Sonication-induced J-aggregation in nonhalogenated solvents boosts exciton delocalization for high-efficiency organic solar cells

Sonication-induced J-aggregation in nonhalogenated solvents boosts exciton delocalization for high-efficiency organic solar cells | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Sonication-induced J-aggregation in nonhalogenated solvents boosts exciton delocalization for high-efficiency organic solar cells Xiaotao Hao, Mingxu Zhou, Xinyue Xu, Dongrui Wang, Jiali Song, and 14 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6915241/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 25 Nov, 2025 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Abstract The commercialization of organic solar cells (OSCs) requires eliminating halogenated solvents from their production. However, the disordered molecular aggregation of nonfullerene acceptors in nonhalogenated solvents hinders the realization of OSCs with high power conversion efficiencies (PCEs). Herein, this problem is addressed using an external physical field modulation strategy involving the sonication of nonfullerene acceptor solutions. The application of ultrasound field induces a transformation from disordered molecular aggregation to ordered J-aggregation via microstreaming and shear stress in nonhalogenated solvents. The aggregation is effectively preserved within films, causing the formation of a fibril network with enhanced π-π stacking interactions. Moreover, sonication promotes the conversion of localized excitons to intra-moiety delocalized excitons and suppresses molecular vibrations, thus favoring charge separation and reducing nonradiative recombination. Consequently, the PCEs of PM6:L8-BO-based binary and ternary devices fabricated using o -xylene as a nonhalogenated solvent are 19.43% and 20.41% (certified 19.84%), which is among the highest values reported for OSCs produced without halogenated solvents. The binary devices also exhibit high thermal stabilities, with the T 80 lifetime exceeding 10000 h. The developed strategy is applicable to various small-molecule acceptor and polymer acceptor systems, facilitating control disordered aggregation in nonfullerene acceptor solutions and thus paving the way for practical industrial applications. Physical sciences/Materials science/Materials for energy and catalysis/Solar cells Physical sciences/Energy science and technology/Renewable energy/Solar energy/Photovoltaics/Solar cells organic solar cells ultrasound field J-aggregation exciton delocalization molecular vibration Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Owing to their intrinsic flexibility, semitransparency, and remarkable cost-effectiveness, organic solar cells (OSCs) hold promise for next-generation photovoltaic technologies 1 – 5 . Rapid advancements in material synthesis, molecular modification, and device engineering have resulted in single-junction OSCs with power conversion efficiencies (PCEs) exceeding 20%, which represents an important milestone on the road to industrial production 6 – 10 . However, these PCEs are strongly affected by spin-coating processes that rely on highly toxic halogenated solvents (such as chloroform) incompatible with large-scale production 11 – 13 . In contrast, the low solubility of numerous nonfullerene acceptors (NFAs) in ecofriendly nonhalogenated solvents leads to substantial aggregation during processing 14 – 17 , with disordered molecular orientations severely hindering exciton dissociation and charge transport. Consequently, a substantial efficiency gap exists between OSCs processed using halogenated and nonhalogenated solvents 18 – 21 , which highlights the importance of controlling NFA preaggregation in nonhalogenated solvents for achieving high photovoltaic performance 22 – 25 . The disordered aggregation state of NFAs in nonhalogenated solvents is typically transformed into more ordered H- or J-aggregation through deliberate induction to eliminate its adverse effects. Compared to H-aggregation (face-to-face stacking), J-aggregation is formed through intermolecular sliding packings, exhibiting distinctive photophysical properties in terms of expanding light absorption and enhancing charge transport 26 . Various approaches have been explored to induce the formation of J-aggregation to improve the PCEs of OSCs. For example, Zhang et al. optimized the π-π stacking and refined the phase separation by promoting donor J-aggregation with highly volatile additives 26 . Li et al. employed molecular modification to weaken the interaction between acceptor molecules and the solvent. This helps to facilitate pronounced J-aggregation, thereby enhancing charge extraction and inhibiting charge recombination 27 . Despite the progress enabled by these approaches, the additive- and molecular-modification-based strategies involve complex synthesis and precise chemical control 28 , 29 . These simplicity- and adaptability-related challenges hinder widespread practical applications of OSCs. External physical fields, such as heat, light, and electric stimuli, enable precise modulation of material solution state, film-formation process, and post-deposition metastable phases through straightforward processing 30 , 31 . Ultrasound emerges as a uniquely powerful tool to selectively cleave noncovalent interaction networks via physical shear effects at the molecular scale by harnessing tunable acoustic cavitation energy 32 , 33 . Concurrently, ultrasound-induced microstreaming can orchestrate molecular rearrangement, offering exceptional control over aggregation pathways. This efficient, ecofriendly approach that remains largely uncharted in organic photovoltaics holds significant potential to address the detrimental preaggregation of NFAs in nonhalogenated solvents. Elucidating the physical mechanism underlying the sonication-induced regulation of NFA aggregation is therefore essential for achieving high-performance OSCs processed by nonhalogenated solvents. Herein, we use sonication to manipulate the preaggregation of NFAs in nonhalogenated solvents. The results of molecular dynamics simulations, quantum chemical computations, and multiple experiments demonstrate that sonication induces a transition from amorphous aggregation to ordered J-aggregation because of the effects of microstreaming and shear stress generated during the sonication process. Morphological studies reveal that sonication enhances intermolecular interactions in neat and blend films and thus promotes ordered molecular stacking. A global fitting model is developed to analyze transient absorption data to quantify the contributions of different charge separation (CS) pathways, and that of the pathways dominated by intra-moiety delocalized excitons is shown to markedly increase after sonication, which indicates a concomitant exciton dissociation enhancement. Sonication is also shown to inhibit molecular vibrations and thereby hinder nonradiative recombination. The binary and ternary OSCs prepared using sonication feature PCEs of 19.43% and 20.41% (certified 19.84%), respectively, along with high thermal stabilities ( T 80 lifetime > 10000 h). The sonication strategy is applicable to other small-molecule and polymer acceptor systems, offering unique insights into the development of high-performance OSCs based on nonhalogenated solvents. Results Mechanism of sonication-regulated NFA aggregation Figure 1a shows the chemical structures of the donor (PM6) and acceptor (L8-BO), and illustrates the sonication-assisted film fabrication processes based on the sequential deposition of these species. Initially, a PM6 layer is constructed on an indium tin oxide (ITO) substrate, and a sonicated solution of L8-BO in o -xylene ( o -xy) is deposited onto the PM6 layer. For numerous organic materials, amorphous aggregate formation due to nanoscale precipitation is observed in low-solubility solvents (such as o -xy), unlike in the case of high-solubility solvents (such as chloroform) 34 . This problem is solved by using ultrasound filed. Sonication causes cavitation, i.e., the formation of tiny bubbles that can rapidly grow and collapse under the periodic pressure changes caused by ultrasonic waves 35 (Supplementary Fig. 1). This typically induces physical and chemical effects. Supplementary Fig. 2 presents the Raman spectra of L8-BO solution. The unaltered peak positions after sonication suggest that no chemical reaction have occurred, but only physical effects exist. Aggregation state changes were probed by conducting in-situ photoluminescence (PL) spectroscopy measurements of the L8-BO solution ( o -xy) during sonication (Fig. 1b and supplementary Fig. 3). We evaluated the in-situ PL spectrum variations across different frequencies and pinpointed the optimal sonication settings at 40 kHz, 720 W, and 3 minutes, as detailed in the Supplementary Information. Upon sonication, the emission peak of L8-BO shifted to substantially longer wavelengths, moving from 793 nm to 805 nm within three minutes, which indicated a concomitant increase in the extent of solution-phase J-aggregation 36 . Moreover, J-aggregation was well maintained in the solution within one minute after sonication was stopped (Fig. 1b), which was sufficient for film preparation. Molecular dynamics simulations were used to verify the effects of sonication on L8-BO molecules dissolved in o -xy (Fig. 1c). Without sonication (in the control system), the L8-BO molecules formed a dispersed state, with the radius of gyration ( R g ) and solvent-accessible surface area (SASA) fluctuating around their mean values. Conversely, in the case of sonication, the R g and SASA values notably decreased before stabilizing, suggesting that sonication increased the degree of aggregation (Fig. 1d and Supplementary Fig. 4). Root mean square deviations (RMSDs) were calculated to observe the overall conformational changes relative to the initial structure throughout the simulation period (Fig. 1e). In both systems, the RMSD initially increased before reaching equilibrium, which reflected rapid convergence toward a stable equilibrium state. The local magnification of simulation snapshots revealed that the control system contained some amorphous structures and H-aggregates, while a marked increase in the number of J-aggregates was observed after sonication, in line with the in-situ PL spectroscopy results. Quantum chemical computations were performed to clarify the mechanisms underlying the sonication-induced J-aggregation (Supplementary Figs. 5a and b). An extended version of the Su–Schrieffer–Heeger tight-binding model was utilized, with a comprehensive description of the calculation process provided in the Supplementary Information 37 . J-aggregation can be categorized into A-to-A and A-to-D types. Supplementary Fig. 5c shows the band gaps of different aggregation modes. The redshift of the PL spectrum is consistent with a reduction in band gap, suggesting a transformation into the A-to-D type J-aggregation. Hence, we focused exclusively on this type of J-aggregation in subsequent analyses. The results (Supplementary Figs. 5d and e) indicated that under various conditions, the aggregation energy of J-aggregation consistently exceeded that of H-aggregation, i.e., the latter aggregation was energetically preferred 38 . Hence, NFAs tended to form H-aggregates in the absence of sonication. Cavitation bubbles are generated during the sonication process. As depicted in Supplementary Fig. 6, the periodic contraction and expansion of cavitation bubbles create a marked velocity gradient between the bubble surface and the surrounding solvent. This gradient drives the generation of microstreaming and shear stress (a pair of parallel reaction forces) 39 . Microstreaming pushes L8-BO molecules to aggregate on the bubble surface. Then the disordered molecules will be oriented and arranged under the effect of shear stress 40 , and subsequently become ordered J-aggregation. Meanwhile, the H-aggregation can overcome the energy barrier after acquiring energy and sliding to ordered J-aggregation. The aggregation behavior of sonicated L8-BO in o -xy was investigated using temperature-dependent PL spectroscopy (Fig. 1f and Supplementary Fig. 7). The long-wavelength shoulder peak (855nm)/main peak (805nm) intensity ratio ( I 0-1 / I 0-0 ) obtained through the double-peak fitting analysis of the PL spectra was employed to quantify the effects of sonication on the aggregation state at different temperatures (Fig. 1g and Supplementary Table 1) 41 . Sonicated L8-BO exhibited more pronounced I 0-1 / I 0-0 variations, along with a stronger dependence of this ratio on the solution temperature compared with the untreated sample, which manifested the enhanced intermolecular interactions and an increased tendency for aggregation. As mentioned above, J-aggregation was well maintained in the solution phase, further influencing the film morphology. Subsequently, we analyzed the ultraviolet (UV)–visible absorption and PL spectra of the corresponding films (Supplementary Figs. 8 and 9). Sonication induced red shifts in the absorption and PL peaks of the L8-BO film, while predictably enhancing its crystallinity 42 . This effect was quantitatively assessed using grazing-incidence wide-angle X-ray scattering (GIWAXS) measurements (Figs. 1h and i, Supplementary Fig. 10, and Supplementary Table 2). Prominent π-π stacking peaks of the control (1.750 Å) and sonicated (1.767 Å) L8-BO films were observed in the out-of-plane (OOP) direction 43 . The decrease in d -spacing from 3.590 Å to 3.556 Å after sonication indicated a closer intermolecular packing. The calculated crystal coherence length (CCL) increased from 19.98 Å in the control film to 21.34 Å in the sonicated film. Thus, the sonicated L8-BO film exhibited a higher crystallinity and more ordered molecular stacking than its nonsonicated counterpart. Time-resolved photoluminescence (TRPL) measurements revealed that sonication increased the exciton lifetime of the L8-BO film from 471 ps to 641 ps, possibly by favoring ordered J-aggregation 44 . Blend film morphology Sonicated L8-BO was combined with PM6 via sequential deposition to construct layer-by-layer (LBL)-type devices and thus assess the role of sonication in determining blend film morphology. According to the corresponding two-dimensional (2D) GIWAXS patterns (Figs. 2a and b), both PM6 and L8-BO exhibited a preferential face-on orientation in blend films, which favored charge transport, as evidenced by prominent (100) lamellar stacking peaks in the in-plane (IP) direction and strong (010) π-π stacking peaks in the OOP direction. After sonication, the CCL of the blend films increased from 21.58 Å to 23.08 Å, and the d -spacing decreased from 3.588 Å to 3.514 Å (Fig. 2c and Supplementary Table 3). These results suggested that sonication facilitated crystallization and ordered molecular stacking within the active layers. Grazing-incidence small-angle X-ray scattering (GISAXS) measurements were performed to characterize the donor and acceptor domain sizes in the blend films and thus investigate phase separation behavior at the nanoscale (Supplementary Fig. 11). The Debye–Anderson–Brumberger and fractal models were used for fitting 45 . The relevant fitting steps are presented in the Supplementary Information, and the relevant fitting parameters are summarized in Supplementary Table 4, with ζ , η , D , and 2 R g representing the average correlation length of the PM6 phase, correlation length, fractal dimension of L8-BO clusters, and size of clustered L8-BO domains, respectively. After sonication, the 2 R g value of the blend films increased, indicating the expansion of acceptor-enriched domains, which facilitated the formation of an optimized morphology and suitable phase separation. To further elucidate the potential mechanisms underlying the effect of sonication on active layer growth, we conducted in-situ UV–visible absorption spectroscopy to monitor the film-formation process of the PM6:L8-BO blends. The characteristic peaks of PM6 and L8-BO were observed within the ranges of 400–660 nm and 660–950 nm, respectively (Supplementary Fig. 12), and peak position changes were analyzed (Figs. 2d and e). Based on peak evolution, the film-formation process was divided into three stages: (i) solvent evaporation and crystal nucleation, (ii) crystallization, and (iii) film formation 46 . Ordered aggregation in the sonicated L8-BO solution allowed the first stage to begin earlier, thus benefiting the prolonged molecular stacking in the second stage. The red shift of the L8-BO molecular peaks after the film- formation stage further corroborates the formation of J-aggregates. Additionally, during the film-formation process of L8-BO, the peak position of PM6 exhibited a notable blue shift, which was attributed to the immediate deposition of the L8-BO layer. The swelling effect resulted in the dissolution of a part of the donor layer during acceptor deposition, thus inducing additional mutual diffusion between them. Importantly, the ordered J-aggregation of the L8-BO molecules markedly reduced the secondary film-formation time of the PM6 molecules, thus minimizing the diffusion from the acceptor layer into the donor layer and thereby hindering unnecessary mixing. The effects of sonication on the surface morphology of thin films were probed by atomic force microscopy 47 . Both the neat L8-BO and blend films exhibited increased root mean square (RMS) roughness values after sonication (Supplementary Fig. 13). The more refined fibril network structure in the blend films facilitated the formation of high-speed pathways for electron and hole transport. To distinguish donor and acceptor domains, the stretching of characteristic bonds was analyzed using photoinduced force microscopy (PiFM) (Supplementary Figs. 14–15, Figs. 2f and g). The line profiles extracted from the corresponding localized near-infrared images are presented in Figs. 2h and i. After sonication, the acceptor domains in the blend films exhibited fibrillar structures with an approximate diameter of 23 nm. The increase in the amount of small fibril structures indicated the formation of a fibril network between the donor and acceptor phases. The overlaid PiFM images of PM6 and L8-BO were used to further explore the sizes of the different phases (Supplementary Figs. 16) 48 , revealing that sonication prevented excessive donor–acceptor mixing. This is beneficial for reducing geminate recombination and increasing the fill factor (FF), which will be discussed below. Film-depth-dependent light absorption spectroscopy was conducted to analyze the effects of sonication on the vertical phase separation of the thin films (Supplementary Fig. 17) 49 . Sonication increased the acceptor concentration near the cathode, resulting in an optimized vertical gradient distribution within the blend films (Fig. 2j). This optimization was conducive to the rapid transfer of excitons to the acceptor phase after dissociation. The effects of sonication on the morphological evolution of the L8-BO molecular aggregates were rationalized as follows (Fig. 2k). Without sonication, the L8-BO molecules spontaneously formed H-type and amorphous aggregates in the solution. This disordered aggregation contributed to an irregular distribution within the solid film and adversely affected charge transfer (CT). Conversely, upon sonication, the L8-BO molecules formed ordered interdigitated J-type aggregates in the solution. During subsequent crystal growth phases, these J-type aggregates promoted ordered molecular stacking and considerably enhanced π-π stacking interactions among molecules. Transient absorption (TA) analysis of exciton delocalization Femtosecond TA spectroscopy measurements were performed to probe the impact of the sonication-optimized microstructure on exciton and charge dynamics. The L8-BO acceptor was selectively excited at 800 nm. The 2D color maps of the TA spectra of the blend films produced from non- and sonicated solutions are presented in Fig. 3a and Supplementary Fig. 18a. The ground-state bleaching (GSB) signal of the PM6 donor at 580 nm was extracted and analyzed using exponential fitting to track hole transfer kinetics (Supplementary Fig. 19a and Supplementary Table 5) 50 . The decrease in sonication-induced lifetime from 0.34 ps to 0.24 ps indicates a concomitant increase in the rate of hole transfer 51 . The signal of the localized exciton (LE) state of the acceptor was extracted at 880 nm (Supplementary Fig. 19b). Sonication considerably accelerated excited state decay (i.e., the lifetime decreased from 3.89 ps to 2.28 ps) and thus increased the efficiency of exciton dissociation (Supplementary Table 6) 52 . In addition, the extended lifetime of the polaron signals observed at 960 nm revealed that sonication suppressed bimolecular recombination in the film (Supplementary Fig. 19c). Supplementary Fig. 19d shows the kinetic curves of the intra-moiety delocalized excitons (i-DE) extracted from TA spectra at 1500 nm. This signal was related to the exciton delocalization and will be discussed in detail below. Global fitting was used to explore the effects of sonication on the contributions of different exciton dissociation pathways (Fig. 3b). The CS process is governed by competition between the LE → i-DE → CS and LE → CT → CS pathways. The former pathway involves a rapid (within ~ 0.2 ps) transformation of LE states into i-DE states in the acceptor domain owing to strong intermolecular interactions among conjugated units. The i-DE states subsequently transform to CS states 53 . The latter pathway resembles traditional models, with LE states diffusing to the donor/acceptor (D/A) interface, transforming to CT states, and further dissociating into CS states. TA spectra recorded using different time delays were analyzed to extract the dynamic evolution signals corresponding to the LE, i-DE, CT, and CS states (Figs. 3c–f and Supplementary Figs. 18b–e), with the associated contributions shown in Fig. 3g. For both the control and the sonicated samples, the transient spectral features evolved on similar time scales. Under 1 ps, GSB features matching the steady-state absorption spectrum of the blend film were prominent in the visible region between 550 nm and 780 nm. A positive excited-state absorption (ESA) feature between 450 nm and 550 nm corresponding to the extracted spectrum of the i-DE state suggested the ultrafast formation of this state within the coherence time (~ 200 fs). This ESA returns to a negative GSB after tens of ps, indicating a progression of the system from the i-DE states to new intermediates (CT/CS). On the timescale of tens of hundreds of ps, whilst the GSB features were still present, the transient spectra adopt broad ESA features spanning from 650 nm to 1500 nm. The broadness of the ESA was consistent with the assignment of a CT type intermediate. The spectrum of the CS state was similar to that of the CT state but with a more flattened ESA feature extending into the NIR region with an absence of a positive peak at ~ 660 nm. On a ~ 200 fs timescale, the sharp ESA of the LE state in the NIR region around 850 nm developed a more prominent shoulder around 960 nm, indicating the ultrafast formation of polarons upon formation of the CT state. The intensity of the NIR signal from 1000 nm to 1500 nm decayed much more slowly compared to the signal at 850–900 nm in the first 10 ps, and through global analysis, this signal was mainly attributed to the LE and the CT states. After sonication, the contribution of the pathway dominated by i-DE states substantially increased (0.54 → 0.6), which indicated that sonication facilitated exciton delocalization. It can effectively reduce the dependence of exciton dissociation on the D/A interface. Figure 3h shows the fitted lifetime for each stage of the two pathways. The lifetime for all three stages after sonication ( τ 1 = 0.23 ps, τ 2 = 2.99 ps, τ 3 = 156 ps) were notably shorter than those observed in the control system ( τ 1 = 0.36 ps, τ 2 = 6.11 ps, τ 3 = 248 ps). Accordingly, the rates of evolution towards successive states in the sonicated film ( k 1 = 4.35 ps − 1 , k 2 = 0.33 ps − 1 , k 3 = 0.0064 ps − 1 ) were faster than that in the control film ( k 1 = 2.78 ps − 1 , k 2 = 0.16 ps − 1 , k 3 = 0.0040 ps − 1 ). These results suggested that sonication promoted efficient CS in both pathways. In addition, the lifetime of the CS state markedly increased after sonication (from 8.6 ns to 10.6 ns), which indicated suppressed charge recombination. Exciton diffusion length ( L D ) plays a crucial role in the CS pathway dominated by the CT state. To investigate the influence of sonication on L D , we recorded the pump-fluence-dependent TA spectra of the L8-BO films (Supplementary Figs. 19e and f) 49 , revealing that the exciton decay rate increased with increasing excitation intensity. Two primary quenching pathways were considered, characterized by the intrinsic exciton decay rate coefficients ( k ) at any given excitation flux and bimolecular decay rate coefficients ( γ ) for exciton–exciton annihilation at high pump fluences 52 . The following diffusion equation was used to globally fit the TA decay curves of the L8-BO acceptor: $$\:\frac{dn\left(t\right)}{dt}=-kn\left(t\right)-\frac{1}{2}\gamma\:{n}^{2}\left(t\right)\:\:\:\:\:\left(1\right)$$ Based on this equation, we derived a function describing the variation in singlet exciton density distribution n with time t : $$\:n\left(t\right)=\frac{n\left(0\right){e}^{-kt}}{1+\frac{\gamma\:}{2k}n\left(0\right)\left[1-{e}^{-kt}\right]}\:\:\:\:\:\left(2\right)$$ where γ = 8π RD ( D is the exciton diffusion coefficient, and R is the annihilation radius of the singlet excitons (assumed to equal 1 nm)). The fitted parameters are listed in Supplementary Table 7. The exciton diffusion length (\(\:{L}_{\text{D}}=\sqrt{D\tau\:}\)) increased from 29.66 nm to 36.56 nm after sonication (Fig. 3i), indicating efficient exciton transport to the D/A interface and supporting the subsequent conversion to CT states. This agrees with the efficient CT suggested by the global fitting results, which is an essential prerequisite for achieving high-efficiency OSCs 54 . Molecular vibration and nonradiative recombination Based on the aforementioned insights into exciton and charge dynamics, transient infrared (TRIR) spectroscopy was used to further examine the effects of sonication on molecular vibration (Figs. 4a, Supplementary Figs. 20a-c). We extracted the polaron dynamic curve within the 1700–1800 cm − 1 range, where the contributions from molecular vibrations were minimal (Fig. 4b). The blend films prepared using sonicated solutions demonstrated accelerated charge generation (i.e., the polaron signal rise time decreased from 9.17 ps to 6.64 ps) and slightly slowed charge recombination, in line with the TAS results. To analyze the transient vibrational spectra and dynamic changes precisely, we used a cubic polynomial function to fit the polaron background and deconvolute its contributions (Supplementary Fig. 21), as illustrated in Fig. 4c and Supplementary Fig. 20d 55 . According to the FTIR spectra (Supplementary Fig. 14), the double-peak GSB absorption feature between 1500 cm -1 and 1550 cm -1 corresponds to L8-BO vibration (C = C bond in ring). These features were sharp at early times and broaden after 10s of ps. The derivative-type feature with a positive peak at 1485 cm -1 (~ 10 cm -1 downshift from the main absorption peak) can most likely be attributed to hot ground states (S 0 ) or hot LE (with minimal structural distortion from the S 0 ). This derivative signal decayed more rapidly in the sonicated sample (8.3 ps) compared to the control blend (10.7 ps, Fig. 4d) and was likely due to the faster heat dissipation in more structurally ordered molecular solids which allowed efficient propagation of phonons. Another noticeable difference was the vibrational peak widths in the TRIR spectra. Figure 4e compares the GSB feature around 1533 cm -1 at 850 ps for the control and the sonicated blend. The peak widths of the control were broader than the sonicated blend, hinting to the existence of a wider variety of relative molecular orientations, thus, a greater degree of disorder. Therefore, it was evident that the structural tuning via sonication successfully reduced the disorder in the film, allowing for efficient propagation of both thermal energy and charge. This result is further discussed with TRPL experimental results in a later section. The removal of the polaron background allowed us to isolate the vibrational mode of PM6 at 1649 cm -1 (see FTIR spectra in Supplementary Fig. 14a). The recovery of this GSB feature was faster in the sonicated blend (5.84 ps), whereas for the control blend (7.13 ps), a small ESA signal at 1670 cm -1 slowly emerged after 10s of ps. This subtle difference suggested that when CT/CS states formed, in the more structurally ordered sonicated film, the PM6 did not undergo much geometry change. However, in the control film, the PM6 was structurally disturbed in the CT/CS states, inducing a frequency shift. Another subtle difference was at 1750 cm -1 . In the control film, an ESA feature was observed after the pulse duration ~ 300 fs and slowly recovered during the scan window. In the sonicated film, this ESA decayed into a clear bleach signal after several hundreds of ps. The FTIR of the sonicated L8-BO neat film compared to the untreated L8-BO hinted that this vibrational mode was likely associated with the presence of J-aggregated L8-BO (π-π stacked geometry). Given that the enhancement of this signal appeared at early times in both films, a possible explanation can be the formation of i-DE states within ~ 200 fs enhanced neighbouring L8-BO coupling through π−π stacking. Temperature-dependent PL and TRPL spectra were recorded to further evaluate the effects of sonication on exciton–vibration coupling and thermally activated processes. With increasing temperature, the PL spectra shifted to shorter wavelengths and broadened (Supplementary Fig. 22). The interactions between excitons and molecular vibrations can be inferred from the peak positions 56 , which are quantitatively described by: $$\:E\left(T\right)=E\left(0\right)-\frac{2\alpha\:}{\text{exp}\left(\theta\:/T\right)-1}\:\:\:\:\:\left(3\right)$$ where E (0) is the bandgap energy at absolute zero (0 K), α is the intensity of the exciton–vibration interaction, and θ is the average phonon temperature. Sonication reduced α from 151.2 meV to 113.3 meV (Fig. 4g and Supplementary Table 8) and was therefore concluded to inhibit exciton–vibration coupling and promote exciton diffusion and polaron transport. By fitting a series of temperature-dependent TRPL spectra (Supplementary Fig. 23), we obtained the temperature dependence of the nonradiative recombination rate, K nr (Fig. 4h). In high-temperature regions, K nr was more sensitive to temperature changes than in low-temperature regions because of the increase in molecular vibrations induced by thermal energy 57 . Moreover, a thermally activated process occurred between the S 0 and excited states (S 1 ) 58 (Fig. 4i). Molecular vibrations led to the formation of an overlapping surface between S 0 and S 1 , which introduced an additional pathway for internal conversion processes 59 . As discussed previously, sonication-induced structural tuning with ordered J-aggregate geometry allows for rapid thermal energy re-distribution and dissipation, which deactivates the thermally-activated non-radiative recombination of excitons. Photovoltaic characteristics To substantiate the benefits of the ordered J-aggregation induced by sonication, we fabricated LBL-type OSCs with an ITO/PEDOT:PSS/active layer/PDINN/Ag structure. The fabrication and optimization details are described in the Supplementary Information, and the J – V curves and photovoltaic parameters of the optimized devices are presented in Fig. 5a and Table 1. The control device achieved a PCE of 17.50%, an open-circuit voltage ( V OC ) of 0.878 V, a short-circuit current ( J SC ) of 25.36 mA cm − 2 , and a FF of 78.56%, in agreement with previously published results 60,61 . By controlling the sonication condition (Supplementary Fig. 24 and Supplementary Tables 9 and 10), we achieved a notably increased PCE of 19.43%, a V OC of 0.881 V, a J SC of 27.05 mA cm − 2 , and a FF of 81.53%. Compared with that of the control, the external quantum efficiency (EQE) spectrum of the sonication-processed device was shifted to longer wavelengths and exhibited a higher full-spectrum response (Fig. 5b), as evidenced by more efficient exciton dissociation and charge extraction (Supplementary Fig. 25 and Supplementary Table 11). Subsequently, we fabricated ternary PM6:D18:L8-BO devices with different active layer thicknesses. The device with a 100-nm-thick active layer featured an increased PCE of 20.41% (certified as 19.84% by the National Institute of Metrology, China) (Fig. 5c and Supplementary Fig. 26), a V OC of 0.899 V, a J SC of 27.43 mA cm − 2 , and a FF of 81.19%; this PCE is one of the highest values reported for OSCs prepared using nonhalogenated solvents (Fig. 5d, Supplementary Tables 12 and 13). The 300-nm-thick device demonstrated a satisfactory PCE of 18.29%, with the FF (76.19%) ranking among the top values reported for an active layer thickness of 300 nm. The corresponding EQE spectra are presented in Supplementary Fig. 27. All J SC values extracted from the J – V curves differed from the integrated current density of the EQE by < 5%, thereby confirming the reliability of the photovoltaic performance. The space charge limited current method was used to evaluate the charge transport which is related to the FF between the devices prepared under control and sonication conditions (Supplementary Fig. 28). The calculated hole and electron mobilities are presented in Fig. 5e and Supplementary Table 14. Sonication increased hole mobility ( µ h ) from 6.27 × 10 − 4 cm 2 V − 1 s − 1 to 8.59 × 10 − 4 cm 2 V − 1 s − 1 and electron mobility ( µ e ) from 5.43 × 10 − 4 cm 2 V − 1 s − 1 to 8.19 × 10 − 4 cm 2 V − 1 s − 1 . The higher mobility and lower µ h / µ e ratio observed for the devices prepared under sonication conditions indicated more efficient and balanced charge transport 43 . The density of trap states ( N t ) was calculated using frequency-dependent capacitance spectra recorded under dark conditions (Fig. 5f). The decrease in N t from 2.02 × 10 16 cm − 3 eV − 1 to 0.98×10 16 cm − 3 eV − 1 indicated that sonication effectively prevented trap state generation 62 . The relationship between J SC and light intensity ( P light ) was fitted as J SC = ( P light ) α (Supplementary Fig. 29a). Sonication increased α from 0.988 to 0.994 and thus effectively inhibited bimolecular recombination processes 63 . The lower slope of the V OC –ln( P light ) plot observed for the sonicated film (Fig. 5g) signified the suppression of trap-assisted recombination, thereby enhancing charge transport in the OSCs 64 . The increase in carrier lifetime from 5.83 µs to 7.44 µs revealed by transient photovoltage (TPV) tests supporting this observation (Supplementary Figs. 29b–c). Transient photocurrent (TPC) tests revealed accelerated charge extraction (i.e., fitting lifetime decreased from 0.35 µs to 0.26 µs) in the devices prepared under sonication conditions (Supplementary Table 15). Fourier transform photocurrent spectroscopy (FTPS-EQE) revealed a decrease in the Urbach energy from 24.16 meV to 23.57 meV after sonication (Supplementary Fig. 30), which suggested a concomitant reduction in energy disorder 65 . These results confirm that sonication can be used to improve OSC performance. Given that sonication can improve OSC efficiency, one can reasonably anticipate that it may also increase device stability. To minimize the adverse impact of organic interfacial layers on stability, we fabricated inverted devices with an ITO/ZnO/active layer/MoO 3 /Ag configuration. Thermal stability was assessed upon continuous annealing in dry nitrogen atmosphere at 80°C (Fig. 5h). Compared with the control devices, the sonication-processed devices (10 independent cells) exhibited a markedly reduced burn-in loss and an improved thermal stability. PCE decay was subjected to linear fitting, and the results indicated that the T 80 lifetime of the optimized devices markedly exceeded 10000 h. We also tested the photothermal stability of OSCs under continuous illumination at 100 mW cm − 2 (50–60°C), revealing that sonication also increased resistance to photothermal stress. The results demonstrate the advantages of the sonication-optimized ordered molecular stacking for improving device stability (Fig. 5i). Application scope of the sonication strategy To explore the application scope of the ultrasound-field-based regulation of ordered J-aggregation in NFAs, we recorded the in-situ PL spectra of other PM6:NFA systems, including BTP-eC9, PY-IT, and PY-DT, during the sonication process (Figs. 6a–c). In addition to the continuous red shift of the PL spectra, we observed the continuous intensification of the long-wavelength shoulder peaks of the PY-IT and PY-DT polymer acceptors, which suggested a concomitant increase in the extent of J-aggregation. We also recorded the temperature-dependent PL spectra of these systems (Supplementary Fig. 31) and extracted the corresponding I 0-1 / I 0-0 peak intensity ratios (Figs. 6d–f and Supplementary Table 16). The results showed enhanced temperature dependence, indicating intermolecular interaction and aggregation enhancement. Subsequently, solar cells based on these three systems were fabricated using a conventional device structure (Supplementary Fig. 32). The corresponding J – V curves are shown in Figs. 6g–i, and the device parameters are summarized in Supplementary Table 17. The sonication-processed PM6:BTP-eC9, PM6:PY-IT, and PM6:PY-DT devices achieved improved PCEs of 19.31%, 18.32%, and 18.71% compared to control systems, respectively. These results demonstrate the broad application scope of the sonication strategy. Discussion The NFA aggregation state in a nonhalogenated solvent was modulated using sonication. The combined results of simulations, computations, and experiments demonstrate that sonication induced the formation of ordered J-aggregates within the NFA solutions and films through microstreaming and shear stress. Morphological characterizations revealed that sonication improved phase separation during film formation, enhancing π-π stacking interactions and optimizing the molecular arrangement. The global fitting of TA data indicated that sonication increased the proportion of CS pathways dominated by i-DE states, thereby enhancing exciton dissociation. TRIR data shows the structural tuning effect on heat and carrier transport in the system, and together with temperature-dependent TRPL data further validated the inhibitory effect of sonication on nonradiative recombination. PM6:L8-BO devices exhibited a high PCE of 19.43% and superior thermal stability ( T 80 lifetime > 10000 h). The addition of D18 to the PM6:L8-BO system resulted in a remarkable PCE of 20.41% (certified PCE of 19.84%), which is among the highest efficiencies reported for OSCs prepared using nonhalogenated solvents. The developed sonication strategy was demonstrated to have a broad application scope, as exemplified by its applicability to small-molecule and polymer acceptor systems. Thus, our work provides a practical approach to the fabrication of efficient, stable, and ecofriendly OSCs, facilitating their large-scale industrial production. Methods Measurements of PL and TRPL spectra The PL spectra and TRPL experiments were performed using a confocal optical microscope, specifically the Nanofinder FLEX2 from Tokyo Instruments, Inc. This setup was equipped with a time-correlated single-photon counting (TCSPC) module, the Becker & Hickl SPC-150. For all PL spectrum measurements, a charge-coupled device (CCD) sensor, model DU420A-OE from Andor, was utilized. The excitation wavelength was consistently set at 400 nm, and the excitation power for both PL spectra and TRPL experiments was maintained at 1 µW. The laser spot had an approximate diameter of 8 µm, resulting in an area calculated as S = 5.02 × 10⁻⁷ cm². Measurements of AFM and PiFM Images AFM measurements were performed on a Bioscope Resolve AFM (Bruker) in a tapping mode under ambient conditions. Photo-induced Force Microscopy (PiFM) measurements were performed by employing Anfatec Instruments AG (Oelsnitz, Germany). Measurement of TA spectra Femtosecond TA spectra were recorded using an optical setup that incorporated a Ti:sapphire femtosecond laser (from Coherent) in conjunction with an optical parametric amplifier (OPA) system. The 800 nm pulse emitted by the laser was divided into two beams via a beam splitter. One of these beams was directed into the optical parametric amplifier (TOPAS, Coherent) to generate pump pulses at 600 nm. The second beam was focused onto both sapphire and YAG plates to produce a white light supercontinuum, which served as the probe beams with a spectral range of 750–1600 nm. The initial pulses were also split into two paths; one path led to the OPA for generating 400 and 750 nm pump pulses utilized in this study, while the other path produced a broad spectrum of probe light spanning from 520–800 nm (visible) and from 850–1300 nm (NIR). Measurement of TRIR spectroscopy TRIR spectra were acquired using pump and probe beams generated through nonlinear optical processes utilizing the 1030 nm fundamental output from a Ytterbium-based laser oscillator (Pharos, Light Conversion). An optical parametric amplifier (OPA, Orpheus, Light Conversion), in conjunction with difference frequency generation (DFG, Lyra, Light Conversion), produced pump pulses with highly tunable wavelengths across the UV-NIR spectrum (210–2660 nm, ~ 150 fs temporal FWHM, ~ 5 nm bandwidth), operating at a base repetition rate of 100 kHz. The laser output was pulse picked to 10kHZ and a mechanical chopper (Thorlabs, MC1F60) was employed to reduce the pump repetition rate to 5 kHz while maintaining a probe rate of 10 kHz. The pump beam traversed a mechanical delay stage (Newport ESP300) to achieve optical delay times ranging from − 50 to 900 ps and was subsequently focused to attain an approximate spatial diameter of FWHM ~ 420 µm at the sample location. Linear polarization devices were utilized to control the relative polarization between the pump and probe beams at the magic angle. The mid-infrared probe spanning wavelengths from 4000 to 13000 nm with a temporal FWHM of less than 180 fs, was generated by an OPA and DFG system (ORPHEUS TWINS, Light Conversion). Given that atmospheric moisture and particulate matter significantly absorb mid-infrared radiation, all components of the probe pathway were enclosed under nitrogen within sealed boxes and conduits. TRIR spectra were captured using a polychromator-based IR spectrometer cooled with liquid nitrogen during experiments (Horiba iHR320 integrated with an FPAS 128-element MCT detector; Infra-Red Systems; spectral acquisition rate: 10 kHz; bandwidth: ∼300 cm⁻¹). The sample was affixed to an X-Z translation stage controlled by two stepper motors for raster scanning purposes. Declarations Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Acknowledgements This work was supported by the National Natural Science Foundation of China (52320105003, 52203333, 32402291). X.T.H acknowledges support from the Taishan Scholars Program (tstp20230610). K.N.Z acknowledges support from the Shandong Provincial Natural Science Foundation (ZR2024QA063) and the Young Scholars Program of Shandong University, China. The authors would like to thank the Analytical Centre for Structural Constituent and Physical Property of Core Facilities Sharing Platform, Shandong University for use of the femtosecond transient absorption spectroscopy system (TAS, 2104573S), and the Instrument Improvement Funds of Shandong University Public Technology Platform (ts20230101). The authors are grateful to the Shanghai Synchrotron Radiation Facility (beamline BL16B1) for support with GIXS measurements. References Luo, S. et al. Auxiliary sequential deposition enables 19%-efficiency organic solar cells processed from halogen-free solvents. Nat. Commun. 14 , 6964 (2023). Ma, L. J. et al. Design of low-cost non-fused ultranarrow-band-gap acceptors for versatile photovoltaic applications. Joule 8 , 2238–2249 (2024). Chen, Z. H. et al. Local dipole modulation toward high fill factor in organic solar cells. Adv. Mater. 34 , 2408858 (2024). Wang, Y. et al. Origins of the open-circuit voltage in ternary organic solar cells and design rules for minimized voltage losses. Nat. Energy 8 , 978–988 (2023). Ding, P. et al. U-shaped dimeric acceptors for balancing efficiency and stability in organic solar cells. Adv. Mater. 42 , 2414080 (2024). Wang, S. et al. Achieving 20% efficiency in organic solar cells through conformationally locked solid additives. Adv. Energy Mater. 25 , 2405205 (2024). Zou, W. et al. Extending exciton diffusion length via an organic-metal platinum complex additive for high-performance thick-film organic solar cells. Adv. Mater. 37 , 2413125 (2024). Wang, J. et al. Isomerism effect of 3D dimeric acceptors for non-halogenated solvent-processed organic solar cells with 20 % efficiency. Angew. Chem. Int. Edit. 64 , e202423562 (2024). Chen, C. et al. Molecular interaction induced dual fibrils towards organic solar cells with certified efficiency over 20%. Nat. Commun. 15 , 6865 (2024). Kong, X. et al. Suppressed non-radiative loss and efficient hole transfer at a small highest occupied molecular orbital offset endows binary organic solar cells with 19.73% efficiency and a small efficiency-cost gap. Energy Environ. Sci. 18 , 386–396 (2025). Lang, Y. et al. Balanced miscibility and crystallinity by 2D acceptors enabled halogen-free solvent-processed organic solar cells to achieve 19.28% efficiency. Adv. Mater. 37 , 2413270 (2025). Zhang, Y. et al. Graded bulk-heterojunction enables 17% binary organic solar cells via nonhalogenated open air coating. Nat. Commun. 12 , 4815 (2021). Chen, H. et al. Heterogeneous nucleating agent for high‐boiling‐point nonhalogenated solvent‐processed organic solar cells and modules. Adv. Mater. 36 , 2402350 (2024). Fan, B. et al. Enabling high efficiency of hydrocarbon-solvent processed organic solar cells through balanced charge generation and non-radiative loss. Adv. Energy Mater. 11 , 2101768 (2021). Fan, B. et al. Optimisation of processing solvent and molecular weight for the production of green-solvent-processed all-polymer solar cells with a power conversion efficiency over 9%. Energy Environ. Sci. 10 , 1243–1251 (2017). Zhao, H. et al. Hot hydrocarbon-solvent slot-die coating enables high-efficiency organic solar cells with temperature-dependent aggregation behavior. Adv. Mater. 32 , 2002302 (2020). Wu, X. et al. Introducing a phenyl end group in the inner side chains of A-DA’D-A acceptors enables high-efficiency organic solar cells processed with nonhalogenated solvent. Adv. Mater. 35 , 2302946 (2023). Xu, X., Yu, L., Yan, H., Li, R. & Peng, Q. Highly efficient non-fullerene organic solar cells enabled by a delayed processing method using a non-halogenated solvent. Energy Environ. Sci. 13 , 4381–4388 (2020). Fan, B. et al. Importance of structural hinderance in performance–stability equilibrium of organic photovoltaics. Nat. Commun. 13 , 5946 (2022). Zhao, W. et al. Vacuum-assisted annealing method for high efficiency printable large-area polymer solar cell modules. J. Mater. Chem. C 7 , 3206–3211 (2019). Zhao, W. et al. Environmentally friendly solvent-processed organic solar cells that are highly efficient and adaptable for the blade-coating method. Adv. Mater. 30 , 1704837 (2018). Li, H. et al. Advances in the device design and printing technology for eco-friendly organic photovoltaics. Energy Environ. Sci. 16 , 76–88 (2023). Chen, H. et al. 17.1 %-efficient eco-compatible organic solar cells from a dissymmetric 3D network acceptor. Angew. Chem. Int. Edit. 60 , 3238–3246 (2021). Zhuo, H. et al. Giant molecule acceptor enables highly efficient organic solar cells processed using non-halogenated solvent. Angew. Chem. Int. Edit. 62 , e202303551 (2023). Pröller, S. et al. Following the morphology formation in situ in printed active layers for organic solar cells. Adv. Energy Mater. 6 , 1501580 (2016). Ge, Y. et al. Aggregation engineering of toluene-processed acceptor layer enables over 19% efficiency of air-blade-coated organic solar cells. Adv. Mater . 22 , 2502579. (2025) Wang, Y. et al. Achieving 20% toluene-processed binary organic solar cells via secondary regulation of donor aggregation in sequential processing. Nano-Micro Lett . 17 , 206 (2025). Han, C. et al. A halogen-free and universally volatile solid additive enables binary organic solar cells to exceed 19% efficiency. Adv. Funct. Mater. 35 , 2416381 (2024). Zhang, R. et al. Equally high efficiencies of organic solar cells processed from different solvents reveal key factors for morphology control. Nat. Energy 10 , 124–134 (2025). Guo, C. et al. Light-induced quinone conformation of polymer donors toward 19.9% efficiency organic solar cells. Energy Environ. Sci. 17 , 2492–2499 (2024). Cui, F. Z. et al. Using an external electric field to tune active layer morphology enabling high-efficiency organic solar cells via ambient blade coating. Sci. Adv. 10 , eado5460 (2024). Wang, W. et al. H-bonded organic frameworks as ultrasound-programmable delivery platform. Nature 638 , 401–410 (2025). Guo, G. et al. Enhanced porosity and permeability of three-dimensional alginate scaffolds via acoustic microstreaming induced by low-intensity pulsed ultrasound. Ultrason. Sonochem. 37 , 279–285 (2017). Sun, M. et al. Overcoming disordered preaggregation in liquid state for highly efficient organic solar cells printed from nonhalogenated solvents. Adv. Energy Mater. 13 , 2203465 (2023). Ashokkumar, et al. The characterization of acoustic cavitation bubbles – an overview. Ultrason. Sonochem. 18 , 864–872 (2011). Liang, H. et al. A rare case of brominated small molecule acceptors for high-efficiency organic solar cells. Nat. Commun. 14 , 4707 (2023). Su, W. P. et al. Solitons in polyacetylene. Phys. Rev. Lett. 42 , 1698–1701 (1979). Zhao, Q. et al. Balancing the H- and J-aggregation in DTS(PTTh2)2/PC70BM to yield a high photovoltaic efficiency. J. Mater. Chem. C 3 , 8183–8192 (2015). Mondal, J. et al. Bubble oscillations at low frequency ultrasound for biological applications. Ultrason. Sonochem . 104 , 106816 (2024). Blanc, F. et al. Rheology of dense suspensions under shear rotation. Phys. Rev. Lett . 130 , 118202 (2023). Clark, J., Silva, C., Friend, R. H. & Spano, F. C. Role of intermolecular coupling in the photophysics of disordered organic semiconductors: aggregate emission in regioregular polythiophene. Phys. Rev. Lett. 98 , 206406 (2007). Zhao, Q. et al. H- and J-aggregation inspiring efficient solar conversion. J. Mater. Chem. A 9 , 1119–1126 (2021). Zhang, G. et al. Delocalization of exciton and electron wavefunction in non-fullerene acceptor molecules enables efficient organic solar cells. Nat. Commun. 11 , 3943 (2020). Qiao, J. et al. Enhanced exciton delocalization in organic near-infrared photodetectors via solid additive-mediated J-aggregation. Adv. Mater. 37 , 2418844 (2024). Wang, J. et al. Two well-compatible acceptors with efficient energy transfer enable ternary organic photovoltaics exhibiting a 13.36% efficiency. Small 15 , 1902602 (2019). Chen, H. et al. A guest-assisted molecular-organization approach for >17% efficiency organic solar cells using environmentally friendly solvents. Nat. Energy 6 , 1045–1053 (2021). Zhou, M. et al. 19.10% efficiency and 80.5% fill factor layer-by-layer organic solar cells realized by 4-bis(2-thienyl)pyrrole-2,5-dione based polymer additives for inducing vertical segregation morphology. Adv. Mater. 35 , 2208279 (2023). Zhu, L. et al. Single-junction organic solar cells with over 19% efficiency enabled by a refined double-fibril network morphology. Nat. Mater. 21 , 656–663 (2022). Cai, Y. et al. Vertically optimized phase separation with improved exciton diffusion enables efficient organic solar cells with thick active layers. Nat. Commun. 13 , 2369 (2022). Cai, Y. et al. A well-mixed phase formed by two compatible non-fullerene acceptors enables ternary organic solar cells with efficiency over 18.6%. Adv. Mater. 33 , 2101733 (2021). Cui, F. et al. Vertical-phase-locking effect in efficient and stable all-polymer-hosted solar cells. ACS Energy Lett. 7 , 3709–3717 (2022). Chen, Z. et al. Trap state induced recombination effects on indoor organic photovoltaic cells. ACS Energy Lett. 6 , 3203–3211 (2021). Wang, R. et al. Charge separation from an intra-moiety intermediate state in the high-performance PM6:Y6 organic photovoltaic blend. J. Am. Chem. Soc. 142 , 12751–12759 (2020). Chen, X. et al. A unified description of non-radiative voltage losses in organic solar cells. Nat. Energy 6 , 799–806 (2021). Zhang, K. et al. Donoracceptor interfacial dipole polarization for efficient and stable thick-film organic photovoltaics. Nano Energy 134 , 110546 (2025). Guha, S. et al. Temperature-dependent photoluminescence of organic semiconductors with varying backbone conformation. Phys. Rev. B 67 , 125204 (2003). Guo, Q. et al. Enhancing the performance of organic solar cells by prolonging the lifetime of photogenerated excitons. Adv. Mater. 32 , 2003164 (2020). Li, Q. et al. A conical intersection model to explain aggregation induced emission in diphenyl dibenzofulvene. Chem. Commun. 49 , 5966–5968 (2013). Gao, Y.-J. et al. Excited-state decay paths in tetraphenylethene derivatives. J. Phys. Chem. A 121 , 2572–2579 (2017). Ma, R. et al. Revealing the underlying solvent effect on film morphology in high-efficiency organic solar cells through combined ex situ and in situ observations. Energy Environ. Sci. 16 , 2316–2326 (2023). Bai, H. et al. Green-solvent processed blade-coating organic solar cells with an efficiency approaching 19% enabled by alkyl-tailored acceptors. Nano-Micro Lett. 15 , 241 (2023). Hughes, M. P. et al. Determining the dielectric constants of organic photovoltaic materials using impedance spectroscopy. Adv. Funct. Mater. 28 , 1801542 (2018). Huang, J. et al. Polymer bulk heterojunction solar cells employing Förster resonance energy transfer. Nat. Photonics 7 , 479–485 (2013). Honda, S. et al . Selective dye loading at the heterojunction in polymer/fullerene solar cells. Adv. Energy Mater. 1 , 588–598 (2011). Wang, J. et al. Increasing donor-acceptor spacing for reduced voltage loss in organic solar cells. Nat. Commun. 12 , 6679 (2021). Table 1 Table 1 is available in the Supplementary Files section. Additional Declarations There is NO Competing Interest. Supplementary Files Supplementaryinformation20250616.docx Supplementary Information for Publication Table1.docx Cite Share Download PDF Status: Published Journal Publication published 25 Nov, 2025 Read the published version in Nature Communications → Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6915241","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":473877471,"identity":"a267cb98-a6e3-4408-81d0-0f2c4c1e61ae","order_by":0,"name":"Xiaotao Hao","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAsklEQVRIiWNgGAWjYBACAzBZwcDYAKQkiNdy4AzJWg62kaLFnP3wMemP8+pkNxxgPnibh8Euj6AWy560ZIOD2w4bbzjAlmzNw5BcTNhhB3IMHxzcdiBxwwEeM2kehgOJDQS1nH9jcODgnDqgFv5vRGq5AbKlgRlkCxuxWp4lG5w5dth45mE2Y8s5BsnEOCz5mERFTZ1s3/HmhzfeVNgR1oIAzGATiFc/CkbBKBgFowAPAADCe0DfiCaS9wAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-0197-6545","institution":"Shandong University","correspondingAuthor":true,"prefix":"","firstName":"Xiaotao","middleName":"","lastName":"Hao","suffix":""},{"id":473877472,"identity":"2eabb428-9832-4f82-9e9a-d22cd114dcde","order_by":1,"name":"Mingxu Zhou","email":"","orcid":"","institution":"Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Mingxu","middleName":"","lastName":"Zhou","suffix":""},{"id":473877473,"identity":"568431ec-305d-49e6-a666-1519bfe91827","order_by":2,"name":"Xinyue Xu","email":"","orcid":"","institution":"The University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Xinyue","middleName":"","lastName":"Xu","suffix":""},{"id":473877474,"identity":"9d4f5826-f816-49e0-923e-a070de4e5d07","order_by":3,"name":"Dongrui Wang","email":"","orcid":"","institution":"Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Dongrui","middleName":"","lastName":"Wang","suffix":""},{"id":473877475,"identity":"86dfb5ac-7fe3-49cf-a9f4-f565d07e3a5f","order_by":4,"name":"Jiali Song","email":"","orcid":"","institution":"Beihang University","correspondingAuthor":false,"prefix":"","firstName":"Jiali","middleName":"","lastName":"Song","suffix":""},{"id":473877476,"identity":"0cafed3e-6006-4dc5-9930-e07011475677","order_by":5,"name":"Jiawei Qiao","email":"","orcid":"","institution":"Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Jiawei","middleName":"","lastName":"Qiao","suffix":""},{"id":473877477,"identity":"278788fe-3969-4802-bd22-a68922fcbaeb","order_by":6,"name":"Ruyue Zhang","email":"","orcid":"","institution":"Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Ruyue","middleName":"","lastName":"Zhang","suffix":""},{"id":473877478,"identity":"0300a91d-2aa9-48fb-9c11-fab0b2b6b7e9","order_by":7,"name":"Yue Wu","email":"","orcid":"","institution":"Nanjing Agricultural University","correspondingAuthor":false,"prefix":"","firstName":"Yue","middleName":"","lastName":"Wu","suffix":""},{"id":473877479,"identity":"fdc1cbf6-0a62-4ab9-843f-44745f4e2e95","order_by":8,"name":"Jiwei Cui","email":"","orcid":"https://orcid.org/0000-0003-1018-4336","institution":"Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Jiwei","middleName":"","lastName":"Cui","suffix":""},{"id":473877480,"identity":"dc0557e2-03ef-4d53-9d6f-c965370bcbf6","order_by":9,"name":"Christopher Hall","email":"","orcid":"https://orcid.org/0000-0003-1267-1277","institution":"The University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Christopher","middleName":"","lastName":"Hall","suffix":""},{"id":473877481,"identity":"8d341432-fd99-4c85-934f-decd5923b3a4","order_by":10,"name":"Hang Yin","email":"","orcid":"https://orcid.org/0000-0003-2600-7238","institution":"Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Hang","middleName":"","lastName":"Yin","suffix":""},{"id":473877482,"identity":"0ebe7f1f-b175-48ee-b0a9-f001fd8c23e4","order_by":11,"name":"Xiaoyan Du","email":"","orcid":"https://orcid.org/0000-0002-0737-2988","institution":"Helmholtz Institute Erlangen-Nürnberg for Renewable Energy (HI ERN), Friedrich‐Alexander‐Universität Erlangen‐Nürnberg, Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyan","middleName":"","lastName":"Du","suffix":""},{"id":473877483,"identity":"16b6e477-73e7-4f4e-b974-1e3f0a19dc86","order_by":12,"name":"Wei Qin","email":"","orcid":"https://orcid.org/0000-0003-4579-0061","institution":"Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Qin","suffix":""},{"id":473877484,"identity":"7ad3c646-f0f5-45ef-a16b-8956783ecdd8","order_by":13,"name":"Kun Gao","email":"","orcid":"https://orcid.org/0000-0002-9146-0090","institution":"School of Physics, State Key Laboratory of Crystal Materials, Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Kun","middleName":"","lastName":"Gao","suffix":""},{"id":473877485,"identity":"19b717f3-6b5c-46f0-8baf-dcb295b5dd61","order_by":14,"name":"Feng Chen","email":"","orcid":"https://orcid.org/0000-0002-9277-9810","institution":"Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Feng","middleName":"","lastName":"Chen","suffix":""},{"id":473877486,"identity":"b1fe03c3-2ef4-4ee7-a745-ca0621e3d153","order_by":15,"name":"Muthupandian Ashokkumar","email":"","orcid":"https://orcid.org/0000-0002-8442-1499","institution":"School of Chemistry, The University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Muthupandian","middleName":"","lastName":"Ashokkumar","suffix":""},{"id":473877487,"identity":"131cbc21-92e4-4f50-9514-aeb5dc791487","order_by":16,"name":"Trevor Smith","email":"","orcid":"https://orcid.org/0000-0003-4453-9713","institution":"The University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Trevor","middleName":"","lastName":"Smith","suffix":""},{"id":473877488,"identity":"791fd019-d657-48d5-8113-e49ecab0479c","order_by":17,"name":"Kangning Zhang","email":"","orcid":"","institution":"School of Physics, State Key Laboratory of Crystal Materials, Shandong University","correspondingAuthor":false,"prefix":"","firstName":"Kangning","middleName":"","lastName":"Zhang","suffix":""},{"id":473877489,"identity":"314ab45a-b6ee-4486-8b1a-05a9e457c8ed","order_by":18,"name":"Yanming Sun","email":"","orcid":"https://orcid.org/0000-0001-7839-3199","institution":"Beihang University","correspondingAuthor":false,"prefix":"","firstName":"Yanming","middleName":"","lastName":"Sun","suffix":""}],"badges":[],"createdAt":"2025-06-17 14:10:15","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6915241/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6915241/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41467-025-65447-y","type":"published","date":"2025-11-25T05:00:00+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":85186499,"identity":"4fb8c058-02eb-49a8-8f29-78e26a5d76dc","added_by":"auto","created_at":"2025-06-23 08:12:51","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":924035,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMechanism underlying sonication effects probed by simulations, calculations and experiments.\u003c/strong\u003e \u003cstrong\u003ea,\u003c/strong\u003e Chemical structures of different materials and schematic of sonication-based processing. \u003cstrong\u003eb,\u003c/strong\u003e Evolution of the photoluminescence (PL) spectra of the L8-BO solution during sonication (inset shows the PL changes at 1 min after stopping sonication). \u003cstrong\u003ec,\u003c/strong\u003e Effects of sonication on the molecular dynamic behavior of L8-BO in \u003cem\u003eo\u003c/em\u003e-xy. \u003cstrong\u003ed,\u003c/strong\u003e Effects of sonication on the change in the radius of gyration (\u003cem\u003eR\u003c/em\u003e\u003csub\u003eg\u003c/sub\u003e) with time. \u003cstrong\u003ee,\u003c/strong\u003e Root mean square deviation–time plots obtained with and without sonication. \u003cstrong\u003ef,\u003c/strong\u003e Temperature-dependent PL spectra of L8-BO-sonicated in \u003cem\u003eo\u003c/em\u003e-xy. \u003cstrong\u003eg,\u003c/strong\u003e Peak intensity ratio extracted from (f). \u003cstrong\u003eh–i, \u003c/strong\u003eTwo-dimensional (2D) grazing-incidence wide-angle X-ray scattering (GIWAXS) patterns of the L8-BO films obtained without and with sonication.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6915241/v1/831ddcd8a1d632de9ee4c7ba.png"},{"id":85184640,"identity":"f661fc4a-95d0-41eb-900c-6425d4cdff37","added_by":"auto","created_at":"2025-06-23 07:56:51","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1585246,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMorphological analysis of blend films.\u003c/strong\u003e \u003cstrong\u003ea–b,\u003c/strong\u003e 2D GIWAXS patterns of PM6:L8-BO (\u003cem\u003eo\u003c/em\u003e-xy) blends obtained without or with sonication. \u003cstrong\u003ec,\u003c/strong\u003e Out-of-plane and in-plane extracted line-cut profiles of the corresponding films. \u003cstrong\u003ed–e,\u003c/strong\u003e UV–vis absorption peak locations of the acceptor and donor for the PM6:L8-BO films prepared without or with sonication. \u003cstrong\u003ef–i,\u003c/strong\u003e Photoinduced force microscopy images acquired using wavenumbers of 1,532 cm\u003csup\u003e−1 \u003c/sup\u003e(representing L8-BO) and 1,648 cm\u003csup\u003e−1 \u003c/sup\u003e(representing PM6) and line profiles recorded along the white arrows to obtain the fibril width for the PM6:L8-BO blend film prepared using sonication. \u003cstrong\u003ej,\u003c/strong\u003e Film-depth-dependent composition profiles extracted from the depth-dependent light absorption spectra of blend films. \u003cstrong\u003ek,\u003c/strong\u003e Schematic illustrating the regulation of film-formation process and crystallization kinetics for the PM6:L8-BO films prepared without or with sonication.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6915241/v1/6342606a9792eb04a2185aca.png"},{"id":85184641,"identity":"fa4ef08d-0146-4313-bd4e-b051ba548755","added_by":"auto","created_at":"2025-06-23 07:56:51","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":1012667,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eGlobal fitting analysis of exciton and charge dynamics.\u003c/strong\u003e \u003cstrong\u003ea,\u003c/strong\u003e Transient absorption (TA) spectral plot of the sample prepared using sonication. \u003cstrong\u003eb,\u003c/strong\u003eModel used in TA global fitting. \u003cstrong\u003ec,\u003c/strong\u003e TA spectra with fitting curves obtained at different time delays for the sonicated sample. \u003cstrong\u003ed-e,\u003c/strong\u003e TA dynamics and related fitting curves obtained for visible and infrared regions. \u003cstrong\u003ef-g,\u003c/strong\u003eLocalized exciton (LE), intra-moiety delocalized exciton (i-DE), charge transfer (CT), and charge separation (CS) signals extracted from TA spectra by global fitting, along with the corresponding lifetime. \u003cstrong\u003eh,\u003c/strong\u003e Lifetime of all signals in PM6:L8-BO films prepared without or with sonication. \u003cstrong\u003ei,\u003c/strong\u003e Pump-fluence-dependent TA kinetics probed at 880 nm using L8-BO films prepared without or with sonication.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6915241/v1/64ed045c344933caad4ca990.png"},{"id":85185093,"identity":"9e6587c2-bd92-492a-9205-7cebf6096c1a","added_by":"auto","created_at":"2025-06-23 08:04:51","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":907526,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExcited-state dynamics revealed by transient infrared (TRIR) spectroscopy.\u003c/strong\u003e \u003cstrong\u003ea,\u003c/strong\u003e 2D color plots of TRIR data for the PM6:L8-BO films prepared using sonication. \u003cstrong\u003eb,\u003c/strong\u003e TRIR spectra obtained after deconvoluting the signals of polarons for the PM6:L8-BO films prepared using sonication. \u003cstrong\u003ec,\u003c/strong\u003e TRIR profiles recorded at 1700–1800 cm\u003csup\u003e−1\u003c/sup\u003e. \u003cstrong\u003ed,\u003c/strong\u003e TRIR profiles obtained at 1485 cm\u003csup\u003e-1\u003c/sup\u003e. \u003cstrong\u003ee,\u003c/strong\u003e TRIR spectra extracted from 850 ps. \u003cstrong\u003ef,\u003c/strong\u003e TRIR profiles obtained at 1649 and 1750 cm\u003csup\u003e-1\u003c/sup\u003e. \u003cstrong\u003eg,\u003c/strong\u003e Gaussian-fitted peak position as a function of temperature in L8-BO films prepared without or with sonication. \u003cstrong\u003eh,\u003c/strong\u003e Nonradiative recombination rate as a function of temperature in the PM6:L8-BO films prepared without or with sonication. \u003cstrong\u003ei,\u003c/strong\u003e Schematic of sonication modulating the thermally activated process between the ground and excited states.\u0026nbsp;\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6915241/v1/276865d79c274c2dfd164f5d.png"},{"id":85186501,"identity":"6efe9ac1-51f7-4e57-af04-70c76b84d179","added_by":"auto","created_at":"2025-06-23 08:12:51","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":586298,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePhotovoltaic properties of organic solar cells. a–b,\u003c/strong\u003e \u003cem\u003eJ–V\u003c/em\u003e curves and external quantum efficiency spectra of the devices prepared without or with sonication. \u003cstrong\u003ec,\u003c/strong\u003e \u003cem\u003eJ–V\u003c/em\u003e curves of the devices with ternary active layers of different thickness prepared using sonication. \u003cstrong\u003ed,\u003c/strong\u003e Positions of different binary and ternary systems in the FF–PCE space relative to those of previously reported systems. \u003cstrong\u003ee,\u003c/strong\u003e Carrier mobilities of the control and sonication-processed devices obtained using space charge limited current (SCLC) measurements. \u003cstrong\u003ef,\u003c/strong\u003e Density of trap states of the blend films and corresponding Gaussian fitting results. \u003cstrong\u003eg,\u003c/strong\u003e Measurements of light-intensity-dependent \u003cem\u003eV\u003c/em\u003e\u003csub\u003eOC\u003c/sub\u003e under different conditions. \u003cstrong\u003eh–i,\u003c/strong\u003e Normalized PCEs obtained for the PM6:L8-BO devices prepared without or with sonication during thermal stability evaluation at 80 °C and photothermal stability evaluation under continuous illumination (100 mW cm\u003csup\u003e−2\u003c/sup\u003e, white light-emitting diode) at 50–60 °C.\u0026nbsp;\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6915241/v1/4e0089cb73071a5beac0c47e.png"},{"id":85185099,"identity":"d478e4b4-05b9-4730-921e-3e998f9ebbb1","added_by":"auto","created_at":"2025-06-23 08:04:51","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":934767,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePhotovoltaic characterization of other sonicated systems.\u003c/strong\u003e \u003cstrong\u003ea–c,\u003c/strong\u003e Evolution of the PL spectra of BTP-eC9, PY-IT and PY-DT solutions during sonication, with inset illustrating the changes in PL spectra at 1 min post sonication. \u003cstrong\u003ed–f,\u003c/strong\u003e Peak ratios extracted from the temperature-dependent PL spectra of BTP-eC9, PY-IT and PY-DT prepared without or with sonication. \u003cstrong\u003eg–i,\u003c/strong\u003e \u003cem\u003eJ\u003c/em\u003e–\u003cem\u003eV\u003c/em\u003ecurves of different systems prepared without or with sonication.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-6915241/v1/3d27b24a4681a2fc2212040f.png"},{"id":96798733,"identity":"d3a92402-2e9d-4313-8e1a-ba65c651b92c","added_by":"auto","created_at":"2025-11-26 08:11:27","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7129278,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6915241/v1/e0b308e1-e5d0-41aa-a147-549223cb6db7.pdf"},{"id":85187022,"identity":"1de653c8-62d9-4969-af4b-675cf5f763a5","added_by":"auto","created_at":"2025-06-23 08:20:51","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":17989791,"visible":true,"origin":"","legend":"Supplementary Information for Publication","description":"","filename":"Supplementaryinformation20250616.docx","url":"https://assets-eu.researchsquare.com/files/rs-6915241/v1/45593efcf854e11fad1fe1d8.docx"},{"id":85184639,"identity":"b393798b-388e-42ef-b8f4-15eb6a313aed","added_by":"auto","created_at":"2025-06-23 07:56:51","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":16547,"visible":true,"origin":"","legend":"","description":"","filename":"Table1.docx","url":"https://assets-eu.researchsquare.com/files/rs-6915241/v1/c9c813cf1ba1efdfdee1b398.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Sonication-induced J-aggregation in nonhalogenated solvents boosts exciton delocalization for high-efficiency organic solar cells","fulltext":[{"header":"Introduction","content":"\u003cp\u003eOwing to their intrinsic flexibility, semitransparency, and remarkable cost-effectiveness, organic solar cells (OSCs) hold promise for next-generation photovoltaic technologies\u003csup\u003e\u003cspan additionalcitationids=\"CR2 CR3 CR4\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e. Rapid advancements in material synthesis, molecular modification, and device engineering have resulted in single-junction OSCs with power conversion efficiencies (PCEs) exceeding 20%, which represents an important milestone on the road to industrial production\u003csup\u003e\u003cspan additionalcitationids=\"CR7 CR8 CR9\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e\u003c/sup\u003e. However, these PCEs are strongly affected by spin-coating processes that rely on highly toxic halogenated solvents (such as chloroform) incompatible with large-scale production\u003csup\u003e\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e. In contrast, the low solubility of numerous nonfullerene acceptors (NFAs) in ecofriendly nonhalogenated solvents leads to substantial aggregation during processing\u003csup\u003e\u003cspan additionalcitationids=\"CR15 CR16\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e, with disordered molecular orientations severely hindering exciton dissociation and charge transport. Consequently, a substantial efficiency gap exists between OSCs processed using halogenated and nonhalogenated solvents\u003csup\u003e\u003cspan additionalcitationids=\"CR19 CR20\" citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e, which highlights the importance of controlling NFA preaggregation in nonhalogenated solvents for achieving high photovoltaic performance\u003csup\u003e\u003cspan additionalcitationids=\"CR23 CR24\" citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe disordered aggregation state of NFAs in nonhalogenated solvents is typically transformed into more ordered H- or J-aggregation through deliberate induction to eliminate its adverse effects. Compared to H-aggregation (face-to-face stacking), J-aggregation is formed through intermolecular sliding packings, exhibiting distinctive photophysical properties in terms of expanding light absorption and enhancing charge transport\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Various approaches have been explored to induce the formation of J-aggregation to improve the PCEs of OSCs. For example, Zhang et al. optimized the π-π stacking and refined the phase separation by promoting donor J-aggregation with highly volatile additives\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e. Li et al. employed molecular modification to weaken the interaction between acceptor molecules and the solvent. This helps to facilitate pronounced J-aggregation, thereby enhancing charge extraction and inhibiting charge recombination\u003csup\u003e\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u003c/sup\u003e. Despite the progress enabled by these approaches, the additive- and molecular-modification-based strategies involve complex synthesis and precise chemical control\u003csup\u003e\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e\u003c/sup\u003e. These simplicity- and adaptability-related challenges hinder widespread practical applications of OSCs. External physical fields, such as heat, light, and electric stimuli, enable precise modulation of material solution state, film-formation process, and post-deposition metastable phases through straightforward processing\u003csup\u003e\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e,\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e. Ultrasound emerges as a uniquely powerful tool to selectively cleave noncovalent interaction networks via physical shear effects at the molecular scale by harnessing tunable acoustic cavitation energy\u003csup\u003e\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e. Concurrently, ultrasound-induced microstreaming can orchestrate molecular rearrangement, offering exceptional control over aggregation pathways. This efficient, ecofriendly approach that remains largely uncharted in organic photovoltaics holds significant potential to address the detrimental preaggregation of NFAs in nonhalogenated solvents. Elucidating the physical mechanism underlying the sonication-induced regulation of NFA aggregation is therefore essential for achieving high-performance OSCs processed by nonhalogenated solvents.\u003c/p\u003e \u003cp\u003eHerein, we use sonication to manipulate the preaggregation of NFAs in nonhalogenated solvents. The results of molecular dynamics simulations, quantum chemical computations, and multiple experiments demonstrate that sonication induces a transition from amorphous aggregation to ordered J-aggregation because of the effects of microstreaming and shear stress generated during the sonication process. Morphological studies reveal that sonication enhances intermolecular interactions in neat and blend films and thus promotes ordered molecular stacking. A global fitting model is developed to analyze transient absorption data to quantify the contributions of different charge separation (CS) pathways, and that of the pathways dominated by intra-moiety delocalized excitons is shown to markedly increase after sonication, which indicates a concomitant exciton dissociation enhancement. Sonication is also shown to inhibit molecular vibrations and thereby hinder nonradiative recombination. The binary and ternary OSCs prepared using sonication feature PCEs of 19.43% and 20.41% (certified 19.84%), respectively, along with high thermal stabilities (\u003cem\u003eT\u003c/em\u003e\u003csub\u003e80\u003c/sub\u003e lifetime\u0026thinsp;\u0026gt;\u0026thinsp;10000 h). The sonication strategy is applicable to other small-molecule and polymer acceptor systems, offering unique insights into the development of high-performance OSCs based on nonhalogenated solvents.\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec3\"\u003e\n \u003ch2\u003eMechanism of sonication-regulated NFA aggregation\u003c/h2\u003e\n \u003cp\u003eFigure 1a shows the chemical structures of the donor (PM6) and acceptor (L8-BO), and illustrates the sonication-assisted film fabrication processes based on the sequential deposition of these species. Initially, a PM6 layer is constructed on an indium tin oxide (ITO) substrate, and a sonicated solution of L8-BO in \u003cem\u003eo\u003c/em\u003e-xylene (\u003cem\u003eo\u003c/em\u003e-xy) is deposited onto the PM6 layer. For numerous organic materials, amorphous aggregate formation due to nanoscale precipitation is observed in low-solubility solvents (such as \u003cem\u003eo\u003c/em\u003e-xy), unlike in the case of high-solubility solvents (such as chloroform)\u003csup\u003e34\u003c/sup\u003e. This problem is solved by using ultrasound filed. Sonication causes cavitation, i.e., the formation of tiny bubbles that can rapidly grow and collapse under the periodic pressure changes caused by ultrasonic waves\u003csup\u003e35\u003c/sup\u003e (Supplementary Fig. 1). This typically induces physical and chemical effects. Supplementary Fig. 2 presents the Raman spectra of L8-BO solution. The unaltered peak positions after sonication suggest that no chemical reaction have occurred, but only physical effects exist.\u003c/p\u003e\n \u003cp\u003eAggregation state changes were probed by conducting in-situ photoluminescence (PL) spectroscopy measurements of the L8-BO solution (\u003cem\u003eo\u003c/em\u003e-xy) during sonication (Fig. 1b and supplementary Fig.\u0026nbsp;3). We evaluated the in-situ PL spectrum variations across different frequencies and pinpointed the optimal sonication settings at 40 kHz, 720 W, and 3 minutes, as detailed in the Supplementary Information. Upon sonication, the emission peak of L8-BO shifted to substantially longer wavelengths, moving from 793 nm to 805 nm within three minutes, which indicated a concomitant increase in the extent of solution-phase J-aggregation\u003csup\u003e36\u003c/sup\u003e. Moreover, J-aggregation was well maintained in the solution within one minute after sonication was stopped (Fig. 1b), which was sufficient for film preparation. Molecular dynamics simulations were used to verify the effects of sonication on L8-BO molecules dissolved in \u003cem\u003eo\u003c/em\u003e-xy (Fig. 1c). Without sonication (in the control system), the L8-BO molecules formed a dispersed state, with the radius of gyration (\u003cem\u003eR\u003c/em\u003e\u003csub\u003eg\u003c/sub\u003e) and solvent-accessible surface area (SASA) fluctuating around their mean values. Conversely, in the case of sonication, the \u003cem\u003eR\u003c/em\u003e\u003csub\u003eg\u003c/sub\u003e and SASA values notably decreased before stabilizing, suggesting that sonication increased the degree of aggregation (Fig. 1d and Supplementary Fig. 4). Root mean square deviations (RMSDs) were calculated to observe the overall conformational changes relative to the initial structure throughout the simulation period (Fig. 1e). In both systems, the RMSD initially increased before reaching equilibrium, which reflected rapid convergence toward a stable equilibrium state. The local magnification of simulation snapshots revealed that the control system contained some amorphous structures and H-aggregates, while a marked increase in the number of J-aggregates was observed after sonication, in line with the in-situ PL spectroscopy results.\u003c/p\u003e\n \u003cp\u003eQuantum chemical computations were performed to clarify the mechanisms underlying the sonication-induced J-aggregation (Supplementary Figs.\u0026nbsp;5a and b). An extended version of the Su–Schrieffer–Heeger tight-binding model was utilized, with a comprehensive description of the calculation process provided in the Supplementary Information\u003csup\u003e37\u003c/sup\u003e. J-aggregation can be categorized into A-to-A and A-to-D types. Supplementary Fig. 5c shows the band gaps of different aggregation modes. The redshift of the PL spectrum is consistent with a reduction in band gap, suggesting a transformation into the A-to-D type J-aggregation. Hence, we focused exclusively on this type of J-aggregation in subsequent analyses. The results (Supplementary Figs. 5d and e) indicated that under various conditions, the aggregation energy of J-aggregation consistently exceeded that of H-aggregation, i.e., the latter aggregation was energetically preferred\u003csup\u003e38\u003c/sup\u003e. Hence, NFAs tended to form H-aggregates in the absence of sonication. Cavitation bubbles are generated during the sonication process. As depicted in Supplementary Fig.\u0026nbsp;6, the periodic contraction and expansion of cavitation bubbles create a marked velocity gradient between the bubble surface and the surrounding solvent. This gradient drives the generation of microstreaming and shear stress (a pair of parallel reaction forces)\u003csup\u003e39\u003c/sup\u003e. Microstreaming pushes L8-BO molecules to aggregate on the bubble surface. Then the disordered molecules will be oriented and arranged under the effect of shear stress\u003csup\u003e40\u003c/sup\u003e, and subsequently become ordered J-aggregation. Meanwhile, the H-aggregation can overcome the energy barrier after acquiring energy and sliding to ordered J-aggregation.\u003c/p\u003e\n \u003cp\u003eThe aggregation behavior of sonicated L8-BO in \u003cem\u003eo\u003c/em\u003e-xy was investigated using temperature-dependent PL spectroscopy (Fig. 1f and Supplementary Fig.\u0026nbsp;7). The long-wavelength shoulder peak (855nm)/main peak (805nm) intensity ratio (\u003cem\u003eI\u003c/em\u003e\u003csub\u003e0-1\u003c/sub\u003e/\u003cem\u003eI\u003c/em\u003e\u003csub\u003e0-0\u003c/sub\u003e) obtained through the double-peak fitting analysis of the PL spectra was employed to quantify the effects of sonication on the aggregation state at different temperatures (Fig. 1g and Supplementary Table\u0026nbsp;1)\u003csup\u003e41\u003c/sup\u003e. Sonicated L8-BO exhibited more pronounced \u003cem\u003eI\u003c/em\u003e\u003csub\u003e0-1\u003c/sub\u003e/\u003cem\u003eI\u003c/em\u003e\u003csub\u003e0-0\u003c/sub\u003e variations, along with a stronger dependence of this ratio on the solution temperature compared with the untreated sample, which manifested the enhanced intermolecular interactions and an increased tendency for aggregation. As mentioned above, J-aggregation was well maintained in the solution phase, further influencing the film morphology. Subsequently, we analyzed the ultraviolet (UV)–visible absorption and PL spectra of the corresponding films (Supplementary Figs. 8 and 9). Sonication induced red shifts in the absorption and PL peaks of the L8-BO film, while predictably enhancing its crystallinity\u003csup\u003e42\u003c/sup\u003e. This effect was quantitatively assessed using grazing-incidence wide-angle X-ray scattering (GIWAXS) measurements (Figs. 1h and i, Supplementary Fig.\u0026nbsp;10, and Supplementary Table\u0026nbsp;2). Prominent π-π stacking peaks of the control (1.750 Å) and sonicated (1.767 Å) L8-BO films were observed in the out-of-plane (OOP) direction\u003csup\u003e43\u003c/sup\u003e. The decrease in \u003cem\u003ed\u003c/em\u003e-spacing from 3.590 Å to 3.556 Å after sonication indicated a closer intermolecular packing. The calculated crystal coherence length (CCL) increased from 19.98 Å in the control film to 21.34 Å in the sonicated film. Thus, the sonicated L8-BO film exhibited a higher crystallinity and more ordered molecular stacking than its nonsonicated counterpart. Time-resolved photoluminescence (TRPL) measurements revealed that sonication increased the exciton lifetime of the L8-BO film from 471 ps to 641 ps, possibly by favoring ordered J-aggregation\u003csup\u003e44\u003c/sup\u003e.\u003c/p\u003e\n\u003c/div\u003e\n\u003ch3\u003eBlend film morphology\u003c/h3\u003e\n\u003cp\u003eSonicated L8-BO was combined with PM6 via sequential deposition to construct layer-by-layer (LBL)-type devices and thus assess the role of sonication in determining blend film morphology. According to the corresponding two-dimensional (2D) GIWAXS patterns (Figs. 2a and b), both PM6 and L8-BO exhibited a preferential face-on orientation in blend films, which favored charge transport, as evidenced by prominent (100) lamellar stacking peaks in the in-plane (IP) direction and strong (010) π-π stacking peaks in the OOP direction. After sonication, the CCL of the blend films increased from 21.58 Å to 23.08 Å, and the \u003cem\u003ed\u003c/em\u003e-spacing decreased from 3.588 Å to 3.514 Å (Fig. 2c and Supplementary Table\u0026nbsp;3). These results suggested that sonication facilitated crystallization and ordered molecular stacking within the active layers. Grazing-incidence small-angle X-ray scattering (GISAXS) measurements were performed to characterize the donor and acceptor domain sizes in the blend films and thus investigate phase separation behavior at the nanoscale (Supplementary Fig.\u0026nbsp;11). The Debye–Anderson–Brumberger and fractal models were used for fitting\u003csup\u003e45\u003c/sup\u003e. The relevant fitting steps are presented in the Supplementary Information, and the relevant fitting parameters are summarized in Supplementary Table 4, with \u003cem\u003eζ\u003c/em\u003e, \u003cem\u003eη\u003c/em\u003e, \u003cem\u003eD\u003c/em\u003e, and 2\u003cem\u003eR\u003c/em\u003e\u003csub\u003eg\u003c/sub\u003e representing the average correlation length of the PM6 phase, correlation length, fractal dimension of L8-BO clusters, and size of clustered L8-BO domains, respectively. After sonication, the 2\u003cem\u003eR\u003c/em\u003e\u003csub\u003eg\u003c/sub\u003e value of the blend films increased, indicating the expansion of acceptor-enriched domains, which facilitated the formation of an optimized morphology and suitable phase separation.\u003c/p\u003e\n\u003cp\u003eTo further elucidate the potential mechanisms underlying the effect of sonication on active layer growth, we conducted in-situ UV–visible absorption spectroscopy to monitor the film-formation process of the PM6:L8-BO blends. The characteristic peaks of PM6 and L8-BO were observed within the ranges of 400–660 nm and 660–950 nm, respectively (Supplementary Fig. 12), and peak position changes were analyzed (Figs. 2d and e). Based on peak evolution, the film-formation process was divided into three stages: (i) solvent evaporation and crystal nucleation, (ii) crystallization, and (iii) film formation\u003csup\u003e46\u003c/sup\u003e. Ordered aggregation in the sonicated L8-BO solution allowed the first stage to begin earlier, thus benefiting the prolonged molecular stacking in the second stage. The red shift of the L8-BO molecular peaks after the film- formation stage further corroborates the formation of J-aggregates. Additionally, during the film-formation process of L8-BO, the peak position of PM6 exhibited a notable blue shift, which was attributed to the immediate deposition of the L8-BO layer. The swelling effect resulted in the dissolution of a part of the donor layer during acceptor deposition, thus inducing additional mutual diffusion between them. Importantly, the ordered J-aggregation of the L8-BO molecules markedly reduced the secondary film-formation time of the PM6 molecules, thus minimizing the diffusion from the acceptor layer into the donor layer and thereby hindering unnecessary mixing.\u003c/p\u003e\n\u003cp\u003eThe effects of sonication on the surface morphology of thin films were probed by atomic force microscopy\u003csup\u003e47\u003c/sup\u003e. Both the neat L8-BO and blend films exhibited increased root mean square (RMS) roughness values after sonication (Supplementary Fig. 13). The more refined fibril network structure in the blend films facilitated the formation of high-speed pathways for electron and hole transport. To distinguish donor and acceptor domains, the stretching of characteristic bonds was analyzed using photoinduced force microscopy (PiFM) (Supplementary Figs. 14–15, Figs. 2f and g). The line profiles extracted from the corresponding localized near-infrared images are presented in Figs. 2h and i. After sonication, the acceptor domains in the blend films exhibited fibrillar structures with an approximate diameter of 23 nm. The increase in the amount of small fibril structures indicated the formation of a fibril network between the donor and acceptor phases. The overlaid PiFM images of PM6 and L8-BO were used to further explore the sizes of the different phases (Supplementary Figs.\u0026nbsp;16)\u003csup\u003e48\u003c/sup\u003e, revealing that sonication prevented excessive donor–acceptor mixing. This is beneficial for reducing geminate recombination and increasing the fill factor (FF), which will be discussed below.\u003c/p\u003e\n\u003cp\u003eFilm-depth-dependent light absorption spectroscopy was conducted to analyze the effects of sonication on the vertical phase separation of the thin films (Supplementary Fig.\u0026nbsp;17)\u003csup\u003e49\u003c/sup\u003e. Sonication increased the acceptor concentration near the cathode, resulting in an optimized vertical gradient distribution within the blend films (Fig. 2j). This optimization was conducive to the rapid transfer of excitons to the acceptor phase after dissociation. The effects of sonication on the morphological evolution of the L8-BO molecular aggregates were rationalized as follows (Fig. 2k). Without sonication, the L8-BO molecules spontaneously formed H-type and amorphous aggregates in the solution. This disordered aggregation contributed to an irregular distribution within the solid film and adversely affected charge transfer (CT). Conversely, upon sonication, the L8-BO molecules formed ordered interdigitated J-type aggregates in the solution. During subsequent crystal growth phases, these J-type aggregates promoted ordered molecular stacking and considerably enhanced π-π stacking interactions among molecules.\u003c/p\u003e\n\u003ch3\u003eTransient absorption (TA) analysis of exciton delocalization\u003c/h3\u003e\n\u003cp\u003eFemtosecond TA spectroscopy measurements were performed to probe the impact of the sonication-optimized microstructure on exciton and charge dynamics. The L8-BO acceptor was selectively excited at 800 nm. The 2D color maps of the TA spectra of the blend films produced from non- and sonicated solutions are presented in Fig. 3a and Supplementary Fig.\u0026nbsp;18a. The ground-state bleaching (GSB) signal of the PM6 donor at 580 nm was extracted and analyzed using exponential fitting to track hole transfer kinetics (Supplementary Fig.\u0026nbsp;19a and Supplementary Table\u0026nbsp;5)\u003csup\u003e50\u003c/sup\u003e. The decrease in sonication-induced lifetime from 0.34 ps to 0.24 ps indicates a concomitant increase in the rate of hole transfer\u003csup\u003e51\u003c/sup\u003e. The signal of the localized exciton (LE) state of the acceptor was extracted at 880 nm (Supplementary Fig.\u0026nbsp;19b). Sonication considerably accelerated excited state decay (i.e., the lifetime decreased from 3.89 ps to 2.28 ps) and thus increased the efficiency of exciton dissociation (Supplementary Table\u0026nbsp;6)\u003csup\u003e52\u003c/sup\u003e. In addition, the extended lifetime of the polaron signals observed at 960 nm revealed that sonication suppressed bimolecular recombination in the film (Supplementary Fig.\u0026nbsp;19c). Supplementary Fig.\u0026nbsp;19d shows the kinetic curves of the intra-moiety delocalized excitons (i-DE) extracted from TA spectra at 1500 nm. This signal was related to the exciton delocalization and will be discussed in detail below.\u003c/p\u003e\n\u003cp\u003eGlobal fitting was used to explore the effects of sonication on the contributions of different exciton dissociation pathways (Fig. 3b). The CS process is governed by competition between the LE → i-DE → CS and LE → CT → CS pathways. The former pathway involves a rapid (within ~ 0.2 ps) transformation of LE states into i-DE states in the acceptor domain owing to strong intermolecular interactions among conjugated units. The i-DE states subsequently transform to CS states\u003csup\u003e53\u003c/sup\u003e. The latter pathway resembles traditional models, with LE states diffusing to the donor/acceptor (D/A) interface, transforming to CT states, and further dissociating into CS states. TA spectra recorded using different time delays were analyzed to extract the dynamic evolution signals corresponding to the LE, i-DE, CT, and CS states (Figs. 3c–f and Supplementary Figs. 18b–e), with the associated contributions shown in Fig. 3g.\u003c/p\u003e\n\u003cp\u003eFor both the control and the sonicated samples, the transient spectral features evolved on similar time scales. Under 1 ps, GSB features matching the steady-state absorption spectrum of the blend film were prominent in the visible region between 550 nm and 780 nm. A positive excited-state absorption (ESA) feature between 450 nm and 550 nm corresponding to the extracted spectrum of the i-DE state suggested the ultrafast formation of this state within the coherence time (~ 200 fs). This ESA returns to a negative GSB after tens of ps, indicating a progression of the system from the i-DE states to new intermediates (CT/CS). On the timescale of tens of hundreds of ps, whilst the GSB features were still present, the transient spectra adopt broad ESA features spanning from 650 nm to 1500 nm. The broadness of the ESA was consistent with the assignment of a CT type intermediate. The spectrum of the CS state was similar to that of the CT state but with a more flattened ESA feature extending into the NIR region with an absence of a positive peak at ~ 660 nm. On a ~ 200 fs timescale, the sharp ESA of the LE state in the NIR region around 850 nm developed a more prominent shoulder around 960 nm, indicating the ultrafast formation of polarons upon formation of the CT state. The intensity of the NIR signal from 1000 nm to 1500 nm decayed much more slowly compared to the signal at 850–900 nm in the first 10 ps, and through global analysis, this signal was mainly attributed to the LE and the CT states.\u003c/p\u003e\n\u003cp\u003eAfter sonication, the contribution of the pathway dominated by i-DE states substantially increased (0.54 → 0.6), which indicated that sonication facilitated exciton delocalization. It can effectively reduce the dependence of exciton dissociation on the D/A interface. Figure 3h shows the fitted lifetime for each stage of the two pathways. The lifetime for all three stages after sonication (\u003cem\u003eτ\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e = 0.23 ps, \u003cem\u003eτ\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e = 2.99 ps, \u003cem\u003eτ\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e = 156 ps) were notably shorter than those observed in the control system (\u003cem\u003eτ\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e = 0.36 ps, \u003cem\u003eτ\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e = 6.11 ps, \u003cem\u003eτ\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e = 248 ps). Accordingly, the rates of evolution towards successive states in the sonicated film (\u003cem\u003ek\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e = 4.35 ps\u003csup\u003e− 1\u003c/sup\u003e, \u003cem\u003ek\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e = 0.33 ps\u003csup\u003e− 1\u003c/sup\u003e, \u003cem\u003ek\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e = 0.0064 ps\u003csup\u003e− 1\u003c/sup\u003e) were faster than that in the control film (\u003cem\u003ek\u003c/em\u003e\u003csub\u003e1\u003c/sub\u003e = 2.78 ps\u003csup\u003e− 1\u003c/sup\u003e, \u003cem\u003ek\u003c/em\u003e\u003csub\u003e2\u003c/sub\u003e = 0.16 ps\u003csup\u003e− 1\u003c/sup\u003e, \u003cem\u003ek\u003c/em\u003e\u003csub\u003e3\u003c/sub\u003e = 0.0040 ps\u003csup\u003e− 1\u003c/sup\u003e). These results suggested that sonication promoted efficient CS in both pathways. In addition, the lifetime of the CS state markedly increased after sonication (from 8.6 ns to 10.6 ns), which indicated suppressed charge recombination.\u003c/p\u003e\n\u003cp\u003eExciton diffusion length (\u003cem\u003eL\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e) plays a crucial role in the CS pathway dominated by the CT state. To investigate the influence of sonication on \u003cem\u003eL\u003c/em\u003e\u003csub\u003eD\u003c/sub\u003e, we recorded the pump-fluence-dependent TA spectra of the L8-BO films (Supplementary Figs.\u0026nbsp;19e and f)\u003csup\u003e49\u003c/sup\u003e, revealing that the exciton decay rate increased with increasing excitation intensity. Two primary quenching pathways were considered, characterized by the intrinsic exciton decay rate coefficients (\u003cem\u003ek\u003c/em\u003e) at any given excitation flux and bimolecular decay rate coefficients (\u003cem\u003eγ\u003c/em\u003e) for exciton–exciton annihilation at high pump fluences\u003csup\u003e52\u003c/sup\u003e. The following diffusion equation was used to globally fit the TA decay curves of the L8-BO acceptor:\u003c/p\u003e\n\u003cdiv id=\"Equa\"\u003e\n \u003cdiv id=\"FileID_Equa\" name=\"EquationSource\"\u003e$$\\:\\frac{dn\\left(t\\right)}{dt}=-kn\\left(t\\right)-\\frac{1}{2}\\gamma\\:{n}^{2}\\left(t\\right)\\:\\:\\:\\:\\:\\left(1\\right)$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003eBased on this equation, we derived a function describing the variation in singlet exciton density distribution \u003cem\u003en\u003c/em\u003e with time \u003cem\u003et\u003c/em\u003e:\u003c/p\u003e\n\u003cdiv id=\"Equb\"\u003e\n \u003cdiv id=\"FileID_Equb\" name=\"EquationSource\"\u003e$$\\:n\\left(t\\right)=\\frac{n\\left(0\\right){e}^{-kt}}{1+\\frac{\\gamma\\:}{2k}n\\left(0\\right)\\left[1-{e}^{-kt}\\right]}\\:\\:\\:\\:\\:\\left(2\\right)$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003ewhere \u003cem\u003eγ\u003c/em\u003e = 8π\u003cem\u003eRD\u003c/em\u003e (\u003cem\u003eD\u003c/em\u003e is the exciton diffusion coefficient, and \u003cem\u003eR\u003c/em\u003e is the annihilation radius of the singlet excitons (assumed to equal 1 nm)). The fitted parameters are listed in Supplementary Table 7. The exciton diffusion length (\\(\\:{L}_{\\text{D}}=\\sqrt{D\\tau\\:}\\)) increased from 29.66 nm to 36.56 nm after sonication (Fig. 3i), indicating efficient exciton transport to the D/A interface and supporting the subsequent conversion to CT states. This agrees with the efficient CT suggested by the global fitting results, which is an essential prerequisite for achieving high-efficiency OSCs\u003csup\u003e54\u003c/sup\u003e.\u003c/p\u003e\n\u003ch3\u003eMolecular vibration and nonradiative recombination\u003c/h3\u003e\n\u003cp\u003eBased on the aforementioned insights into exciton and charge dynamics, transient infrared (TRIR) spectroscopy was used to further examine the effects of sonication on molecular vibration (Figs. 4a, Supplementary Figs.\u0026nbsp;20a-c). We extracted the polaron dynamic curve within the 1700–1800 cm\u003csup\u003e− 1\u003c/sup\u003e range, where the contributions from molecular vibrations were minimal (Fig. 4b). The blend films prepared using sonicated solutions demonstrated accelerated charge generation (i.e., the polaron signal rise time decreased from 9.17 ps to 6.64 ps) and slightly slowed charge recombination, in line with the TAS results. To analyze the transient vibrational spectra and dynamic changes precisely, we used a cubic polynomial function to fit the polaron background and deconvolute its contributions (Supplementary Fig. 21), as illustrated in Fig. 4c and Supplementary Fig.\u0026nbsp;20d\u003csup\u003e55\u003c/sup\u003e. According to the FTIR spectra (Supplementary Fig.\u0026nbsp;14), the double-peak GSB absorption feature between 1500 cm\u003csup\u003e-1\u003c/sup\u003e and 1550 cm\u003csup\u003e-1\u003c/sup\u003e corresponds to L8-BO vibration (C = C bond in ring). These features were sharp at early times and broaden after 10s of ps. The derivative-type feature with a positive peak at 1485 cm\u003csup\u003e-1\u003c/sup\u003e (~ 10 cm\u003csup\u003e-1\u003c/sup\u003e downshift from the main absorption peak) can most likely be attributed to hot ground states (S\u003csub\u003e0\u003c/sub\u003e) or hot LE (with minimal structural distortion from the S\u003csub\u003e0\u003c/sub\u003e).\u003c/p\u003e\n\u003cp\u003eThis derivative signal decayed more rapidly in the sonicated sample (8.3 ps) compared to the control blend (10.7 ps, Fig. 4d) and was likely due to the faster heat dissipation in more structurally ordered molecular solids which allowed efficient propagation of phonons. Another noticeable difference was the vibrational peak widths in the TRIR spectra. Figure 4e compares the GSB feature around 1533 cm\u003csup\u003e-1\u003c/sup\u003e at 850 ps for the control and the sonicated blend. The peak widths of the control were broader than the sonicated blend, hinting to the existence of a wider variety of relative molecular orientations, thus, a greater degree of disorder. Therefore, it was evident that the structural tuning via sonication successfully reduced the disorder in the film, allowing for efficient propagation of both thermal energy and charge. This result is further discussed with TRPL experimental results in a later section.\u003c/p\u003e\n\u003cp\u003eThe removal of the polaron background allowed us to isolate the vibrational mode of PM6 at 1649 cm\u003csup\u003e-1\u003c/sup\u003e (see FTIR spectra in Supplementary Fig. 14a). The recovery of this GSB feature was faster in the sonicated blend (5.84 ps), whereas for the control blend (7.13 ps), a small ESA signal at 1670 cm\u003csup\u003e-1\u003c/sup\u003e slowly emerged after 10s of ps. This subtle difference suggested that when CT/CS states formed, in the more structurally ordered sonicated film, the PM6 did not undergo much geometry change. However, in the control film, the PM6 was structurally disturbed in the CT/CS states, inducing a frequency shift. Another subtle difference was at 1750 cm\u003csup\u003e-1\u003c/sup\u003e. In the control film, an ESA feature was observed after the pulse duration ~ 300 fs and slowly recovered during the scan window. In the sonicated film, this ESA decayed into a clear bleach signal after several hundreds of ps. The FTIR of the sonicated L8-BO neat film compared to the untreated L8-BO hinted that this vibrational mode was likely associated with the presence of J-aggregated L8-BO (π-π stacked geometry). Given that the enhancement of this signal appeared at early times in both films, a possible explanation can be the formation of i-DE states within ~ 200 fs enhanced neighbouring L8-BO coupling through π−π stacking.\u003c/p\u003e\n\u003cp\u003eTemperature-dependent PL and TRPL spectra were recorded to further evaluate the effects of sonication on exciton–vibration coupling and thermally activated processes. With increasing temperature, the PL spectra shifted to shorter wavelengths and broadened (Supplementary Fig.\u0026nbsp;22). The interactions between excitons and molecular vibrations can be inferred from the peak positions\u003csup\u003e56\u003c/sup\u003e, which are quantitatively described by:\u003c/p\u003e\n\u003cdiv id=\"Equc\"\u003e\n \u003cdiv id=\"FileID_Equc\" name=\"EquationSource\"\u003e$$\\:E\\left(T\\right)=E\\left(0\\right)-\\frac{2\\alpha\\:}{\\text{exp}\\left(\\theta\\:/T\\right)-1}\\:\\:\\:\\:\\:\\left(3\\right)$$\u003c/div\u003e\n\u003c/div\u003e\n\u003cp\u003ewhere \u003cem\u003eE\u003c/em\u003e(0) is the bandgap energy at absolute zero (0 K), \u003cem\u003eα\u003c/em\u003e is the intensity of the exciton–vibration interaction, and \u003cem\u003eθ\u003c/em\u003e is the average phonon temperature. Sonication reduced \u003cem\u003eα\u003c/em\u003e from 151.2 meV to 113.3 meV (Fig. 4g and Supplementary Table 8) and was therefore concluded to inhibit exciton–vibration coupling and promote exciton diffusion and polaron transport. By fitting a series of temperature-dependent TRPL spectra (Supplementary Fig. 23), we obtained the temperature dependence of the nonradiative recombination rate, \u003cem\u003eK\u003c/em\u003e\u003csub\u003enr\u003c/sub\u003e (Fig. 4h). In high-temperature regions, \u003cem\u003eK\u003c/em\u003e\u003csub\u003enr\u003c/sub\u003e was more sensitive to temperature changes than in low-temperature regions because of the increase in molecular vibrations induced by thermal energy\u003csup\u003e57\u003c/sup\u003e. Moreover, a thermally activated process occurred between the S\u003csub\u003e0\u003c/sub\u003e and excited states (S\u003csub\u003e1\u003c/sub\u003e)\u003csup\u003e58\u003c/sup\u003e (Fig. 4i). Molecular vibrations led to the formation of an overlapping surface between S\u003csub\u003e0\u003c/sub\u003e and S\u003csub\u003e1\u003c/sub\u003e, which introduced an additional pathway for internal conversion processes\u003csup\u003e59\u003c/sup\u003e. As discussed previously, sonication-induced structural tuning with ordered J-aggregate geometry allows for rapid thermal energy re-distribution and dissipation, which deactivates the thermally-activated non-radiative recombination of excitons.\u003c/p\u003e\n\u003ch3\u003ePhotovoltaic characteristics\u003c/h3\u003e\n\u003cp\u003eTo substantiate the benefits of the ordered J-aggregation induced by sonication, we fabricated LBL-type OSCs with an ITO/PEDOT:PSS/active layer/PDINN/Ag structure. The fabrication and optimization details are described in the Supplementary Information, and the \u003cem\u003eJ\u003c/em\u003e–\u003cem\u003eV\u003c/em\u003e curves and photovoltaic parameters of the optimized devices are presented in Fig. 5a and Table 1. The control device achieved a PCE of 17.50%, an open-circuit voltage (\u003cem\u003eV\u003c/em\u003e\u003csub\u003eOC\u003c/sub\u003e) of 0.878 V, a short-circuit current (\u003cem\u003eJ\u003c/em\u003e\u003csub\u003eSC\u003c/sub\u003e) of 25.36 mA cm\u003csup\u003e− 2\u003c/sup\u003e, and a FF of 78.56%, in agreement with previously published results\u003csup\u003e60,61\u003c/sup\u003e. By controlling the sonication condition (Supplementary Fig. 24 and Supplementary Tables 9 and 10), we achieved a notably increased PCE of 19.43%, a \u003cem\u003eV\u003c/em\u003e\u003csub\u003eOC\u003c/sub\u003e of 0.881 V, a \u003cem\u003eJ\u003c/em\u003e\u003csub\u003eSC\u003c/sub\u003e of 27.05 mA cm\u003csup\u003e− 2\u003c/sup\u003e, and a FF of 81.53%. Compared with that of the control, the external quantum efficiency (EQE) spectrum of the sonication-processed device was shifted to longer wavelengths and exhibited a higher full-spectrum response (Fig. 5b), as evidenced by more efficient exciton dissociation and charge extraction (Supplementary Fig.\u0026nbsp;25 and Supplementary Table\u0026nbsp;11). Subsequently, we fabricated ternary PM6:D18:L8-BO devices with different active layer thicknesses. The device with a 100-nm-thick active layer featured an increased PCE of 20.41% (certified as 19.84% by the National Institute of Metrology, China)\u003c/p\u003e\n\u003cp\u003e(Fig. 5c and Supplementary Fig. 26), a \u003cem\u003eV\u003c/em\u003e\u003csub\u003e\u003cem\u003eOC\u003c/em\u003e\u003c/sub\u003e of 0.899 V, a \u003cem\u003eJ\u003c/em\u003e\u003csub\u003e\u003cem\u003eSC\u003c/em\u003e\u003c/sub\u003e of 27.43 mA cm\u003csup\u003e− 2\u003c/sup\u003e, and a FF of 81.19%; this PCE is one of the highest values reported for OSCs prepared using nonhalogenated solvents (Fig. 5d, Supplementary Tables 12 and 13). The 300-nm-thick device demonstrated a satisfactory PCE of 18.29%, with the FF (76.19%) ranking among the top values reported for an active layer thickness of 300 nm. The corresponding EQE spectra are presented in Supplementary Fig. 27. All \u003cem\u003eJ\u003c/em\u003e\u003csub\u003eSC\u003c/sub\u003e values extracted from the \u003cem\u003eJ\u003c/em\u003e–\u003cem\u003eV\u003c/em\u003e curves differed from the integrated current density of the EQE by \u0026lt; 5%, thereby confirming the reliability of the photovoltaic performance.\u003c/p\u003e\n\u003cp\u003eThe space charge limited current method was used to evaluate the charge transport which is related to the FF between the devices prepared under control and sonication conditions (Supplementary Fig. 28). The calculated hole and electron mobilities are presented in Fig. 5e and Supplementary Table\u0026nbsp;14. Sonication increased hole mobility (\u003cem\u003eµ\u003c/em\u003e\u003csub\u003eh\u003c/sub\u003e) from 6.27 × 10\u003csup\u003e− 4\u003c/sup\u003e cm\u003csup\u003e2\u003c/sup\u003e V\u003csup\u003e− 1\u003c/sup\u003e s\u003csup\u003e− 1\u003c/sup\u003e to 8.59 × 10\u003csup\u003e− 4\u003c/sup\u003e cm\u003csup\u003e2\u003c/sup\u003e V\u003csup\u003e− 1\u003c/sup\u003e s\u003csup\u003e− 1\u003c/sup\u003e and electron mobility (\u003cem\u003eµ\u003c/em\u003e\u003csub\u003ee\u003c/sub\u003e) from 5.43 × 10\u003csup\u003e− 4\u003c/sup\u003e cm\u003csup\u003e2\u003c/sup\u003e V\u003csup\u003e− 1\u003c/sup\u003e s\u003csup\u003e− 1\u003c/sup\u003e to 8.19 × 10\u003csup\u003e− 4\u003c/sup\u003e cm\u003csup\u003e2\u003c/sup\u003e V\u003csup\u003e− 1\u003c/sup\u003e s\u003csup\u003e− 1\u003c/sup\u003e. The higher mobility and lower \u003cem\u003eµ\u003c/em\u003e\u003csub\u003eh\u003c/sub\u003e/\u003cem\u003eµ\u003c/em\u003e\u003csub\u003ee\u003c/sub\u003e ratio observed for the devices prepared under sonication conditions indicated more efficient and balanced charge transport\u003csup\u003e43\u003c/sup\u003e. The density of trap states (\u003cem\u003eN\u003c/em\u003e\u003csub\u003et\u003c/sub\u003e) was calculated using frequency-dependent capacitance spectra recorded under dark conditions (Fig. 5f). The decrease in \u003cem\u003eN\u003c/em\u003e\u003csub\u003et\u003c/sub\u003e from 2.02 × 10\u003csup\u003e16\u003c/sup\u003e cm\u003csup\u003e− 3\u003c/sup\u003e eV\u003csup\u003e− 1\u003c/sup\u003e to 0.98×10\u003csup\u003e16\u003c/sup\u003e cm\u003csup\u003e− 3\u003c/sup\u003e eV\u003csup\u003e− 1\u003c/sup\u003e indicated that sonication effectively prevented trap state generation\u003csup\u003e62\u003c/sup\u003e. The relationship between \u003cem\u003eJ\u003c/em\u003e\u003csub\u003eSC\u003c/sub\u003e and light intensity (\u003cem\u003eP\u003c/em\u003e\u003csub\u003elight\u003c/sub\u003e) was fitted as \u003cem\u003eJ\u003c/em\u003e\u003csub\u003eSC\u003c/sub\u003e = (\u003cem\u003eP\u003c/em\u003e\u003csub\u003elight\u003c/sub\u003e)\u003csup\u003e\u003cem\u003eα\u003c/em\u003e\u003c/sup\u003e (Supplementary Fig. 29a). Sonication increased \u003cem\u003eα\u003c/em\u003e from 0.988 to 0.994 and thus effectively inhibited bimolecular recombination processes\u003csup\u003e63\u003c/sup\u003e. The lower slope of the \u003cem\u003eV\u003c/em\u003e\u003csub\u003eOC\u003c/sub\u003e–ln(\u003cem\u003eP\u003c/em\u003e\u003csub\u003elight\u003c/sub\u003e) plot observed for the sonicated film (Fig. 5g) signified the suppression of trap-assisted recombination, thereby enhancing charge transport in the OSCs\u003csup\u003e64\u003c/sup\u003e. The increase in carrier lifetime from 5.83 µs to 7.44 µs revealed by transient photovoltage (TPV) tests supporting this observation (Supplementary Figs. 29b–c). Transient photocurrent (TPC) tests revealed accelerated charge extraction (i.e., fitting lifetime decreased from 0.35 µs to 0.26 µs) in the devices prepared under sonication conditions (Supplementary Table 15). Fourier transform photocurrent spectroscopy (FTPS-EQE) revealed a decrease in the Urbach energy from 24.16 meV to 23.57 meV after sonication (Supplementary Fig. 30), which suggested a concomitant reduction in energy disorder\u003csup\u003e65\u003c/sup\u003e. These results confirm that sonication can be used to improve OSC performance.\u003c/p\u003e\n\u003cp\u003eGiven that sonication can improve OSC efficiency, one can reasonably anticipate that it may also increase device stability. To minimize the adverse impact of organic interfacial layers on stability, we fabricated inverted devices with an ITO/ZnO/active layer/MoO\u003csub\u003e3\u003c/sub\u003e/Ag configuration. Thermal stability was assessed upon continuous annealing in dry nitrogen atmosphere at 80°C (Fig. 5h). Compared with the control devices, the sonication-processed devices (10 independent cells) exhibited a markedly reduced burn-in loss and an improved thermal stability. PCE decay was subjected to linear fitting, and the results indicated that the \u003cem\u003eT\u003c/em\u003e\u003csub\u003e80\u003c/sub\u003e lifetime of the optimized devices markedly exceeded 10000 h. We also tested the photothermal stability of OSCs under continuous illumination at 100 mW cm\u003csup\u003e− 2\u003c/sup\u003e (50–60°C), revealing that sonication also increased resistance to photothermal stress. The results demonstrate the advantages of the sonication-optimized ordered molecular stacking for improving device stability (Fig. 5i).\u003c/p\u003e\n\u003cdiv id=\"Sec8\"\u003e\n \u003ch2\u003eApplication scope of the sonication strategy\u003c/h2\u003e\n \u003cp\u003eTo explore the application scope of the ultrasound-field-based regulation of ordered J-aggregation in NFAs, we recorded the in-situ PL spectra of other PM6:NFA systems, including BTP-eC9, PY-IT, and PY-DT, during the sonication process (Figs. 6a–c). In addition to the continuous red shift of the PL spectra, we observed the continuous intensification of the long-wavelength shoulder peaks of the PY-IT and PY-DT polymer acceptors, which suggested a concomitant increase in the extent of J-aggregation. We also recorded the temperature-dependent PL spectra of these systems (Supplementary Fig. 31) and extracted the corresponding \u003cem\u003eI\u003c/em\u003e\u003csub\u003e0-1\u003c/sub\u003e/\u003cem\u003eI\u003c/em\u003e\u003csub\u003e0-0\u003c/sub\u003e peak intensity ratios (Figs. 6d–f and Supplementary Table 16). The results showed enhanced temperature dependence, indicating intermolecular interaction and aggregation enhancement. Subsequently, solar cells based on these three systems were fabricated using a conventional device structure (Supplementary Fig. 32). The corresponding \u003cem\u003eJ\u003c/em\u003e–\u003cem\u003eV\u003c/em\u003e curves are shown in Figs. 6g–i, and the device parameters are summarized in Supplementary Table\u0026nbsp;17. The sonication-processed PM6:BTP-eC9, PM6:PY-IT, and PM6:PY-DT devices achieved improved PCEs of 19.31%, 18.32%, and 18.71% compared to control systems, respectively. These results demonstrate the broad application scope of the sonication strategy.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe NFA aggregation state in a nonhalogenated solvent was modulated using sonication. The combined results of simulations, computations, and experiments demonstrate that sonication induced the formation of ordered J-aggregates within the NFA solutions and films through microstreaming and shear stress. Morphological characterizations revealed that sonication improved phase separation during film formation, enhancing π-π stacking interactions and optimizing the molecular arrangement. The global fitting of TA data indicated that sonication increased the proportion of CS pathways dominated by i-DE states, thereby enhancing exciton dissociation. TRIR data shows the structural tuning effect on heat and carrier transport in the system, and together with temperature-dependent TRPL data further validated the inhibitory effect of sonication on nonradiative recombination. PM6:L8-BO devices exhibited a high PCE of 19.43% and superior thermal stability (\u003cem\u003eT\u003c/em\u003e\u003csub\u003e80\u003c/sub\u003e lifetime\u0026thinsp;\u0026gt;\u0026thinsp;10000 h). The addition of D18 to the PM6:L8-BO system resulted in a remarkable PCE of 20.41% (certified PCE of 19.84%), which is among the highest efficiencies reported for OSCs prepared using nonhalogenated solvents. The developed sonication strategy was demonstrated to have a broad application scope, as exemplified by its applicability to small-molecule and polymer acceptor systems. Thus, our work provides a practical approach to the fabrication of efficient, stable, and ecofriendly OSCs, facilitating their large-scale industrial production.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eMeasurements of PL and TRPL spectra\u003c/h2\u003e \u003cp\u003eThe PL spectra and TRPL experiments were performed using a confocal optical microscope, specifically the Nanofinder FLEX2 from Tokyo Instruments, Inc. This setup was equipped with a time-correlated single-photon counting (TCSPC) module, the Becker \u0026amp; Hickl SPC-150. For all PL spectrum measurements, a charge-coupled device (CCD) sensor, model DU420A-OE from Andor, was utilized. The excitation wavelength was consistently set at 400 nm, and the excitation power for both PL spectra and TRPL experiments was maintained at 1 µW. The laser spot had an approximate diameter of 8 µm, resulting in an area calculated as S = 5.02 × 10⁻⁷ cm².\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eMeasurements of AFM and PiFM Images\u003c/h2\u003e \u003cp\u003eAFM measurements were performed on a Bioscope Resolve AFM (Bruker) in a tapping mode under ambient conditions. Photo-induced Force Microscopy (PiFM) measurements were performed by employing Anfatec Instruments AG (Oelsnitz, Germany).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eMeasurement of TA spectra\u003c/h2\u003e \u003cp\u003eFemtosecond TA spectra were recorded using an optical setup that incorporated a Ti:sapphire femtosecond laser (from Coherent) in conjunction with an optical parametric amplifier (OPA) system. The 800 nm pulse emitted by the laser was divided into two beams via a beam splitter. One of these beams was directed into the optical parametric amplifier (TOPAS, Coherent) to generate pump pulses at 600 nm. The second beam was focused onto both sapphire and YAG plates to produce a white light supercontinuum, which served as the probe beams with a spectral range of 750–1600 nm. The initial pulses were also split into two paths; one path led to the OPA for generating 400 and 750 nm pump pulses utilized in this study, while the other path produced a broad spectrum of probe light spanning from 520–800 nm (visible) and from 850–1300 nm (NIR).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eMeasurement of TRIR spectroscopy\u003c/h2\u003e \u003cp\u003eTRIR spectra were acquired using pump and probe beams generated through nonlinear optical processes utilizing the 1030 nm fundamental output from a Ytterbium-based laser oscillator (Pharos, Light Conversion). An optical parametric amplifier (OPA, Orpheus, Light Conversion), in conjunction with difference frequency generation (DFG, Lyra, Light Conversion), produced pump pulses with highly tunable wavelengths across the UV-NIR spectrum (210–2660 nm, ~ 150 fs temporal FWHM, ~ 5 nm bandwidth), operating at a base repetition rate of 100 kHz. The laser output was pulse picked to 10kHZ and a mechanical chopper (Thorlabs, MC1F60) was employed to reduce the pump repetition rate to 5 kHz while maintaining a probe rate of 10 kHz. The pump beam traversed a mechanical delay stage (Newport ESP300) to achieve optical delay times ranging from − 50 to 900 ps and was subsequently focused to attain an approximate spatial diameter of FWHM ~ 420 µm at the sample location. Linear polarization devices were utilized to control the relative polarization between the pump and probe beams at the magic angle. The mid-infrared probe spanning wavelengths from 4000 to 13000 nm with a temporal FWHM of less than 180 fs, was generated by an OPA and DFG system (ORPHEUS TWINS, Light Conversion). Given that atmospheric moisture and particulate matter significantly absorb mid-infrared radiation, all components of the probe pathway were enclosed under nitrogen within sealed boxes and conduits. TRIR spectra were captured using a polychromator-based IR spectrometer cooled with liquid nitrogen during experiments (Horiba iHR320 integrated with an FPAS 128-element MCT detector; Infra-Red Systems; spectral acquisition rate: 10 kHz; bandwidth: ∼300 cm⁻¹). The sample was affixed to an X-Z translation stage controlled by two stepper motors for raster scanning purposes.\u003c/p\u003e \u003c/div\u003e "},{"header":"Declarations","content":"\u003ch3\u003eReporting summary\u003c/h3\u003e\n\u003cp\u003eFurther information on research design is available in the Nature Portfolio Reporting Summary linked to this article.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the National Natural Science Foundation of China (52320105003, 52203333, 32402291). X.T.H acknowledges support from the Taishan Scholars Program (tstp20230610). K.N.Z acknowledges support from the Shandong Provincial Natural Science Foundation (ZR2024QA063) and the Young Scholars Program of Shandong University, China. The authors would like to thank the Analytical Centre for Structural Constituent and Physical Property of Core Facilities Sharing Platform, Shandong University for use of the femtosecond transient absorption spectroscopy system (TAS, 2104573S), and the Instrument Improvement Funds of Shandong University Public Technology Platform (ts20230101). The authors are grateful to the Shanghai Synchrotron Radiation Facility (beamline BL16B1) for support with GIXS measurements.\u003c/p\u003e\n"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLuo, S. \u003cem\u003eet al.\u003c/em\u003e Auxiliary sequential deposition enables 19%-efficiency organic solar cells processed from halogen-free solvents. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 6964 (2023).\u003c/li\u003e\n\u003cli\u003eMa, L. J. \u003cem\u003eet al.\u003c/em\u003e Design of low-cost non-fused ultranarrow-band-gap acceptors for versatile photovoltaic applications. \u003cem\u003eJoule\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 2238\u0026ndash;2249 (2024).\u003c/li\u003e\n\u003cli\u003eChen, Z. H. \u003cem\u003eet al.\u003c/em\u003e Local dipole modulation toward high fill factor in organic solar cells. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e34\u003c/strong\u003e, 2408858 (2024).\u003c/li\u003e\n\u003cli\u003eWang, Y. \u003cem\u003eet al.\u003c/em\u003e Origins of the open-circuit voltage in ternary organic solar cells and design rules for minimized voltage losses. \u003cem\u003eNat. Energy\u003c/em\u003e \u003cstrong\u003e8\u003c/strong\u003e, 978\u0026ndash;988 (2023).\u003c/li\u003e\n\u003cli\u003eDing, P. \u003cem\u003eet al.\u003c/em\u003e U-shaped dimeric acceptors for balancing efficiency and stability in organic solar cells. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, 2414080 (2024).\u003c/li\u003e\n\u003cli\u003eWang, S. \u003cem\u003eet al.\u003c/em\u003e Achieving 20% efficiency in organic solar cells through conformationally locked solid additives. \u003cem\u003eAdv. Energy Mater.\u003c/em\u003e \u003cstrong\u003e25\u003c/strong\u003e, 2405205 (2024).\u003c/li\u003e\n\u003cli\u003eZou, W. \u003cem\u003eet al.\u003c/em\u003e Extending exciton diffusion length via an organic-metal platinum complex additive for high-performance thick-film organic solar cells. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 2413125 (2024).\u003c/li\u003e\n\u003cli\u003eWang, J. \u003cem\u003eet al.\u003c/em\u003e Isomerism effect of 3D dimeric acceptors for non-halogenated solvent-processed organic solar cells with 20 % efficiency. \u003cem\u003eAngew. Chem. Int. Edit.\u003c/em\u003e \u003cstrong\u003e64\u003c/strong\u003e, e202423562 (2024).\u003c/li\u003e\n\u003cli\u003eChen, C. \u003cem\u003eet al.\u003c/em\u003e Molecular interaction induced dual fibrils towards organic solar cells with certified efficiency over 20%. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 6865 (2024).\u003c/li\u003e\n\u003cli\u003eKong, X. \u003cem\u003eet al.\u003c/em\u003e Suppressed non-radiative loss and efficient hole transfer at a small highest occupied molecular orbital offset endows binary organic solar cells with 19.73% efficiency and a small efficiency-cost gap. \u003cem\u003eEnergy Environ. Sci.\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 386\u0026ndash;396 (2025).\u003c/li\u003e\n\u003cli\u003eLang, Y. \u003cem\u003eet al.\u003c/em\u003e Balanced miscibility and crystallinity by 2D acceptors enabled halogen-free solvent-processed organic solar cells to achieve 19.28% efficiency. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 2413270 (2025).\u003c/li\u003e\n\u003cli\u003eZhang, Y. \u003cem\u003eet al.\u003c/em\u003e Graded bulk-heterojunction enables 17% binary organic solar cells via nonhalogenated open air coating. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 4815 (2021).\u003c/li\u003e\n\u003cli\u003eChen, H. \u003cem\u003eet al.\u003c/em\u003e Heterogeneous nucleating agent for high‐boiling‐point nonhalogenated solvent‐processed organic solar cells and modules. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e36\u003c/strong\u003e, 2402350 (2024).\u003c/li\u003e\n\u003cli\u003eFan, B. \u003cem\u003eet al.\u003c/em\u003e Enabling high efficiency of hydrocarbon-solvent processed organic solar cells through balanced charge generation and non-radiative loss. \u003cem\u003eAdv. Energy Mater.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 2101768 (2021).\u003c/li\u003e\n\u003cli\u003eFan, B. \u003cem\u003eet al.\u003c/em\u003e Optimisation of processing solvent and molecular weight for the production of green-solvent-processed all-polymer solar cells with a power conversion efficiency over 9%. \u003cem\u003eEnergy Environ. Sci.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 1243\u0026ndash;1251 (2017).\u003c/li\u003e\n\u003cli\u003eZhao, H. \u003cem\u003eet al.\u003c/em\u003e Hot hydrocarbon-solvent slot-die coating enables high-efficiency organic solar cells with temperature-dependent aggregation behavior. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 2002302 (2020).\u003c/li\u003e\n\u003cli\u003eWu, X. \u003cem\u003eet al.\u003c/em\u003e Introducing a phenyl end group in the inner side chains of A-DA\u0026rsquo;D-A acceptors enables high-efficiency organic solar cells processed with nonhalogenated solvent. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 2302946 (2023).\u003c/li\u003e\n\u003cli\u003eXu, X., Yu, L., Yan, H., Li, R. \u0026amp; Peng, Q. Highly efficient non-fullerene organic solar cells enabled by a delayed processing method using a non-halogenated solvent. \u003cem\u003eEnergy Environ. Sci.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 4381\u0026ndash;4388 (2020).\u003c/li\u003e\n\u003cli\u003eFan, B. \u003cem\u003eet al.\u003c/em\u003e Importance of structural hinderance in performance\u0026ndash;stability equilibrium of organic photovoltaics. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 5946 (2022).\u003c/li\u003e\n\u003cli\u003eZhao, W. \u003cem\u003eet al.\u003c/em\u003e Vacuum-assisted annealing method for high efficiency printable large-area polymer solar cell modules. \u003cem\u003eJ. Mater. Chem. C\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 3206\u0026ndash;3211 (2019).\u003c/li\u003e\n\u003cli\u003eZhao, W. \u003cem\u003eet al.\u003c/em\u003e Environmentally friendly solvent-processed organic solar cells that are highly efficient and adaptable for the blade-coating method. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e30\u003c/strong\u003e, 1704837 (2018).\u003c/li\u003e\n\u003cli\u003eLi, H. \u003cem\u003eet al.\u003c/em\u003e Advances in the device design and printing technology for eco-friendly organic photovoltaics. \u003cem\u003eEnergy Environ. Sci.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 76\u0026ndash;88 (2023).\u003c/li\u003e\n\u003cli\u003eChen, H. \u003cem\u003eet al.\u003c/em\u003e 17.1 %-efficient eco-compatible organic solar cells from a dissymmetric 3D network acceptor. \u003cem\u003eAngew. Chem. Int. Edit.\u003c/em\u003e \u003cstrong\u003e60\u003c/strong\u003e, 3238\u0026ndash;3246 (2021).\u003c/li\u003e\n\u003cli\u003eZhuo, H. \u003cem\u003eet al.\u003c/em\u003e Giant molecule acceptor enables highly efficient organic solar cells processed using non-halogenated solvent. \u003cem\u003eAngew. Chem. Int. Edit.\u003c/em\u003e \u003cstrong\u003e62\u003c/strong\u003e, e202303551 (2023).\u003c/li\u003e\n\u003cli\u003ePr\u0026ouml;ller, S. \u003cem\u003eet al.\u003c/em\u003e Following the morphology formation in situ in printed active layers for organic solar cells. \u003cem\u003eAdv. Energy Mater.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 1501580 (2016).\u003c/li\u003e\n\u003cli\u003eGe, Y. et al. Aggregation engineering of toluene-processed acceptor layer enables over 19% efficiency of air-blade-coated organic solar cells. \u003cem\u003eAdv. Mater\u003c/em\u003e.\u003cstrong\u003e 22\u003c/strong\u003e, 2502579. (2025)\u003c/li\u003e\n\u003cli\u003eWang, Y. et al. Achieving 20% toluene-processed binary organic solar cells via secondary regulation of donor aggregation in sequential processing. \u003cem\u003eNano-Micro Lett\u003c/em\u003e. \u003cstrong\u003e17\u003c/strong\u003e, 206 (2025).\u003c/li\u003e\n\u003cli\u003eHan, C. \u003cem\u003eet al.\u003c/em\u003e A halogen-free and universally volatile solid additive enables binary organic solar cells to exceed 19% efficiency. \u003cem\u003eAdv. Funct. Mater.\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 2416381 (2024).\u003c/li\u003e\n\u003cli\u003eZhang, R. \u003cem\u003eet al.\u003c/em\u003e Equally high efficiencies of organic solar cells processed from different solvents reveal key factors for morphology control. \u003cem\u003eNat. Energy\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, 124\u0026ndash;134 (2025).\u003c/li\u003e\n\u003cli\u003eGuo, C. \u003cem\u003eet al.\u003c/em\u003e Light-induced quinone conformation of polymer donors toward 19.9% efficiency organic solar cells. \u003cem\u003eEnergy Environ. Sci.\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 2492\u0026ndash;2499 (2024).\u003c/li\u003e\n\u003cli\u003eCui, F. Z. \u003cem\u003eet al.\u003c/em\u003e Using an external electric field to tune active layer morphology enabling high-efficiency organic solar cells via ambient blade coating. \u003cem\u003eSci. Adv.\u003c/em\u003e \u003cstrong\u003e10\u003c/strong\u003e, eado5460 (2024).\u003c/li\u003e\n\u003cli\u003eWang, W. \u003cem\u003eet al.\u003c/em\u003e H-bonded organic frameworks as ultrasound-programmable delivery platform. \u003cem\u003eNature\u003c/em\u003e \u003cstrong\u003e638\u003c/strong\u003e, 401\u0026ndash;410 (2025).\u003c/li\u003e\n\u003cli\u003eGuo, G. \u003cem\u003eet al.\u003c/em\u003e Enhanced porosity and permeability of three-dimensional alginate scaffolds via acoustic microstreaming induced by low-intensity pulsed ultrasound. \u003cem\u003eUltrason. Sonochem.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 279\u0026ndash;285 (2017).\u003c/li\u003e\n\u003cli\u003eSun, M. \u003cem\u003eet al.\u003c/em\u003e Overcoming disordered preaggregation in liquid state for highly efficient organic solar cells printed from nonhalogenated solvents. \u003cem\u003eAdv. Energy Mater.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 2203465 (2023).\u003c/li\u003e\n\u003cli\u003eAshokkumar, \u003cem\u003eet al.\u003c/em\u003e The characterization of acoustic cavitation bubbles \u0026ndash; an overview. \u003cem\u003eUltrason. Sonochem.\u003c/em\u003e \u003cstrong\u003e18\u003c/strong\u003e, 864\u0026ndash;872 (2011).\u003c/li\u003e\n\u003cli\u003eLiang, H. \u003cem\u003eet al.\u003c/em\u003e A rare case of brominated small molecule acceptors for high-efficiency organic solar cells. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e14\u003c/strong\u003e, 4707 (2023).\u003c/li\u003e\n\u003cli\u003eSu, W. P. \u003cem\u003eet al.\u003c/em\u003e Solitons in polyacetylene. \u003cem\u003ePhys. Rev. Lett.\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, 1698\u0026ndash;1701 (1979).\u003c/li\u003e\n\u003cli\u003eZhao, Q. \u003cem\u003eet al.\u003c/em\u003e Balancing the H- and J-aggregation in DTS(PTTh2)2/PC70BM to yield a high photovoltaic efficiency. \u003cem\u003eJ. Mater. Chem. C\u003c/em\u003e \u003cstrong\u003e3\u003c/strong\u003e, 8183\u0026ndash;8192 (2015).\u003c/li\u003e\n\u003cli\u003eMondal, J. et al. Bubble oscillations at low frequency ultrasound for biological applications. \u003cem\u003eUltrason. Sonochem\u003c/em\u003e. \u003cstrong\u003e104\u003c/strong\u003e, 106816 (2024).\u003c/li\u003e\n\u003cli\u003eBlanc, F. et al. Rheology of dense suspensions under shear rotation. \u003cem\u003ePhys. Rev. Lett\u003c/em\u003e. \u003cstrong\u003e130\u003c/strong\u003e, 118202 (2023).\u003c/li\u003e\n\u003cli\u003eClark, J., Silva, C., Friend, R. H. \u0026amp; Spano, F. C. Role of intermolecular coupling in the photophysics of disordered organic semiconductors: aggregate emission in regioregular polythiophene. \u003cem\u003ePhys. Rev. Lett.\u003c/em\u003e \u003cstrong\u003e98\u003c/strong\u003e, 206406 (2007).\u003c/li\u003e\n\u003cli\u003eZhao, Q.\u003cem\u003e et al.\u003c/em\u003e H- and J-aggregation inspiring efficient solar conversion. \u003cem\u003eJ. Mater. Chem. A\u003c/em\u003e \u003cstrong\u003e9\u003c/strong\u003e, 1119\u0026ndash;1126 (2021).\u003c/li\u003e\n\u003cli\u003eZhang, G. \u003cem\u003eet al.\u003c/em\u003e Delocalization of exciton and electron wavefunction in non-fullerene acceptor molecules enables efficient organic solar cells. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e11\u003c/strong\u003e, 3943 (2020).\u003c/li\u003e\n\u003cli\u003eQiao, J. \u003cem\u003eet al.\u003c/em\u003e Enhanced exciton delocalization in organic near-infrared photodetectors via solid additive-mediated J-aggregation. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e37\u003c/strong\u003e, 2418844 (2024).\u003c/li\u003e\n\u003cli\u003eWang, J. \u003cem\u003eet al.\u003c/em\u003e Two well-compatible acceptors with efficient energy transfer enable ternary organic photovoltaics exhibiting a 13.36% efficiency. \u003cem\u003eSmall\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1902602 (2019).\u003c/li\u003e\n\u003cli\u003eChen, H. \u003cem\u003eet al.\u003c/em\u003e A guest-assisted molecular-organization approach for \u0026gt;17% efficiency organic solar cells using environmentally friendly solvents. \u003cem\u003eNat. Energy\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 1045\u0026ndash;1053 (2021).\u003c/li\u003e\n\u003cli\u003eZhou, M. \u003cem\u003eet al.\u003c/em\u003e 19.10% efficiency and 80.5% fill factor layer-by-layer organic solar cells realized by 4-bis(2-thienyl)pyrrole-2,5-dione based polymer additives for inducing vertical segregation morphology. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e35\u003c/strong\u003e, 2208279 (2023).\u003c/li\u003e\n\u003cli\u003eZhu, L. \u003cem\u003eet al.\u003c/em\u003e Single-junction organic solar cells with over 19% efficiency enabled by a refined double-fibril network morphology. \u003cem\u003eNat. Mater.\u003c/em\u003e \u003cstrong\u003e21\u003c/strong\u003e, 656\u0026ndash;663 (2022).\u003c/li\u003e\n\u003cli\u003eCai, Y. \u003cem\u003eet al.\u003c/em\u003e Vertically optimized phase separation with improved exciton diffusion enables efficient organic solar cells with thick active layers. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e13\u003c/strong\u003e, 2369 (2022).\u003c/li\u003e\n\u003cli\u003eCai, Y. \u003cem\u003eet al.\u003c/em\u003e A well-mixed phase formed by two compatible non-fullerene acceptors enables ternary organic solar cells with efficiency over 18.6%. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e33\u003c/strong\u003e, 2101733 (2021).\u003c/li\u003e\n\u003cli\u003eCui, F. \u003cem\u003eet al.\u003c/em\u003e Vertical-phase-locking effect in efficient and stable all-polymer-hosted solar cells. \u003cem\u003eACS Energy Lett.\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 3709\u0026ndash;3717 (2022).\u003c/li\u003e\n\u003cli\u003eChen, Z. \u003cem\u003eet al.\u003c/em\u003e Trap state induced recombination effects on indoor organic photovoltaic cells. \u003cem\u003eACS Energy Lett.\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 3203\u0026ndash;3211 (2021).\u003c/li\u003e\n\u003cli\u003eWang, R. \u003cem\u003eet al.\u003c/em\u003e Charge separation from an intra-moiety intermediate state in the high-performance PM6:Y6 organic photovoltaic blend. \u003cem\u003eJ. Am. Chem. Soc.\u003c/em\u003e \u003cstrong\u003e142\u003c/strong\u003e, 12751\u0026ndash;12759 (2020).\u003c/li\u003e\n\u003cli\u003eChen, X. \u003cem\u003eet al.\u003c/em\u003e A unified description of non-radiative voltage losses in organic solar cells. \u003cem\u003eNat. Energy\u003c/em\u003e \u003cstrong\u003e6\u003c/strong\u003e, 799\u0026ndash;806 (2021).\u003c/li\u003e\n\u003cli\u003eZhang, K. \u003cem\u003eet al.\u003c/em\u003e Donoracceptor interfacial dipole polarization for efficient and stable thick-film organic photovoltaics. \u003cem\u003eNano Energy\u003c/em\u003e \u003cstrong\u003e134\u003c/strong\u003e, 110546 (2025).\u003c/li\u003e\n\u003cli\u003eGuha, S. \u003cem\u003eet al.\u003c/em\u003e Temperature-dependent photoluminescence of organic semiconductors with varying backbone conformation. \u003cem\u003ePhys. Rev. B\u003c/em\u003e \u003cstrong\u003e67\u003c/strong\u003e, 125204 (2003).\u003c/li\u003e\n\u003cli\u003eGuo, Q. \u003cem\u003eet al.\u003c/em\u003e Enhancing the performance of organic solar cells by prolonging the lifetime of photogenerated excitons. \u003cem\u003eAdv. Mater.\u003c/em\u003e \u003cstrong\u003e32\u003c/strong\u003e, 2003164 (2020).\u003c/li\u003e\n\u003cli\u003eLi, Q. \u003cem\u003eet al.\u003c/em\u003e A conical intersection model to explain aggregation induced emission in diphenyl dibenzofulvene. \u003cem\u003eChem. Commun.\u003c/em\u003e \u003cstrong\u003e49\u003c/strong\u003e, 5966\u0026ndash;5968 (2013).\u003c/li\u003e\n\u003cli\u003eGao, Y.-J. \u003cem\u003eet al.\u003c/em\u003e Excited-state decay paths in tetraphenylethene derivatives. \u003cem\u003eJ. Phys. Chem. A\u003c/em\u003e \u003cstrong\u003e121\u003c/strong\u003e, 2572\u0026ndash;2579 (2017).\u003c/li\u003e\n\u003cli\u003eMa, R. \u003cem\u003eet al.\u003c/em\u003e Revealing the underlying solvent effect on film morphology in high-efficiency organic solar cells through combined ex situ and in situ observations. \u003cem\u003eEnergy Environ. Sci.\u003c/em\u003e \u003cstrong\u003e16\u003c/strong\u003e, 2316\u0026ndash;2326 (2023).\u003c/li\u003e\n\u003cli\u003eBai, H. et al. Green-solvent processed blade-coating organic solar cells with an efficiency approaching 19% enabled by alkyl-tailored acceptors. \u003cem\u003eNano-Micro Lett.\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 241 (2023).\u003c/li\u003e\n\u003cli\u003eHughes, M. P. \u003cem\u003eet al.\u003c/em\u003e Determining the dielectric constants of organic photovoltaic materials using impedance spectroscopy. \u003cem\u003eAdv. Funct. Mater.\u003c/em\u003e \u003cstrong\u003e28\u003c/strong\u003e, 1801542 (2018).\u003c/li\u003e\n\u003cli\u003eHuang, J. \u003cem\u003eet al.\u003c/em\u003e Polymer bulk heterojunction solar cells employing F\u0026ouml;rster resonance energy transfer. \u003cem\u003eNat. Photonics\u003c/em\u003e \u003cstrong\u003e7\u003c/strong\u003e, 479\u0026ndash;485 (2013).\u003c/li\u003e\n\u003cli\u003eHonda, S. \u003cem\u003eet al\u003c/em\u003e. Selective dye loading at the heterojunction in polymer/fullerene solar cells. \u003cem\u003eAdv. Energy Mater.\u003c/em\u003e \u003cstrong\u003e1\u003c/strong\u003e, 588\u0026ndash;598 (2011).\u003c/li\u003e\n\u003cli\u003eWang, J. \u003cem\u003eet al.\u003c/em\u003e Increasing donor-acceptor spacing for reduced voltage loss in organic solar cells. \u003cem\u003eNat. Commun.\u003c/em\u003e \u003cstrong\u003e12\u003c/strong\u003e, 6679 (2021).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Table 1","content":"\u003cp\u003eTable 1 is available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"nature-portfolio","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"","title":"Nature Portfolio","twitterHandle":"","acdcEnabled":false,"dfaEnabled":false,"editorialSystem":"ejp","reportingPortfolio":"","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"organic solar cells, ultrasound field, J-aggregation, exciton delocalization, molecular vibration","lastPublishedDoi":"10.21203/rs.3.rs-6915241/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6915241/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe commercialization of organic solar cells (OSCs) requires eliminating halogenated solvents from their production. However, the disordered molecular aggregation of nonfullerene acceptors in nonhalogenated solvents hinders the realization of OSCs with high power conversion efficiencies (PCEs). Herein, this problem is addressed using an external physical field modulation strategy involving the sonication of nonfullerene acceptor solutions. The application of ultrasound field induces a transformation from disordered molecular aggregation to ordered J-aggregation via microstreaming and shear stress in nonhalogenated solvents. The aggregation is effectively preserved within films, causing the formation of a fibril network with enhanced π-π stacking interactions. Moreover, sonication promotes the conversion of localized excitons to intra-moiety delocalized excitons and suppresses molecular vibrations, thus favoring charge separation and reducing nonradiative recombination. Consequently, the PCEs of PM6:L8-BO-based binary and ternary devices fabricated using \u003cem\u003eo\u003c/em\u003e-xylene as a nonhalogenated solvent are 19.43% and 20.41% (certified 19.84%), which is among the highest values reported for OSCs produced without halogenated solvents. The binary devices also exhibit high thermal stabilities, with the \u003cem\u003eT\u003c/em\u003e\u003csub\u003e80\u003c/sub\u003e lifetime exceeding 10000 h. The developed strategy is applicable to various small-molecule acceptor and polymer acceptor systems, facilitating control disordered aggregation in nonfullerene acceptor solutions and thus paving the way for practical industrial applications.\u003c/p\u003e","manuscriptTitle":"Sonication-induced J-aggregation in nonhalogenated solvents boosts exciton delocalization for high-efficiency organic solar cells","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-06-23 07:56:46","doi":"10.21203/rs.3.rs-6915241/v1","editorialEvents":[],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"nature-communications","isNatureJournal":true,"hasQc":false,"allowDirectSubmit":false,"externalIdentity":"NCOMMS","sideBox":"Learn more about [Nature Communications](http://www.nature.com/ncomms/)","snPcode":"","submissionUrl":"https://mts-ncomms.nature.com/","title":"Nature Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"ejp","reportingPortfolio":"Nature Communications","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"5946e8c4-627e-40cd-8d77-801780220f66","owner":[],"postedDate":"June 23rd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":50326247,"name":"Physical sciences/Materials science/Materials for energy and catalysis/Solar cells"},{"id":50326248,"name":"Physical sciences/Energy science and technology/Renewable energy/Solar energy/Photovoltaics/Solar cells"}],"tags":[],"updatedAt":"2025-11-26T08:11:19+00:00","versionOfRecord":{"articleIdentity":"rs-6915241","link":"https://doi.org/10.1038/s41467-025-65447-y","journal":{"identity":"nature-communications","isVorOnly":false,"title":"Nature Communications"},"publishedOn":"2025-11-25 05:00:00","publishedOnDateReadable":"November 25th, 2025"},"versionCreatedAt":"2025-06-23 07:56:46","video":"","vorDoi":"10.1038/s41467-025-65447-y","vorDoiUrl":"https://doi.org/10.1038/s41467-025-65447-y","workflowStages":[]},"version":"v1","identity":"rs-6915241","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6915241","identity":"rs-6915241","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}