Molecular interaction regulation by adding a third component with high miscibility suppresses the energetic disorder and reduces energy loss for efficient ternary solar cells

Abstract

In the advancement of organic solar cells (OSCs), the ternary strategy has emerged as an effective approach for fabricating devices with high photovoltaic performance. In this contribution, we have introduced a novel wide bandgap donor, PBTz-Cl, into the D18:L8-BO binary system to address the excessive aggregation of D18. PBTz-Cl exhibits excellent miscibility with D18 in ternary films due to their similar building blocks. Our findings show that the addition of PBTz-Cl forms a molecular alloy within the amorphous regions of D18. This not only boosts additional exciton generation in D18 through Förster resonance energy transfer, but also suppresses the non-radiative recombination energy loss (ΔE₃) due to the reduced crystallinity difference between D18 and L8-BO by enhancing the crystallinity of L8-BO. The optimized ternary blend film exhibits superior microstructure morphology and enhanced charge dynamics, leading to enhanced photovoltaic performance and remarkable stability. The resulting OSCs show a remarkable increase in performance with a J_SC of 25.30 mA cm⁻², a V_OC of 0.911 V, and an FF of 78.33%. Our study highlights the effectiveness of the strategy, which derives from the synergistic effects of compatible polymer donors to precisely regulate molecular packing and optimize film morphology for improved photovoltaic performance.

---

Introduction

Organic solar cells (OSCs) have garnered considerable attention and hold promising prospects for applications due to their lightweight nature, flexibility, and enormous potential for high-throughput roll-to-roll manufacturing.^1–4 Over the past three decades, various polymer donors, such as P3HT, PM6, and D18, along with non-fullerene electron acceptors like ITIC and Y6, have been successfully developed.^4–13 Moreover, advancements in device and interface engineering enable the comprehensive exploration of new materials to enhance photovoltaic performance.^14–18 In summary, the remarkable advancements in material design and device fabrication processes have resulted in the rapid progress of organic photovoltaic technology. Currently, state-of-the-art OSCs have achieved power conversion efficiencies (PCE) exceeding 19% in single-junction devices, gradually approaching the performance level of their inorganic counterparts.^19–27

The ternary strategy has been recognized as an effective way to boost the efficiency while maintaining a simple device fabrication process.^28–34 Ternary solar cells (TSCs) typically consist of two acceptors and one donor or two donors and one acceptor in the active layer. By introducing a third component to the host binary blend, molecular packing, crystallinity, interface distribution of donors and acceptors, and light absorption spectra can be effectively manipulated,^35–38 leading to enhanced exciton and charge dynamics, suppressed charge recombination, and thereby elevated short-circuit current (J_SC) and fill factor (FF).^39,40 For instance, adding a highly crystalline molecular donor, DRTB-T-C4, into the PM6:Y6 host binary system significantly improved charge transport, reduced recombination loss, and increased the FF to 81.3%.⁴¹ Similarly, adding a polymer acceptor, N2200, into the PBDB-T:IT-M binary blend effectively regulated vertical phase distribution for enhanced charge transport and collection and reduced trap states.⁴² Moreover, by constructing alloy-like components, the charge transfer state in ternary films could be finely tuned, resulting in reduced energy loss and elevated open-circuit voltage (V_OC).⁴³ Utilizing the fluorescence resonance energy transfer (FRET) of materials in ternary films has also been found to effectively convert high-energy excitons into charges, leading to a significant improvement in J_SC. For example, the efficient FRET between the wide bandgap acceptor IDTT-M and Y6 provides a non-radiative pathway for IDTT-M in the ternary system as well as an additional pathway to facilitate exciton separation and charge collection.⁴⁴ However, most studies have focused on the effects of energy and charge transfer in ternary blend films, with limited research on the miscibility and aggregation behavior, which are closely linked to the materials' compatibility in multi-component active layers and the energetic disorder of blend films.^45–48 Recently reported studies have demonstrated that inhibiting the aggregation of small molecule electron acceptors can effectively enhance the photovoltaic performance.⁴⁹ For example, by introducing the 3D-shaped guest acceptor TPA-4PDI into the PM6:Y6 binary system, the excessive aggregation behavior of Y6 was effectively suppressed. As a result, highly efficient ternary devices were achieved, demonstrating simultaneously increased open-circuit voltage (V_OC), J_SC and FF.⁵⁰

D18 is a highly crystalline polymer and commonly used as a seeding agent in ternary systems to control the crystallinity and phase separation of blend films.^51–53 However, its poor solubility and excessive aggregation often result in unfavorable morphology when processed from blend solutions. The key to elevating the photovoltaic performance of D18 based OSCs lies in alleviating the excessive aggregation of the D18 polymer at the molecular level. From the perspective of molecular thermodynamics and kinetics, molecular interactions play a significant role in crystallization, phase separation, and microstructure morphology. In conjugated organic materials, π–π interactions, dictated by their chemical structures, are key to the miscibility of materials. High miscibility within a ternary blend system is likely to form a mixed amorphous phase, which in turn improves solubility.^54,55 Building on this understanding, by carefully designing a third component that mixes well with D18, it is feasible to obtain a blend film with a suitable microstructure for fabrication of high-efficiency D18 based TSCs. In this contribution, we introduced PBTz-Cl, a new wide bandgap donor with good chloroform solubility, and successfully used it as the third component in the D18:L8-BO binary system.⁵⁶ The polymer PBTz-Cl has the same benzo[1,2-b:4,5-b′]dithiophene (BDT) electron donor units as D18 and when added in an appropriate amount, it significantly boosts the photovoltaic performance of the TSCs. The contact angle measurements and grazing-incidence wide-angle X ray scattering (GIWAXS) study suggested the outstanding miscibility and strong molecular interactions between PBTz-Cl and D18. This good compatibility leads to the formation of a mixed amorphous phase, which improves the solubility of D18. Consequently, the optimized ternary film displayed an advantageous microstructure morphology. In addition, the enhanced crystallinity of the L8-BO acceptor in the ternary blend also indicated the reduced crystallinity difference. As a result, the enhanced miscibility and reduced crystallinity difference synergistically contribute to the suppressed energetic disorder and reduced energy loss (ΔE₃), resulting in a high V_OC for ternary solar cells. Moreover, PBTz-Cl also functioned as a Förster resonance energy transfer (FRET) donor, generating additional D18 excitons for high photocurrent. This was conducive to markedly improved exciton dynamics and more balanced charge transfer, strongly corroborated by transient photovoltage (TPV), transient photocurrent (TPC), and femtosecond transient absorption spectroscopy (fs-TAS) investigations. Ultimately, the resultant ternary device exhibited a remarkable elevation in photovoltaic performance, with a significant increase in a J_SC value of 25.30 mA cm⁻², a V_OC value of 0.911 V and an FF of 78.33%.

Results and discussion

The molecular structures of D18, PBTz-Cl, and acceptor L8-BO employed in this study are depicted in Fig. 1a. We determined the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) energy levels for D18, the D18:PBTz-Cl blend, PBTz-Cl and L8-BO using cyclic voltammetry (CV) measurements (Fig. S1, ESI†). The derived energy level diagram is showcased in Fig. 1b. The calculated HOMO/LUMO levels for D18 and L8-BO are –5.35/–3.39 eV and –5.67/–3.84 eV, respectively, whereas PBTz-Cl has HOMO/LUMO levels at –5.60 eV/–3.61 eV. Notably, PBTz-Cl exhibits a deeper HOMO level compared to D18, which is advantageous for promoting a higher V_OC in TSCs. In the D18:PBTz-Cl blend, the HOMO level is reduced, aligning well with L8-BO, which might reduce voltage losses and enhance V_OC in TSCs.

Fig. 1 (a) The chemical structures of D18, PBTz-Cl and L8-BO. (b) Energy level alignment of D18, D18 : PBTz-Cl (with a weight ratio of 0.975 : 0.025), PBTz-Cl and L8-BO. (c) UV-vis-NIR absorption spectra of D18 and PBTz-Cl neat films and their blend films. (d) UV-vis-NIR absorption spectra of binary films based on D18:L8-BO and PBTz-Cl:L8-BO and the optimized ternary blend film of D18:PBTz-Cl:L8-BO.

We investigated the optical absorption of D18, PBTz-Cl, and their blend (weight ratio of D18 : PBTz-Cl of 0.975 : 0.025). The UV-vis-NIR absorption spectra, shown in Fig. 1c, exhibit distinct optical absorption peaks at 545 nm and 580 nm for D18 and 510 nm for PBTz-Cl. The D18:PBTz-Cl blend film demonstrates an enhanced absorption coefficient across nearly the entire spectral region without peak shifts relative to the neat D18 film. As depicted in Fig. 1d, the binary and ternary blend films exhibit similar absorption trends to the characteristic peaks of D18. The absorption peak of L8-BO in the D18:L8-BO binary film is located at 795 nm, which experiences a slight blue-shift to 790 nm in the ternary blend. Importantly, the introduction of PBTz-Cl enhances the vibrational 0–1 peak of L8-BO, resulting in a lower I_0–0/I_0–1 ratio (1.52) in the ternary film compared to the binary film (1.60). This suggests a modification in the molecular packing and aggregation properties of L8-BO. Such changes are likely to foster more robust intermolecular interactions, enabling L8-BO to aggregate through face-to-face self-assembly. According to the reported study, the changes in molecular packing modes and molecular aggregation significantly impacted the non-radiative recombination of charge transfer states and V_OC in OSCs.⁵⁷ These changes in molecular packing not only enhance the films' optical properties but also their charge transport and collection capacity.^58–60 Steady-state photoluminescence (PL) spectra for the neat D18 film, the D18:L8-BO binary film, and the D18:PBTz-Cl:L8-BO ternary film were recorded, with the corresponding spectra illustrated in Fig. S2 (ESI†). In these films, D18 was excited at 550 nm. Both binary and ternary films exhibit notable fluorescence quenching, especially in the ternary film, indicating more effective charge transfer. Additionally, the absorption spectrum of D18 and the PL emission spectrum of PBTz-Cl overlapped (Fig. 1e), leading to a complete quenching of PBTz-Cl's PL emission and an increase in theD18's emission as weight ratios of their blends changed (Fig. 1f). This suggests a strong FRET between D18 and PBTz-Cl, implying that they are closely packed (within 10 nm) and show strong intermolecular interactions.^32,61 This non-radiative energy transfer from PBTz-Cl to D18 boosts the production of additional D18 excitons, potentially offering a new route to the increased photocurrent seen in ternary devices.

Then, we fabricated OSC devices with a conventional structure of ITO/PEDOT:PSS/active layer/PDIN/Ag. The photovoltaic performance of these devices, with various solution concentrations and the weight ratios of PBTz-Cl ranging from 0 to 20 wt%, is detailed in Fig. S3 and Tables S1, S2 (ESI†). Fig. 2a displays the representative current density–voltage (J–V) curves of D18:L8-BO, PBTz-Cl:L8-BO, and D18:PBTz-Cl:L8-BO binary and ternary OSCs, with photovoltaic parameters summarized in Table 1. As illustrated in Fig. S4 (ESI†), good reproducibility in photovoltaic performance was confirmed for 22 D18:L8-BO based binary and D18:PBTz-Cl:L8-BO based ternary devices. Specifically, D18:L8-BO based binary devices demonstrated a fill factor (FF) of 76.69%, a J_SC of 24.37 mA cm⁻², and a moderate PCE of 16.87%. Devices based on PBTz-Cl:L8-BO, however, exhibited a lower PCE of 10.19% due to a reduced FF of 58.17% and a J_SC of 19.39 mA cm⁻². Interestingly, adding 2.5 wt% PBTz-Cl into the D18:L8-BO enabled TSC device with an impressive PCE exceeding 18%, along with an improved FF of 78.33%, a J_SC of 25.30 mA cm⁻², and a V_OC of 0.911 V. The increased J_SC could be attributed to the enhanced absorption, more efficient charge transfer and the existence of non-radiative energy transfer. The optimized TSC device delivered a V_OC of 0.911 V, exceeding that of the D18:L8-BO based binary device (V_OC = 0.903 V) and even the PBTz-Cl:L8-BO based binary device (V_OC = 0.904 V), despite its deep HOMO energy level of –5.60 eV. Fig. 2b presents the correlation between the device performance and the PBTz-Cl content, showing an enhancement in J_SC and FF at 2.5 wt% PBTz-Cl. However, higher weight ratios of PBTz-Cl diminished the photovoltaic performance. Specifically, for 10 wt% PBTz-Cl, both the FF and J_SC significantly reduced to 24.13 mA cm⁻² and 75.61%, respectively, and the PCE dropped to 16.74%. Despite this, the V_OC value in TSCs experienced a monotonic increase with the increase of PBTz-Cl. The observed enhancement in V_OC suggests that the incorporation of PBTz-Cl might promote an alloy-like structure. This change is associated with a diminished energy offset and reduced energy loss.^62,63 To comprehensively understand the superior J_SC in TSCs, we performed external quantum efficiency (EQE) measurements, and the results are depicted in Fig. 2c. The D18:L8-BO-based binary device shows a significant photoresponse between 515–585 nm and 650–800 nm, which aligns with the absorption profiles of D18 and L8-BO, respectively. Conversely, the PBTz-Cl:L8-BO based binary device achieves a low EQE across the entire wavelength range due to the inferior morphology that hinders exciton dissociation and collection. However, the optimized ternary device demonstrates superior EQE values for almost the entire spectrum, with a notable peak of 86% at 550 nm. This indicates that excitons, especially those directly generated by D18 and the additional D18 exciton from FRET, are effectively separated and collected by the electrodes. The calculated current densities (J_cal) from EQE curves are 23.62 mA cm⁻² for D18:L8-BO, 24.64 mA cm⁻² for D18:PBTZ-Cl:L8-BO, and 19.30 mA cm⁻² for PBTz-Cl:L8-BO based devices, consistent with the results obtained from J–V measurements.

Fig. 2 (a) The representative J–V curves of binary and ternary devices. (b) The ternary device parameter trends as a function of the weight ratio of PBTz-Cl. (c) EQE curves of optimized binary and ternary devices. (d) Storage stability of D18:L8-BO based binary and optimized ternary devices.

Table 1 Summary data of photovoltaic performance based on binary devices and ternary devices

Active layer	V OC (V)	J SC/JCal (mA cm^−2)	FF (%)	PCE (%)
a Average PCE values were obtained from 22 devices. b Average PCE values were obtained from 10 devices.
D18:L8-BO	0.903	24.37/23.62	76.69	16.87 (16.33 ± 0.302)^a
D18:PBTz-Cl:L8-BO	0.911	25.30/24.64	78.33	18.05 (17.73 ± 0.271)^a
PBTz-Cl:L8-BO	0.904	19.39/19.30	58.17	10.19 (9.72 ± 0.325)^b

---

To further study the reason for a monotonic increase in V_OC for TSCs, the total energy losses (E_loss) in TSCs with different ratios of PBTz-Cl were analyzed through highly sensitive EQE (s-EQE) and external electroluminescence quantum efficiency (EQE_EL) measurements, which are depicted in Fig. 3a–e. The detail data of energy losses are listed in Table 2. The E_loss can be divided into three parts as follows:⁶⁴

E_loss = E_g − qV_OC = (E_g − qV^SQ_OC) + (qV^SQ_OC − qV^rad_OC) + (qV^rad_OC − qV_OC) = ΔE₁ + ΔE₂ + ΔE₃ (1)

where ΔE₁ is the inherent energy difference between the optical gap edge E_g and the Shockley–Queisser (SQ) limit qV_OC (qV^SQ_OC); ΔE₂ is the energy loss from additional radiative recombination below the optical gap, which is the energy difference between qV^SQ_OC and radiative limit qV_OC (qV^rad_OC); ΔE₃ is the energy difference between qV^rad_OC and qV_OC, quantified by EQE_EL, representing nonradiative recombination losses of devices: ΔE₃ = −k_BT ln(EQE_EL) (k_B is Boltzmann's constant and T is the absolute temperature). When 2.5%wt PBTz-Cl was added to the devices, E_loss decreased from 0.556 eV to 0.547 eV, leading to an elevated V_OC. This decline continued with more PBTz-Cl, dropping to 0.544 eV at 10%wt and reaching 0.522 eV at 20%wt. These reductions suggested that energy transfer between D18 and PBTz-Cl minimizes energy disorder and thus energy loss. The material-dependent ΔE₁ and donor–acceptor ratio-dependent ΔE₂ remained relatively stable (∼0.265 eV and ∼0.056 eV, respectively), indicating that non-radiative recombination energy loss ΔE₃ is the main contributor to the total energy loss in TSCs. A decrease in ΔE₃ from 0.236 eV to 0.205 eV was noted with increasing PBTz-Cl content, up to 20%wt. However, the highest PBTz-Cl content also led to a higher charge-separated state energy (E_CS), suggesting a deeper HOMO level of the D18:PBTz-Cl donor and a reduced charge generation force (E_g − E_CS), which could lower the photocurrent.⁶⁵ Conversely, a 2.5%wt addition of PBTz-Cl achieved a balance between non-radiative recombination and charge generation, enhancing both J_SC and V_OC. To assess the impact of FRET between D18 and PBTz-Cl on energy disorder, we measured the Urbach energy (E_U) to assess the density of states (DoS) distribution, which characterizes the width of the DoS tail.⁶⁶ By fitting the edge of s-EQE with the equation α_hv = α₀ exp(hv/E_U) (α_hv is the absorption coefficient, h is Planck's constant, v is the incident photon frequency, and α₀ is the absorption coefficient at E_g),⁶⁷ we determined that ternary devices with PBTz-Cl exhibited a lower E_U (25.97 meV compared to 26.71 meV in the D18:L8-BO based binary device), indicating a narrower DoS distribution, reduced energy disorder, and subsequently less non-radiative recombination and higher V_OC.

Fig. 3 (a–d) s-EQE and EL spectra of different ratios of PBTz-Cl (0–20 wt%). (e) EQE_EL spectra of TSC devices at different injected currents. (f) A histogram of three part energy loss (ΔE₁, ΔE₂, and ΔE₃) in different PBTz-Cl contents.

Table 2 Detailed values of energy losses of OSCs based on D18:PBTz-Cl

D18 : PBTz-Cl : L8-BO	E g (eV)	qV OC (eV)	E loss (eV)	qV ^SQOC (eV)	qV ^radOC (eV)	ΔE1 (eV)	ΔE2 (eV)	ΔE3 (eV)	EQEEL (%)
1 : 0 : 1.4	1.461	0.905	0.556	1.196	1.141	0.265	0.055	0.236	0.345
0.975 : 0.025 : 1.4	1.461	0.914	0.547	1.196	1.139	0.264	0.057	0.226	10.894
0.9 : 0.1 : 1.4	1.461	0.917	0.544	1.196	1.141	0.265	0.056	0.223	23.370
0.8 : 0.2 : 1.4	1.457	0.931	0.526	1.192	1.136	0.265	0.056	0.205	3.506

---

Stability is also a very important indicator for evaluating the photovoltaic performance of OSCs. We measured the storage stability and thermal stability of the devices. Notably, the TSC device impressively retained more than 97% of its initial PCE after 1000 hours, showcasing remarkable storage stability. However, the D18:L8-BO based binary device only maintained 89% of its initial PCE (Fig. 2d). This reduction is mainly due to a drop in the FF. When subjected to 100 °C for an hour, the TSC device still held onto over 86% of its initial PCE, outperforming the D18:L8-BO device which only managed to maintain 80% of its initial PCE (Fig. S5, ESI†). This suggests that the D18:L8-BO binary device suffers from greater morphological degradation. Atomic force microscopy (AFM) images (Fig. S6a–f, ESI†) reveal that the root-mean-square (RMS) roughness of the D18:L8-BO film significantly decreased from 1.17 to 0.99 nm after thermal annealing, while the RMS roughness of both PBTz-Cl:L8-BO and D18:PBTz-Cl:L8-BO films remained stable, changing by an insignificant 0.05 nm. This indicates that the ternary films remain morphologically stable even after thermal annealing at high temperature.

To elucidate the mechanisms for the improved photovoltaic performance, hole-only and electron-only devices were fabricated to evaluate transport properties utilizing the space charge-limited current method. The corresponding J–V curves in the dark are illustrated in Fig. S7 and S8 (ESI†), with hole mobility (μ_h) and electron mobility (μ_e) values listed in Table S3 (ESI†). The D18:L8-BO film's μ_h (2.71 × 10⁻⁴ cm² V⁻¹ s⁻¹) was much lower than its μ_e (4.39 × 10⁻⁴ cm² V⁻¹ s⁻¹), leading to an imbalance in charge transport (μ_e/μ_h ratio of 1.61) that could induce charge accumulation and promote charge recombination. The D18:PBTz-Cl:L8-BO based ternary film, however, displayed a significantly enhanced μ_e of 8 × 10⁻⁴ cm² V⁻¹ s⁻¹ and μ_h of 7.42 × 10⁻⁴ cm² V⁻¹ s⁻¹, respectively. Moreover, the μ_e/μ_h ratio decreases to 1.33, indicating more balanced charge transport conducive to higher J_SC and FF. Charge transfer between D18 and PBTz-Cl was further confirmed by the higher J_SC in OSCs fabricated with a D18:PBTz-Cl mixture compared to neat D18, as shown in Fig. S9 (ESI†) and listed in Table S4 (ESI†).

For further understanding the factors that contribute to the enhanced photovoltaic performance in ternary devices, we examined the exciton dissociation and charge collection in both binary and ternary OSCs. We measured the photocurrent density as a function of effective applied voltage (J_ph–V_eff) under short-circuit conditions, as depicted in Fig. 4a. Under a high V_eff, the J_ph could be saturated (J_sat), reflecting the photon absorption capacity of the devices. As listed in Table S5 (ESI†), the increased J_sat suggests better light absorption in ternary films. The exciton dissociation efficiency (P_diss) and charge collection efficiency (P_coll) of D18:L8-BO based devices are estimated to be 95.84% and 87.32%, respectively. An optimized ternary device improved these efficiencies to 96.39% (P_diss) and 89.09% (P_coll), demonstrating that adding PBTz-Cl could facilitate exciton dissociation at the donor–acceptor interface and charge transport and charge collection, which consist with enhanced charge generation forces as mentioned above. Additionally, we investigated charge recombination behavior by examining J–V curves of devices under varying light intensity (P_light). The dependence of J_SC on P_light can be expressed as J_SC ∝ (P_light)^S, where S is the exponential factor for evaluating the bimolecular recombination. The curves, presented in Fig. 4b, show S values of 0.98 for D18:L8-BO, 0.98 for D18:PBTz-Cl:L8-BO, and 0.96 for PBTz-Cl:L8-BO, indicating negligible bimolecular recombination in ternary systems and implying that adding 2.5 wt% PBTz-Cl would not create harmful recombination centers that are detrimental to device performance.

Fig. 4 (a) The J_ph–V_eff curves and (b) J_SCversus light intensity plots of the D18:L8-BO based binary and optimized ternary devices. (c) TPC and (d) TPV curves of binary films and optimized ternary devices.

In addition to steady-state measurements, time-resolved experiments were also used to investigate the charge dynamics. The time-resolved photoluminescence (TRPL) spectra (Fig. S10, ESI†) showed that the ternary film with D18:PBTz-Cl:L8-BO had a fluorescence lifetime (τ) of 398 ps, shorter than the 404 ps of the binary D18:L8-BO film, indicating more efficient charge transfer. Additionally, transient photovoltage (TPV) and transient photocurrent (TPC) measurements revealed that the ternary device had a longer charge carrier lifetime (τ_TPV) of 5.57 μs (versus 5.51 μs for the binary device) (Fig. 4c) and a faster charge extraction time (τ_TPC) of 0.170 μs (versus 0.246 μs for the binary device) (Fig. 4d), suggesting reduced recombination and enhanced charge sweep-out in the ternary device. Evidently, adding PBTz-Cl actually has multiple advantages, including enhancing the charge transport and extraction, as well as suppressing the charge recombination, as confirmed by the highest J_SC and FF observed in the ternary devices. Finally, femtosecond transient absorption (fs-TA) spectroscopy was further employed to investigate the excited-state dynamics process. Fig. 5a–c depict the fs-TA spectra of D18:L8-BO, D18:PBTz-Cl:L8-BO and PBTz-Cl:L8-BO blend films at selected time delays. A pump laser of 800 nm selectively excites the acceptor L8-BO in the blend films. Immediately after excitation, a strong positive excited state absorption (ESA) peak appeared around 900 nm while a negative ground state bleach (GSB) peak is observed at around 820 nm, with additional GSB peaks at ∼600 nm and ∼525 nm for D18 and PBTz-Cl, reaching their maximum within 100 ps and 4 ps, respectively. This suggests a rapid hole-transfer from L8-BO to the other components, generating a charge transfer state. The hole transfer kinetics was further analyzed using a biexponential model with a fast process (τ₁) and a lower process (τ₂), corresponding to the direct exciton dissociation of the charge transfer states at the donor/acceptor interfaces (exciton lifetime) and diffusion-mediated exciton dissociation at the donor/acceptor interfaces (charge carrier lifetime), respectively. As shown in Fig. 5d and Table S6 (ESI†), the PBTz-Cl:L8-BO blend showed the quickest charge transfer with a τ₁ of only 0.8 ps, but also a fast decay (τ₂ of 321 ps) to the ground state, leading to significant charge recombination and lower device performance. In contrast, the D18:L8-BO and D18:PBTz-Cl:L8-BO based blend films had longer τ₁ of 9.4 ps and 8.9 ps, respectively. The D18 GSBs (524 nm) for the D18:L8-BO based binary film and the D18:PBTz-Cl:L8-BO ternary film continuously increase and reach a maximum after the initial rise. This indicates a diffusion-mediated charge process that occurs after the ultrafast charge transfer at the donor/acceptor interfaces. Therefore, these two films exhibited a longer τ₂, 1763 ps for the ternary film and 1402 ps for the binary film. This result is also in accordance with the above TPV measurement. According to the reported study,⁶⁸ the poor performance of the PBTz-Cl:L8-BO based device was attributed to the excessive mixing of the donor and acceptor, along with undersized donor/acceptor phase domains, causing short diffusion-mediated charge transfer and the fast charge recombination. Conversely, the D18:L8-BO based blend film benefitted from larger phase domains, resulting in slower charge transfer and recombination. The D18:PBTz-Cl:L8-BO based ternary blend film struck a balance between fast charge transfer and slow charge recombination, leading to a higher charge transfer state yield and a longer charge lifetime, aligning with their better device performance. This also agrees well with the energy loss analysis mentioned above, suggesting that the ternary blend manages to balance between charge generation and non-radiative recombination effectively.

Fig. 5 (a–c) fs-TA spectra of D18:L8-BO, D18:PBTz-Cl:L8-BO and PBTz-Cl:L8-BO blend films at selected time delays. (d) TA kinetics and fitting curves showing the hole transfer process. The excitation wavelength is ∼800 nm, and the excitation fluence is 3.6 × 10¹³ cm⁻² pulse⁻¹.

To gain a better understanding of the enhanced exciton and charge dynamics, as well as the resulting boost in photovoltaic efficiency due to film morphology enhancements, we first examined the film surfaces using AFM. Fig. S11 (ESI†) illustrates the surface morphology of the pristine D18 film, which exhibits a fibrous network with a RMS roughness of 1.38 nm. This structure is advantageous for hole transport, but possibly hindering L8-BO permeation, resulting in unbalanced charge transport. The PBTz-Cl film, however, had a smoother and more uniform surface with a RMS of 0.85 nm, likely due to its high solubility and minimal aggregation leading to small phase domains. The D18:PBTz-Cl blend film maintained the fibrous structure with a slightly smoother RMS of 1.24 nm. Among three blend films, the D18:PBTz-Cl:L8-BO based ternary film had the roughest surface at 1.25 nm, compared to 1.17 nm for D18:L8-BO and 0.88 nm for PBTz-Cl:L8-BO (Fig. S6a–c, ESI†). This suggests that, when the donor blend (D18:PBTz-Cl) initially precipitates, L8-BO integrates more effectively into the forming ternary film during ternary film formation, promoting orderly packing and crystallinity, thereby enhancing charge transport, suppressing recombination and increasing photocurrent.

The fundamental origin of morphological evolution with the addition of PBTz-Cl was investigated by conducting the contact angle measurements to assess the compatibility between D18, PBTz-Cl, D18:PBTz-Cl, and L8-BO, and is shown in Fig. 6. The calculated surface energy (γ) and the Flory–Huggins interaction parameters (χ) between D18, PBTz-Cl and L8-BO are listed in Table 3. The γ values for D18, PBTz-Cl and L8-BO films are found to be 19.89, 20.29 and 23.22 mN m⁻¹, respectively. The close γ values for D18 and PBTz-Cl imply favorable miscibility of these two polymers. Furthermore, the χ also supported this compatibility,^69–71 with χ values of 0.002κ for D18 and PBTz-Cl and 0.037κ for PBTz-Cl and L8-BO, both of which are lower than 0.056κ for D18 and L8-BO. This suggested the excellent miscibility between D18/PBTz-Cl and PBTz-Cl/L8-BO, which would facilitate the formation of a D18:PBTz-Cl alloy-like phase and benefit L8-BO permeating into the polymer network for the orderly packing and crystallization during the film formation. According to the reported studies,^62,72–77 such compatibility can lead to the formation of the alloy-like phase between two donors accounting for the monotonically elevated V_OC of ternary devices and variations in the energy level of polymer blends as the PBTz-Cl content increases (Table S1 (ESI†) and Fig. 1).

Fig. 6 The contact angle images of D18, PBTz-Cl, D18:PBTz-Cl, and L8-BO films tested by using water and glycerol liquid drops.

Table 3 The surface energy (γ) and Flory–Huggins interaction parameters (χ) of different films obtained from water and glycerol contact angle measurements

Film	γ [mN m^−1]	χ with D18 [mN m^−1]	χ with L8-BO [mN m^−1]
D18	19.89	—	0.056κ
PBTz-Cl	20.29	0.002κ	0.037κ
L8-BO	22.07	0.056κ	—

---

Finally, we performed grazing-incidence wide-angle X-ray scattering (GIWAXS) measurements to study the molecular crystallinity and orientation in neat, binary blend, and the optimized ternary films. Fig. 7a–f illustrate the 2D diffraction patterns and the extracted line-cut profiles in both in-plane (IP) and out-of-plane (OOP) directions. The accompanying data for d-spacing and the crystal coherence length (CCL) are found in Tables S7 and S8 (ESI†). The neat films of D18 and PBTz-Cl exhibit a preferred face-on orientation, with intense π–π stacking (010) diffraction peaks at 1.66 Å⁻¹ and 1.56 Å⁻¹ and lamellar packing (100) peaks at 0.31 Å⁻¹ and 0.32 Å⁻¹, respectively. A higher d-spacing of 4.03 Å for neat PBTz-Cl indicates a more loosely arranged π–π stacking in the OOP direction compared to that of 3.78 Å for the D18 film. In the D18:PBTz-Cl binary blend film, the π–π stacking (010) diffraction peak remains at 1.66 Å⁻¹ in the OOP direction, but with a narrower full-width-half-maximum (FWHM) of 0.434 Å⁻¹, signifying an increase in CCL. These findings demonstrate that the molecular packing of the D18:PBTz-Cl blend adopts the structure of D18, with PBTz-Cl well-integrated into the D18 framework without disrupting its molecular packing. This potentially forms a molecular alloy within the amorphous regions of the D18 polymer. The featured distributions of PBTz-Cl in ternary films effectively optimize the film morphology, leading to better charge transfer, lower energy losses, and consequently elevated V_OC and PCE. All blend films displayed clear π–π stacking (010) diffraction peaks, contributed by the overlapping π–π stacking from the polymer donors (D18 or PBTz-Cl) and L8-BO. In binary blends, the π–π stacking (010) peak was at 1.73 Å⁻¹, while in ternary blends, it appeared at 1.71 Å⁻¹ with a d-spacing of 3.67 Å. More importantly, ternary blends also showed strong (100) peak scattering and a reduced FWHM for the L8-BO (100) peak from 0.314 Å⁻¹ to 0.304 Å⁻¹, with an increased CCL from 18.00 Å to 19.34 Å. These GIWAXS results implied the enhanced crystallinity of L8-BO in the ternary film, reducing the crystallinity difference between D18 and L8-BO that could inhibit the non-radiative recombination and achieve smaller ΔE₃ for higher V_OC in OSCs, which correlates well with the improved charge mobility, suppressed non-radiative recombination and decreased ΔE₃ in TSCs mentioned above.

Fig. 7 (a) 2D GIWAXS patterns of neat D18, PBTz-Cl and their blend films. (b) 2D GIWAXS patterns of two binary blend films and optimized ternary blend films. GIWAXS line-cuts of neat D18, PBTz-Cl and their blend films (c) in the in-plane direction and (d) in the out-of-plane direction. GIWAXS line-cuts of two binary blend films and optimized ternary blend films (c) in the in-plane direction and (d) in the out-of-plane direction.

Conclusions

In this contribution, we introduced a new soluble polymer donor, PBTz-Cl, into the D18:L8-BO host system to fabricate high-efficiency TSCs. The structural synergy between PBTz-Cl and D18 led to an alloy-like structure, which could fine-tune the blend film morphology and subsequently enhanced the charge dynamics while reducing energy loss from non-radiative recombination. Additionally, the addition of PBTz-Cl facilitated the orderly packing of L8-BO molecules. As a result, we successfully developed a robust TSC device based on D18 : PBTz-Cl : L8-BO with a weight ratio of 0.975 : 0.025 : 1.4. This device achieved an impressive PCE of 18.05%, with a V_OC of 0.911 V, a J_SC of 25.30 mA cm⁻², and an FF of 78.33%. Our approach demonstrated the potential of combining compatible polymers to control the energy offset, regulate molecular packing, and optimize film morphology, which could effectively contribute to the suppressed energetic disorder and reduced non-radiative recombination energy loss. This approach holds great promise for the fabrication of high-performance TSCs and should promote the development of OSC technology.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

L. X. acknowledges the financial support from the National Natural Science Foundation of China (51903057) and the Guangzhou Basic and Applied Basic Research Foundation (2023A04J0349). We also acknowledge the support from the National Key Research and Development Program of China (2020YFB0408100). L. X. would like to thank the Analysis and Test Center of the Guangdong University of technology for the UV-vis-NIR and PL spectroscopy measurements.

References

1. Y. Liang, D. Zhang, Z. Wu, T. Jia, L. Lüer, H. Tang, L. Hong, J. Zhang, K. Zhang, C. J. Brabec, N. Li and F. Huang, Nat. Energy, 2022, 7, 1180–1190.
2. M. Kaltenbrunner, M. S. White, E. D. Głowacki, T. Sekitani, T. Someya, N. S. Sariciftci and S. Bauer, Nat. Commun., 2012, 3, 770.
3. K. Fukuda, K. Yu and T. Someya, Adv. Energy Mater., 2020, 10, 2000765.
4. X. Meng, L. Zhang, Y. Xie, X. Hu, Z. Xing, Z. Huang, C. Liu, L. Tan, W. Zhou, Y. Sun, W. Ma and Y. Chen, Adv. Mater., 2019, 31, 1903649.
5. K. Jin, Z. Xiao and L. Ding, J. Semicond., 2021, 42, 060502.
6. G. Zhang, F. R. Lin, F. Qi, T. Heumüller, A. Distler, H.-J. Egelhaaf, N. Li, P. C. Y. Chow, C. J. Brabec, A. K. Y. Jen and H.-L. Yip, Chem. Rev., 2022, 122, 14180–14274.
7. J. Yuan and Y. Zou, Org. Electron., 2022, 102, 106436.
8. Z. Yao, X. Cao, X. Bi, T. He, Y. Li, X. Jia, H. Liang, Y. Guo, G. Long, B. Kan, C. Li, X. Wan and Y. Chen, Angew. Chem., Int. Ed., 2023, 62, e202312630.
9. J.-W. Lee, J. S. Park, H. Jeon, S. Lee, D. Jeong, C. Lee, Y.-H. Kim and B. J. Kim, Chem. Soc. Rev., 2024, 53, 4674–4706.
10. A. Mishra, Energy Environ. Sci., 2020, 13, 4738–4793.
11. P. Ding, D. Yang, S. Yang and Z. Ge, Chem. Soc. Rev., 2024, 53, 2350–2387.
12. A. Mishra and G. D. Sharma, Angew. Chem., Int. Ed., 2023, 135, e202219245.
13. B. Liu, C. Li, X. Gu, Y. Han, Z. Wei, Y. Cai, X. Zhang, H. Huang and Y. Huo, Chin. J Chem., 2024, 42, 485–490.
14. Z. He, B. Xiao, F. Liu, H. Wu, Y. Yang, S. Xiao, C. Wang, T. P. Russell and Y. Cao, Nat. Photonics, 2015, 9, 174–179.
15. K. Zhang, Z. Hu, C. Sun, Z. Wu, F. Huang and Y. Cao, Chem. Mater., 2017, 29, 141–148.
16. T. Liu, L. Huo, S. Chandrabose, K. Chen, G. Han, F. Qi, X. Meng, D. Xie, W. Ma, Y. Yi, J. M. Hodgkiss, F. Liu, J. Wang, C. Yang and Y. Sun, Adv. Mater., 2018, 30, 1707353.
17. L. Guo, Q. Li, J. Ren, Y. Xu, J. Zhang, K. Zhang, Y. Cai, S. Liu and F. Huang, Energy Environ. Sci., 2022, 15, 5137–5148.
18. X. Wu, Y. Wu, S. Peng, L. Xiao, Z. Xiao, W. Zhang, G. Ren, Y. Min and Y. Liu, Sol. RRL, 2023, 7, 2300136.
19. W. Gao, F. Qi, Z. Peng, F. R. Lin, K. Jiang, C. Zhong, W. Kaminsky, Z. Guan, C.-S. Lee, T. J. Marks, H. Ade and A. K. Y. Jen, Adv. Mater., 2022, 34, 2202089.
20. Y. Wei, Z. Chen, G. Lu, N. Yu, C. Li, J. Gao, X. Gu, X. Hao, G. Lu, Z. Tang, J. Zhang, Z. Wei, X. Zhang and H. Huang, Adv. Mater., 2022, 34, 2204718.
21. L. Kong, Z. Zhang, N. Zhao, Z. Cai, J. Zhang, M. Luo, X. Wang, M. Chen, W. Zhang, L. Zhang, Z. Wei and J. Chen, Adv. Energy Mater., 2023, 13, 2300763.
22. S. Luo, C. Li, J. Zhang, X. Zou, H. Zhao, K. Ding, H. Huang, J. Song, J. Yi, H. Yu, K. S. Wong, G. Zhang, H. Ade, W. Ma, H. Hu, Y. Sun and H. Yan, Nat. Commun., 2023, 14, 6964.
23. L. Zhu, M. Zhang, J. Xu, C. Li, J. Yan, G. Zhou, W. Zhong, T. Hao, J. Song, X. Xue, Z. Zhou, R. Zeng, H. Zhu, C.-C. Chen, R. C. I. MacKenzie, Y. Zou, J. Nelson, Y. Zhang, Y. Sun and F. Liu, Nat. Mater., 2022, 21, 656–663.
24. D. Li, N. Deng, Y. Fu, C. Guo, B. Zhou, L. Wang, J. Zhou, D. Liu, W. Li, K. Wang, Y. Sun and T. Wang, Adv. Mater., 2023, 35, 2208211.
25. L. Wang, C. Chen, Y. Fu, C. Guo, D. Li, J. Cheng, W. Sun, Z. Gan, Y. Sun, B. Zhou, C. Liu, D. Liu, W. Li and T. Wang, Nat. Energy, 2024, 9, 208–218.
26. Y. Sun, L. Wang, C. Guo, J. Xiao, C. Liu, C. Chen, W. Xia, Z. Gan, J. Cheng, J. Zhou, Z. Chen, J. Zhou, D. Liu, T. Wang and W. Li, J. Am. Chem. Soc., 2024, 146, 12011–12019.
27. 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.
28. D. Qiu, H. Zhang, C. Tian, J. Zhang, L. Zhu, Z. Wei and K. Lu, Adv. Mater., 2023, 35, 2307398.
29. T. N. Duan, W. Y. Feng, Y. L. Li, Z. X. Li, Z. Zhang, H. Z. Liang, H. B. Chen, C. Zhong, S. Jeong, C. D. Yang, S. S. Chen, S. R. Lu, O. A. Rakitin, C. X. Li, X. J. Wan, B. Kan and Y. S. Chen, Angew. Chem., Int. Ed., 2023, 62, e202308832.
30. Z. Chen, J. Zhu, D. Yang, W. Song, J. Shi, J. Ge, Y. Guo, X. Tong, F. Chen and Z. Ge, Energy Environ. Sci., 2023, 16, 3119–3127.
31. C. Zhang, J. Li, W. Deng, J. Dai, J. Yu, G. Lu, H. Hu and K. Wang, Adv. Funct. Mater., 2023, 33, 2301108.
32. L. Xiao, B. He, Q. Hu, L. Maserati, Y. Zhao, B. Yang, M. A. Kolaczkowski, C. L. Anderson, N. J. Borys, L. M. Klivansky, T. L. Chen, A. M. Schwartzberg, T. P. Russell, Y. Cao, X. Peng and Y. Liu, Joule, 2018, 2, 2154–2166.
33. X. Ma, A. Zeng, J. Gao, Z. Hu, C. Xu, J. H. Son, S. Y. Jeong, C. Zhang, M. Li, K. Wang, H. Yan, Z. Ma, Y. Wang, H. Y. Woo and F. Zhang, Natl. Sci. Rev., 2021, 8, nwaa305.
34. S. Cheng, J. Qiao, P. Lu, W. Qin and X. Hao, Chin. J Chem., 2024, 42, 1275–1283.
35. N. Zhang, K. Jiang, F. R. Lin, Y. An, G. Du, T. Xia, A. K. Y. Jen and H.-L. Yip, Mater. Today Energy, 2024, 42, 101548.
36. K. N'Konou, S. Y. Kim and N. Y. Doumon, Carbon Energy, 2024, e579.
37. N. Y. Doumon, F. V. Houard, J. Dong, P. Christodoulis, M. V. Dryzhov, G. Portale and L. J. A. Koster, J. Mater. Chem. C, 2019, 7, 5104–5111.
38. S. Jeong, A. Rana, J.-H. Kim, D. Qian, K. Park, J.-H. Jang, J. Luke, S. Kwon, J. Kim, P. S. Tuladhar, J.-S. Kim, K. Lee, J. R. Durrant and H. Kang, Adv. Sci., 2023, 10, 2206802.
39. N. Gasparini, A. Salleo, I. McCulloch and D. Baran, Nat. Rev. Mater., 2019, 4, 229–242.
40. Q. An, F. Zhang, J. Zhang, W. Tang, Z. Deng and B. Hu, Energy Environ. Sci., 2016, 9, 281–322.
41. R. Ma, T. Liu, Z. Luo, K. Gao, K. Chen, G. Zhang, W. Gao, Y. Xiao, T.-K. Lau, Q. Fan, Y. Chen, L.-K. Ma, H. Sun, G. Cai, T. Yang, X. Lu, E. Wang, C. Yang, A. K. Y. Jen and H. Yan, ACS Energy Lett., 2020, 5, 2711–2720.
42. H. Liu, Z. Chen, R. Peng, Y. Qiu, J. Shi, J. Zhu, Y. Meng, Z. Tang, J. Zhang, F. Chen and Z. Ge, Chem. Eng. J., 2023, 474, 145807.
43. W. Peng, Y. Lin, S. Y. Jeong, Z. Genene, A. Magomedov, H. Y. Woo, C. Chen, W. Wahyudi, Q. Tao, J. Deng, Y. Han, V. Getautis, W. Zhu, T. D. Anthopoulos and E. Wang, Nano Energy, 2022, 92, 106681.
44. D. Li, L. Wang, C. Guo, Y. Liu, B. Zhou, Y. Fu, J. Zhou, D. Liu, W. Li and T. Wang, ACS Mater. Lett., 2023, 5, 2065–2073.
45. D. Li, L. Zhu, X. Liu, W. Xiao, J. Yang, R. Ma, L. Ding, F. Liu, C. Duan, M. Fahlman and Q. Bao, Adv. Mater., 2020, 32, 2002344.
46. P. Bi, S. Zhang, T. Xiao, M. Cui, Z. Chen, J. Ren, C. Qin, G. Lu, X. Hao and J. Hou, Sci. China: Chem., 2021, 64, 599–607.
47. J. Gao, N. Yu, Z. Chen, Y. Wei, C. Li, T. Liu, X. Gu, J. Zhang, Z. Wei, Z. Tang, X. Hao, F. Zhang, X. Zhang and H. Huang, Adv. Sci., 2022, 9, 2203606.
48. L. Xiao, X. Wu, G. Ren, M. A. Kolaczkowski, G. Huang, W. Tan, L. Ma, Y. Liu, X. Peng, Y. Min and Y. Liu, Adv. Funct. Mater., 2021, 31, 2105304.
49. T. Wang, X. Wang, R. Yang and C. Li, Sol. RRL, 2021, 5, 2100496.
50. D. Lee, H. Hwang, D. H. Sin, C. Park, S. G. Han, J. Mun, J. Noh, S. H. Kim, H. Kim, H. Lee, C. Lee, J. Rho, K. Cho and M. S. Jeong, ACS Energy Lett., 2021, 6, 2610–2618.
51. A. A. Mohapatra, V. Tiwari and S. Patil, Energy Environ. Sci., 2021, 14, 302–319.
52. S. Peng, J. Luo, P. Li, L. Xiao, C. Yang, Y. Jin, R. Lin, Y. Huo, Y. Liu and Y. Min, Chem. Eng. J., 2023, 474, 145501.
53. H. Lu, G. Ran, Y. Liu, Z. Pei, W. Liu, Y. Liu, Z. Tang, W. Zhang and Z. Bo, Adv. Funct. Mater., 2023, 33, 2301866.
54. M. Liu, X. Ge, X. Jiang, F. Guo, S. Gao, Q. Peng, L. Zhao and Y. Zhang, Adv. Funct. Mater., 2023, 33, 2300214.
55. Q. Liu, Y. Jiang, K. Jin, J. Qin, J. Xu, W. Li, J. Xiong, J. Liu, Z. Xiao and K. Sun, Sci. Bull., 2020, 65, 272–275.
56. J. Luo, S. Peng, J. Zhang, Y. Jin, D. Chen, Y. Mu, L. Xiao, D. Hu, Q.-D. Yang, S. Ji, Y. Min and Y. Huo, Org. Electron., 2023, 120, 106860.
57. R. Xu, Y. Jiang, F. Liu, W. Su, W. Liu, S. Xu, H. Fan, C. Jiang, Q. Zong, W. Zhang and X. Zhu, Chem. Eng. J., 2023, 464, 142507.
58. N. Yi, Q. Ai, W. Zhou, L. Huang, L. Zhang, Z. Xing, X. Li, J. Zeng and Y. Chen, Chem. Mater., 2019, 31, 10211–10224.
59. X. Xu, Y. Li and Q. Peng, Adv. Mater., 2022, 34, 2107476.
60. J. Luo, S. Peng, J. Zhang, Y. Jin, D. Chen, Y. Mu, L. Xiao, D. Hu, Q.-D. Yang, S. Ji, Y. Min and Y. Huo, Org. Electron., 2023, 120, 106860.
61. X. K. Chen, M. K. Ravva, H. Li, S. M. Ryno and J. L. Brédas, Adv. Energy Mater., 2016, 6, 1601325.
62. S. Cai, L. Chen, D. Zha and Y. Chen, J. Polym. Sci., Part A: Polym. Chem., 2013, 51, 624–634.
63. R. Ma, H. Li, T. A. D. Peña, X. Xie, P. W.-K. Fong, Q. Wei, C. Yan, J. Wu, P. Cheng, M. Li and G. Li, Adv. Mater., 2023, 35, 2304632.
64. Q. Zhao and F. He, J. Energy Chem., 2024, 93, 174–192.
65. A. R. Clapp, I. L. Medintz, J. M. Mauro, B. R. Fisher, M. G. Bawendi and H. Mattoussi, J. Am. Chem. Soc., 2004, 126, 301–310.
66. L. Xiao, T. Liang, K. Gao, T. Lai, X. Chen, F. Liu, T. P. Russell, F. Huang, X. Peng and Y. Cao, ACS Appl. Mater. Interfaces, 2017, 9, 29917–29923.
67. K. Yu, W. Song, Y. Li, Z. Chen, J. Ge, D. Yang, J. Zhang, L. Xie, C. Liu and Z. Ge, Small Struct., 2021, 2, 2100099.
68. Y. Yan, Y. Zhang, Y. Liu, Y. Shi, D. Qiu, D. Deng, J. Zhang, B. Wang, M. A. Adil and K. Amin, Adv. Energy Mater., 2022, 12, 2200129.
69. K. Nakano, Y. Chen, B. Xiao, W. Han, J. Huang, H. Yoshida, E. Zhou and K. Tajima, Nat. Commun., 2019, 10, 2520.
70. T. Kirchartz, B. E. Pieters, J. Kirkpatrick, U. Rau and J. Nelson, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 83, 115209.
71. F. Urbach, Phys. Rev., 1953, 92, 1324.
72. H. Zhuo, X. Li, J. Zhang, C. Zhu, H. He, K. Ding, J. Li, L. Meng, H. Ade and Y. Li, Nat. Commun., 2023, 14, 7996.
73. X. Liao, Q. Xie, Y. Guo, Q. He, Z. Chen, N. Yu, P. Zhu, Y. Cui, Z. Ma, X. Xu, H. Zhu and Y. Chen, Energy Environ. Sci., 2022, 15, 384–394.
74. Z. Zhang, D. Deng, Y. Li, J. Ding, Q. Wu, L. Zhang, G. Zhang, M. J. Iqbal, R. Wang, J. Zhang, X. Qiu and Z. Wei, Adv. Energy Mater., 2022, 12, 2102394.
75. H. Liang, Y. Wang, X. Guo, D. Yang, X. Xia, J. Wang, L. Zhang, Y. Shi, X. Lu and M. Zhang, J. Mater. Chem. A, 2022, 10, 10926–10934.
76. F. Liu, L. Zhou, W. Liu, Z. Zhou, Q. Yue, W. Zheng, R. Sun, W. Liu, S. Xu, H. Fan, L. Feng, Y. Yi, W. Zhang and X. Zhu, Adv. Mater., 2021, 33, 2100830.
77. X. Wang, J. Wang, P. Wang, C. Han, F. Bi, J. Wang, N. Zheng, C. Sun, Y. Li and X. Bao, Adv. Mater., 2023, 35, 2305652.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4tc03741c

‡ S. P and R. L contributed equally to this work.

---