Porphyrins with di-alkoxy-thiophene side chains for organic solar cells

Abstract

The crystallinity of electron donors and their compatibility with electron acceptors play important roles for the performance of bulk-heterojunction organic solar cells. Considering that large electronegative atoms such as O and S atoms are often introduced into active materials to enhance intermolecular interactions and poly(3,4-ethylenedioxythiophene) (PEDOT) is a highly conductive polymer, here in, we introduce dimethoxythiophene (DMOT) and ethylene dioxyl-thiophene (EDOT) side chains at two of the meso positions of a porphyrin core to synthesize two small molecular donors, ZnP-DMOT and ZnP-EDOT, and PCEs of 9.63% and 9.06%, respectively, are achieved for the OSCs with PC₆₁BM as the acceptor. These PCEs are very close to the state-of-the-art among fullerene-based OSCs. In addition, when non-fullerene acceptor 6TIC was employed, the ZnP-DMOT and the ZnP-EDOT binary devices showed PCEs of 11.55% and 10.86%, respectively. Experimental results show that di-alkoxy groups improve not only the crystallinity of the porphyrins but also their compatibility with PC₆₁BM and 6TIC. Furthermore, the introduction of PC₆₁BM as the third component into 6TIC based binary layers significantly improves the electron mobility and the balance of hole and electron transport. As a result, the ternary ZnP-DMOT:6TIC:PC₆₁BM and ZnP-EDOT:6TIC:PC₆₁BM devices achieve PCEs of 13.32% and 12.96%, respectively.

---

New concepts

Side-chain engineering is an underexplored yet powerful strategy for small-molecule porphyrin donors. Here, we introduce oxygenated thiophene side chains—3,4-dimethoxythiophene (DMOT) and 3,4-ethylenedioxythiophene (EDOT)—at two of the meso positions of a DPP-bridged porphyrin core to simultaneously couple the ordered molecular packing of the donor molecules with the miscibility of the donor with acceptor molecules. This modification enables near-“theoretical” fullerene performance (9.66%/9.10% with PC₆₁BM) and competitive non-fullerene efficiencies (11.52%/10.89% with 6TIC), clearly outperforming the alkyl-thiophene analogue (8.08%) mainly through fill-factor improvements. Mechanistic insights reveal a direct structure–property–performance relationship: DFT shows a smaller dihedral angle for EDOT in solution, while DMOT promotes stronger film-state ordering, manifested in shorter π–π stacking (3.72 Å) and longer coherence length, leading to a deeper HOMO (−5.20 eV), a higher V_OC, and an improved FF. Furthermore, incorporating a small fraction of PC₆₁BM into the 6TIC binary further enhances electron transport, elevating EQE above 80% and boosting the PCEs to 13.36%/12.91%. Collectively, these results demonstrate that oxygenated thiophene side chains offer a general and scalable approach to reconcile processability with intermolecular order in porphyrin donors.

Introduction

Solar energy is one of the renewable, cleanest and most promising energy sources. Organic solar cells (OSCs), which convert solar energy into electricity, have drawn much attention due to their advantages such as low cost, flexibility and lightness.^1–4 To date, much progress has been achieved with the highest power conversion efficiencies (PCEs) up to 21%.^5–8 The active materials in OSCs can be divided into polymers and small molecules.^9,10 Compared with polymers, small-molecule materials exhibiting small batch-to-batch variations, well-defined structures and ease of purification,^11–14 therefore, have been widely applied in organic optoelectronics. Among small-molecule donor materials, porphyrin-based small-molecule donors offer several advantages. First, the porphyrin macrocycle provides strong and broad absorption, which can be extended to the near-infrared region to allow effective harvesting of the solar light.^15–17 Second, they can be structurally well-defined: functionalization at meso or β positions and the choice of side chains or a central metal allow fine tuning of absorption spectra, frontier orbital energy levels, and molecular packing properties.^18–21 Specifically, the porphyrins linked with electron acceptor (A) units via ethynylene at the meso positions can make the whole molecules conjugated, extending the conjugation and enhancing the intramolecular charge transfer, thereby improving the performance of their organic electronic devices and making the porphyrin unit one of the widely used building blocks in organic photovoltaic materials. Third, porphyrins often exhibit good chemical and thermal stability, which is beneficial for achieving stable devices. In previous studies, porphyrin derivatives have been shown to retain performance under thermal stress and have more tolerant morphologies. These properties demonstrate the great potential of porphyrin small molecule donors in all small molecule organic solar cells.

Constructing D–A molecules (D: electron donor and A: electron acceptor) is a common strategy to design organic photovoltaic active materials.^22,23 Many D and A units have been widely used for constructing D–A active materials. Especially, high PCEs are often achieved for OSCs with diketopyrrolopyrrole (DPP) incorporated D–A porphyrins because the introduction of the DPP groups with suitable side chains can not only effectively regulate their electronic properties but also enhance the π–π stacking interactions.^24–27 Namely, side chain engineering is also very important for optimizing the photovoltaic properties of active materials because side chains impact their solubility and crystallinity and the miscibility of active materials.^28–31 In order to improve the intermolecular interactions, charge mobility and device performance, large electronegative atoms such as O and S atoms are often introduced into the side chains of the active materials to optimize the film morphology.^32–35

Poly(3,4-ethylenedioxythiophene) (PEDOT) is known for its high electrical conductivity and is commonly used as the hole transport layer in OSCs.^36,37 Its high conductivity is attributed in part to enhanced intermolecular interactions conferred by the 3,4-ethylenedioxy groups. Therefore, di-alkoxy thiophene side chains are sometimes employed to design photovoltaic active materials. For example, Peng et al. developed two novel low-bandgap copolymers, PTB-EDOT and PTB-EDOTS, by introducing EDOT side chains, and achieved a PCE of 10.18% for the binary cells based on PTB-EDOTS:ITIC-Th active materials.³⁸ Bo et al. designed and synthesized a series of non-fullerene acceptor materials with 3,4-propylenedioxythiophene (PDOT) and EDOT bridging units and achieved a power conversion efficiency (PCE) of 11.32% for IDT-EDOT-based devices though a PCE of only 2.18% was achieved for IDT-PDOT-based cells when poly[(2,6-(4,8-bis(5-(2-ethylhexydppl)thiophen-2-yl)-benzo(1,2-b:4,5-b′)dithiophene))-alt-(5,5-(1′,3′-di-2-thienyl-5′,7′-bis(2-ethylhexyl)benzo(1′,2′-c:4′,5′-c′)dithiophene-4,8-dione))] (PBDB-T) was employed as the electron donor.³⁹

However, the reports of active materials with di-alkoxy-thiophene side chains are very limited, and it is still necessary to investigate the impacts of these kinds of side chains on the performance of OSCs. Especially, there are no reports on small molecular materials with 3,4-dimethoxythiophene (DMOT) and EDOT side chains. Herein, we linked two DPP electron acceptor units with ethynylene to a porphyrin core as the conjugated main chain to regulate their electronic properties. In order to further regulate the solubility, π–π stacking interactions, crystallinity and miscibility with electron acceptors, we introduced two DMOT and EDOT with branched alky chains at two of the meso positions of a porphyrin core as the side chains to synthesize two porphyrin-based small-molecule donor materials, ZnP-DMOT and ZnP-EDOT, respectively. When PC₆₁BM is employed as the electron accepting material, the devices based on ZnP-DMOT and ZnP-EDOT show PCEs of 9.63% and 9.06%, respectively. These PCEs are higher than that (8.08%) of the devices based on the porphyrin of DPPEZnP-TEH with the same main chain but alkyl-thiophene side chains, reported previously. Among the OSCs based on small molecule donors and fullerene acceptors, the PCEs in the range of 9–10% are rare, and the 9.63% PCE is approaching the upper bound of what have been achieved for such OSCs. On the other hand, the PCEs of 11.55% and 10.86% are achieved for the ZnP-DMOT and ZnP-EDOT devices, respectively, when the two porphyrins are blended with non-fullerene acceptor 6TIC. Furthermore, we also fabricated ternary OSCs with ZnP-DMOT:PC₆₁BM:6TIC and ZnP-EDOT:PC₆₁BM:6TIC as the active layers, which show PCEs of 13.32% and 12.96%, respectively.

Results and discussion

The chemical structures of ZnP-DMOT and ZnP-EDOT are shown in Scheme 1. As seen in the SI, the intermediates 6a and 6b were synthesized according to reported procedures, and the target products ZnP-DMOT and ZnP-EDOT were synthesized by first removing the TIPS protecting group from the intermediates 6a and 6b and then performing Sonogashira coupling reactions with di-bromo-dithieno-diketopyrrolopyrrole (DPP-Br) under an oxygen-free atmosphere and in the presence of the catalysts Pd(PPh₃)₄ and CuI with reaction yields of 60% and 77%, respectively. We estimate a cost of about $77.07/g (Table S1, SI) for ZnP-DMOT under our lab-scale conditions. While this is a rough estimate, the lower estimated material cost of our porphyrin donors suggests a favorable cost advantage relative to Y6 (material cost for Y6 on the order of $146.9/g and a small-scale-market price of $9226.8/g) under comparable scales.^40,41 The chemical structures of the important intermediates and the two final products were characterized with nuclear magnetic resonance (NMR) spectroscopy and matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) (Fig. S1–S6, SI). In order to obtain ¹H NMR spectra, 5% pyridine-d5 (C₅D₅N) is necessary to be added to their CDCl₃ solutions to coordinate with the Zn atom of the two porphyrins to prevent the strong π–π stacking of these two molecules due to the large conjugation and good planarity.

Scheme 1 The chemical structures of ZnP-DMOT, ZnP-EDOT, 6TIC and PC₆₁BM.

The UV-vis absorption spectra of ZnP-DMOT and ZnP-EDOT in chloroform solution and in thin film state are shown in Fig. 1, and the corresponding absorption parameters are summarized in Table 1. In chloroform solutions, ZnP-DMOT and ZnP-EDOT show the same Soret band at 471 nm, while the two Q bands of ZnP-EDOT are slightly red-shifted from 570 and 728 nm of ZnP-DMOT to 571 and 731 nm, respectively. The shapes and intensities of the peaks are similar, indicating that the electronic structures and conjugation degrees of the two porphyrins are similar. Compared with the absorption of ZnP-DOMT, the slight red-shifts of ZnP-EDOT stem from the more effective conjugation between EDOT side chains and the porphyrin core due to their less steric hindrance, which is consistent with the DFT calculation results that its dihedral angle (68.33°) is smaller than that between DOMT and the porphyrin unit (83.95°). The smaller dihedral angle facilitates the conjugation between the thiophene group and the porphyrin core.

Fig. 1 UV-vis absorption spectra of ZnP-DMOT and ZnP-EDOT (a) in chloroform solution and (b) in thin films.

Table 1 Optical and electrochemical parameters of the small molecule donors ZnP-DMOT and ZnP-EDOT

Molecules	λ max (nm)^a	λ max (nm)^b	λ onset (nm)	E ^optg (eV)^c	E ox (V)	E HOMO (eV)^d	E LUMO (eV)^e
a In chloroform. b In films. c Calculated from 1240/λ^filmonset. d Calculated from cyclic voltammetry. e E LUMO = EHOMO + E^optg.
ZnP-DMOT	728	803	884	1.40	0.80	−5.20	−3.80
ZnP-EDOT	731	793	885	1.40	0.73	−5.13	−3.73

---

In films, ZnP-DMOT shows a Soret band at 487 nm and Q bands at 575 and 803 nm. For the ZnP-EDOT film, the Soret band is at 483 nm and its Q bands are at 576 and 793 nm while both films show shoulders at 732 nm. The 732 nm shoulders can be ascribed to the non-ordered molecules because the peaks are almost the same as those in solutions. Compared with those in solutions, the maximum absorption peak of ZnP-DMOT is redshifted by 75 nm while that of ZnP-EDOT is redshifted by 62 nm, indicating that there are more ordered self-assembly for the ZnP-DMOT film though the π–π stackings of the two compounds are both strong in film states, which is conducive to the inter-molecular charge transfer of their OSCs. The offsets of the absorption spectra of the two films are almost the same of 884 and 885 nm, from which their optical band gaps (E^opt_g) were both calculated to be 1.40 eV for both ZnP-DMOT and ZnP-EDOT.

In order to determine the energy levels of their frontier molecular orbitals of ZnP-DMOT and ZnP-EDOT, we measured their cyclic voltammogram characteristics using cyclic voltammetry (CV) with ferrocene (Fc/Fc⁺) as the internal standard. The obtained curves are shown in Fig. 2a, and the corresponding electrochemical parameters are summarized in Table 1. We observed two oxidation peaks in the cyclic voltammetry curves of ZnP-DMOT, likely arising from its aggregation. To mitigate this, we supplemented the CV measurements by adding pyridine as a coordinating additive (Fig. S16). The CV shoulder is significantly reduced by the pyridine additive, demonstrating that the shoulder peaks are ascribed to the molecular aggregation and partial aggregation still remains even when pyridine is introduced. In light of this, we define the main oxidation potential as that of the peak with higher intensity. The initial oxidation potentials of ZnP-DMOT and ZnP-EDOT are 0.80 and 0.73 V, respectively. Since the oxidation potential of ferrocene relative to the silver/silver chloride electrode was measured to be 0.40 V and its theoretical vacuum energy level is 4.8 eV, the highest occupied molecular orbital (HOMO) energy level (E_HOMO) of ZnP-DMOT was calculated to be −5.20 eV, and the lowest unoccupied molecular orbital (LUMO) energy level (E_LUMO) was calculated to be −3.80 eV according to the formula of E_LUMO = E_HOMO+ E^opt_g. Accordingly, the E_HOMO and E_LUMO values of ZnP-EDOT were calculated to be −5.13 and −3.73 eV, respectively. As seen in Fig. 2b, the energy levels of the both materials are well matched with those of the acceptor materials PC₆₁BM and 6TIC. Both the HOMO and LUMO energy levels of ZnP-DMOT are deeper than those of ZnP-EDOT, respectively, which is benificial for obtaining higher open circuit voltages (V_OC) for the ZnP-DMOT OSCs compared with those of ZnP-EDOT-based devices.

Fig. 2 (a) Cyclic voltammetric curves of ZnP-DMOT and ZnP-EDOT. (b) Energy level diagrams of the porphyrin donor, 6TIC and PC₆₁BM acceptor.

To study the spatial geometries and the electron distribution of the frontier molecular orbitals of the small molecules ZnP-DMOT and ZnP-EDOT impacted by the side chains, we conducted density functional theory (DFT) simulations using Gaussian 09 with B3LYP/(6-31G(d,p)) basis sets.⁴² To facilitate the calculations, we replaced the long alkyl chains of the molecules with methyl groups, and the results are shown in Fig. 3. From the side and top views, it can be seen that the DPP units of the both small molecules are almost coplanar with the porphyrin core, indicating that DPP units can conjugate with the porphyrin core to form good intramolecular charge transfer. The dihedral angle between the DMOT side chains and the porphyrin unit is 83.95°, and that between the EDOT side chains and the porphyrin unit is 68.33°. The samller dihedral angle between EDOT side chains and porphyrin unit suggests that EDOT side chains can slightly conjugate more with the porphyrin units, which is in agreement with the slight redshift of the absorption of ZnP-EDOT in solution. From DFT calculations, the HOMO and LUMO energy levels of ZnP-DMOT were calculated to be −5.06 and −2.89 eV while those of ZnP-EDOT were −4.96 and −2.81 eV, respectively, which are consistent with the trends of the optical and electrochemical measurement results.

Fig. 3 DFT calculations of ZnP-DMOT and ZnP-EDOT.

To investigate the photovoltaic performance of ZnP-DMOT and ZnP-EDOT, we first fabricated organic solar cells using them as the donors and PC₆₁BM as the acceptor with a device structure of ITO/PEDOT:PSS/active layer/PFN/Ag. The mass ratios of the two porphyrins to PC₆₁BM were both optimized to be 1 : 1, and 2% pyridine (Py) was added as the solvent additive.⁴³ After the active layers were casted, solvent vapor annealing (SVA) treatment of the active layers with dichloromethane was further employed to optimize the device performance. In addition, we have included the detailed process of ZnP-DMOT in Table S2 (SI), and the optimization approach of ZnP-EDOT is the same as that of ZnP-DMOT. The current density–voltage (J–V) curves of the devices were measured as shown in Fig. 4a, and the photovoltaic parameters are presented in Table 2. Under the as-cast conditions, both devices show low current densities (J_SC) and fill factors (FF), resulting in lower PCEs. After Py + SVA treatment, the ZnP-EDOT:PC₆₁BM devices achieved a PCE of 9.06% with an open circuit voltage (V_OC) of 0.73 V, a J_SC of 18.20 mA cm⁻² and an FF of 68.21%. After replacing the EDOT with DMOT side chains, the PCE of the devices based on ZnP-DMOT:PC₆₁BM was improved to 9.63%, with a V_OC of 0.77 V, a slightly reduced J_SC to 17.91 mA cm⁻², and an FF of 69.82%. SVA treatment significantly increased the J_SC and FF values due to the enhanced exciton separation efficiency, mobility, and crystallinity, while the V_OC values slightly decreased owing to the increased non-radiative recombination.⁴⁴ Although the ZnP-DMOT device has no advantage in J_SC, the higher V_OC and FF lead to a higher PCE. It should be noted that the devices based on the porphyrin of DPPEZnP-TEH with the same main chain but alkyl-thiophene side chain reported previously only exhibited a PCE of 8.08% with a V_OC of 0.78 V, a J_SC of 16.76 mA cm⁻² and an FF of 61.80%.⁴⁵ The performances of both ZnP-DMOT and ZnP-EDOT devices are higher. Especially, the FF values of ZnP-DMOT and ZnP-EDOT devices are remarkably improved, which is contributed by the di-alkoxy groups at the thiophene side chains.

Fig. 4 (a) J–V curves and (b) EQE curves of ZnP-EDOT:PC₆₁BM and ZnP-DMOT:PC₆₁BM. (c) J–V curves (d) and EQE curves of ZnP-EDOT:6TIC and ZnP-DMOT:6TIC. (e) and (f): Absorption spectra of ZnP-EDOT:PC₆₁BM and ZnP-DMOT:PC₆₁BM and ZnP-EDOT:6TIC and ZnP-DMOT:6TIC.

Table 2 Binary device performance parameters of ZnP-DMOT and ZnP-EDOT OSCs

Active layer	Conditions	V OC (V)	J SC (mA cm^−2)	FF (%)	PCE^a (%)
a The maximum PCE with the average value (in bracket) based on 20 devices. b The values for DPPEZnP-TEH:PC61BM are sourced from ref. 43.
ZnP-DMOT:PC61BM	As cast	0.83 (0.82 ± 0.01)	11.61 (11.36 ± 0.26)	36.12 (35.82 ± 0.3)	3.48 (3.12 ± 0.26)
ZnP-DMOT:PC61BM	Py + SVA	0.77 (0.76 ± 0.01)	17.91 (17.59 ± 0.34)	69.82 (69.53 ± 0.4)	9.63 (9.32 ± 0.21)
ZnP-EDOT:PC61BM	As cast	0.78 (0.76 ± 0.02)	12.12 (12.01 ± 0.16)	34.46 (34.20 ± 0.3)	3.26 (2.74 ± 0.32)
ZnP-EDOT:PC61BM	Py + SVA	0.73 (0.72 ± 0.01)	18.20 (17.82 ± 0.30)	68.21 (67.93 ± 0.4)	9.06 (8.83 ± 0.18)
DPPEZnP-TEH:PC61BM	PY + TA	0.78	16.76	61.80	8.08^b
ZnP-DMOT:6TIC	As cast	0.80 (0.79 ± 0.01)	14.83 (14.67 ± 0.14)	49.67 (49.23 ± 0.3)	5.89 (5.54 ± 0.34)
ZnP-DMOT:6TIC	DIO + SVA	0.77 (0.75 ± 0.02)	20.83 (20.67 ± 0.19)	72.00 (71.68 ± 0.3)	11.55 (11.29 ± 0.12)
ZnP-EDOT:6TIC	As-cast	0.77 (0.76 ± 0.01)	15.23 (15.04 ± 0.24)	43.84 (43.73 ± 0.2)	5.14 (4.80 ± 0.28)
ZnP-EDOT:6TIC	DIO + SVA	0.73 (0.71 ± 0.02)	21.55 (21.21 ± 0.37)	69.05 (68.73 ± 0.2)	10.86 (10.59 ± 0.16)

---

To further investigate the photovoltaic properties, we also used ZnP-DMOT and ZnP-EDOT to fabricate OSCs with non-fullerene acceptor 6TIC. The mass ratios of the small molecule donor material to 6TIC were also 1 : 1, and 0.5% volume of 1,8-diodononane (DIO) was employed as the solvent additive. After the active layers were casted, they were also treated with dichloromethane SVA. The J–V curves of the devices were measured as shown in Fig. 4c, and the photovoltaic parameters are shown in Table 2. Under the as-cast conditions, both the ZnP-EDOT and ZnP-DMOT devices had low J_SC and FF values, leading to low PCEs of 5.14% and 5.89%, respectively. After DIO + SVA treatment, the devices based on ZnP-EDOT:6TIC achieved a PCE of 10.86%, with a V_OC of 0.73 V, a J_SC of 21.55 mA cm⁻², and an FF of 69.05%, and the PCE of the devices based on ZnP-DMOT:6TIC was improved to 11.55% with an increased V_OC to 0.77 V, a slightly decreased J_SC of 20.83 mA cm⁻², and an improved FF to 72.00%. The changes in the performance parameters are consistent with previous literature reports, indicating that SVA treatment induces multiscale morphology in porphyrin/non-fullerene systems.⁴⁶

To verify the light response and J_SC values of the devices, we measured the external quantum efficiency (EQE) curves of the devices and the absorption spectra of the active layers after Py + SVA treatment, as shown in Fig. 4. For the blend films with PC₆₁BM, both show similar absorption spectra. However, the ZnP-EDOT:PC₆₁BM film shows a slightly red-shifted absorption in the range of 700–850 nm. Therefore, after the same Py + SVA treatment, the ZnP-EDOT:PC₆₁BM devices show slightly wider EQE response in the near-infrared region compared to the ZnP-DMOT:PC₆₁BM device.

We also recorded the EQE curves of the devices and absorption spectra of the active layers using non-fullerene acceptor 6TIC under as-cast and DIO + SVA treatment conditions, as shown in Fig. 4d and f, respectively. Both the as-cast and DIO + SVA-treated devices of ZnP-EDOT show slightly stronger light responses than ZnP-DMOT cells throughout the absorption range, which is in line with the slightly higher J_SC values of ZnP-EDOT devices. It is noted that all the EQE values are more than 50% and the maximum EQE value of 76% is obtained at 815 nm after DIO + SVA treatment. The higher external quantum efficiencies are ascribed to the higher conversion efficiencies of photogenerated carriers in both ZnP-EDOT and ZnP-DMOT devices.

Furthermore, the J_SC values obtained by integrating the EQE curves of ZnP-EDOT:PC₆₁BM, ZnP-DMOT:PC₆₁BM devices under as-casted and Py + SVA treatment are 11.09, 11.78, 17.69 and 17.25 mA cm⁻², respectively, and the integrated J_SC values of the as-casted and DIO + SVA-treated ZnP-EDOT:6TIC and ZnP-DMOT:6TIC devices are 14.67, 14.36, 20.75 and 19.83 mA cm⁻², respectively, which are all consistent with the measured J_SC values with less than 5% errors.

To further enhance the efficiencies of the devices, we fabricated ternary devices with the two porphyrin small molecules as the donors and PC₆₁BM and 6TIC as the acceptors. The mass ratios of the porphyrins to 6TIC and PC₆₁BM are 1 : 0.8 : 0.2. The 0.5% DIO additive and SVA treatment with dichloromethane were employed to optimize the device performance. The J–V curves of the devices were measured as shown in Fig. 5a, and the photovoltaic parameters are shown in Table 3. Compared with the corresponding binary cells, all the photovoltaic parameters of the two ternary cells are improved. The PCEs of the ternary devices based on ZnP-DMOT and ZnP-EDOT are enhanced to 13.32% and 12.96% with improved J_SC values of 23.14 and 23.89 mA cm⁻², FF values of 73.82 and 72.31, and V_OC values 0.78 and 0.75 V, respectively. ZnP-DMOT has the deeper HOMO compared to ZnP-EDOT, which can contribute to its higher V_OC of ZnP-DMOT devices. Similar to the related binary cells, though the ZnP-DMOT:6TIC:PC₆₁BM devices still show a slightly lower J_SC, the larger V_OC and FF enabled the devices to achieve a higher PCE of 13.36%.

Fig. 5 (a) J–V curves and (b) EQE curves of the ternary OSCs under 100 mW cm⁻² AM 1.5, and (c) absorption spectra of the ternary active layers.

Table 3 Photovoltaic parameters of the ternary cells

Active layer	V OC (V)	J SC (mA cm^−2)	FF (%)	PCE^a (%)
a The maximum PCE with the average value (in bracket) based on 20 devices.
ZnP-DMOT: 6TIC:PC61BM	0.78 (0.76 ± 0.02)	23.14 (23.02 ± 0.16)	73.82 (73.72 ± 0.2)	13.32 (13.14 ± 0.21)
ZnP-EDOT: 6TIC:PC61BM	0.75 (0.74 ± 0.01)	23.89 (23.75 ± 0.21)	72.31 (71.95 ± 0.3)	12.96 (12.61 ± 0.33)

---

In addition, we also recorded the EQE curves of the ternary devices and the absorption spectra of the ternary films prepared under DIO + SVA conditions, as shown in Fig. 5b and c, respectively. The EQE values of the two ternary devices are higher with the maximum EQE values exceeding 80%, indicating that both the ternary cells show good photoresponse. In addition, the EQE values of ZnP-EDOT:6TIC:PC₆₁BM cells are slightly higher almost in the whole photoresponse, which is similar to those binary cells and in line with the measured higher J_SC values of ZnP-EDOT-based cells. The integrated J_SC values from the EQE curves of ZnP-EDOT:6TIC:PC₆₁BM and ZnP-DMOT:6TIC:PC₆₁BM devices are 22.65 and 21.57 mA cm⁻², respectively, which are consistent with the measured values within an allowable error range of less than 5%.

To investigate the charge transport performance of ZnP-DMOT and ZnP-EDOT devices, we employed the space charge limited current (SCLC) method to measure the hole and electron mobilities of the blend films with hole only and electron only device structures of ITO/PEDOT:PSS/active layer/MoO₃/Ag and ITO/ZnO/active layer/PFN–Br/Ag, respectively.⁴⁷ The corresponding J–V curves are shown in Fig. 6, and the data are summarized in Table 4. When blended with PC₆₁BM, the ZnP-DMOT devices show a hole mobility (μ_h) of 4.69 × 10⁻⁴ cm² V⁻¹ s⁻¹ and an electron mobility (μ_e) of 2.12 × 10⁻⁴ cm² V⁻¹ s⁻¹, and the μ_h and μ_e values of ZnP-EDOT devices are 3.78 × 10⁻⁴ and 2.05 × 10⁻⁴ cm² V⁻¹ s⁻¹, respectively. Although the μ_h of ZnP-DMOT device is higher, its ratio of hole mobility to electron mobility (μ_h/μ_e) is larger, indicating that the hole transport and electron transport are more unbalanced and there is severe carrier recombination in the ZnP-DMOT-based devices, which is not beneficial for the J_SC.

Fig. 6 J–V curves of (a) and (c) hole only and (b) and (d) electron only devices of ZnP-DMOT and ZnP-EDOT blended with the acceptors.

Table 4 Mobility of the donors ZnP-DMOT and ZnP-EDOT blended with acceptors

Active layer	μ h (cm^2 V^−1 s^−1)	μ e (cm^2 V^−1 s^−1)	μ h/μe
ZnP-DMOT:PC61BM	4.69 × 10^−4	2.12 × 10^−4	2.21
ZnP-EDOT:PC61BM	3.78 × 10^−4	2.05 × 10^−4	1.84
ZnP-DMOT:6TIC:PC61BM	6.98 × 10^−4	6.21 × 10^−4	1.12
ZnP-EDOT:6TIC:PC61BM	5.83 × 10^−4	4.53 × 10^−4	1.29

---

For the ternary devices of ZnP-DMOT:6TIC:PC₆₁BM and ZnP-EDOT:6TIC:PC₆₁BM, the μ_h values are 6.98 × 10⁻⁴ and 5.83 × 10⁻⁴ cm² V⁻¹ s⁻¹, and the μ_e values are 6.21 × 10⁻⁴ and 4.53 × 10⁻⁴ cm² V⁻¹ s⁻¹, respectively. The μ_h/μ_e values for ZnP-DMOT and ZnP-EDOT are calculated to be 1.12 and 1.29, respectively. These results indicate that the mixing of the fullerene acceptor PC₆₁BM and non-fullerene acceptor 6TIC with the porphyrins not only significantly improves the hole and electron mobilities but also the mobility balance of the devices, which is conducive to enhance the charge transport and suppress the carrier recombination, leading to enhanced J_SC and FF.⁴⁸

To further elucidate the mechanism on the performance enhancement by PC₆₁BM incorporation into the porphyrin:6TIC system, we investigated the charge transport dynamics in binary and ternary cells based on two porphyrin donors. As shown in Fig S17 (SI), from transient photovoltage (TPV) measurements, the carrier lifetimes for the binary blends of ZnP-DMOT:6TIC and ZnP-EDOT:6TIC are 13.94 and 10.99 μs, respectively. Upon the introduction of PC₆₁BM, the lifetimes of the ternary blends are increased to 19.87 and 21.37 μs, respectively. Transient photocurrent (TPC) measurements reveal that the carrier extraction times are 0.57 and 0.63 μs for ZnP-DMOT:6TIC and ZnP-EDOT:6TIC binary blends, which are decreased to 0.34 and 0.48 μs for the ternary films, respectively. The enhanced carrier mobility, increased carrier lifetime, and accelerated charge extraction contributed to the higher J_SC and FF values for the ternary device.

Electrochemical impedance spectroscopy (EIS) was employed to analyze the performance enhancement of devices incorporating PC₆₁BM. We recorded Nyquist plots (Fig. S18, SI) under appropriate bias conditions and fitted them with an equivalent circuit comprising R_S (series resistance) in series with a parallel combination of R_CT (charge transfer resistance). The fitting results are shown in Table S3 (SI). For the ZnP-DMOT: 6TIC system, R_CT decreases from 22 528 Ω in the binary device to 14 322 Ω in the ternary device. For ZnP-EDOT: 6TIC, R_CT reduces from 24 956 Ω (binary) to 16 792 Ω (ternary). The reduction in R_CT indicates less impedance against charge transfer in ternary devices, consistent with more efficient charge transport and lower recombination losses.

To investigate the surface morphology of the active layers under the optimal conditions, first we obtained atomic force microscopy (AFM) images of the binary and ternary blend films, as shown in Fig. 7. The root mean square roughness (R_q) values of ZnP-DMOT:PC₆₁BM and ZnP-EDOT:PC₆₁BM surfaces are 0.90 and 0.94 nm, and those of ZnP-DMOT:6TIC and ZnP-EDOT:6TIC are 0.89 and 0.93 nm, respectively. For the ternary films, the R_q values of ZnP-DMOT:6TIC:PC₆₁BM and ZnP-EDOT:6TIC:PC₆₁BM are decreased to 0.78 and 0.89 nm, respectively. The overall surfaces of all the films are relatively smooth. It is speculated that the di-alkoxy thienyl side chains not only improve the intermolecular interactions between the donor molecules but also enhance their compatibility with PC₆₁BM and 6TIC, making the blend films smooth. Compared with the binary films, the R_q values of the ternary films are further reduced, indicating that the co-introduction of PC₆₁BM and 6TIC enhances the compatibility between the donors and the acceptors. All the blend films show a relatively smooth surface with appropriate phase separation and strong molecular aggregation, which are beneficial for the charge dissociation and transport in the OSCs.^49,50

Fig. 7 The AFM height and phase images of the binary and ternary blend films.

In order to deeply explore the impacts of DMOT and EDOT side chains in the porphyrin unit on the molecular orientation and crystallization of the active layers, we measured the grazing incident wide-angle X-ray scattering (GIWAXs) patterns of the blend films prepared under the optimal conditions.⁵¹ Their two-dimensional GIWAXs diffraction patterns and one-dimensional diffraction integral curves are shown in Fig. 8, and the corresponding data are summarized in Table 5. From the two-dimensional graphs, it can be seen that both the binary blend films of ZnP-DMOT:PC₆₁BM and ZnP-EDOT:PC₆₁BM show a clear ring-like peak belonging to PC₆₁BM at 1.38 Å⁻¹, indicating that PC₆₁BM in the binary active layers has no obvious crystalline orientation. However, though both the (100) layer stacking peaks appear at 0.30 Å⁻¹ in the out-of-plane (OOP) direction, the crystal coherence lengths (CCL) of the (100) plane are calculated to be 47.1 and 35.3 Å for ZnP-DMOT: PC₆₁BM and ZnP-EDOT PC₆₁BM, respectively, according to the Scherrer formula,⁵² indicating that ZnP-DMOT has better crystallization properties with PC₆₁BM than ZnP-EDOT.

Fig. 8 (a) 2D GIWAXs images and (b) the corresponding 1D integral curves of the binary and ternary blend films.

Table 5 GIWAXs parameters of small molecule donor ZnP-DMOT and ZnP-EDOT blended with acceptors

Samples	q z (Å)	Out of plane		q xy (Å)	In plane
		d-spacing (Å)	CCL (Å) (FWHM)		d-spacing (Å)	CCL(Å) (FWHM)
ZnP-DMOT:PC61BM	0.30	20.94	47.1 (0.12)	0.32	19.63	62.8 (0.09)
	1.38	4.55	25.7 (0.22)
ZnP-EDOT:PC61BM	0.30	20.94	35.3 (0.16)	0.32	19.63	56.5 (0.10)
	1.38	4.55	24.6 (0.23)
ZnP-DMOT:6TIC	1.69	3.72	11.1 (0.51)	0.34	18.48	56.5 (0.10)
ZnP-EDOT:6TIC	1.63	3.85	9.3 (0.61)	0.35	17.95	33.3 (0.17)
ZnP-DMOT:6TIC:PC61BM	1.62	3.88	8.2 (0.69)	0.35	17.95	35.3 (0.16)
ZnP-EDOT:6TIC:PC61BM	1.50	4.19	7.2 (0.79)	0.35	17.95	35.3 (0.16)

---

For ZnP-DMOT:6TIC and ZnP-EDOT:6TIC films, it can be observed that distinct (010) plane π–π stacking peaks appear at 1.63 and 1.69 Å⁻¹ in the OOP direction, and the distinct (100) layered stacking peaks appear at 0.35 and 0.34 Å⁻¹, respectively, in the in-plane (IP) direction, indicating that the introduction of 6TIC is beneficial for adjusting the morphology of the active layers to be parallel to the substrate in the face-on orientation. According to the Bragg equation, the π–π stacking distances of the (010) planes are 3.72 and 3.85 Å, respectively, suggesting that the DMOT side chains play a better role in shortening the π–π stacking distance. Calculations show that the CCLs of the (010) planes of the films are 11.1 and 9.3 Å, respectively, indicating that the DMOT side chains improve the crystallinity of the film than EDOT. Therefore, the devices based on ZnP-DMOT exhibit better face-on orientation and better crystallization performance, which is beneficial for improving the charge transport and achieving higher V_OC and other device parameters.

For ZnP-DMOT:6TIC:PC₆₁BM and ZnP-EDOT:6TIC:PC₆₁BM ternary blend films, the distinct (010) plane π–π stacking peaks appear at 1.62 and 1.50 Å⁻¹ in the OOP direction, respectively, and (100) layered stacking peaks appear at 0.35 Å⁻¹ in the IP direction. According to the Bragg equation, the π–π stacking distances of the (010) planes are 3.88 and 4.19 Å, and the CCLs of the (010) planes are 8.2 and 7.2 Å, respectively. These results show that ZnP-DMOT in the ternary blend film with 6TIC and PC₆₁BM still maintains the relatively superior face-on orientation and crystallization. However, compared with the binary films with 6TIC, the π–π stacking distances are slightly increased while the CCLs are decreased, indicating that the introduction of PC₆₁BM has negative effects on the stacking orientation and crystallization of the active layers, and the improvement of the efficiencies of the ternary devices is ascribed to the enhancement of the mobilities and more balanced hole and electron transport induced by PC₆₁BM.

Finally, we compared the thermal stability performance of two porphyrin-based devices. We placed the devices in a nitrogen-filled glove box under dark conditions, continuously heated at 80 °C for 72 hours without encapsulation and tested them at irregular intervals during this period (Fig. S19, SI). The ZnP-EDOT-based devices retained 71.32% of their initial PCE, whereas the ZnP-DMOT devices retained 70.24%. Thus, the EDOT side chain shows a marginally better thermal stability under the tested conditions, though the difference is small and within experimental uncertainty. This suggests that while both side chains provide reasonable morphological and electronic stability, the side chain rigidity may have only a modest effect on stability under the tested conditions.

Conclusions

We introduced dimethoxythiophene (DMOT) and ethylene dioxyl-thiophene (EDOT) side chains at two of the meso positions of a porphyrin core to synthesize two small molecular donors, ZnP-DMOT and ZnP-EDOT, and PCEs of 9.63% and 9.06%, respectively, are achieved for the OSCs with PC₆₁BM as the acceptor. These PCEs are very close to the state-of-the-art among fullerene-based OSCs, and are significantly higher than that (PCE: 8.08%) of the OSCs based on PC₆₁BM and the porphyrin with the same main chain but alky thiophene side chains reported previously. In addition, when non-fullerene acceptor 6TIC was employed, the ZnP-DMOT and ZnP-EDOT devices show PCEs of 11.55% and 10.86%, respectively. Experimental results show that di-alkoxy groups improve not only the crystallization of the porphyrins but also the compatibility with PC₆₁BM and 6TIC. Furthermore, the introduction of PC₆₁BM as the third component into 6TIC based binary layers significantly improves the mobilities and the balance of hole and electron transport. As a result, the ternary ZnP-DMOT:6TIC:PC₆₁BM and ZnP-EDOT:6TIC:PC₆₁BM devices achieve PCEs of 13.32% and 12.96%, respectively. These results demonstrate that the introduction of oxygen-containing groups is a promising approach for designing highly effective small-molecule donor materials for OSCs.

Author contributions

Jifa Wu: conceptualization, data curation, formal analysis, investigation, methodology, validation, writing – original draft, and writing – review and editing. Ziqin He: conceptualization, data curation, formal analysis, investigation, methodology, validation, and writing – original draft. Xiaobin Peng: conceptualization, methodology, funding acquisition, project administration, supervision, and writing – review and editing.

Conflicts of interest

The authors declare no competing financial interest.

Data availability

The authors confirm that the data supporting the findings of this study are available within the article and its supplementary information (SI). The Supplementary information includes additional experimental details, characterization data, NMR spectra, and supporting figures. See DOI: https://doi.org/10.1039/d5mh01605c.

Acknowledgements

The authors acknowledge the National Natural Science Foundation of China (52173162, 51773065 and 51861145301), the Guangdong Basic and Applied Basic Research Foundation (2022A1515011846) and the National Key Research and Development Program of China (2017YFA0206602) for the financial support.

References

1. H. Li, S. Liu, X. Wu, S. Yao, X. Hu and Y. Chen, Energy Environ. Sci., 2023, 16, 76–88.
2. L. Hu, J. Quan, J. Li, Z. Li, S. Lan, M. Yu, X. Liao, Y. Jin, X. Yin, J. Song, D. Zhou, Z. Li and Y. Chen, Adv. Mater., 2024, 36, 2413232.
3. S. Liang, B. Liu, S. Karuthedath, J. Wang, Y. He, W. L. Tan, H. Li, Y. Xu, N. Li, J. Hou, Z. Tang, F. Laquai, C. R. McNeill, C. J. Brabec and W. Li, Angew. Chem., Int. Ed., 2022, 134, e202209316.
4. X. Duan, W. Song, J. Qiao, X. Li, Y. Cai, H. Wu, J. Zhang, X. Hao, Z. Tang, Z. Ge, F. Huang and Y. Sun, Energy Environ. Sci., 2022, 15, 1563–1572.
5. C. Li, Y. Cai, P. Hu, T. Liu, L. Zhu, R. Zeng, F. Han, M. Zhang, M. Zhang, J. Lv, Y. Ma, D. Han, M. Zhang, Q. Lin, J. Xu, N. Yu, J. Qiao, J. Wang, X. Zhang, J. Xia, Z. Tang, L. Ye, X. Li, Z. Xu, X. Hao, Q. Peng, F. Liu, L. Guo and H. Huang, Nat. Mater., 2025, 24, 1626–1634.
6. Z. Chen, J. Ge, W. Song, X. Tong, H. Liu, X. Yu, J. Li, J. Shi, L. Xie, C. Han, Q. Liu and Z. Ge, Adv. Mater., 2024, 36, 2406690.
7. Y. Jiang, S. Sun, R. Xu, F. Liu, X. Miao, G. Ran, K. Liu, Y. Yi, W. Zhang and X. Zhu, Nat. Energy, 2024, 9, 975–986.
8. X. Gu, Y. Wei, R. Zeng, J. Lv, Y. Hou, N. Yu, S. Tan, Z. Wang, C. Li, Z. Tang, Q. Peng, F. Liu, Y. Cai, X. Zhang and H. Huang, Angew. Chem. Int. Ed., 2024, 64, e202418926.
9. D. Hu, H. Tang, J. Lv, Z. Liao, Q. Chen, H. Chen and S. Lu, Sustainable Energy Fuels, 2021, 5, 3593–3597.
10. R. Zeng, L. Zhu, M. Zhang, W. Zhong, G. Zhou, J. Zhuang, T. Hao, Z. Zhou, L. Zhou, N. Hartmann, X. Xue, H. Jing, F. Han, Y. Bai, H. Wu, Z. Tang, Y. Zou, H. Zhu, C.-C. Chen, Y. Zhang and F. Liu, Nat. Commun., 2023, 14, 4148.
11. R. Zhou, Z. Jiang, C. Yang, J. Yu, J. Feng, M. A. Adil, D. Deng, W. Zou, J. Zhang, K. Lu, W. Ma, F. Gao and Z. Wei, Nat. Commun., 2019, 10, 5393.
12. J. Qin, Z. Chen, P. Bi, Y. Yang, J. Zhang, Z. Huang, Z. Wei, C. An, H. Yao, X. Hao, T. Zhang, Y. Cui, L. Hong, C. Liu, Y. Zu, C. He and J. Hou, Energy Environ. Sci., 2021, 14, 5903–5910.
13. B. Qiu, Z. Chen, S. Qin, J. Yao, W. Huang, L. Meng, H. Zhu, Y. Yang, Z.-G. Zhang and Y. Li, Adv. Mater., 2020, 32, 1908373.
14. V. Piradi, X. Xu, Z. Wang, J. Ali, Q. Peng, F. Liu and X. Zhu, ACS Appl. Mater. Interfaces, 2019, 11, 6283–6291.
15. Y. Guo, J. Wu, Z. Lin, F. Tang, L. Yuan, H. Wu and X. Peng, ACS Appl. Mater. Interfaces, 2024, 16, 17531–17539.
16. X. J. Pan, J. F. Wu, L. G. Xiao, B. Yap, R. D. Xia and X. B. Peng, ChemSusChem, 2021, 14, 3614–3621.
17. L. Yuan, J. Wu, H. Wu, Y. Li, Y. Guo and X. Peng, Asian J. Org. Chem., 2025, 14, e202400789.
18. H. Qin, L. Li, F. Guo, S. Su, J. Peng, Y. Cao and X. Peng, Energy Environ. Sci., 2014, 7, 1397–1401.
19. T. Lai, X. Chen, L. Xiao, L. Zhang, T. Liang, X. Peng and Y. Cao, Chem. Commun., 2017, 53, 5113–5116.
20. L. Xiao, T. Lai, X. Liu, F. Liu, T. P. Russell, Y. Liu, F. Huang, X. Peng and Y. Cao, J. Mater. Chem. A, 2018, 6, 18469–18478.
21. T. Lai, L. Xiao, K. Deng, T. Liang, X. Chen, X. Peng and Y. Cao, ACS Appl. Mater. Interfaces, 2018, 10, 668–675.
22. S. Zhou, C. Li, J. Ma, Y. Guo, J. Zhang, Y. Wu and W. Li, J. Mater. Chem. C, 2018, 6, 717–721.
23. Y. Guo, Y. Liu, Q. Zhu, C. Li, Y. Jin, Y. Puttisong, W. Chen, F. Liu, F. Zhang, W. Ma and W. Li, ACS Appl. Mater. Interfaces, 2018, 10, 32454–32461.
24. R. Shivhare, G. J. Moore, A. Hofacker, S. Hutsch, Y. Zhong, M. Hambsch, T. Erdmann, A. Kiriy, S. C. B. Mannsfeld, F. Ortmann and N. Banerji, Adv. Mater., 2022, 34, 2101784.
25. X. Zhou, W. Tang, P. Bi, Z. Liu, W. Lu, X. Wang, X. Hao, W.-K. Wong and X. Zhu, J. Mater. Chem. C, 2019, 7, 380–386.
26. V. Cuesta, M. Vartanian, P. Malhotra, S. Biswas, P. de la Cruz, G. D. Sharma and F. Langa, J. Mater. Chem. A, 2019, 7, 11886–11894.
27. H. Shen, Y. Xu, W. Zou, W. Zhang, J. Li, P. Cai, Y. Guo, H. Xu, X. Hao, Y. Sun, Y. Kan, Y. Yang and K. Gao, Chem. Eng. J., 2023, 469, 144063.
28. Z. Luo, T. Xu, C. E. Zhang and C. Yang, Energy Environ. Sci., 2023, 16, 2732–2758.
29. K. Ma, W. Feng, H. Liang, H. Chen, Y. Wang, X. Wan, Z. Yao, C. Li, B. Kan and Y. Chen, Adv. Funct. Mater., 2023, 33, 2214926.
30. C. Li, J. Zhou, J. Song, J. Xu, H. Zhang, X. Zhang, J. Guo, L. Zhu, D. Wei, G. Han, J. Min, Y. Zhang, Z. Xie, Y. Yi, H. Yan, F. Gao, F. Liu and Y. Sun, Nat. Energy, 2021, 6, 605–613.
31. Y. Cui, H. Yao, J. Zhang, K. Xian, T. Zhang, L. Hong, Y. Wang, Y. Xu, K. Ma, C. An, C. He, Z. Wei, F. Gao and J. Hou, Adv. Mater., 2020, 32, 1908205.
32. J. Hai, Y. Song, L. Li, X. Liu, X. Shi, Z. Huang, G. Qian, Z. Lu, J. Yu, H. Hu and S. Chen, Adv. Funct. Mater., 2023, 33, 2213429.
33. H. Yang, C. Cui and Y. Li, Acc. Mater. Res., 2021, 2, 986–997.
34. A. M. H. Cheung, H. Yu, S. Luo, Z. Wang, Z. Qi, W. Zhou, L. Arunagiri, Y. Chang, H. Yao, H. Ade and H. Yan, J. Mater. Chem. A, 2020, 8, 23239–23247.
35. H. Bin, Z.-G. Zhang, L. Gao, S. Chen, L. Zhong, L. Xue, C. Yang and Y. Li, J. Am. Chem. Soc., 2016, 138, 4657–4664.
36. T. L. H. Mai, Z. Sun, S. Kim, S. Jeong, S. Lee, J. Park and C. Yang, Energy Environ. Sci., 2024, 17, 7435–7444.
37. S. Kashani, Z. Wang, C. Risko and H. Ade, Mater. Horiz., 2023, 10, 443–453.
38. C. Liao, M. Zhang, X. Xu, F. Liu, Y. Li and Q. Peng, J. Mater. Chem. A, 2019, 7, 716–726.
39. P. Jiang, S. Ming, Q.-Q. Jia, Y. Liu, H. Lu, M. Li, X. Xu, H.-B. Li and Z. Bo, J. Mater. Chem. A, 2018, 6, 21335–21340.
40. M. Casademont-Viñas, D. Capolat, A. Quesada-Ramírez, M. Reinfelds, G. Trimmel, M. Sanviti, J. Martín, A. R. Goñi, T. Kirchartz and M. Campoy-Quiles, J. Mater. Chem. A, 2024, 12, 16716–16728.
41. S. Tsuzuki, R. Ono, S. Inoue, S. Matsuoka and T. Hasegawa, Commun. Chem., 2024, 7, 253.
42. X. Wang, Z. Zhang, L. Kong, M. Luo, M. Chen, L. Zhang and J. Chen, J. Mater. Chem. A, 2024, 12, 5731–5739.
43. L. Li, Y. Huang, J. Peng, Y. Cao and X. Peng, J. Mater. Chem. A, 2013, 1, 2144–2150.
44. K. Gao, W. Deng, L. Xiao, Q. Hu, Y. Kan, X. Chen, C. Wang, F. Huang, J. Peng, H. Wu, X. Peng, Y. Cao, T. P. Russell and F. Liu, Nano Energy, 2016, 30, 639–648.
45. K. Gao, L. S. Li, T. Q. Lai, L. G. Xiao, Y. Huang, F. Huang, J. B. Peng, Y. Cao, F. Liu, T. P. Russell, R. A. J. Janssen and X. B. Peng, J. Am. Chem. Soc., 2015, 137, 7282–7285.
46. K. Gao, S. B. Jo, X. Shi, L. Nian, M. Zhang, Y. Kan, F. Lin, B. Kan, B. Xu, Q. Rong, L. Shui, F. Liu, X. Peng, G. Zhou, Y. Cao and A. K. Y. Jen, Adv. Mater., 2019, 31, 1807842.
47. J. Chen, J. Cao, L. Liu, L. Xie, H. Zhou, J. Zhang, K. Zhang, M. Xiao and F. Huang, Adv. Funct. Mater., 2022, 32, 2200629.
48. W. Shockley and H. J. Queisser, J. Appl. Phys., 1961, 32, 510–519.
49. R. Qian, J. C. Liao, G. P. Luo and H. B. Wu, Org. Electron., 2022, 108, 106614.
50. R. Lin, Z. Luo, Y. Wang, J. Wu, T. Jia, W. Zhang, X. Gu, Y. Liu, L. Xiao and Y. Min, Chem. Eng. J., 2025, 503, 158635.
51. P. Müller-Buschbaum, Adv. Mater., 2014, 26, 7692–7709.
52. J. Rivnay, S. C. B. Mannsfeld, C. E. Miller, A. Salleo and M. F. Toney, Chem. Rev., 2012, 112, 5488–5519.

---