Introduction

Organic single crystals with molecularly ordered stacking exhibit superior multidimensional optoelectronic characteristics compared to their amorphous counterparts, demonstrating enhanced carriers’ mobility1,2, exceptional photo response characteristics3, and precise crystal structure determination and analysis4, exhibiting promising applications in micro/nano electronic devices5, and providing unique platforms for fundamental investigations of light-matter interactions6. The growth of organic single crystals primarily relies on liquid- and vapor-phase methods. The liquid-phase approach involves solvent evaporation and molecular supersaturation to induce crystallization. For example, Grzybowski and collaborators reported enhanced single crystal growth methods by polyelectrolyte solutions and shear flow, a process that occurs in a solvent7. The vapor-phase method involves high-temperature evaporation and deposition growth. For example, Thomas J. Kempa and collaborators reported the direct growth of high-quality metal-organic framework (MOF) single crystals via chemical vapor deposition (CVD)8. In fact, combined liquid- and vapor-phase methods have also been developed to synthesize 2D organic lateral heterojunction crystals9.

However, for certain complex organic molecules designed to integrate multiple functions, conventional growth methods (including liquid- and vapor-phase) face challenges due to their low decomposition temperatures and long side chains which hinder crystallization. The Y6, a star non-fullerene acceptor (NFA) in photovoltaics, is derived from the TPBT central unit (TPBT refers to 2, 1, 3-benzothiadiazole (BT)-core-based fused-unit dithie-nothiophen [3.2-b]-pyrrolobenzothiadiazole)10. The intricate molecular structure of Y6 is designed to integrate multiple functionalities, including a narrow optical gap, high absorption coefficient, and good solubility in common solvents. Inspired by these advantages, numerous Y6-like NFAs have been reported, exhibiting exceptional optoelectronic properties11. However, the optoelectronic potential of Y6 and Y6-like single crystals remains unexplored due to challenges in growing high-quality and large-scale single crystals. The conventional methods face two limitations: (i) vapor-phase growth is hindered by thermal decomposition at elevated temperatures12, and (ii) liquid-phase growth is impeded by steric interference from long side chains13.

To address this challenge, the cocrystal growth method—mostly used for pharmaceutical crystal engineering—has been employed to facilitate single crystal growth14. Cocrystals consist of two or more distinct molecular components, stabilized by supramolecular interactions such as hydrogen bonds and π–π stacking. The cocrystal growth method has emerged as a powerful strategy for growing high-quality and large-scale organic single crystals, particularly for structurally complex molecules such as Y6 and Y6-like NFAs. Key techniques such as solvent evaporation, melt crystallization, and mechanochemical synthesis are widely employed for cocrystallization. For instance, Dominik Cinčić and colleagues demonstrated the growth of cocrystals via halogen bonding to phosphorus, arsenic, and antimony, combining experimental observations with theoretical analysis to elucidate solid-state assembly mechanisms15. Xutang Tao and colleagues developed a micro-spacing in-air sublimation method for growing organic cocrystals, enabling precise morphology control and rapid crystal growth16,17.

These successful examples demonstrate that the cocrystal growth method is an effective strategy for obtaining organic single crystals with complex structures and multifunctional properties. Building on this approach, a strategy for growing Y6 and Y6-like NFAs was raised through engineered cocrystal design. The single crystal structure of Y6-additive cocrystals (Y6-Additive Cocrystals, named YACs) is determined by Micro Electron Diffraction Technology at first time, forming two distinct morphologies: (ⅰ) elongated strip-like crystals with lengths up to ~450 μm (the largest length reaches 1.5 mm) and (ⅱ) ultrathin sheet-like crystals with monolayer thicknesses of ~18 nm (tunable from 18 nm to 341 nm). This strategy demonstrates broad applicability, as evidenced by successful extension to an additional 10 Y6-like NFAs and 2 kinds of effective additives. Strikingly, the photodetector based on YACs demonstrates exceptional polarized and helical light response and realizing single-pixel imaging. The excellent second harmonic generation (SHG) responses appear on most of the YACs. These findings establish a novel pathway for growing organic single crystals of complex molecular architectures and promote the optoelectrical properties research.

Results and discussion

Cocrystal Growth Strategy for Non-fullerene Acceptors (NFAs)

Y6 ((2,2’-((2Z,2’Z)-((12,13-bis(2-ethylhexyl)−3,9-diundecyl-12,13-dihydro-[1,2,5]thiadiazolo[3,4-e]thieno[2”,3’‘:4’,5’]thieno[2’,3’:4,5]pyrrolo[3,2-g]thieno[2’,3’:4,5]thieno[3,2-b]indole-2,10-diyl)bis(methanylylidene))bis(5,6-difluoro-3-oxo-2,3-dihydro-1H-indene-2,1-diylidene))dimalononitrile)), a prominent NFA in organic photovoltaics, was selected as target molecule for cocrystal growth due to its optimal optical bandgap, low exciton binding energy, and high carriers’ mobility18 (Fig. 1a). Using 4,8-Bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5-b’]dithiophene as a structure-directing additive (Fig. 1b), Y6-based cocrystals were grown successfully to further investigate their optoelectronic properties. This approach differs fundamentally from conventional organic crystal growth methods, as the additive actively participates in crystal formation through π–π stacking interactions at a precise 1:1 molar ratio with Y6: additive, becoming a component of integral structure, and forming a new class of single crystals (Y6-Additive Cocrystals, named YACs).

Fig. 1: Schematic illustration of cocrystal growth strategy for non-fullerene acceptors (NFAs).
Fig. 1: Schematic illustration of cocrystal growth strategy for non-fullerene acceptors (NFAs).The alternative text for this image may have been generated using AI.
Full size image

The molecular structures of Y6 (a) and the additive (b). Conventional vapor- and liquid-phase growth methods prove ineffective for Y6 and Y6-like NFA due to their low decomposition temperatures and steric hindrance from long side chains. By employing an additive, a cocrystal is grown successfully through the establishment of novel configuration coupling interactions, enabling the growth of high-quality organic single crystals (c).

Conventional single-crystal growth of organic materials primarily relies on vapor-phase and liquid-phase methods. However, these approaches face fundamental limitations for NFA molecules used in photovoltaics, which typically exhibit low thermal decomposition temperatures and sterically hindered long side chains10. Vapor-phase growth is precluded by thermal instability, as exemplified by Y6 with a thermal decomposition temperature of only approximately 320 °C (Fig. S1), which is far too low for successful thermal evaporation. Meanwhile, liquid-phase growth is impeded by disordered molecular packing caused by side chain interference during solvent evaporation and supersaturation. Regardless of using low or high precursor concentrations, it is impossible to grow high-quality single crystals (Fig. S2). Although successful growth of single crystals for the non-fullerene acceptor molecule L8-BO has been reported in the literature19(Fig. S3), it has proven effective only for specific molecules. To overcome these challenges, a cocrystal growth strategy was developed by introducing a structure-directing additive. This approach establishes new configuration coupling interactions while simultaneously mitigating the hindrance effects of side chains on molecular ordering (Fig. 1c). Y6 belongs to the A-D-A’-D-A class of NFAs, featuring a fused-ring central core with alternating electron-donating (D) and electron-accepting (A/A’) units10. While the extended π-conjugated system enables multifunctional integration, several intrinsic molecular characteristics impede crystal growth: (ⅰ) bulky side chains that enhance solubility but disrupt ordered packing, and (ⅱ) limited thermal stability (decomposition temperature ~320 °C). These factors collectively preclude the formation of high-quality and large-scale single crystals via thermal evaporation methods. The extended alkyl side chains in Y6 introduce significant steric hindrance, disrupting the π–π stacking and long-range molecular ordering required for single-crystal growth via liquid-phase crystallization. The 4,8-bis[(2-ethylhexyl)oxy]benzo[1,2-b:4,5-b’]dithiophene was selected as additive for cocrystallization due to its dual functionality: (ⅰ) a planar benzo ring in site of core that facilitates π–π stacking interactions with Y6 molecule, and (ⅱ) suitable physical state such as oiliness to provide suitable condition of molecular self-assembly (Viscosity: 17 mPa·s at 80 °C, as shown in Fig. S4). Both Y6 and additive demonstrate excellent mutual solubility in chloroform (CF), forming a homogeneous precursor solution essential for controlled cocrystal growth. (See section of materials and methods). The growth results demonstrate that the additive is indispensable for cocrystal formation, as confirmed by control experiments (Fig. S5). Upon determining the crystal structure, it’s unexpectedly discovered that the additive actively participates in the crystal lattice, forming a 1:1 cocrystal (YACs) with Y6. This cocrystallization behavior fundamentally differs from conventional organic crystal growth methods. Single crystal structure analysis reveals that the additive serves as a bridging template, directing the ordered stacking of Y6 molecules through a newly formed configuration coupling of Y6-additive π–π stacking. Simultaneously, the additive’s participation creates expanded intermolecular spacing, which accommodates Y6’s long side chains while reducing their steric hindrance effects. Furthermore, the additive’s oily nature at both room temperature and 80 °C (Growth temperature) constitutes another critical factor, as it provides an optimal dynamic environment that enables Y6 molecules to freely migrate and self-assemble, ultimately facilitating YACs formation.

Structural and morphological characterizations of YACs

Polarized optical microscopy reveals well-defined striated YACs grown on SiO2/Si substrates, exhibiting chromatic variations due to anisotropic light scattering20 (Fig. 2a). The observed smooth surfaces and straight edges indicate high crystalline quality, with a measured thickness of 327 nm (Fig. 2b). Statistical analysis shows the length distribution centers at approximately 450 μm (Lorentzian fitting), while the maximum observed length extends to 1500 μm (Fig. 2c). Additionally, theoretical X-ray Diffraction (XRD) spectra derived from the YACs via analog computation match the experimental data, with the strong (100) diffraction peak indicating preferred orientation along the (100) plane and high crystallinity. (Fig. 2d). The high-density YACs distribution and regularly oriented growth radiating from a single nucleation point to multiple directions observed in polarized microscopy images demonstrate exceptionally high growth productivity (See sample movie 1in Supplementary Materials). The morphology and shape diversity of YACs can be precisely controlled by adjusting the Y6-to-additive ratio and precursor concentration. The primary morphological evolution follows two main trends: (ⅰ) Increasing the additive proportion transforms the shape of YACs from sheet-like to strip-like forms, and (ⅱ) decreasing the precursor concentration reduces the overall thickness (Figs. S6–22). The YACs’ shapes predominantly manifest as either two-dimensional sheets or one-dimensional strips. The sheets exhibit nm-level thickness through layer-by-layer growth (18 nm to 341 nm), with each monolayer measuring approximately 2.0 nm in thickness.

Fig. 2: Structural and morphological characterizations of YACs.
Fig. 2: Structural and morphological characterizations of YACs.The alternative text for this image may have been generated using AI.
Full size image

a Polarized optical microscopy images of YACs with corresponding enlarged optical and AFM images (b). c Statistical length distribution of YACs with Lorentzian fitting. Inset: Schematic illustrating the length measurement methodology. d Comparison of experimental and simulated XRD patterns for YACs. e TEM image of YACs. f–i 3D reciprocal lattice of YACs. j, k Stacking mode of Y6 and the additive. l Unit cell structure of YACs and corresponding schematic diagram (m). n Multi-period crystal schematic diagram of YACs.

The YACs structure was determined using Micro Electron Diffraction Technology21,22,23 (Figs. S23–28, Tables S1-6). Data visualization was performed using the REDp program, and processing was conducted using DiffProAnalyze software (developed by ReadCrystal Tech Inc.). The single-crystal structure was determined, belonging to the P\(\bar{1}\) space group. The diffraction pattern is consistent with the extinction rules of P\(\bar{1}\), permitting all reflections without restrictions on h, k, or l indices. Any combination of h, k, and l (positive, negative, or zero) can give rise to a reflection, provided it is structurally feasible based on the crystal’s atomic arrangement (Fig. 2e–i). The analysis confirms that unit lattice parameters of YACs are a = 14.640(3) Å, b = 19.170(4) Å, c = 19.730(4) Å, and angles α (∠boc) = 75.00(3)°, β (∠aoc) = 70.00(3)°, γ (∠aob) = 87.00(3)° (CIF information of YACs in Supplementary Material). Y6 and the additive form π–π stacking between the terminal benzene ring attached to two fluorine atoms of Y6 and the centrally located benzene ring of the additive, and the distance of π–π stacking is 3.446 Å (Fig. 2j, k). The YACs possesses an inversion center as its symmetry element (Fig. 2l). The Y6 and the additive couple to form a basic unit, which then stacks in a back-to-back, step-like fashion (Fig. 2m, n).

Growth mechanism of YACs

Our observations reveal that stick YACs exhibit radial growth originating from a central nucleation point, with crystalline domains expanding uniformly in all directions. This growth behavior was characterized through in situ optical microscopy during thermal annealing at 80 °C, capturing the complete crystallization process from nucleation (t = 0 s) to crystal development (t = 1040 s) (Fig. 3a). The time-resolved imaging demonstrates distinct stages of nucleation initiation followed by directional crystal growth. (See growth movie 2in Supplementary Material). In situ imaging identifies molecular aggregation sites as nucleation centers, with both AFM and SEM images revealing subsequent crystal growth emanating from these nucleation sites, and growing layer by layer (Figs. 3b, S29). Other shapes, including lamellar and foliate YACs, also exhibit radial growth patterns, expanding outward from a single nucleation center (Fig. 3c) as 1:1 cocrystallization of Y6 and additive molecule growth (Fig. 3d). To elucidate the Y6-additive interaction mechanism, the additive’s influence was investigated systematically through both chemical structure and physical state analyses. A series of derivative additives containing isolated functional groups from the parent additive were designed for controlled experiments, enabling the precise determination of each functional group’s role. The primary additive was structurally deconvoluted into two key components: 3-methoxyheptane, benzo[1,2-b:4,5-b’]dithiophene, and further fragmented into its constituent basic units (benzene, thiophene, and thianaphthene). This systematic fragmentation enabled controlled investigation of individual functional group contributions through comparative experiments. Notably, the extended alkyl side chains were found to substantially enhance molecular solubility and facilitate solid-to-liquid phase transitions in small-molecule systems24. Control experiments reveal distinct growth outcomes: (ⅰ) unable to crystallize occurs with benzene, thiophene, and 3-methoxyheptane (all liquids, fast volatility), whereas (ⅱ) small crystal grains form with benzo[1,2-b:4,5-b′]dithiophene and thianaphthene (both solids) (Fig. 3e).

Fig. 3: Growth mechanism of YACs.
Fig. 3: Growth mechanism of YACs.The alternative text for this image may have been generated using AI.
Full size image

a Time-resolved optical microscopy images tracking YACs growth from 0 s to 1040 s. b The AFM image of YACs’ surface. c Radial growth pattern of stick YACs from a central nucleation point, with inset showing the 1:1 co-crystallization of Y6 and additive molecules (d). e Control experiments evaluating crystal growth with fragmented additive derivatives, demonstrating that only the original additive, benzo[1,2-b:4,5-b’]dithiophene and thianaphthene promote crystal growth. f A molecular dynamics simulation shows that π–π stacking appeared between the fluorinated terminal benzene ring of Y6 and the central benzene ring of the additive and hydrogen bonding at specific sites. g–i Molecular dynamics simulations demonstrate the self-assembly of Y6, additive, and chloroform solvent into cocrystal structures at 80 °C.

Further, comparative experiments were conducted using a combination of solid additive and oily alkane. Employing 1,8-diiodooctane as the oily alkane and benzo[1,2-b:4,5-b′]dithiophene as the solid additive yielded crystalline materials. The resulting crystals exhibited superior quality compared to those grown without additives or with only a solid additive, though they were inferior to those obtained with the oily additive in terms of crystal quality (Fig. S30). This outcome is attributed to the favorable growth environment provided by the oily alkane, which facilitates molecular stacking and self-assembly. However, the inhomogeneous distribution of the solid additive within the oily alkane, due to its tendency to self-aggregate, limits the continuous growth of high-quality cocrystals, resulting in lower effectiveness than molecularly defined oily additives. Chloroform is widely employed as a solvent for non-fullerene acceptor molecules due to its excellent solvation capability and high volatility, which promote homogeneous mixing with additives and facilitate rapid, complete solvent evaporation without disrupting cocrystal formation. Nevertheless, solvent selection exhibits limited impact on crystallization outcomes, as demonstrated by the successful growth of well-defined crystals using alternative solvents such as acetone and chlorobenzene (Fig. S31).

These results demonstrate that effective cocrystal growth requires effective packing sites and an optimal physical state. (Table S7) The YACs growth is influenced by the physical state of the parent additive and its functional groups critically. The viscous liquid phase (oiliness) of the parent additive creates an optimal environment for crystallization, as evidenced by the growth of YACs under additive-rich conditions (Fig. S10). In contrast, fast-volatility additives fail to support crystallization due to rapid evaporation, which prevents sufficient time for molecular self-assembly. Based on the structural analysis of YACs with molecular dynamics simulations, π–π stacking interactions are identified as the key determinant for growth. The computational results reveal that π–π stacking occurs between the terminal fluorinated benzene ring of Y6 and the central benzene ring of the additive and hydrogen bonding at specific sites25. The molecular dynamics simulation illustrates the three-dimensional structural evolution and dynamic conformational changes of the complex. The molecular components-Y6 (blue), additive (red), and chloroform solvent (cyan)-are depicted in stick or space-filling representations against a transparent background to enhance clarity. Sequential snapshots captured at 80 °C from a consistent viewing angle visually track the conformational progression during the simulation, with the final configuration aligning with the experimentally observed YACs structure. (Figs. 3f–i, S32–38).

Generalized growth approach of YACs

The versatility of this approach is demonstrated through successful YACs growth on diverse substrates, including quartz, glass, ITO-coated glass, polyimide (PI), and aluminum (Al) foil, indicating its potential for device integration and scalable fabrication (Figs. 4a, S39). The YACs morphology and growth behavior remain consistent across substrates, predominantly forming stick-like and sheet-like structures similar to those observed on SiO2/Si substrate, demonstrating excellent substrate compatibility. Notably, YACs also grow conformally along the inner walls of capillary tubes with diameters of 1.12 mm and 0.50 mm, maintaining their crystalline structure even in curved configurations (Fig. 4b). This behavior demonstrates two key advantages: (i) the ability to grow in confined micro-spaces, and (ii) inherent mechanical flexibility of YACs26 (Fig. S40). Aligned growth of YACs in parallel arrangements is achieved on Si substrates patterned with parallel nanograting, which directs precursor distribution via micro/nanostructure confinement (Figs. 4c, S41). In addition, the light-controlled growth of YACs has been realized. The experiment results reveal that YACs selectively grow in non-illuminated regions while remaining absent in light-exposed ( < about 900 nm, which precisely aligns with the edge of Y6 absorption spectrum) areas (Figs. S42, S43). The controlled growth of patterned arrays can be achieved using a light-masking, where a specific optical array is projected to define the growth pattern. In this process, the illuminated regions correspond to the non-growth zones.

Fig. 4: Generalized growth approach of YACs.
Fig. 4: Generalized growth approach of YACs.The alternative text for this image may have been generated using AI.
Full size image

a Universal substrate compatibility demonstrated on quartz, glass, ITO, polyimide (PI), and aluminum (Al) foil. b Confined growth within capillary tubes. c Directed assembly on Si substrates with parallel nanograting. General growth to other NFAs, including typical axially symmetric (A-D-A’-D-A) for COTIC-4F, Y6-BO, BTP-eC9, Y7 (d) and centrally symmetric (A-D-A) for ITIC, ITIC-M, ITIC-4F, ITIC-4Cl, ITIC-Th, IDIC (e). Co-crystal growth of Y6 molecules via novel additives A1 (f) and A2 (g).

To systematically evaluate the general applicability of additive-directed crystallization for NFAs, this growth method was successfully extended to multiple material systems, including: A-D-A’-D-A type of axial symmetry (COTIC-4F, Y6-BO, BTP-eC9, Y7) (Figs. 4d, S44) and A-D-A type of central symmetry (ITIC, ITIC-M, ITIC-4F,Y7 ITIC-4Cl, ITIC-Th, IDIC) (Figs. 4e, S45, and Table S8). These crystals exhibit varied morphologies (blocks, strips, and sheets), attributed to distinct molecular packing configurations and intermolecular coupling effects27,28. According to the above-mentioned principle for the selection of additives, 2,6-dibromo-4,8-bis[(2-butyl-n-octyl)oxy]benzo[1,2-b:4,5-b’]dithiophene (A1) and benzo[1,2-b:4,5-b’]dithiophene-4,8-dione (A2) are also selected as additives for YACs growth under identical conditions. The A1 exhibits a viscous liquid state at both room temperature and 80 °C, and A2 transforms from a solid at room temperature to a viscous liquid state at 80 °C. Both of them have suitable packing sites. The growth results indicate a high-quality single crystal of YACs (Figs. 4f, g, S46).

Optoelectronic Performance of YACs

The optoelectronic properties of Y6 and YACs were focused on absorption, photoluminescence (PL) spectra, and polarized optoelectronic response. Due to the localized nature of electronic density in organic molecules29,30,31, their optoelectronic characteristics at the molecular level remain largely unaffected after crystallization. However, a key consideration here is whether the introduction of the additive in YACs modifies the electronic distribution through charge transfer interactions between the two molecules32. The nearly identical absorption and PL spectra of Y6 and YACs suggest minimal changes between the amorphous and crystal states (Fig. S47a, b). As a hallmark property of organic crystalline, the polarized optoelectronic response has been extensively studied and holds significant promise for applications in multidimensional light detection technologies33. The angular-dependent PL spectra were measured under linearly polarized excitation at room temperature. The PL peak intensity exhibits a pronounced angular dependence, demonstrating clear dichroic behavior characteristic of anisotropic optical transitions (Fig S47c, d). The observed similarity in absorption between YACs and pristine Y6 (Fig. S47a), despite the disrupted Y6–Y6 stacking in the cocrystal. Spectral analysis reveals a broadened full width at half maximum (FWHM) in YACs, indicative of disrupted J-aggregation due to the formation of Y6–additive–Y6 alternating stacks (3.446 Å). Additionally, the pristine Y6 film under measured conditions is predominantly amorphous with minimal π–π stacking (Fig. S2a), resulting in an absorption profile that differs insignificantly from the well-ordered YACs crystal. The representative YACs (Y6) cocrystal exhibited a dielectric constant ranging from 2.10 to 2.34 (Fig. S48) and a charge carrier mobility of 0.5551 cm²/(V·s) (Fig. S49). Thermogravimetric analysis revealed that the decomposition temperature of the cocrystal is primarily determined by those of its individual components, Y6 and the additive (Fig. S50).

Organic nonlinear optical (NLO) materials offer advantages like low cost and ease of synthesis over inorganic counterparts, enabling applications in wavelength conversion and telecom34. Second harmonic generation (SHG), a key NLO process doubling incident light frequency, provides efficient frequency conversion and high-power laser generation35. Here, Second-order nonlinear optical responses in YACs were characterized using back-reflection SHG measurements36 (Fig. 5a). A strong emission peak appears at 515 nm from the YACs (Based on IT-M) sample when excited by a 1030 nm femtosecond laser (Fig. 5b). A plot of SHG intensity versus pump power yields a slope of 2.08 ± 0.15 on a log-log scale, confirming the characteristic quadratic dependence of high-quality SHG in YACs (Fig. 5c). The SHG response also exhibits polarization dependence at 1030 nm and 9.6 μW, attributed to the structural anisotropy of YACs (Fig. 5d). The polarization-dependent SHG response, characterized by four peaks (two strong and two weak) spaced 90° apart, indicates a multicomponent nature of SHG process from YACs. We analyzed that this may be related to the crystal structure of organic co-crystals, particularly single crystals composed of two molecules coupling37. Strict frequency doubling is observed across a series of fundamental wavelengths from 994 nm to 1030 nm, generating SHG signals from 497 nm to 515 nm. The SHG intensity decreases with blue-shifting fundamental wavelength, due to weakening by intrinsic optical absorption38 (Fig. 5e). SHG measurements conducted on all YACs at 1030 nm and 4.6 μW indicate that YACs exhibit standard SHG responses (Fig. 5f). While a non-centrosymmetric crystal structure is a necessary condition for SHG, we note that YACs based on Y6—which belongs to the centrosymmetric space group P\(\bar{1}\) with centrosymmetric properties—also exhibit strong SHG. According to previous studies, the local disorder in molecular orientation caused by long side chains39, along with charge distortion and local dipole chains induced by charge transfer40, can break the centrosymmetric properties of YACs, thereby enabling SHG. Based on comparative analysis with reported literature (Table S9), the SHG performance of the present work demonstrates notable advantages, particularly in terms of laser damage threshold. The cocrystal strategy endows YAC single crystals with both a high laser damage threshold and a strong nonlinear optical response.

Fig. 5: SHG response of YACs.
Fig. 5: SHG response of YACs.The alternative text for this image may have been generated using AI.
Full size image

a SHG measurement geometry. b SHG intensity versus pump power. c Pump-power-dependent SHG intensity with linear fit (slope = 2.08 ± 0.15). d Polarization-dependent SHG showing high anisotropy. e Wavelength-dependent SHG at 8 nW pump power. f SHG responses of different YACs systems.

The photodetector based on YACs (Y6) was fabricated by direct crystal growth on Au interdigitated electrodes (Fig. S51), exhibiting efficient photoresponse. Angular-dependent photocurrent measurements revealed strong dichroism, indicating well-aligned molecular orientation (Fig. S52). The device also showed distinct optical rotation under circularly polarized light and was successfully applied in a single-pixel imaging system, reconstructing a high-contrast lotus pattern (Fig. S53). To achieve near-infrared photo response, COTIC-4F was selected for device fabrication due to its longest absorption wavelength among all successfully grown crystals, enabling photodetection at 940 nm (Fig. S54). Given the broad application of non-fullerene acceptors in photovoltaics, molecular stacking critically influences charge transport. This proof-of-concept experiment validates the feasibility of YACs-based devices for future optoelectrical technologies.

In summary, we have developed a generalizable additive-directed cocrystal strategy to grow YACs, a previously unattainable feat due to steric hindrance from long side chains. Structural characterization confirms the formation of YACs in the triclinic P\(\bar{1}\) space group, where additive molecules enable cocrystallization through π–π stacking via configuration coupling. This approach provides unprecedented control over YACs morphology (stick/sheet) and dimensions (18 nm-341 nm thick; central and longest lengths to 450 μm and 1.5 mm). Through pattern summarization, this growth method can be extended to single-crystal growth of ten NFA molecules and the selection of two novel additives. The optoelectrical properties were focused on the SHG response. Most of the YACs exhibit a strong SHG response, including polarization dependence. Our work establishes a paradigm for single crystal growth for structurally complex functional NFAs molecules, unlocking their potential for advanced optoelectronic applications.

Methods

Materials

ITIC-TH, IDIC, Y7, Y6, IT-4CL, ITIC, Y6-BO, COTIC-4F, L8-BO, IT-M, BTP-EC9, IT-4F purchased from Shanghai Weizhu Chemical Technology Co., Ltd., purity 99%. 2,6-Dibromo-4,8-bis[(2-butyl-n-octyl)oxy]benzo[1,2-b:4,5-b’]dithiophene (A1) purchased from Shanghai Hao Hong Biological Pharmaceutical Technology Co., LTD., purity ≥97%. Benzo[1,2-b:4,5-b’]dithiophene-4,8-dione (A2) was purchased from Shanghai Maclin Biochemical Technology Co., LTD. Purity 98%., purity ≥99%. Trichloromethane (also known as chloroform, CF, CAS No. 67-66-3, formula: CHCl₃, Analyzable pure) purchased from Shanghai Lingfeng Chemical Reagent Co., LTD)

Growth methods

Solution preparation: Dissolve 4 mg of the target material (or 3 mg of Y6 in the light control growth test) in 100 μL chloroform and add different volumes of additives. The solution was dissolved by magnetic stirring at a speed of 1200 rpm. Spin coating films: Take 10 μL solution and add it to pre-treated SiO2/Si substrate (or quartz, glass, grating and other substrates), spin coating for 60 s at 4000 rpm to form a uniform film. Solvent annealing: The coated substrate is placed on a constant temperature table at 80 °C to promote solvent volatilization and crystal self-assembly. After growth is completely, rinse with ethanol to get clean YACs. Y6 light-controlled growth: After spin coating, the substrate was placed on the heated microscope stage (80 °C), and the light source of the microscope was used for local irradiation for 3 h, and the crystal morphology was observed immediately after the heating was turned off. Y6 multi-substrate adaptability test: Repeat the above steps on quartz, glass, grating and capillary surfaces (capillary tubes are sucked into the inner tube wall by capillary effect and heated directly). YACs growth based on different additives: Y6 and additives A1, A2 were dissolved in chloroform (3 mg/mL) at a mass ratio of 1:1, stirred at room temperature for 2 h and then spin-coated on SiO₂/Si substrate (4000 rpm, 60 s). After solvent annealing at 80 °C for 18 h, high-quality single crystals were formed and observed under the microscope with different magnifications.

Characterizations and tests

The PL spectra were carried on the Horiba LabRAM HR800. The absorption spectra were carried out on the Nicolet iS50. AFM images were carried on Bruker Dimension Ico, Optical and polarized images were carried on OLYMPUS microscope, SEM images were carried on HITACHI SU8010, XRD spectra were carried on D8 Advance. SHG measurements were performed using a custom reflection-mode microscope. A tunable femtosecond laser (pulse width: 200 fs; repetition rate: 100 kHz) served as the excitation source. Polarized excitation was achieved using a 1030 nm half-wave plate. Emitted signals were collected via a multimode fiber coupled to a spectrometer. All experiments were conducted at room temperature.