Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
Advertisement
Nature Communications volume 17, Article number: 2105 (2026) Cite this article
2719 Accesses
Metrics details
Solid additives (SAs) improve organic solar cells (OSCs) morphology and performance, yet, their efficiency in all-small-molecule (ASM) OSCs is limited by strong donor-acceptor crystallinity, and the underlying mechanisms remain unclear. Herein, structurally resembling the MPhS-C2 donor, 3-allylrhodanine (3-AR) as a SA enables precise control over the active layer morphology. Interesting, combined theoretical calculations and thin-film morphology analyses reveal 3-AR preferentially interacts with acceptor N3, thereby directing morphology evolution. Moreover, 3-AR effectively narrows the film formation window, suppressing lateral over-aggregation and crystallization, while promoting vertical donor–acceptor homogeneity. This optimized morphology synergistically facilitates efficient charge dissociation and collection, suppresses both bimolecular and trap-assisted recombination, and extends the lifetimes of excitons and charge carriers. Therefore, the 3-AR-treated MPhS-C2:N3 device achieves a remarkable power conversion efficiency (PCE) of 18.43% (certified 18.16%)—the highest reported for binary ASM-OSCs. This work shows additive efficacy is governed by balanced selective interactions, not by simplistic structural similarity.
Printable organic solar cells (OSCs), featuring newspaper-like manufacturability along with ultralight/ultrathin/flexible characteristics, are ideally suited for deployment in distributed photovoltaic systems^1,2,3,4. Recent breakthroughs in device engineering have propelled polymer-based organic solar cells (P-OSCs) beyond the 21% power conversion efficiency (PCE) milestone; however, all-small-molecule OSCs (ASM-OSCs) still remain, with PCEs currently capped at around 18%^5,6,7. In contrast to the batch-dependent performance of P-OSCs, ASM-OSCs offer superior material homogeneity, defined molecular structures, simplified purification, and excellent batch-to-batch consistency, positioning them as highly competitive candidates for industrial-scale manufacturing^8,9,10. However, the strong crystallization tendency of small-molecule donors and acceptors leads to large phase separation domains (>60 nm) with diminishing donor−acceptor interface¹¹, which impedes exciton diffusion (10–30 nm)^1,12, thereby reducing short-circuit current density (J_sc) in ASM-OSCs. Besides, their high molecular-ordering features prevent the formation of a bi-continuous interpenetrating network morphology, instead developing island-like structures that act as traps for excitons and charges, inducing recombination and counterproductively degrading the fill factor (FF) parameter in ASM-OSCs^{13,14,15,16,17}.
As a result, precisely controlling the active layer morphology in ASM-OSCs becomes significantly more challenging. Consequently, the delicate balance between molecular crystallization and phase separation within the ASM-OSCs remains difficult to regulate, hindering the achievement of high PCEs (>20%). To date, various strategies have been developed to tune the active layer morphology in ASM-OSCs, including optimization of the donor/acceptor ratio¹⁸, use of processing additives^19,20,21, thermal and/or solvent vapor annealing post-treatments^22,23, incorporation of ternary components^24,25, etc. Among these strategies, solid additives (SAs) have proven particularly effective in finely tuning the nucleation and growth dynamics within ASM-OSCs (Fig. 1). The dual-additive strategy, combined a SA and a solvent additive through their synergistic effects to achieve optimal ASM-OSC active layer morphology. A representative case is DIB and DIM engineered by Lu’s group, which achieved a solvent annealing-free PCE of 15.2% in BTR-Cl:Y6-based ASM-OSCs via a dual-additive strategy, where volatile solid additive (VSA) DIB induced a Y6-based eutectic phase to enhance intermolecular interactions and modulated acceptor phase separation, while solvent additive DIM volatilized to suppress BTR-Cl over-aggregation during film formation²⁶. Additionally, the large π-ring strategy, demonstrated by Gao’s group through combining graphene oxide methyl ester (GOMe) as a SA with a layer-by-layer deposition technique, enabled precise control vertical phase separation and crystallization kinetics in the active layer, achieving a PCE of 17.18% in ZnP-TSEH:4TIC:6TIC-based ASM-OSCs²⁷. Moreover, the polymer additive strategy, implemented by Li’s group through incorporating the insulating polymer styrene-ethylene-butene-styrene (SEBS) as a morphology stabilizer, achieved enhanced stability in BM-ClEH:BO-4Cl-based ASM-OSCs. By suppressing donor molecules degradation and mitigating crystallization/aggregation in the blend film, SEBS significantly enhanced the operational stability under continuous illumination (T80 > 15,000 h)²⁸. Furthermore, VSA strategy, developed by Sun’s group using 1,1-diaminomethyl-3-indolone (IC-FI), structurally analogous to the terminal group of the N3 acceptor, achieved a PCE of 14.43% in BTR-Cl:N3-based ASM-OSCs. This enhancement resulted from strengthened π-π interactions among N3 molecules and dense molecular self-assembly induced by IC-FI volatilization²⁹. What’s more, ternary strategy, demonstrated by Min’s group, achieving the highest PCE of 18.1% reported for ternary MPhS-C2:L8-BO:L8-S9-based ASM-OSCs though co-utilization with the VSA DIB⁸. As demonstrated in the examples above, SAs show significant potential in regulating crystallization and phase separation in ASM-OSCs. Nevertheless, most SAs have been optimized from those originally developed for P-OSCs, and the underlying interaction mechanisms between SAs and active layer components in ASM-OSCs remain poorly understood.
Four strategies—dual-additive, large π-ring, polymer, and volatile solid additives (VSAs)—are illustrated the reported PCE of all-small-molecule organic solar cells (ASM-OSCs).
In this study, we introduce 3-allylrhodanine (3-AR) as a VSAs. While its structure resembles the terminal group of the MPhS‑C2 donor, its functional groups (carbonyl, -C=S, and -C-H, etc.) can form selective interactions with the donor (e.g., Dithienobenzodithiophene (DTBDT) core units) and acceptor (e.g., 2-(5,6-difluoro-3-oxo-2,3-dihydro-1H-inden-1-ylidene)ma-lononitrile (2FIC) units and 6,7-dihydropyrrolo[3,2-g][1,2,5]thiadiazolo[3,4-e]indole core units)^30,31, thereby potentially regulating the morphology and improving photovoltaic performance of ASM-OSCs. Theoretical calculations result clearly indicates that, thermodynamically, 3-AR has a greater tendency to interact with N3 molecules. Combined thin-film structural characterization, we found that the thermodynamic differences in intermolecular interactions, which directly dictate the evolution of the final morphology. Moreover, the 3-AR VSA narrows the film formation window, thereby reducing film formation time. It suppresses excessive crystallization of the blend film, promoting the formation of smaller, well-suited phase separation domains laterally, while simultaneously enhancing vertical donor/acceptor homogeneity within the active layer. Thus, the 3-AR VSA enhances charge dissociation and collection kinetics, mitigates bimolecular and trap-assisted recombination, and prolongs exciton and charge lifetimes. Consequently, the 3-AR-optimized ASM-OSCs attained a superior PCE of 18.43%, with an improved J_sc of 27.32 mA/cm² and an unprecedented FF of 78.19%, representing a 0.97-percentage-point gain in PCE over the VSA-free control one (17.46%). This work not only establishes a new efficiency benchmark for binary ASM-OSCs, achieving a third-party certified PCE of 18.16% among single-junction devices to date, but also advances the development of high-performance ASM-OSCs.
The chemical structures of small molecule donor MPhS-C2¹⁰, acceptor N3³², and 3-AR VSA used in this study are shown in Fig. 2a. To evaluate the effect of the 3-AR additive on the photovoltaic performance of the devices, ASM-OSCs with a conventional structure of indium tin oxides (ITO)/(2-(9H-Carbazol-9-yl) ethyl) phosphonic acid (2PACz)/active layer/H75/Ag were fabricated, where the recently developed H75 serves as the electron transport layer³³. The chemical structures of 2PACz and H75 are shown in Supplementary Fig. 1. The processing parameters, including donor-to-acceptor ratio, solution concentration, spin-coating speed, annealing temperature, and additive concentration, were systematically optimized through iterative screening, as detailed in Supplementary Figs. 2–6 and Supplementary Tables 1–5. The optimal device performance is at 2.5 wt.% 3-AR (weight ratios relative to MPhS-C2:N3), with PCEs progressively declining at higher concentrations. Table 1 summarizes the key performance metrics of the control and optimized devices and Fig. 2b displays their current density−voltage (J–V) characteristics. The control device based on MPhS-C2:N3 displayed a PCE of 17.46%, with an open-circuit voltage (V_oc) of 0.863 V, a J_sc of 27.03 mA/cm², and an FF of 74.85%. With 2.5 wt% 3-AR VSA, the device attained a champion PCE of 18.43%, propelled by a notable enhancement in FF of 78.19%, accompanied by a modest increase in J_sc of 27.32 mA/cm², while maintaining a V_oc of 0.863 V. Furthermore, the 3-AR-processed devices garnered independently certified PCEs of 18.16% from the National Center of Inspection on Solar Photovoltaic Products Quality (CPVT), with comprehensive certification reports provided in Supplementary Fig. 7. Crucially, comprehensive statistical analysis of state-of-the-art ASM-OSCs with PCEs exceeding 15% reveals that our 3-AR-processed device exhibits the high J_sc and exceptional FF, synergistically achieving a remarkable PCE of 18.43%, which is the highest value for binary ASM-OSCs reported to date. The comparative data are presented in Fig. 2c and Supplementary Table 6. Supplementary Fig. 8 compares the PCE distributions of 50 devices with/without 3-AR, revealing improved reproducibility and reduced performance variation upon 3-AR treatment⁸. These results confirm that trace amounts of the 3-AR enhance the J_sc and FF, thereby boosting overall photovoltaic performance of the device. Additionally, 3-AR not only effectively enhances the storage, photo-, and thermal stability of ASM-OSCs but also demonstrates excellent universality by being applicable to both representative all-small-molecule (e.g., B1:L8-BO and BTR-Cl:N3) and polymer systems (e.g., D18-Cl:N3 and PM6:L8-BO), as shown in Supplementary Figs. 9–11 and Supplementary Tables 7–13.
a Chemical structure of MPhS-C2, N3, and 3-AR. b J–V curves of the devices. c Plots of the PCE versus J_sc for the efficient ASM-OSCs reported in the literature. d External quantum efficiency (EQE) spectra of the devices. e The binding energies of MPhS-C2@3-AR and N3@3-AR. Source data are provided as a Source Data file.
The external quantum efficiency (EQE) curves of MPhS-C2:N3-based devices processed with and without 3-AR additives are shown in Fig. 2d, illustrating their wavelength-dependent photon-to-electron conversion efficiency across the 300–1000 nm range. The 3-AR-processed device exhibits higher EQE across the 510–830 nm compared to the control device, achieving a peak value of 88% (vs. 86% for the control device). The integrated J_sc values calculated from control and 3-AR-processed devices are 26.01 and 26.28 mA/cm², respectively, closely matching the J_scs derived from J–V measurements³⁴. The internal quantum efficiency (IQE) spectra were derived from reflection (R) and EQE measurements (Supplementary Fig. 12). Compared to the control device, the 3-AR treated device exhibits significantly improved IQE values across the range of 510–830 nm, which contributes to achieving a higher J_sc. This improvement in photo-electron conversion efficiency can be attributed to enhanced exciton dissociation and more efficient charge transport within the 3-AR-processed active layer³⁵.
As shown in Supplementary Fig. 13, the thermogravimetric analysis (TGA) curve of 3-AR solid powder, heated from 25 to 300 °C at a scanning rate of 5 °C/min, shows a weight loss of 5% at 150 °C and a complete weight loss of 100% at about 225 °C. As seen from Supplementary Fig. 14, 3-AR exhibits limited standalone film formation with a non-uniform coverage on the quartz substrate, and thermal annealing at 100 °C induces sequential phase transitions: liquefaction within 1 min, partial evaporation by 6 min, and complete elimination after 10 min, confirming its superior volatility. Fourier transform infrared spectroscopy (FTIR) spectra reveals the 3-AR’s characteristic peak at 1734 cm⁻¹ in the MPhS-C2:N3 blend film, which vanishes after 100 °C thermal annealing, further confirming complete volatilization of 3-AR (Supplementary Fig. 15). Given the high volatility of 3-AR, its complete absence was further verified by nuclear magnetic resonance (NMR) spectroscopy (Supplementary Fig. 16). Comparative analysis of the films before and after annealing showed that the characteristic signals of the 3-AR disappeared entirely after the thermal annealing treatment. In short, the residue-free volatilization of 3-AR is critical for achieving stable and additive-free active layers in OSCs³⁶.
We calculated the molecular electrostatic potential (ESP) distribution and dipole moments of MPhS-C2, N3, and 3-AR by using density functional theory (DFT) to investigate the intermolecular interactions between 3-AR and photoactive materials. As illustrated in Supplementary Fig. 17, MPhS-C2 and N3 exhibited distinct ESP characteristics. The MPhS-C2 showed a more negative ESP distribution, with fewer positive charges on the molecular surface. In contrast, the N3 predominantly displayed a positive ESP distribution, with localized negative regions at its carbonyl, cyano, and fluorine substituents. For the 3-AR, the strong electronegativity of the sulfinyl and carbonyl groups results in a weak negative ESP distribution across its main molecular surface, while positive charge regions were localized on the hydrogen and propylene groups of the thiazolidine ring. Based on Coulomb’s law, 3-AR exhibits preferential interactions with the N3. To further elucidate the interaction mechanism, complexes of 3-AR with either the end groups or core units of MPhS-C2 and N3 were constructed, and their respective binding energies (ΔE_b) were calculated, as illustrated in Fig. 2e. The ΔE_b are calculated to be −18.44 (C–H···S and C=O···H) and −18.99 (C–H···H) kcal/mol for the 3-AR with the end groups of MPhS-C2 and N3, respectively, as determined by the Eq. (1) (see details in Methods). 3-AR is found to have more negative ΔE_b values of −19.61 (C=S···H) and −21.07 (C–N···H) kcal/mol when interact with the core units of MPhS-C2 and N3 molecules, respectively. Generally, the more negative ΔE_b values indicate that the stronger intermolecular interactions. The results suggest that 3-AR preferentially tends to stack on the core units of MPhS-C2 and N3 molecules rather than on their end groups.
Two-dimensional ¹H-¹H nuclear overhauser effect spectroscopy (NOESY) can determine spatial proximity between protons within 5 Å and was thus used to probe intermolecular interactions between additives and photoactive materials³⁷. Although the NOESY spectrum of 3-AR with MPhS-C2 (Supplementary Fig. 18a) showed no crossover resonance peaks, the possibility of other types of interactions could not be ruled out. Therefore, NMR titration was performed (Supplementary Fig. 19a), which showed chemical shift changes (δ = 7.8) in the ¹H NMR spectra with increasing proportion of 3-AR, confirming an interaction between MPhS-C2 and 3-AR. On the other hand, the NOESY spectrum of 3-AR with N3 solution (Supplementary Fig. 18b) displayed weak but discernible cross resonance peaks (marked by the green box) between the terminal proton of N3 (Ha, δ = 8.40) and a proton of 3-AR (Hb, δ = 4.0), indicating the spatial proximity and interaction between N3 and 3-AR³⁸. This result was further supported by NMR titration experiments (Supplementary Fig. 19b). Collectively, these NMR analyses and DFT calculations demonstrate that 3-AR participates in distinct intermolecular interactions with MPhS-C2 and N3 through different functional groups, thereby effectively modulating the morphology of the active layer.
In-situ ultraviolet-visible (UV–Vis) absorption spectroscopy was used to track morphology evolution kinetics during film formation, revealing how 3-AR modulates crystallization via solvent evaporation control. The time-resolved color mapping of UV–Vis spectra for MPhS-C2, N3 and MPhS-C2:N3 films prepared without and with 3-AR are displayed in Fig. 3a and Supplementary Fig. 20a, b, while the corresponding spectra are shown in Supplementary Fig. 21. The film formation process evolves through three distinct stages: (I) solvent evaporation, (II) nucleation and crystallization (phase transition process), and (III) film solidification³⁹. The duration of stage II (Δt) plays a critical role in molecular self-assembly kinetics. Typically, prolonged Δt facilitates molecular rearrangement for ordered packing, whereas shortened Δt restricts intermolecular interactions, thereby suppressing molecular crystallization^40,41. For neat films, Supplementary Fig. 20c reveals that 3-AR modulates the film formation kinetics of MPhS-C2 and N3 divergently during stage II: the MPhS-C2’s processing window shortens from 0.20 to 0.10 s, likely suppressing excessive crystallization of MPhS-C2, while the N3’s duration extends from 0.097 to 0.191 s, facilitating ordered N3 molecular packing, as confirmed by grazing-incidence wide-angle X-ray scattering (GIWAXS) analysis. Analysis of MPhS-C2: N3 blend film formation kinetic reveals three key insights (Fig. 3b): (1) MPhS-C2 preferentially nucleation and crystallization from solution (t = 0.39 s), then N3 starts to crystallization, indicating that the crystallization of MPhS-C2 is the driving force for the phase separation; (2) 3-AR additives extend the temporal gap between MPhS-C2 finishing crystallization and N3 starting nucleation (Δt from 0.38 to 0.48 s). This extended temporal gap may improve donor–acceptor miscibility and suppress the early nucleation of N3⁴²; (3) 3-AR additives shorten stage II processing time (MPhS-C2: Δτ from 0.097 to 0.095 s; N3: Δτ from 0.190 to 0.097 s), effectively inhibiting excessive MPhS-C2:N3 blend crystallization, which is helpful to promote an optimal nanoscale phase morphology (as evidenced by GIWAXS). 3-AR additives moderately reduce the active layer film formation time, effectively suppressing excessive donor/acceptor crystallization through kinetic modulation, which facilitates an optimal and small nanoscale phase separation morphology.
a 2D time-resolved color mapping of in-situ UV–Vis absorption. b The maximum peak variations of MPhS-C2:N3 during spin coating. c 2D grazing-incidence wide-angle X-ray scattering (GIWAXS) diffraction patterns of the blend films. d The corresponding line-cut profiles of the blend films. e Atomic force microscopy (AFM) height images of the blend films. f Transmission electron microscope (TEM) images of the blend films. Source data are provided as a Source Data file.
The surface free energy (γ) of the active layer components treated without and with 3-AR additives was investigated by measuring the contact angle of the films with deionized water and formamide. The contact angle diagram and calculation results are shown in Supplementary Fig. 22 and Supplementary Tables 14–15. The γ values of MPhS-C2, MPhS-C2:3-AR, N3, and N3:3-AR films are estimated to be 18.57, 30.45, 18.04, and 27.78 mJ/m², respectively. According to literature reports, the γ of 2PACz is approximately 48 mJ/m² ⁴³. The increased surface free energies after 3-AR treatment indicate improved wettability and reduced interfacial energy mismatch with the 2PACz hole transport layer, facilitating formation of intimate Ohmic contact during spin coating and lowering interfacial resistance. This enhanced interfacial compatibility also contributes to multidimensional morphology optimization. The synergistic effects arising from enhancements the lateral nanostructure and vertical phase uniformity will be analyzed in the following section^44,45.
GIWAXS was employed to probe the impact of 3-AR on film morphology and crystallinity. For neat MPhS-C2 films, the in-plane (IP) (010) peak at q = 1.799 Å⁻¹ and out-of-plane (OOP) (100/200/300) peaks at q = 0.249/0.526/0.793 Å⁻¹ indicate edge-on molecular packing⁴⁶. The 3-AR treatment preserves this orientation but reduces crystalline coherence length (CCL), calculated from the full width at half-maximum (FWHM) of these peaks, indicating suppressed MPhS-C2 crystallization, as detailed in Supplementary Fig. 23 and Supplementary Table 16. Neat N3 films exhibit face-on orientation with IP (100) lamellar peak at q = 0.278 Å⁻¹ and OOP (010) π−π stacking at q = 1.763 Å⁻¹. The 3-AR treatment reduces the lamellar CCL from 9.32 to 9.14 nm but increases the π−π stacking CCL from 3.37 to 3.46 nm, suggesting enhanced molecular packing orderliness of N3, as detailed in Supplementary Fig. 24 and Supplementary Table 17. The MPhS-C2 transitions from edge-on to face-on molecular orientation when blended with N3, as indicated by the overlapping (010) OOP peak (d-spacing ≈ 3.45–3.60 Å), which arises from co-planar π–π stacking of MPhS-C2 and N3 components, as detailed in Fig. 3c and d and Supplementary Table 18⁴⁷. In the MPhS-C2:N3 blend film, 3-AR treatment reduces the CCL values in both orientations. Specifically, the CCLs for the (010) OOP and (100) IP peaks decrease from 4.30 to 4.06 nm and from 15.43 to 15.10 nm, respectively. Although 3-AR treatment suppresses crystallization of the donor MPhS-C2 while promoting that of the N3 acceptor, in-situ absorption measurements reveal that the MPhS-C2 donor precipitates first from the solution. This suggests that the early-precipitating MPhS-C2, with inhibited crystallization, drives phase separation while simultaneously constraining the crystallization of the acceptor N3. This effect ultimately results in overall weakened crystallization within the blend film. The 3-AR additive concurrently suppresses the crystallization of both MPhS-C2 and N3 components, acting as a dual crystallization inhibitor—a rare characteristic among OSC additives. This unconventional mechanism facilitates optimized nanoscale phase separation.
Atomic force microscopy (AFM) and transmission electron microscopy (TEM) were employed to characterize the surface morphology and nanoscale phase separation of the active layer with and without 3-AR. The AFM height image of neat MPhS-C2 film shows a root-mean-square (RMS) surface roughness of 3.61 nm (Supplementary Fig. 25a), whereas the 3-AR-processed MPhS-C2 film exhibits slightly reduced roughness (RMS = 3.37 nm), suggesting effective suppression of excessive donor crystallization. This observation well-aligns with the reduced crystallinity evidenced by GIWAXS. However, the 3-AR-processed N3 film exhibits a slight increase in surface roughness (RMS = 1.93 nm vs. 1.63 nm), likely due to increased CCL observed in GIWAXS analysis, as depicted in Supplementary Fig. 25b. The 3-AR-processed MPhS-C2:N3 blend film exhibited a reduced RMS value of 1.80 nm compared to 1.90 nm for the untreated film, indicating suppressed crystallinity and improved surface homogeneity through additive-mediated crystallization control, as shown in Fig. 3e. In addition, AFM phase images exhibit that 3-AR mitigates MPhS-C2 over-aggregation, resulting in optimally phase-separation domains with ordered and uniform nanoscale protrusions in the blend film, as shown in Supplementary Fig. 26. Further TEM observations of the MPhS-C2:N3 blend film, shown in Fig. 3f, reveal that the 3-AR-treated film exhibits significantly reduced nanoscale phase separation, featuring a distinct whisker-like network uniformly interpenetrating smaller domains compared to the untreated counterpart. The incorporation of the 3-AR effectively modulates the film’s nano-structure morphology by suppressing excessive phase separation, thereby facilitating efficient exciton dissociation and enabling superior charge transport properties.
Film-depth-dependent light absorption spectroscopy (FLAS) was employed to probe 3-AR’s modulation effects on vertical phase separation, with mapping of component distribution gradients, exciton population profiles, and spatially exciton generation rates. The absorption spectra of MPhS-C2:N3 blend films at varying film depths are shown in Fig. 4a, revealing three critical observations: (1) The donor and acceptor exhibit characteristic absorption peaks at 570/623 nm for MPhS-C2 and 830 nm for N3, consistent with their respective single-component spectra (Supplementary Fig. 27). (2) The absorption peaks are sharper near the film surface, which is attributed to rapid surface crystallization as the chloroform solvent evaporates first at the interface. In contrast, slower solvent evaporation in the bulk promotes intermixing of MPhS-C2 and N3, suppressing crystallization in the bottom layers. (3) The reduced 0-0/0-1 vibronic peak intensity ratio correlates with decreased MPhS-C2 molecular crystallization under 3-AR-processed blend film compared to control film, which is consistent with GIWAXS results.
a The film-depth-dependent light absorption spectra of the blend films. b Component distribution gradients across film depths of the blend films. c Exciton population profiles of the blend films. d Spatially exciton generation rates of the blend films.
The component distribution gradients across the film depths exhibited a flattened characteristic in the 3-AR-processed blend film, indicating improved vertical homogeneity of component distribution within the active layer, as shown in Fig. 4b. Furthermore, the exciton population profiles and spatially exciton generation rates, derived from FLAS data simulations, as shown in Fig. 4c, d. In the control blend film, the maximum exciton generation rate (G_max) is localized within the 40−60 nm region. This localization increases the charge transport distance and raises the risk of recombination. In contrast, 3-AR-processed blend film shows a uniform exciton generation distribution, concentrated within the top 20 nm and bottom 40 nm regions. This optimized spatial distribution enables dissociated charges to quickly migrate to the electrodes, thereby enhancing charge transport and collection efficiency, along with a notable improvement in J_sc and FF²³.
To elucidate the 3-AR’s modulation effects on charge transport characteristics in MPhS-C2:N3-based devices, the dark J–V characteristics of unipolar devices (electron-transport: ITO/ZnO/Active layer/H75/Ag; hole-transport: ITO/2PACz/Active layer/MoO₃/Ag) were analyzed via space-charge-limited current (SCLC) method. As shown in Supplementary Fig. 28 and Supplementary Table 19, thelectron mobility (μ_e = 1.28 × 10⁻⁴ cm² V⁻¹s⁻¹) and hole mobility (μ_h = 0.93 × 10⁻⁴ cm² V⁻¹ s⁻¹) are improved in 3-AR-processed devices compared to the control devices (μ_e = 1.03 × 10⁻⁴, μ_h = 0.69 × 10⁻⁴ cm² V⁻¹ s⁻¹), due to synergistic morphology optimization. Besides, the more balanced μ_e/μ_h ratio of 1.37 (vs. 1.50 in control devices) indicates effective suppression of space-charge accumulation, which in turn can enhance FF value^2,7.
We further analyze the photocurrent density (J_ph) and effective voltage (V_eff) to reveal enhanced exciton dissociation (P_diss: 97.48 to 98.06%) and charge collection efficiency (P_coll: 86.94 to 88.48%) in 3-AR additives-treated devices, demonstrating optimized carrier dynamics for improved J_sc, as illustrated in Supplementary Fig. 29^4,15. Charge recombination analysis reveals suppressed bimolecular (α value from 0.975 to 0.996) and trap-assisted recombination (n value from 1.07 to 1.04) in 3-AR additives-treated devices via light-intensity-dependent J_sc/V_oc measurements, as shown in Supplementary Fig. 30, demonstrating enhanced charge transport^19,48. Furthermore, we quantified the trap density by deep-level transient spectroscopy (DLTS). As shown in Supplementary Fig. 31, comparing with control device, the trap density of the 3-AR treated device decreased from 3.14 × 10¹⁷ to 3.07 × 10¹⁷cm⁻³. The reduced trap density indicates suppressed trap-assisted recombination, which is beneficial for improving FF and PCE of the device⁴⁹. In addition, steady-state photoluminescence (PL) spectroscopy (Supplementary Fig. 32) reveals that the 3-AR-treated MPhS-C2:N3 blend film achieves enhanced PL quenching efficiencies of 95.01%/97.86%, respectively (94.37%/97.17% for control film). This improvement promotes exciton dissociation and facilitates highly efficient bidirectional charge separation^50,51.
Transient photocurrent (TPC) and transient photovoltage (TPV) characterization techniques were employed to probe the 3-AR’s modulation effects on the charge extraction kinetics and recombination dynamics in the devices⁵². As shown in Supplementary Fig. 33a, TPC decay curves reveal accelerated charge extraction in 3-AR-processed device (τ = 0.24 μs) compared with control counterparts (τ = 0.32 μs), correlating with enhanced carrier collection efficiency. Besides, as shown in Supplementary Fig. 33b, TPV decay curves show prolonged charge carrier lifetimes (τ = 5.80 μs) in the 3-AR-processed device, compared with the control counterparts (τ = 3.47 μs). This enhanced carrier longevity significantly suppresses recombination losses. The results validate the 3-AR-processed device possesses multiscale morphology, ranging from nanoscale phase separation to microscale spatial distribution, which synergistically enhance charge transport characteristics, ultimately leading to higher J_sc and FF values.
The charge transfer dynamics within MPhS-C2:N3 blend films were systematically investigated using femtosecond transient absorption (fs-TA) spectroscopy. The 2D fs-TA spectra of the control and 3-AR-processed blend films are shown in Fig. 5a, while Fig. 5b displays representative transient absorption spectra at selected delay times. Upon selective excitation of the N3 acceptor at 825 nm, characteristic excited-state signatures of the acceptor were observed at early time delays (Supplementary Fig. 34), featuring ground state bleaching (GSB) signals at 710 nm and 850 nm, along with an excited state absorption (ESA) at 920 nm. As charge transfer progressed, the signals associated with the excited state of the acceptor decayed rapidly, while GSB signals attributed to the donor MPhS-C2 appear around 580 nm and 635 nm, indicating the occurrence of a hole transfer process from N3 to MPhS-C2. Concurrently, a new ESA signal emerged at 1010 nm, which can be attributed to the charge transfer state of N3. Comparative analysis of the fs-TA kinetics at 1010 nm revealed distinct trends between control and 3-AR-treated films. As illustrated in Fig. 5c, the kinetics at 1010 nm exhibited significantly slower decay in 3-AR-treated films, demonstrating effective suppression of charge recombination and extended charge carrier lifetime. To quantitatively deconvolute the temporal evolution of distinct species involved in the photoelectric conversion process, we performed global analysis based on a consecutive kinetic model, obtaining evolution-associated difference spectra (Fig. 5d, Supplementary Fig. 35 and Supplementary Table 20)⁵³. This analysis resolved three key time constants: τ₁ (ultrafast interfacial charge transfer), τ₂ (exciton diffusion-mediated charge transfer), and τ₃ (charge recombination). Relative to the control film, the 3-AR-treated blend exhibited a reduction in τ₁ from 1.8 to 1.5 ps and in τ₂ from 51.9 to 40.7 ps, alongside an increase in τ₃ from 4595 to 5335 ps. The shortened τ₁ and τ₂ constants indicate enhanced exciton diffusion to the donor-acceptor interface and more efficient dissociation into free charges, consistent with an optimized phase-separated morphology and improved interfacial mixing induced by 3-AR treatment. The prolonged τ₃ signifies an extended charge-separated state lifetime, reflecting reduced trap-assisted non-radiative recombination. Further investigation into the photoexcitation transfer kinetics⁵⁴ (Supplementary Fig. 36 and Supplementary Table 21) revealed that 3-AR treatment increased the exciton lifetime (τ_e) from 17.22 to 17.44 ps and the charge lifetime (τ_c) from 569.85 to 595.61 ps. The concurrent extension of both exciton and charge lifetimes contributes to reduced charge recombination losses, improved exciton dissociation efficiency, and enhanced charge carrier mobility⁵⁵. Collectively, these results underscore that 3-AR treatment effectively optimizes the blend film morphology, facilitating exciton dissociation and charge transport, which ultimately leads to the observed enhancements in J_sc and FF.
a 2D transient absorption spectra of the blend films. b Femtosecond-transient absorption (fs-TA) spectra of the blend films at selected time delays. c Kinetic traces of charge transfer state excited state absorption (ESA) probing at 1010 nm for the blend films. d The line graph of the global fitting results.
Combined DFT calculation, morphology analysis, and exciton and charge carrier dynamics, we established a work mechanism for 3-AR SA, as schematically illustrated in Fig. 6. During film formation, MPhS-C2 precipitated first from the solution. The incorporation of 3-AR selectively interacted with the core units of MPhS-C2 through intermolecular interactions, and shortened the Stage II duration, thus suppressed excessive crystallization of the MPhS-C2. The incorporation of 3-AR delays the precipitation of the acceptor N3 from the solution. Concurrently, the early-precipitating MPhS-C2—with its crystallization suppressed—drives phase separation while constraining the crystallization of N3. The increased surface free energies after 3-AR treatment with improved wettability with the 2PACz hole transport layer, so enhanced homogeneity spatial distribution. After thermal annealing, 3-AR can completely volatilize from the active layer, resulting in a smaller and suitable phase separation. The 3-AR additive-induced inhibition of photoactive’s crystallization and improved wettability with the 2PACz effectively modulates the hierarchical morphology in ASM-OSCs, as demonstrated by its enhanced photovoltaic performance, optimized morphology, and improved exciton/carrier dynamics. It should be noted that while 3-AR is structurally similar to the MPhS-C2 donor, its binding energy with the acceptor N3 is more negative (–21.07 kcal/mol) than that with MPhS-C2 (–19.61 kcal/mol). This indicates a thermodynamic preference for interacting with N3. The result differs from the commonly reported trend that structurally similar additives tend to interact with like components^29,56. Furthermore, GIWAXS results reveal that the addition of 3-AR significantly enhances the crystallinity and molecular ordering of N3, while the CCL of MPhS-C2 decreases, suggesting a moderate suppression of its crystallization. These observations corroborate, at the film morphology level, a coherent alignment between thermodynamic mechanisms and microstructural evolution. This finding profoundly reveals that the effectiveness of an additive molecule depends on its selective balance of interactions with various components within the complex blend system, rather than being solely governed by the simplistic principle of “structural similarity”.
Nucleation and crystallization process of the MPhS-C2:N3 blend tuned by volatile additive 3-AR.
In summary, by employing 3-AR VSA, we achieved optimized hierarchical morphology in MPhS-C2:N3-based ASM-OSCs. Interesting, theoretical calculations result clearly indicates that, thermodynamically, 3-AR has a greater tendency to interact with N3 molecules. The selective intermolecular interactions of 3-AR at the molecular level manifest as the differential crystallinity of the donor and acceptor at the macroscopic scale. Moreover, in situ absorption spectroscopy analysis demonstrates that the 3-AR additive accelerates the film formation process of both donor and acceptor components, effectively suppressing their excessive crystallization. The 3-AR treatment not only led to the optimized phase separation–slightly reduced CCLs and smaller domains, but also enhanced vertical homogeneity through the optimized hierarchical morphology, as evidenced by various characterizations (e.g., GIWAXS, TEM, and FLAS). These advantageous features synergistically improved charge mobility, boosted exciton dissociation efficiency, and prolonged exciton and charge lifetimes (e.g., SCLC, TPC, TPV, and fs-TA). As a result, the 3-AR-processed device yielded an outstanding PCE of 18.43%, with a certified value of 18.16%, which marks the highest reported PCE for binary ASM-OSCs. This study showcases hierarchical morphology control in high-efficiency ASM-OSCs and revealing that the role of VSAs in multi-component blend systems extends far beyond the simplistic principle of “structural similarity”. It establishes a design paradigm for VSA engineering, thereby collectively driving significant progress toward achieving higher PCEs in ASM-OSCs.
The small molecule donor MPhS-C2 was purchased from Solarmer Materials (Beijing) Inc. The acceptor N3 and hole interface layer (2-(9H-Carbazol-9-yl) ethyl) phosphonic acid (2PACz) were purchased from Derthon Materials, Inc. Solid additive 3-allylrhodanine (3-AR) was bought from Aladdin Scientific Corp. H75 was synthesized as previously reported (Adv. Funct. Mater. 33, 2303386 (2023)). Chloroform (CF) was bought from Sigma-Aldrich. All the materials were used as received without further and purification.
All devices were fabricated with a conventional structure (ITO/2PACz/Active layer/H75/Ag). Pre-patterned ITO-glass substrates were cleaned by sequential sonication in detergent, deionized water, acetone, and isopropanol for 15 min, then dried by Nitrogen gun, and subsequent ultraviolet-ozone treatment for 20 min. Then the ethyl alcohol solution of 2PACz (0.27 mg/mL) was spin-coated at 3000 rpm on top of the cleaned ITO substrates and annealed at 100 °C for 10 min. For the active layer, the weight ratio of donor and acceptor was kept at 1.4:1.0 (w/w), and the blend solution had a total concentration of 20 mg/mL in CF. After stirred at 50 °C for 1 h, different weight ratio of 3-AR was added to the blend solution, and the blend solution was spin-coated at around 2000 rpm for 30 s. The as-cast films were solvent (CF) annealed for 30 s and then thermally annealed at 100 °C for 10 min. The electron transporting layer of H75 was deposited on the top of the active layer from a 1.0 mg/ml methanol solution via spin-coating at 3000 rpm for 20 s. Finally, the Ag cathode ( ~ 100 nm) was thermally evaporated on top of the substrates under high vacuum (<2 × 10^–4 Pa).
The J–V curves of all devices were measured under AM 1.5 G solar illumination at 100 mW/cm² using a Keysight B2901BL Precision Source/Measure Unit and an SS-X50 solar simulator (Enli Technology Co., Ltd., Taiwan) with a standard Si solar cell as a calibration standard. The EQE curves were measured by QE-R solar cell spectral response measurement system (Enli Technology Co., Ltd., Taiwan) in ambient air.
All quantum chemical calculations are performed using Gaussian 16. The optimized geometries, ESP distribution, and dipole moment of molecules are calculated using the B3LYP density functional with the 6-311 G (d, p) basis set. For the active layer system, under the same density functional and basis set, and the solvent effect was additionally considered. The binding energies between 3-AR additives and MPhS-C2 donor (or N3 acceptor) the components were calculated by the following Eq. (1):
In-situ UV–vis absorption spectra were acquired by a spectrometer (DU-300, Shaanxi Puguang Weishi Co. Ltd.). The spectrometer consists of a tungsten halogen light source, detector and spin coating instrument. The detector collects the transmission spectra ranged from 200 to 1050 nm during coating. The light source and detector were turned on before spin-coating the film, so time zero is the point at which the detector collected the first solution absorption spectrum. Before time zero, there is only noise in the absorption spectrum (Adv. Mater. 36, 2313105 (2024), Adv. Mater. 36, 2313532 (2024)).
The FT-IR absorption spectra were measured on Nicolet iS 5 (Thermo Fisher Scientific) with wavenumber ranging from 400 to 4000 cm⁻¹.
The electron and hole mobility were measured by using the method of space-charge limited current (SCLC) for electron-only devices with the structure of ITO/ZnO/active layers/H75/Ag and hole-only devices with the structure of ITO/2PACz/active layers/MoO₃/Ag. The charge carrier mobility was determined by fitting the dark current to the model of a single carrier SCLC according to the Eq. (2) (Adv. Funct. Mater. 33, 2305450 (2023)):
where (J) is the current density, (L) is the film thickness of the active layer, (mu) is the charge carrier mobility, ({varepsilon }_{r}) is the relative dielectric constant of the transport medium, and ({varepsilon }_{0}) is the permittivity of free space. (V) is the internal voltage in the device and (V={V}_{{{mathrm{appl}}}}-{V}_{{{mathrm{br}}}}-{V}_{{{mathrm{bi}}}}), where ({V}_{{{mathrm{appl}}}}) is the applied voltage to the device, ({V}_{{{mathrm{br}}}}) is the voltage drop due to contact resistance and series resistance across the electrodes, and ({V}_{{{mathrm{bi}}}}) is the built-in voltage due to the relative work function difference of the two electrodes. The carrier mobilities can be calculated from the slope of the ({J}^{1/2} sim V) curves.
The GIWAXS measurement was carried out at the PLS-II 6D U-SAXS and 9 A beamline of the Pohang Accelerator Laboratory in Korea. The scattering signal was recorded using a 2-D CCD detector (Rayonix SX165). The X-ray light had an energy of 11.24 KeV. The incidence angle of X-rays was adjusted to 0.09–0.12° to maximize the signal-to-background ratio.
AFM measurements were performed by using a Scanning Probe Microscope Dimension iCon in tapping mode under atmosphere conditions at room temperature. All film samples were spin-cast on glass/ITO substrates.
TEM images were performed using a JEOL USA JEM-1400 transmission electron microscope. The specific image system for EM (CCD camera).
The FLAS was performed by a PDU-400 spectrometer from Shaanxi Puguangweishi Technology Co. Ltd. The FLAS is in-situ measured by a setup which is composed of a halogen light source, a quasivacuum chamber generating soft plasma (pressure below 30 Pa, the flow rate of oxygen is 4 L/S), and a CCD spectrometer, which are connected by optical fibers. The film surface was incrementally etched by the soft ion source, without damage to the materials underneath the surface, which was in situ monitored by a spectrometer. From the evolution of the spectra and the Beer–Lambert’s Law, film-depth-dependent absorption spectra were extracted. The composition distribution along the film-depth direction was obtained from the film-depth-dependent spectra by fitting the sub-layer absorption using the absorption of the pure components. The exciton generation contour is numerically simulated upon inputting sub-layer absorption spectra into a modified optical transfer-matrix approach (Adv. Funct. Mater. 35, 2415499 (2024). Rev. Sci. Instrum. 94, 023907(2023)).
Contact angle measurements were carried out by an OCA20 Contact-angle System (Dataphysics, Germany), using water and formamide by sessile drop analysis. The interfacial surface energy values of materials can be obtained to Young’s equation.
The fs-TA measurements were performed on an ultrafast fs-TA spectrometer (HARPIA, Light Conversion). The output of a femtosecond laser (repetition rate of 40 kHz, 1030 nm central wavelength, and pulse duration of ~120 fs, PHAROS, Light Conversion) was split into two parts. One went through optical parametric amplification (ORPHEUS twins, Light Conversion) and then became a pump to excite thin films. The other went through the delay stage and produced white light via sapphire acting as the probe light. The pump and probe were focused on the sample. The optical path difference between the pump light and the probe light, which was controlled by the motorized optical delay-line, was used to get the relative time delay between the probe light and the pump light. The collimated and concentrated probe beams are collected by a fiber-coupled multi-channel spectrometer with a CCD sensor. As the delay between the pump light and the detector light varies continuously, the signal strength of the detector light received through the sample to be measured changes (Chem 8, 3051 (2022). Adv. Mater. 36, 2406690 (2024)).
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
All the data generated in this study are provided in the Article file, Supplementary Information file, and Source Data file. Source data are provided in this paper. Source data are provided with this paper.
Chen, H. et al. Organic solar cells with 20.82% efficiency and high tolerance of active layer thickness through crystallization sequence manipulation. Nat. Mater. 24, 444–453 (2025).
Article  ADS  CAS  PubMed  Google Scholar 
Jiang, Y. et al. Non-fullerene acceptor with asymmetric structure and phenyl-substituted alkyl side chain for 20.2% efficiency organic solar cells. Nat. Energy 9, 975–986 (2024).
Article  ADS  CAS  Google Scholar 
Wang, L. et al. Donor–acceptor mutually diluted heterojunctions for layer-by-layer fabrication of high-performance organic solar cells. Nat. Energy 9, 208–218 (2024).
Article  ADS  CAS  Google Scholar 
Liu, K. et al. 19.7% efficiency binary organic solar cells achieved by selective core fluorination of nonfullerene electron acceptors. Joule 8, 835–851 (2024).
Article  CAS  Google Scholar 
Li, H. et al. Ethanol processable inorganic-organic hybrid hole transporting layers enabled 20.12 % efficiency organic solar cells. Angew. Chem. Int. Ed 64, e202416866 (2024).
Article  Google Scholar 
Zhu, L. et al. Achieving 20.8% organic solar cells via additive-assisted layer-by-layer fabrication with bulk p-i-n structure and improved optical management. Joule 8, 3153–3168 (2024).
Article  CAS  Google Scholar 
Wang, L. et al. Diluted ternary heterojunctions to suppress charge recombination for organic solar cells with 21% efficiency. Adv. Mater. 37, 2419923 (2025).
Article  CAS  Google Scholar 
Gao, Y. et al. All-small-molecule organic solar cells with 18.1% efficiency and enhanced stability enabled by improving light harvesting and nanoscale microstructure. Joule 7, 2845–2858 (2023).
Article  CAS  Google Scholar 
Gao, Y. et al. High-performance small molecule organic solar cells enabled by a symmetric-asymmetric alloy acceptor with a broad composition tolerance. Adv. Mater. 35, 2300531 (2023).
Article  CAS  Google Scholar 
Yang, X. et al. Efficient and stable all-small-molecule solar cells enabled by incorporating a designed giant molecule acceptor. Energy Environ. Sci. 17, 5962–5971 (2024).
Article  CAS  Google Scholar 
Zhang, L. et al. High miscibility compatible with ordered molecular packing enables an excellent efficiency of 16.2% in all-small-molecule organic solar cells. Adv. Mater. 34, 2106316 (2022).
Article  CAS  Google Scholar 
Jiang, Y. et al. 20.6% Efficiency organic solar cells enabled by incorporating a lower bandgap guest nonfullerene acceptor without open-circuit voltage loss. Adv. Mater. 37, 2500282 (2025).
Article  CAS  Google Scholar 
Chen, H. Y. et al. All-small-molecule organic solar cells with an ordered liquid crystalline donor. Joule 3, 3034–3047 (2019).
Article  CAS  Google Scholar 
Gao, W. et al. Intramolecular chloro–sulfur interaction and asymmetric side-chain isomerization to balance crystallinity and miscibility in all-small-molecule solar cells. Angew. Chem. Int. Ed. 61, e20225168 (2022).
Article  Google Scholar 
Nian, L. et al. Approaching 16% efficiency in all-small-molecule organic solar cells based on ternary strategy with a highly crystalline acceptor. Joule 4, 2223–2236 (2020).
Article  CAS  Google Scholar 
Ye, W. et al. Nonfullerene all-small-molecule organic solar cells: prospect and limitation. Sol. RRL 4, 2000258 (2020).
Article  CAS  Google Scholar 
Tang, H. et al. Delicate morphology control triggers 14.7% efficiency all-small-molecule organic solar cells. Adv. Energy Mater. 10, 2001076 (2020).
Article  CAS  Google Scholar 
Huang, W. et al. High-performance organic bulk-heterojunction solar cells based on multiple-donor or multiple-acceptor components. Adv. Mater. 30, 1705706 (2018).
Article  Google Scholar 
Zhong, L. et al. Solid additive delicately controls morphology formation and enables high-performance in organic solar cells. Adv. Funct. Mater. 33, 2305450 (2023).
Article  CAS  Google Scholar 
Zeng, R. et al. Achieving 19% efficiency in non-fused ring electron acceptor solar cells via solubility control of donor and acceptor crystallization. Nat. Energy 9, 1117–1128 (2024).
ADS  CAS  Google Scholar 
Qin, J. et al. Volatile solid additive-assisted sequential deposition enables 18.42% efficiency in organic solar cells. Adv. Sci. 9, 2105347 (2022).
Article  CAS  Google Scholar 
Wang, Y. et al. A novel upside-down thermal annealing method toward high-quality active layers enables organic solar cells with efficiency approaching 20%. Adv. Mater. 36, 2411957 (2024).
Article  CAS  Google Scholar 
Sun, X. et al. Efficiency boost in all-small-molecule organic solar cells: Insights from the re-ordering kinetics. Adv. Energy Mater. 14, 2302731 (2024).
Article  CAS  Google Scholar 
Fu, J. et al. Rational molecular and device design enables organic solar cells approaching 20% efficiency. Nat. Commun. 15, 6865 (2024).
ADS  Google Scholar 
Xu, Y. et al. Suppression of energy disorder by incorporating a small-molecule acceptor into binary all-polymer solar cells. Energy Environ. Sci. 16, 5863–5870 (2023).
Article  CAS  Google Scholar 
Liu, H. et al. Dual-additive-driven morphology optimization for solvent-annealing-free all-small-molecule organic solar cells. Adv. Funct. Mater. 33, 2303307 (2023).
Article  CAS  Google Scholar 
Sun, Y. et al. Rational control of sequential morphology evolution and vertical distribution toward 17.18% efficiency all-small-molecule organic solar cells. Joule 6, 2835–2848 (2022).
Article  CAS  Google Scholar 
Ma, R. et al. Insulator polymer matrix construction on all-small-molecule photoactive blend towards extrapolated 15000 hour T80 stable devices. Adv. Mater. 36, 2405005 (2024).
Article  CAS  Google Scholar 
Chen, S. et al. Molecular ordering and phase segregation induced by a volatile solid additive for highly efficient all-small-molecule organic solar cells. J. Mater. Chem. A 9, 2857–2863 (2021).
Article  ADS  CAS  Google Scholar 
Zhang, H. et al. Mitigating exciton recombination losses in organic solar cells by engineering nonfullerene molecular crystallization behavior. Adv. Energy Mater. 13, 2302063 (2023).
Article  CAS  Google Scholar 
Wei, Z. et al. High-efficiency, ultra-flexible organic solar cells enabled by chloroprene rubber as both a non-volatile solid additive and plasticizer. Joule 9, 101996 (2025).
Article  CAS  Google Scholar 
Jiang, K. et al. Alkyl chain tuning of small molecule acceptors for efficient organic solar cells. Joule 3, 3020–3033 (2019).
Article  CAS  Google Scholar 
Mai, T. L. H. et al. Over 18.2%-efficiency organic solar cells with exceptional device stability enabled by bay-area benzamide-functionalized perylene diimide interlayer. Adv. Funct. Mater. 33, 2303386 (2023).
Article  CAS  Google Scholar 
Song, X. et al. Frenkel and charge-transfer excitonic couplings strengthened by thiophene-type solvent enables binary organic solar cells with 19.8 % efficiency. Angew. Chem. Int. Ed. 63, e202411512 (2024).
Article  CAS  Google Scholar 
Yu, X. et al. Self-assembled molecules with asymmetric backbone for highly stable binary organic solar cells with 19.7 % efficiency. Angew. Chem. Int. Ed. 63, e202401518 (2024).
Article  CAS  Google Scholar 
Zhong, L. et al. Naphthalene as a thermal-annealing-free volatile solid additive in non-fullerene polymer solar cells with improved performance and reproducibility. Adv. Funct. Mater. 32, 2201080 (2022).
Article  CAS  Google Scholar 
Li, C. et al. Achieving record-efficiency organic solar cells upon tuning the conformation of solid additives. J. Am. Chem. Soc. 144, 14731–14739 (2022).
Article  ADS  CAS  PubMed  PubMed Central  Google Scholar 
Chen, H. et al. Heterogeneous nucleating agent for high-boiling-point nonhalogenated solvent-processed organic solar cells and modules. Adv. Mater. 36, 2402350 (2024).
Article  CAS  Google Scholar 
Bo, Z. et al. Boosting organic solar cells to over 18% efficiency through dipole-dipole interactions in fluorinated nonfused ring electron acceptors. Angew. Chem. Int. Ed. 63, e202412854 (2024).
Article  Google Scholar 
Wupur, A. et al. Unravel the distinctive roles of liquid and solid additives in blade-coated active layer for organic solar cell modules. Adv. Energy Mater. 16, 2403132 (2024).
Xie, J. et al. Synergistic regulation of crystallization kinetics of donor/acceptor by new volatile additives for high performance organic solar cells. Adv. Funct. Mater. 34, 2402281 (2024).
Article  CAS  Google Scholar 
Qiu, W. et al. Breaking 20% efficiency of all-polymer solar cells via benzo[1,2-d:4,5-d’]bisthiazole-based terpolymer donor strategy for fine morphology optimization. Adv. Funct. Mater. 35, 2503009 (2025).
Article  CAS  Google Scholar 
Mou, H. et al. Transient dipole strategy boosts highly oriented self-assembled monolayers for organic solar cells approaching 21% efficiency. J. Am. Chem. Soc. 147, 21241–21251 (2025).
Article  ADS  PubMed  Google Scholar 
Lv, J. et al. Additive-induced miscibility regulation and hierarchical morphology enable 17.5% binary organic solar cells. Energy Environ. Sci. 14, 3044–3052 (2021).
Article  CAS  Google Scholar 
He, Z. et al. Wide-bandgap polymers with a C(sp3)─F polyfluoride backbone enable high-efficient ternary organic solar cells. Adv. Funct. Mater. 34, 2315650 (2024).
Article  CAS  Google Scholar 
Liu, J. et al. Thermodynamic and kinetic insights for regulating molecular orientation in nonfullerene all-small-molecule solar cells. Battery Energy 1, 20220013 (2022).
Article  CAS  Google Scholar 
Zhang, L. et al. Donor end-capped alkyl chain length dependent non-radiative energy loss in all-small-molecule organic solar cells. Adv. Mater. 34, 2207020 (2022).
Article  CAS  Google Scholar 
Wang, S. et al. Achieving 20% efficiency in organic solar cells through conformationally locked solid additives. Adv. Energy Mater. 16, 2405205 (2025).
Bai, L. et al. Stepwise control of molecular packing orientation and order in non-fullerene acceptors using dual additives for high-efficiency organic solar cells. J. Energy Chem. 110, 962–972 (2025).
Article  CAS  Google Scholar 
Ran, X. et al. Cyanobenzene-modified quinoxaline-based acceptors with optimal excitonic behavior enable efficient organic solar cells. Adv. Mater. 37, 2504805 (2025).
Article  CAS  Google Scholar 
Dai, T. et al. Optimizing molecular crystallinity and suppressing electron-phonon coupling in completely non-fused ring electron acceptors for organic solar cells. Angew. Chem. Int. Ed. 63, e20240305 (2024).
Article  Google Scholar 
Zhang, G. et al. Volatile additive strategy triggering 17.48% efficient post-treatment-free organic solar cells. Sol. RRL 7, 2200994 (2023).
Article  ADS  CAS  Google Scholar 
Wang, Y.-T. et al. Integrated omnidirectional design of non-volatile solid additive enables binary organic solar cells with efficiency exceeding 19.5. Angew. Chem. Int. Ed. 64, e202417643 (2025).
Article  CAS  Google Scholar 
Liao, X. et al. Regulating crystallinity mismatch between donor and acceptor to improve exciton/charge transport in efficient organic solar cells. Angew. Chem. Int. Ed. 63, e202318595 (2024).
Article  CAS  Google Scholar 
Kong, L. et al. In situ removable additive assisted organic solar cells achieving efficiency over 19% and fill factor exceeding 81%. Adv. Energy Mater. 13, 2300763 (2023).
Article  CAS  Google Scholar 
Yu, R. et al. Design and application of volatilizable solid additives in non-fullerene organic solar cells. Nat. Commun. 9, 4645 (2018).
Article  ADS  PubMed  PubMed Central  Google Scholar 
Download references
This work was supported by the National Natural Science Foundation of China (no. 62304149 and no.52473318), the Zhejiang Provincial Natural Science Foundation of China (no. LY24E030008), and the Key Research Program of Chinese Academy of Sciences (E4226101).
Zhejiang Key Laboratory for Island Green Energy and New Materials, Taizhou University, Taizhou, China
Duoling Cao, Lian Zhong, Jintong Sun, Qianguang Yang, Weijie Ding & Shirong Lu
School of Materials Science and Engineering, Taizhou University, Taizhou, China
Duoling Cao, Lian Zhong, Jintong Sun, Qianguang Yang, Weijie Ding & Shirong Lu
Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing, China
Duoling Cao & Jing Li
Chongqing School, University of Chinese Academy of Sciences, Chongqing, China
Duoling Cao & Jing Li
Department of Energy Engineering, School of Energy and Chemical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, South Korea
Zhe Sun, Huyen Thi Le Mai & Changduk Yang
Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Organic Solids, Institute of Chemistry, Chinese Academy of Sciences, Beijing, China
Jinyuan Zhang
School of Physical Science and Technology, Guangxi University, Nanning, China
Jingjing Zhao & Zhipeng Kan
Graduate School of Carbon Neutrality, Ulsan National Institute of Science and Technology (UNIST), Ulsan, South Korea
Changduk Yang
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
Search author on:PubMed Google Scholar
D.C. and L.Z. conceived the research concept. D.C. fabricated the devices, conducted performance testing, and drafted the manuscript. L.Z. drafted and revised the manuscript. S.Z. measured and analyzed the GIWAXS. J.S. conducted in-situ absorption spectroscopy, UV–vis spectroscopy, and contact angle measurements. T.L.H.M. provided the H75 electron transport layer materials. J.Y.Z. conducted fs-TA Characterization. J.J.Z. and Z.K. contributed to conducting TPC, TPV, PL and DLTS measurements. Q.Y. and W.D. participated in manuscript critique. C.Y., J.L., and S.L. supervised the research design, experimental execution, and data interpretation.
Correspondence to Lian Zhong, Changduk Yang, Jing Li or Shirong Lu.
The authors declare no competing interests.
Nature Communications thanks Yanming Sun, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Reprints and permissions
Cao, D., Zhong, L., Sun, Z. et al. Allylrhodanine-processed all-small-molecule organic solar cell achieves an 18.43% efficiency breakthrough. Nat Commun 17, 2105 (2026). https://doi.org/10.1038/s41467-026-68924-0
Download citation
Received: 30 May 2025
Accepted: 21 January 2026
Published: 28 January 2026
Version of record: 02 March 2026
DOI: https://doi.org/10.1038/s41467-026-68924-0
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative
Advertisement
Nature Communications (Nat Commun)
ISSN 2041-1723 (online)
© 2026 Springer Nature Limited
Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

source