XW58), compared with that of XW4 (6.49%) using the cobalt electrolyte. By combining these two approaches, we further designed and synthesized XW59. The simultaneous improvement of J_sc and V_oc has been achieved through red-shifted absorption and suppressed dye aggregation. Finally, the DSSCs based on XW59 exhibit the highest photovoltaic efficiency of 7.34% using the cobalt electrolyte.

  Figure 1

  

  Figure 1.  Molecular structures of porphyrin dyes XW4 and XW57-XW59.

  DownLoad: Full-Size Img PowerPoint

  The synthetic routes for the porphyrin dyes XW57-XW59 are depicted in Scheme S1 (Supporting information). The donor and acceptor moieties were successively introduced to the porphyrin framework via Sonogashira cross-coupling reactions, and the target products were finally obtained by hydrolysis of the ester to generate the anchoring carboxylic groups. All the synthetic products have been fully characterized by NMR and mass spectra (Figs. S1-S13 in Supporting information).

  The UV-vis absorption spectra of XW4 and XW57-XW59 in THF are shown in Fig. 2a, with the corresponding data listed in Table S1 (Supporting information). XW58 displays a typical intense Soret band around 459 nm and a less intense Q band around 664 nm. Compared with XW4, the bulky dihexyloxyphenyl groups introduced in XW58 cause negligible red-shift of both Soret and Q bands. Similarly, slightly red-shifted absorption spectrum is observed for XW59, compared with that of XW57. On the other hand, incorporation of the BTD groups into XW57 and XW59 leads to dramatically broadened absorption and red-shifted Q-bands centered around 683 nm with a striking absorption onset at ca. 734 nm. This observation can be rationalized by the strength-ened intramolecular charge transfer (ICT) effect due to the strong electron-deficient character of the BTD unit. The broadened and red-shifted absorption spectra are favorable for light-harvesting and are expected to afford better cell performance. Upon anchoring to 3 μm thick TiO₂ films, all three dyes exhibit blue-shifted Q bands relative to those in THF (Table S1 and Fig. S14 in Supporting information) because of the combined effect of deprotonation and aggregation [42]. Notably, insertion of the auxiliary BTD unit in XW57 results in a blue shift of ca. 19 nm, more obvious than that of 16 nm observed for XW4, denoting aggravated H-type aggregation [15] resulting from BTD unit induced π-conjugation framework extension. A similar trend is also observed between XW58 and XW59.

  Figure 2

  

  Figure 2.  Absorption spectra of dyes XW4 and XW57-XW59. (a) in THF solutions, (b) on TiO₂ films (3 μm in thickness).

  DownLoad: Full-Size Img PowerPoint

  On the other hand, the corresponding blue-shift of ca. 14 nm observed for XW58 is smaller than that observed for XW4 (16 nm), indicating that the bulky groups attached on the carbazole donor effectively suppress the aggregation. Similar observations could be found between XW57 and XW59. Finally, the trend of blue shifts on TiO₂ films lie in the order of XW58 < XW4 < XW59 < XW57, which is consistent with the sequence of the adsorption amounts, i.e., 3.84×10 ^-8, 4.02×10 ^-8, 4.31×10 ^-8 and 4.55×10 ^-8 mol/cm² for XW58, XW4, XW59, and XW57, respectively (Table S4 in Supporting information). On the other hand, both the Soret band and Q-bands are broadened due to interactions between dye molecules and/or between the dyes and nanocrystalline TiO₂, (Fig. 2b and Fig. S14), which is beneficial for further improving light-harvesting ability and enhancing J_sc.

  Cyclic voltammetry was employed to investigate the oxidation potentials and evaluate the possibility of electron injection and dye regeneration from the thermodynamic point of view (Table S1 and Fig. S15 in Supporting information). The estimated ground-state oxidation potentials (E_ox) of XW57-XW59 were found to be 0.88 V, 0.85 V and 0.87 V, respectively (Table S1) vs. normal hydrogen electrode (NHE), which are obviously more positive than the potential of the Co³⁺/Co²⁺ redox couple (~0.56 V), indicative of enough driving forces for regenerating the dyes. In addition, the excited-state oxidation potentials (E_ox^*) of XW57-XW59 were calculated to be -0.92 V, -1.00 V and -0.92 V, respectively. All these E_ox^* values are more negative than that of the TiO₂ conduction band (~-0.5 V), energetically allowing efficient electron injection into TiO₂ from the excited sensitizers [43].

  To further understand the structure dependent electron distribution in the frontier molecular orbitals and the absorption spectra, we employed density functional theory (DFT), and time-dependent density functional theory (TDDFT) calculations using the Gaussian 09 program package [44-47]. The electrons of the HOMOs of XW57-XW59 mainly distribute over the carbazole donor moiety and porphyrin, while the LUMOs predominantly delocalized over the carboxyphenyl acceptor moiety and porphyrin (Fig. S17 in Supporting information). Thus, electron transfer from the HOMO to the LUMO can be easily accessible from the donor to the anchoring carboxyphenyl moiety, enabling electron injection from the LUMO to the conduction band of TiO₂. The simulated absorption spectra are shown in Fig. S18 (Supporting information), and the corresponding data are listed in Tables S2 and S3 (Supporting information). The trend of theoretical results for XW57-XW59 is in good agreement with that obtained from the absorption spectra.

  The photovoltaic performances of XW57-XW59 were tested using the cobalt-based electrolyte. The corresponding photocur-rent density-voltage (J-V) curves and monochromatic incident photon to current conversion efficiency (IPCE) spectra are shown in Fig. 3, with the corresponding photovoltaic parameters (J_sc, V_oc, FF, PCE) listed in Table 1. The IPCE spectra for all three dyes exhibit a broad spectral response within a large wavelength range (Fig. 3b), indicating that all the dyes can efficiently convert the visible light into photocurrent. Notably, remarkably broadened IPCE action spectra, absorbing in the panchromatic visible region and part of the NIR region, were achieved for XW57 and XW59 owing to the presence of the BTD unit in the molecules. Both XW57 and XW59 exhibit dramatically red-shifted IPCE onset wavelengths of ca. 820 nm relative to that of 740 nm for XW58. Introduction of the BTD unit leads to stronger intramolecular charge transfer effect, resulting in the broadened visible region absorption. As a result, J_sc values of 13.72 and 13.19 mA/cm² were obtained for XW57 and XW59, respectively, in comparison to an obviously lower J_sc value of 11.79 mA/cm² for XW58 (Fig. 3a). Compared with XW4 (12.82 mA/cm², 781 mV), XW57 shows a higher J_sc and a lower V_oc. These observations may be rationalized by the introduction of the BTD unit which gives the broadened IPCE but more severe dye aggregation. Finally, a higher efficiency of 7.15% was achieved for XW57, compared with that of 6.49% obtained for XW4. On the other hand, XW58 exhibits the highest V_oc of 844 mV owing to its relatively small π-conjugation size and the presence of the bulky alkoxy chains on the carbazole unit, which is favorable for suppressing the dye aggregation and preventing the Co³⁺/Co²⁺ ions in the electrolyte from approaching the TiO₂ surface [48-50]. However, a dramatic increase of V_oc for XW58 is accompanied with the sacrifice of J_sc.

  Figure 3

  

  Figure 3.  (a) J-V curves, (b) IPCE action spectra, plots of (c) C_μ and (d) t versus potential bias for DSSCs based on XW57-XW59.

  DownLoad: Full-Size Img PowerPoint

  Table 1

  

  Table 1.  Photovoltaic parameters of the solar cells sensitized by XW4 and XW57-XW59 under simulated AM 1.5 G full sunlight (100 mW/cm²). The active area is 0.12 cm².

  DownLoad: CSV

  As a result, XW58 exhibits an efficiency of 6.54%, slightly higher than that of XW4 (6.49%), but lower than that of XW57 in spite of its higher V_oc, indicating that synergistic enhancement of V_oc and J_sc is necessary for achieving good photovoltaic performance. Thus, XW59 contains both additional alkoxy chains/bulky groups and a BTD unit, which may effectively suppress dye aggregation and charge recombination as well as broadening the absorption. As expected, the J_sc and V_oc values of XW59 are simultaneously enhanced to 793 mV and 13.19 mA/cm², respectively, afford the highest efficiency of 7.34%. Finally, the PCEs follow the order of XW4 (6.49%) XW58 (6.54%) < XW57 (7.15%) < XW59 (7.34%).

  Generally, alteration of the photovoltage originates from a shift of the TiO₂ electron quasi-Fermi-level, which may be related to two main reasons: (i) a shift in the TiO₂ conduction band edge (E_CB), which can be evidenced from the chemical capacitance (C_μ), and (ii) fluctuation of electron density, which is related to the electron lifetime (τ) governed by the charge recombination rate [51, 52]. Thus, electrochemical impedance spectroscopy (EIS) was recorded in the dark at a series of bias voltages to obtain the corresponding C_μ and t values. As shown in Fig. 3c, the C_μ values of the cells at a fixed bias voltage rank in the order of XW58 < XW59 < XW57, indicative of more positive shifts of the TiO₂ conduction band, consistent with the order of the decreasing V_oc values. In addition, the electron lifetimes of XW58 and XW59 -based cells are longer than those based on XW57 (Fig. 3d), indicating that XW57 suffers from more serious dye-aggregation/charge recombination induced by the absence of the bulky groups, which leads to the lowest observed V_oc value (vide supra). On the other hand, the electron lifetime for XW58 is longer than that of XW59, which may be related to its relatively small conjugation framework. These results indicate that the V_oc values of the porphyrin dye-sensitized cells are governed by both the TiO₂ conduction band position and the charge recombination rate.

  In summary, three porphyrin dyes XW57-XW59 have been rationally designed and synthesized by introducing bulky sub-stituents into the carbazole-based donor as well as an auxiliary electron-accepting BTD group. As expected, introduction of a strong electron accepting BTD unit in XW57 resulted in a remarkably red-shifted and broadened Soret band in the range of 400-500 nm. While the onset wavelength of absorption was red-shifted from 695 nm (XW4) to 730 nm due to the larger π-conjugation framework and the stronger intramolecular charge transfer complex induced by the BTD unit, which will be favorable for harvesting sunlight in the NIR region.Thus, the IPCE of XW57 shows extremely red-shifted onset wavelength of ca. 820 nm, resulting in a higher J_sc value of 13.72 mA/cm² than that of 12.82 mA/cm² for XW4. However, the V_oc value of XW57 was undesirably decreased due to the more severe dye aggregation caused by the larger conjugation framework. As a result of the contradictory effects on J_sc and V_oc, XW57 exhibits an efficiency of 7.15%, higher than that of 6.49% obtained for XW4. On the other hand, XW58 exhibits the highest V_oc value of 844 mV, which may be ascribed to the least aggregation owing to both the absence of the BTD unit and the presence of additional bulky groups attached at the carbazole unit against π-π interaction and approaching of Co³⁺/Co²⁺ in the electrolyte to the TiO₂ surface. At last, accompanied with a sacrifice of J_sc, XW58 exhibits an efficiency of 6.54%. In the sensitizer XW59, a BTD unit and the alkoxy chains/ bulky groups were simultaneously introduced, and thus the highest PCE value of 7.34% was achieved, with simultaneously enhanced J_sc (13.19 mA) and V_oc (793 mV), when compared to those of XW4. It is noteworthy that the rational modulation of the carbazole-based donor by introducing the bulky groups together with the introduction of a benzothiadiazole unit as an extra auxiliary electron acceptor can simultaneously broaden the absorption and suppress the dye aggregation/charge recombina-tion. These results may provide new strategy to develop efficient DSSCs by synergistically enhanced photovoltage and photocurrent.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  This work was financially supported by Central Public-interest Scientific Institution Basal Research Fund, ECSFR, CAFS (No. 2018T02), the National Natural Science Foundation of China (Nos. 21772041, 21811530005, 21971063, U1707602), and the Fundamental Research Funds for the Central Universities (Nos. WK1616004, 222201717003).

  Appendix A. Supplementary data

  Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.cclet.2019.12.038.

  
  
• 1. [1]

     B. O'Regan, M. Grätzel, Nature 353(1991) 737-740. doi: 10.1038/353737a0

  2. [2]

     H. Tian, S.C. Huang, S.C. Zhou, Acta Phys. Chim. Sin. 4(1988) 314-319. doi: 10.3866/PKU.WHXB19880317

  3. [3]

     A. Yella, H.W. Lee, H.N. Tsao, et al., Science 334(2011) 629-634. doi: 10.1126/science.1209688

  4. [4]

     M.K. Nazeeruddin, A. Kay, I. Rodicio, et al., J. Am. Chem. Soc. 115(1993) 6382-6390. doi: 10.1021/ja00067a063

  5. [5]

     Y.J. Xie, Z. Li, Mater. Chem. Front. 4(2020) 317-331. doi: 10.1039/C9QM00580C

  6. [6]

     T. Hua, Z.S. Huang, K. Cai, et al., Electrochim. Acta 302(2019) 225-233. doi: 10.1016/j.electacta.2019.02.013

  7. [7]

     B. Pan, Y.Z. Zhu, D. Ye, et al., New J. Chem. 42(2018) 4133-4141. doi: 10.1039/C7NJ04629D

  8. [8]

     Y. Wang, B. Chen, W. Wu, et al., Angew. Chem. Int. Ed. 53(2014) 10779-10783. doi: 10.1002/anie.201406190

  9. [9]

     Y. Xie, Y. Tang, W. Wu, et al., J. Am. Chem. Soc. 137(2015) 14055-14058. doi: 10.1021/jacs.5b09665

  10. [10]

      Y. Tang, Y. Wang, X. Li, et al., ACS Appl. Mater. Interfaces 7(2015) 27976-27985. doi: 10.1021/acsami.5b10624

  11. [11]

      K. Zeng, Y. Lu, W. Tang, et al., Chem. Sci. 10(2019) 2186-2192. doi: 10.1039/C8SC04969F

  12. [12]

      Y. Lu, H. Song, X. Li, et al., ACS Appl. Mater. Interfaces 11(2019) 5046-5054. doi: 10.1021/acsami.8b19077

  13. [13]

      Y. Xie, W. Wu, H. Zhu, et al., Chem. Sci. 7(2016) 544-549. doi: 10.1039/C5SC02778K

  14. [14]

      Y. Lu, Y. Cheng, C. Li, et al., Sci. China Chem. 62(2019) 994-1000. doi: 10.1007/s11426-019-9471-y

  15. [15]

      H. Song, Q. Liu, Y. Xie, Chem. Commun. 54(2018) 1811-1824. doi: 10.1039/C7CC09671B

  16. [16]

      M. Urbani, M. Grätzel, M.K. Nazeeruddin, Chem. Rev. 114(2014) 12330-12396. doi: 10.1021/cr5001964

  17. [17]

      Y.Q. Wang, L. Xu, X.D. Wei, et al., New J. Chem. 38(2014) 3227-3235. doi: 10.1039/C4NJ00651H

  18. [18]

      Y. Lu, Q. Liu, J. Luo, et al., ChemSusChem 12(2019) 2802-2809. doi: 10.1002/cssc.201900139

  19. [19]

      K.W. Zeng, W.Q. Tang, C.J. Li, et al., J. Mater. Chem. A 7(2019) 20854-20860. doi: 10.1039/C9TA06911A

  20. [20]

      C.J. Li, J.L. Zhang, J.X. Song, et al., Sci. China Chem. 61(2018) 511-514. doi: 10.1007/s11426-018-9245-6

  21. [21]

      H.L. Song, J. Zhang, J.M. Jin, J. Mater. Chem. C 6(2018) 3927-3936. doi: 10.1039/C8TC00610E

  22. [22]

      A. Yella, C.L. Mai, S.M. Zakeeruddin, et al., Angew. Chem. Int. Ed. 53(2014) 2973-2977. doi: 10.1002/anie.201309343

  23. [23]

      Y.C. Chang, C.L. Wang, T.Y. Pan, Chem. Commun. 47(2011) 8910-8912. doi: 10.1039/c1cc12764k

  24. [24]

      M.K. Nazeeruddin, F. De Angelis, S. Fantacci, et al., J. Am. Chem. Soc. 127(2005) 16835-16847. doi: 10.1021/ja052467l

  25. [25]

      C.H. Wu, M.C. Chen, P.C. Su, et al., J. Mater. Chem. A 2(2014) 991-999. doi: 10.1039/C3TA14208F

  26. [26]

      V. Piradi, X. Xu, Z. Wang, et al., ACS Appl. Mater. Interfaces 11(2019) 6283-6291. doi: 10.1021/acsami.8b19240

  27. [27]

      Z. Yao, H. Wu, Y. Ren, et al., Energy Environ. Sci. 8(2015) 1438-1442. doi: 10.1039/C4EE03934C

  28. [28]

      P. Gautam, R. Misra, S.A. Siddiqui, et al., ACS Appl. Mater. Interfaces 719(2015) 10283-10292.

  29. [29]

      H. Zhu, W. Li, Y. Wu, et al., ACS Sustain. Chem. Eng. 2(2014) 1026-1034. doi: 10.1021/sc500035j

  30. [30]

      J.P. Hill, Angew. Chem. Int. Ed. 55(2016) 2976-2978. doi: 10.1002/anie.201511506

  31. [31]

      H.P. Wu, Z.W. Ou, T.Y. Pan, et al., Energy Environ. Sci. 5(2012) 9843-9848. doi: 10.1039/c2ee22870j

  32. [32]

      M. Zhang, Y. Wang, M. Xu, et al., Energy Environ. Sci. 6(2013) 2944-2949. doi: 10.1039/c3ee42331j

  33. [33]

      H.L. Jia, M.D. Zhang, W. Yan, et al., J. Mater. Chem. A 4(2016) 11782-11788. doi: 10.1039/C6TA03740B

  34. [34]

      W.I. Hung, Y.Y. Liao, T.H. Lee, et al., Chem. Commun. 51(2015) 2152-2155. doi: 10.1039/C4CC09294E

  35. [35]

      X. Zhang, J. Mao, D. Wang, et al., ACS Appl. Mater. Interfaces 7(2015) 2760-2771. doi: 10.1021/am507824h

  36. [36]

      Z. Wang, M. Liang, Y. Tan, et al., J. Mater. Chem. A 3(2015) 4865-4874. doi: 10.1039/C4TA06705C

  37. [37]

      X. Li, Z. Zheng, W. Jiang, et al., Chem. Commun. 51(2015) 3590-3592. doi: 10.1039/C4CC08539F

  38. [38]

      S.A. Sapp, C.M. Elliott, C. Contado, et al., J. Am. Chem. Soc. 124(2002) 11215-11222. doi: 10.1021/ja027355y

  39. [39]

      Y. Hao, Y. Saygili, J. Cong, et al., ACS Appl. Mater. Interfaces 8(2016) 32797-32804. doi: 10.1021/acsami.6b09671

  40. [40]

      H.N. Tsao, T. Moehl, J.H. Yum, et al., ChemSusChem 4(2011) 591-594. doi: 10.1002/cssc.201100120

  41. [41]

      S.M. Feldt, E.A. Gibson, E. Gabrielsson, et al., J. Am. Chem. Soc. 132(2010) 16714-16724. doi: 10.1021/ja1088869

  42. [42]

      L. Zhang, J.M. Cole, J. Mater. Chem. A 5(2017) 19541-19559. doi: 10.1039/C7TA05632J

  43. [43]

      H.H. Chou, K.S.K. Reddy, H.P. Wu, et al., ACS Appl. Mater. Interfaces 8(2016) 3418-3427. doi: 10.1021/acsami.5b11554

  44. [44]

      M.J. Frisch, G.W. Trucks, H.B. Schlegel, et al., Gaussian 09 Revision A.2, Gaussian Inc., Wallingford CT, 2009.

  45. [45]

      A.D. Becke, J. Chem. Phys. 98(1993) 5648-5652. doi: 10.1063/1.464913

  46. [46]

      C. Lee, W. Yang, R.G. Parr, Phys. Rev. B 37(1988) 785-789. doi: 10.1103/PhysRevB.37.785

  47. [47]

      W.J. Hehre, R. Ditchfield, J.A. Pople, J. Chem. Phys. 56(1972) 2257-2261. doi: 10.1063/1.1677527

  48. [48]

      P. Dai, H. Dong, M. Liang, et al., ACS Sustain. Chem. Eng. 5(2017) 97-104. doi: 10.1021/acssuschemeng.6b00706

  49. [49]

      Y.K. Eom, S.H. Kang, I.T. Choi, et al., J. Mater. Chem. A 5(2017) 2297-2308. doi: 10.1039/C6TA09836C

  50. [50]

      Y. Ren, D. Sun, Y. Cao, et al., J. Am. Chem. Soc. 140(2018) 2405-2408. doi: 10.1021/jacs.7b12348

  51. [51]

      T. Wei, X. Sun, X. Li, et al., ACS Appl. Mater. Interfaces 7(2015) 21956-21965. doi: 10.1021/acsami.5b06610

  52. [52]

      G. Yang, Y. Tang, X. Li, et al., ACS Appl. Mater. Interfaces 9(2017) 36875-36885. doi: 10.1021/acsami.7b12066

• 

  Figure 1  Molecular structures of porphyrin dyes XW4 and XW57-XW59.

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Absorption spectra of dyes XW4 and XW57-XW59. (a) in THF solutions, (b) on TiO₂ films (3 μm in thickness).

  下载: 全尺寸图片 幻灯片
  

  Figure 3  (a) J-V curves, (b) IPCE action spectra, plots of (c) C_μ and (d) t versus potential bias for DSSCs based on XW57-XW59.

  下载: 全尺寸图片 幻灯片

  Table 1.  Photovoltaic parameters of the solar cells sensitized by XW4 and XW57-XW59 under simulated AM 1.5 G full sunlight (100 mW/cm²). The active area is 0.12 cm².

  下载: 导出CSV
•