the hole transporting materials, along with forming a better contact between the perovskite and the spiro-OMeTAD layers.
The FF, thereby, has been improved from 75% to80%, leading to an enhancement of the PCE from 18.99% to the highest 19.96%. Moreover, due to the higher hydrophobic properties of the interfacial layer, PSCs usingF3BnCz,F5BnCzandF4BnCz2 showed an enhanced long-term stability over 20 days.
Results and discussion
The synthetic scheme and the chemical structure ofF3BnCz, F5BnCzandF4BnCz2are shown in Scheme 1. Details of the synthetic procedures are given in the ESI.†These three mole- cules have been obtained in a single step reaction by coupling 3,6-bis(4,40-dimethoxydiphenylaminyl)-carbazole (Cz-DMPA) and commercially availableuorobenzene derivatives.
F3BnCzhas been prepared by a modied procedure reported by S. Benhattabet al.20A Buchwald coupling ofCz-DMPAand 5- bromo-1,2,3-triuorobenzene in the presence of Pd(OAc)2, (t- Bu)3P andt-BuoNa in toluene at 110C allows the synthesis of F3BnCz with a yield of 62% aer 6 hours. The synthesis of F5BnCzby the same procedure did not occur. Ullmann condi- tions using 1,2,3,4,5-pentauoro-6-iodobenzene with K2CO3, Cu(0) and 18-crown-6 at 170C have already been shown in the literature,21 leading to the replacement of the iodine by a hydrogen atom. Even at low temperature (50C), the reaction led to N3,N3,N6,N6-tetrakis(4-methoxyphenyl)-9-(2,3,5,6-tetra-
uoro-4-iodophenyl)-9H-carbazole-3,6-diamine (53% yield) instead ofF5BnCz. Using a nucleophilic aromatic substitution with sodium hydride,22 we could afford the target molecule.
F5BnCzhas been synthesized using 5 equivalents ofCz-DMPA in the presence of NaH at 0 C (56% yield). Under the same conditions, using only 3 equivalents of Cz-DMPA at room temperature,F4BnCz2has been obtained (80% yield). The three
molecules are well soluble in common organic solvents such as toluene, dichloromethane and chlorobenzene. These uori- nated molecules have been fully characterized by1H,19F and
13C NMR spectroscopy and HRMS. All the analytical data are consistent with the target structures (ESI†).
The thermal properties of the three compounds have been investigated by thermogravimetric analysis (TGA) and differ- ential scanning calorimetry (DSC) (in Fig. S1, ESI†). The mole- cules show a high thermal stability since the decomposition temperatures are 281C, 265C and 295C forF3BnCz,F5BnCz and F4BnCz2, respectively. Considering the DSC measure- ments, it can be seen that F3BnCz and F5BnCz exhibit a molecular glass behavior with respectively a glass transition temperature of 77 C and 44 C. Neither crystallization nor melting transition peaks have been observed upon several heating/cooling cycles. Comparing with CzPF and CzP20 (chemical structures in Fig. S2, ESI†), it can be seen that the values ofTgare lower (82C forCzPFand 97C forCzP). The introduction of additionaluorine atoms on the benzene ring reduces theTg. The moreuorine atoms on the benzene ring, the more theTgdecreases. These results conrm the trend that has been observed in the literature.23 ForF4BnCz2, no glass transition temperature has been observed in the range of measurements (25–200C).
The normalized UV-Vis absorption spectra of F3BnCz, F5BnCzandF4BnCz2in solution and as thinlms are shown in Fig. 1a. In the solid state, the spectra exhibit a bathochromic shi due to p-stacking. It is noticeable that the red shi is signicantly more pronounced as the number ofuorine atom increases, suggesting that the presence of uorine atoms induces stronger intermolecular packing.20 In solution, the main peaks at 303–307 nm and 370–375 nm, which shiabout 10–15 nm in solid state thinlms, could be attributed to local carbazole and diphenylamine transitions. The optical band
Scheme 1 Synthesis route and chemical structure of the interfacial moleculesF3BnCz,F5BnCzandF4BnCz2.
gaps ofF3BnCzandF5BnCz, as determined from the absorp- tion onset of thinlms, are similar to each other (2.75 eV).
However, the introduction of a secondCz-DMPAmoiety induces a decrease of 40 meV (Eoptg ¼2.71 eV), which is characteristic of additionalp–pstacking interactions and increased conjugation length.
The electrochemical properties of these glassy molecules were then studied using cyclic voltammetry (CV) in order to determine their energy levels (Fig. 1b). The three molecules exhibit two quasi-reversible oxidation waves, attributed to the formation of radical cations and dications of the carbazole moiety.24 The rst half-wave oxidation potential is lower for F3BnCz(+0.12 Vvs.Fc/Fc+) than forF5BnCz(+0.19 V). Consid- ering the potential of the reference ferrocene/ferrocenium couple at 5.1 eVvs.vacuum, the highest occupied molecular orbital (HOMO) levels ofF3BnCzandF5BnCzare determined to be 5.22 eV and 5.29 eV, respectively. As expected, the HOMO energy level is slightly lowered upon addition of strongly elec- tronegative F atoms.23F4BnCzshows a HOMO level at 5.27 eV, indicating that the graing of a secondCz-DMPAunit does not signicantly increase the HOMO level. The lowest unoccupied molecular orbital (LUMO) levels of F3BnCz, F5BnCz, and F4BnCz, extrapolated from each optical band gap, are 2.47 eV,
2.54 eV and 2.56 eV, respectively.
As shown in Fig. 1c, the HOMO energy levels of F3BnCz, F5BnCz, andF4BnCzall lie in-between MAPbI3( 5.43 eV) and spiro-OMeTAD ( 5.20 eV) energy levels. Thus, all three mole- cules are expected to be suitable for their utilization as an
interface layer, as they should allow efficient hole transfer from the perovskite lms to the spiro-OMeTAD hole transporting material layer. Moreover, each LUMO level of the interfacial layers is sufficiently high to block electrons transferring from the perovskite to the spiro-OMeTAD.25Hence, attempting to use these synthesized materials as interfacial materials, we have fabricated planar n–i–p type perovskite solar cells with the device conguration uorine doped tin oxide (FTO)/SnO2/ MAPbI3/FxBnCzy/spiro-OMeTAD/Ag (Fig. 1d). The thickness of the interfacial layer has been optimized by changing the spin speed during the spin-coating process as shown in Fig. S3 and Table S1.†It was found that at 5000 rpm, the device gives the best performance. This is possibly due to the formation of a continuouslm with appropriate thickness. Devices without interfacial layers were also built and characterized as reference devices.
The current density–voltage (J–V) curves of the best- performing devices are shown in Fig. 2a, with the details of specic parameters exhibited in Table 1. The best-performing devices based on F3BnCz, F5BnCz, and F4BnCz2 interfacial layers obtained PCEs of 19.86% (VOC¼1.13 V,JSC¼22.10 mA cm2, and FF¼79.83%), 19.86% (VOC¼1.11 V,JSC¼22.52 mA cm2, and FF¼79.40%), and 19.96% (VOC¼1.12 V,JSC¼22.65 mA cm2, and FF¼78.77%), respectively. For comparison, the best control device afforded a PCE of 18.99% (VOC¼1.12 V,JSC
¼ 22.56 mA cm2, and FF ¼ 75.17%). It is noted that the interfacial layers are benecial for the enhancement of the FF for the all devices, thus leading to higher PCEs. The Fig. 1 (a) Normalized UV-Vis absorption spectra in THF and infilms, (b) cyclic voltammetry in solution (10 3M) and CH2Cl2/TBAP (0.1 M) at 50 mV s 1ofF3BnCz,F5BnCzandF4BnCz2. (c) Illustration of the energy diagrams of the electrode, interfacial layer and HTL, and the charge transfer processes in the PSCs. (d) Schematic architecture of the PSCs used in this study.
corresponding external quantum efficiency (EQE) measure- ments were conducted as shown in Fig. 2b. The PSCs with these interfacial layers show similar EQE values around 80% in the 450 to 650 nm wavelength range, much higher than that of the control device, demonstrating the benecial effect of the interfacial materials on the dynamics of charges. The integrated JSCwas improved from 16.61 to20 mA cm2.
Meanwhile, Fig. 2c shows the results of steady-state current density and PCE, measured at constant bias voltages of 0.91 V, 0.94 V, 0.92 V and 0.90 V, respectively. The stabilized PCEs were 12.49%, 16.17%, 16.70% and 15.06%, with stabilized current densities of 13.73, 17.20, 18.15 and 15.06 mA cm2, for the devices w/oF3BnCz, F5BnCzand F4BnCz2 interfacial layers, correspondingly. This is consistent with the trend based on the parameters obtained from the J–V curves. In addition, 20 independent devices prepared under the same conditions were fabricated, and the statistical distribution of the PCE,VOC,JSC and FF is exhibited in Fig. 2d (and Fig. S4, ESI†). The PSCs with
F3BnCz,F5BnCzandF4BnCz2give a better performance than interfacial layer-free cells, with an average PCE of 18.36%, 18.26% and 18.20%, respectively, compared with only 17.19%
for the reference devices. As expected, the enhancement comes from the higher FF from 72% toca. 75–76% as indicated in Fig. S4.†
To explain the higher FF of the devices with interfacial layers, the morphology was rst studied with a scanning electron microscope (SEM) as shown in Fig. 3. The MAPbI3lm in Fig. 3a was highly crystalized with the crystal size ranging from 600 nm to 1mm, with clear boundaries. From Fig. 3b–d, it can be seen that the interfacial layers basically formed aat and homoge- neous lm on the perovskite. While aer the deposition of spiro-OMeTAD, the morphological difference of perovskite/
interfacial layer/spiro-OMeTAD (Fig. 3f–h) was even negligible.
Therefore, it is considered that these comparable morphologies make no difference to the photovoltaic performance.
Fig. 2 Performances of the PSCs with and without various interfacial layers. (a)J Vcurves. (b) EQE and integratedJSC. (c) Steady-state current density and PCE at a constant bias voltage andVmax. (d) PCE distributions.
Table 1 Best and average photovoltaic parameters of PSCs
Device
VOC[V] JSC[mA cm 2] FF [%] PCE [%]
Best Average Best Average Best Average Best Average
Without 1.12 1.09 22.56 21.69 75.17 72.97 18.99 17.19
F3BnCz 1.13 1.10 22.10 21.84 79.83 76.46 19.86 18.36
F5BnCz 1.11 1.10 22.52 22.07 79.40 75.48 19.86 18.26
F4BnCz2 1.12 1.11 22.65 21.62 78.77 75.81 19.96 18.20
To further explore the origin of the improved FF, we studied the charge transfer dynamics by conducting electrochemical impedance spectrometry (EIS) in the dark. Fig. 4a shows the Nyquist tting results of experimental data based on various PSCs. The series resistances (Rs) are determined to be 13.40, 11.00, 12.05 and 10.91U for PSCs with F3BnCz, F5BnCz and F4BnCz2and w/o interfacial layers, respectively. Thearccan be assigned to the charge transfer process of the devices,26,27which gives a charge transfer resistance (Rtr) of 253.14, 147.98, 167.95, and 200.09U, accordingly. The prominently decreasedRsandRtr suggest a better contact at the interfaces of perovskite/interfacial layer/spiro-OMeTAD, which could favor the efficient hole transfer and transport processes and lead to an increased FF.28
The charge transfer efficiency was further investigated by steady-state photoluminescence emission spectroscopy (PL) on the lms of perovskite, perovskite/spiro-OMeTAD and perovskite/interfacial layer/spiro-OMeTAD on quartz, with an excitation wavelength of 450 nm. As shown in Fig. 4b, the pure perovskite lm displays a strong emission peak at around 780 nm, which was signicantly quenched by the deposition of spiro-OMeTAD. By comparison, upon the insertion of the interfacial layer, the photoluminescence was further sup- pressed, revealing the efficient extraction of photo-generated holes from the perovskite to the composite interfacial/spiro- OMeTAD hole-transporting layer. In addition, the correspond- ing time-resolved photoluminescence (TRPL) was examined as shown in Fig. 4c. The biexponential decay lifetimes1, attributed to the trap-assisted recombination,29is reduced from 24.71 ns
to 21.53 ns aer depositing spiro-OMeTAD on the perovskite
lm, and the values are further decreased to 4.39 ns, 12.86 ns and 18.87 ns aer applying F3BnCz, F5BnCz and F4BnCz2 interfacial layers between the perovskite and spiro-OMeTAD
lms, which demonstrates the signicant suppression of traps. At the same time, the lifetimes2, representative of free carrier recombination, is decreased from 59.97 ns to 51.02 ns, 12.70 ns, 47.50 ns, and 49.95 ns, respectively, consistent with the electron blocking effect induced by the higher LUMO levels of these interfacial materials. The details of the TRPL data and thetting equation are shown in Table S2.†
Hence, from the EIS, PL and TRPL results, it is concluded that the insertion of these three interfacial layers could not only lead to a better interfacial contact due to the trap passivation, but also block electron back-transfer effectively due to the higher LUMO energy levels of these dopant-free interfacial layers. Thus, the insertion of our designed interfacial materials improves the charge extraction and transfer efficiency, leading to the high FF. Overall, the inuence ofF3BnCz,F5BnCzand F4BnCz2 interfacial layers on the photovoltaic performances was quite similar.
Fig. 5 shows the stability of the PSCs with and without inter- facial layers at a relative humidity around 35%. The devices with Fig. 3 Morphological characterization. Top-view SEM image of (a)
pure perovskitefilm, (b)F3BnCz/perovskite, (c)F5BnCz/perovskite, (d) F4BnCz2/perovskite, (e) spiro-OMeTAD/perovskite, (f) perovskite/
F3BnCz/spiro-OMeTAD, (g) perovskite/F5BnCz/spiro-OMeTAD, and (h) perovskite/F4BnCz2/spiro-OMeTAD. Scale bar: 500 nm.
Fig. 4 Characterization of charge transfer dynamics. (a) Nyquist plots of the devices, measured under dark ambient conditions. The inset shows the equivalent circuit. (b) Steady-state PL and (c) TRPL spectra.
Fig. 5 Stability of the PSCs with and without interfacial layers. The insets show the water contact angle of thefilms of spiro-OMeTAD, F3BnCz,F5BnCzandF4BnCz2.
the interfacial layers maintain 80% of the initial efficiency aer 23 day storage, in sharp contrast with the PCE of the pristine device,ca.40% of the initial PCE remains. The water-contact- angle test further conrmed the higher hydrophobicity of the
uorinated molecules, with values of F3BnCz, F5BnCz and F4BnCz2of 108, 108 and 101, respectively, which are remarkably larger than that of spiro-OMeTAD (85). Hence, the uorine- substituted interfacial layers are more water-resistant and protect the perovskite from moisture efficiently. The thermal stability of the solar cells has also been investigated as shown in Fig. S5.†Aer the continuous thermal annealing of the devices at 85 C, the devices with interfacial layers show a better PCE stability than the interfacial layer free device.
Conclusions
In summary, aiming at the enhancement of the FF, we have designed and synthesized a series of novel uorinated small molecules as interfacial layer materials to improve the contact between the perovskite and spiro-OMeTAD. Through the investigation of their optical, electrochemical and photovoltaic properties, we found that the inuence of the small variation on the molecular structure, such as the number of F atoms or the dimer derivative, is highly limited. Finally, the perovskite solar cells with these interfacial layers show a better PCE owing to a signicant increase of the FF, up to 80% based on the MAPbI3 system, while maintaining JSC and VOC unchanged.
Furthermore, the device stability has also been improved due to their stronger hydrophobicity.
Author contributions
R. W. conducted the device fabrication and characterization.
R. N. designed and synthesized the molecules. R. W., R. N., Y. J.
and J. W. G. wrote the manuscript. All the authors discussed the results and made contributions to the manuscript. J. W. G.
directed the research.
Con fl icts of interest
There are no conicts to declare.
Acknowledgements
We thank the nancial support from NSFC-Guangdong Joint Fund (No. U1801256), National Key R&D Program of China (No.
2016YFA0201002), Guangdong International Science and Technology Cooperation Project (No. 2020A0505100054), Inno- vation Team of Guangdong Education Department (No.
2016KCXTD009) and Program for Chang Jiang Scholars and Innovative Research Teams in Universities (No. IRT 17R40). R.
Nakar, N. Berton and B. Schmaltz acknowledge the R´egion Centre-Val de Loire (France) for itsnancial support through the CELEZ project.
References
1 X. Liu, E. Rezaee, H. Shan, J. Xu, Y. Zhang, Y. Feng, J. Dai, Z. K. Chen, W. Huang and Z. X. Xu, J. Mater. Chem. C, 2018,6, 4706–4713.
2 J. Tian, Q. Xue, X. Tang, Y. Chen, N. Li, Z. Hu, T. Shi, X. Wang, F. Huang, C. J. Brabec, H. L. Yip and Y. Cao,Adv.
Mater., 2019,31, 1901152.
3 Z. Gan, X. Wen, W. Chen, C. Zhou, S. Yang, G. Cao, K. P. Ghiggino, H. Zhang and B. Jia, Adv. Energy Mater., 2019,9, 1900185.
4 A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am.
Chem. Soc., 2009,131, 6050–6051.
5Best Research-Cell Efficiencies, https://www.nrel.gov/pv/cell- efficiency.html.
6 X. Wu, Y. Jiang, C. Chen, J. Guo, X. Kong, Y. Feng, S. Wu, X. Gao, X. Lu, Q. Wang, G. Zhou, Y. Chen, J. M. Liu, K. Kempa and J. Gao,Adv. Funct. Mater., 2019,30, 1908613.
7 C. Chen, Y. Jiang, J. Guo, X. Wu, W. Zhang, S. Wu, X. Gao, X. Hu, Q. Wang and G. Zhou,Adv. Funct. Mater., 2019,29, 1900557.
8 N. J. Jeon, H. Na, E. H. Jung, T. Y. Yang, Y. G. Lee, G. Kim, H. W. Shin, S. Il Seok, J. Lee and J. Seo,Nat. Energy, 2018, 3, 682–689.
9 M. Kim, G. H. Kim, T. K. Lee, I. W. Choi, H. W. Choi, Y. Jo, Y. J. Yoon, J. W. Kim, J. Lee, D. Huh, H. Lee, S. K. Kwak, J. Y. Kim and D. S. Kim,Joule, 2019,3, 2179–2192.
10 I. Mora-Ser´o,Joule, 2018,2, 585–587.
11 C. M. Wolff, P. Caprioglio, M. Stolterfoht and D. Neher,Adv.
Mater., 2019,31, 1902762.
12 X. Zhao, L. Tao, H. Li, W. Huang, P. Sun, J. Liu, S. Liu, Q. Sun, Z. Cui, L. Sun, Y. Shen, Y. Yang and M. Wang,Nano Lett., 2018,18, 2442–2449.
13 W. Shockley and H. J. Queisser,J. Appl. Phys., 1961,32, 510–
519.
14 X. Meng, X. Cui, M. Rager, S. Zhang, Z. Wang, J. Yu, Y. W. Harn, Z. Kang, B. K. Wagner, Y. Liu, C. Yu, J. Qiu and Z. Lin,Nano Energy, 2018,52, 123–133.
15 B. Wang, M. Zhang, X. Cui, Z. Wang, M. Rager, Y. Yang, Z. Zou, Z. L. Wang and Z. Lin,Angew. Chem., Int. Ed., 2020, 59, 1611–1618.
16 S. Wang, Z. Huang, X. Wang, Y. Li, M. Gunther, S. Valenzuela, P. Parikh, A. Cabreros, W. Xiong and Y. S. Meng,J. Am. Chem. Soc., 2018,140, 16720–16730.
17 F. Qi, X. Deng, X. Wu, L. Huo, Y. Xiao, X. Lu, Z. Zhu and A. K. Y. Jen,Adv. Energy Mater., 2019,9, 1902600.
18 Y. Bai, X. Meng and S. Yang, Adv. Energy Mater., 2018, 8, 1701883.
19 S. Feng, Y. Yang, M. Li, J. Wang, Z. Cheng, J. Li, G. Ji, G. Yin, F. Song, Z. Wang, J. Li and X. Gao,ACS Appl. Mater. Interfaces, 2016,8, 14503–14512.
20 S. Benhattab, A. N. Cho, R. Nakar, N. Berton, F. Tran-Van, N. G. Park and B. Schmaltz,Org. Electron., 2018,56, 27–30.
21 Z. Zhao, B. Xu, Z. Yang, H. Wang, X. Wang, P. Lu and W. Tian,J. Phys. Chem. C, 2008,112, 8511–8515.
22 S. Qiao, W. Huang, H. Wei, T. Wang and R. Yang,Polymer, 2015,70, 52–58.
23 Z. Li, B. Jiao, Z. Wu, P. Liu, L. Ma, X. Lei, D. Wang, G. Zhou, H. Hu and X. Hou,J. Mater. Chem. C, 2013,1, 2183–2192.
24 J. F. Ambrose, L. L. Carpenter and R. F. Nelson, J.
Electrochem. Soc., 1975,122, 876.
25 N. Arora, S. Orlandi, M. I. Dar, S. Aghazada, G. Jacopin, M. Cavazzini, E. Mosconi, P. Gratia, F. De Angelis, G. Pozzi, M. Graetzel and M. K. Nazeeruddin, ACS Energy Lett., 2016,1, 107–112.
26 F. Giordano, A. Abate, J. P. Correa Baena, M. Saliba, T. Matsui, S. H. Im, S. M. Zakeeruddin, M. K. Nazeeruddin, A. Hagfeldt and M. Graetzel,Nat. Commun., 2016,7, 10379.
27 B. Roose, K. C. G¨odel, S. Pathak, A. Sadhanala, J. P. C. Baena, B. D. Wilts, H. J. Snaith, U. Wiesner, M. Gr¨atzel, U. Steiner and A. Abate,Adv. Energy Mater., 2016,6, 1501868.
28 H. D. Kim and H. Ohkita,Sol. RRL, 2017,1, 1700027.
29 B. Cai, X. Yang, X. Jiang, Z. Yu, A. Hagfeldt and L. Sun,J.
Mater. Chem. A, 2019,7, 14835–14841.
