Revealing Pathway Complexity and Helical Inversion in Supramolecular Assemblies Through Solvent‐Induced Radical Disparities

Abstract New insights are raised to interpret pathway complexity in the supramolecular assembly of chiral triarylamine tris‐amide (TATA) monomer. In cosolvent systems, the monomer undergoes entirely different assembly processes depending on the chemical feature of the two solvents. Specifically, 1,2‐dichloroethane (DCE) and methylcyclohexane (MCH) cosolvent trigger the cooperative growth of monomers with M helical arrangement, and hierarchical thin nanobelts are further formed. But in DCE and hexane (HE) combination, a different pathway occurs where monomers go through isodesmic growth to generate twisted nanofibers with P helical arrangement. Moreover, the two distinct assemblies exhibit opposite excited‐state chirality. The driving force for both assemblies is the formation of intermolecular hydrogen bonds between amide moieties. However, the mechanistic investigation indicates that radical and neutral triarylamine species go through distinct assembly phases by changing solvent structures. The neutralization of radicals in MCH plays a critical role in pathway complexity, which significantly impacts the overall supramolecular assembly process, giving rise to inversed supramolecular helicity and distinct morphologies. This differentiation in pathways affected by radicals provides a new approach to manipulate chiral supramolecular assembly process by facile solvent–solute interactions.


Figure
Figure S14.CD and UV spectra of 1S assembly in DCE/MCH before and after adding TFA to dissociate hydrogen bonds.....

1 H
NMR spectra were recorded on a nuclear magnetic resonance instrument (300 MHz, Bruker, Germany) using DMSO-d6 and CDCl3 as the solvent and tetramethylsilane (TMS) as the internal standard at 25 °C.UV-vis and CD spectra were recorded on a JASCO J-1500 spectropolarimeter equipped with a Peltier-controlled unit using an SQ-grade cuvette, a single accumulation, a path length of 10 mm, a bandwidth of 2 nm, a scanning rate of 200 nm/min, and a response time of 2 s.The samples were measured at 25 °C unless otherwise declared.The UV-vis and CD data were analyzed using Spectra Manager program.TEM images were taken with a HITACHI HT7700 instrument operated at an accelerating voltage of 120 kV.AFM images were captured with peak force quantitative nanomechanical mapping scan mode on Multimode 8 microscope (Bruker, Germany), AFM samples were prepared by depositing droplets of the solution without dilution onto a silicon wafer until it was fully covered (about 10 μL), followed by slow evaporation under the poor solvent atmosphere.The thermal behaviors were measured by TA instrument DSC 250 (New Castle, DE, USA).The heating rate and cooling rate were both 10 °C/min.Fourier transform infrared (FT-IR) spectra were recorded on a NICOLET-6700 FT-IR spectrometer with Φ 20 mm × 4 mm potassium bromide tablets.The IR samples were prepared by adding 200 μL solution dropwise and then used Argon purging to evaporate the solvent of the samples.The ESR measurement were performed on a JES-X320 electron spin resonance spectrometer (JEOL, USA) with samples sealed in capillaries.All assemblies were prepared by adding poor solvent into the monomer solution solvated by good solvent DCE.We employed a standardized solution preparation method for all our samples.Monomer solution dissolved in good solvent was added into the cuvette, then poor solvent was slowly added along the walls of cuvette.Clear stratification between the two solvents could be observed at the bottom of the cuvette.
Figure S1.Characterization of the 1S.(a)Chemical structure and (b) 1 H NMR spectrum of 1S.

Figure S3 .
Figure S3.DSC measured curves of 1S.(a) DSC measured curves of 1S, the heating and cooling rates were both 10 K/min.(b) The order-to-disorder transition is analyzed using TRIOS program in the second heating curve, which is around 300 K (27 °C).

Figure S4 .
Figure S4.UV-vis spectra of 1S monomer in various single solvents with fixed 20 μM concentration.

Figure S6 .
Figure S6.The CD spectra of 1S assembly in (a) DCE/MCH and (b) DCE/HE with various water content in poor solvent.The poor solvents with different water content were prepared by mixing a water-saturated solvent with a drying solvent.The water content was estimated based on the recorded solubility of water inhydrocarbons at 20 ℃[2] .

Figure S7 .
Figure S7.(a) Temperature-dependent CD spectra of 1S assembly in DCE/MCH during a slow cooling process from 363 K to 268 K (cooling rate: 0.3 K/min, concentration: 10 μM).(b) Temperature-dependent CD spectra of 1S assembly in DCE/HE during a slow cooling process from 353 K to 268 K (cooling rate: 0.3 K/min, concentration: 10 μM).(c) The variation in the degree of aggregation (determined by intensities of CD bands) as a function of temperature during the slow cooling process with different cooling rates.(d) The variation in the degree of aggregation (determined by absorbance) as a function of temperature during the slow cooling process with different cooling rates.

Figure S8 .
Figure S8.CD spectra of the 1S assemblies in DCE/MCH (3%, v/v) solution with 13.6 μ M and 16.8 μM concentrations.Depolymerization experiments of 1S assemblies in DCE/MCH (3%, v/v) solution by measuring CD intensities at the single wavelength of 311 nm with (a) 13.6 μM and (b) 16.8 μM concentrations.The curves are fitted with mass balance model.

Figure S9 .
Figure S9.CD spectra of the 1S assemblies in DCE/MCH (3%, v/v) solution with 9.7 μM and 12.9 μM concentrations.Depolymerization experiments of 1S assemblies in DCE/HE (3%, v/v) solution by measuring CD intensities at the single wavelength of 311 nm with (a) 9.7 μM and (b) 12.9 μM concentrations.The curves are fitted with mass balance model.

Figure S10 .
Figure S10.TEM images of 1S aggregates assembled in DCE/MCH (3%, v/v) with 32 μM concentration.The samples were prepared by directly drop-casting assembly solutions onto carbon film supported copper grids.

Figure S11 .
Figure S11.AFM images of 1S aggregates assembled in DCE/MCH (3%, v/v) with 32 μM concentration.The samples were prepared by drop-casting assembly solutions onto freshly cleaned silicon wafers.

Figure S12 .
Figure S12.TEM images of 1S aggregates assembled in DCE/HE (3%, v/v) with 32 μM concentration.The samples were prepared by directly drop-casting assembly solutions onto carbon film supported copper grids.

Figure S13 .
Figure S13.AFM images of 1S aggregates assembled in DCE/HE (3%, v/v) with 32 μM concentration.The samples were prepared by drop-casting assembly solutions onto freshly cleaned silicon wafers.

Figure S14 .
Figure S14.CD and UV spectra of 1S assembly in DCE/MCH before and after adding TFA to dissociate hydrogen bonds.(a) CD and (b) UV spectra.The disappearance of CD signals and increase of absorbance indicate the disassembly of aggregates to loss the entire supramolecular nanostructures.

Figure S15 .
Figure S15.CD and UV spectra of 1S assembly in DCE/HE before and after adding TFA to dissociate hydrogen bonds.(a) CD and (b) UV spectra.Same to that in MCH, the disappearance of CD signals and increase of absorbance indicate the disassembly of aggregates to loss the entire supramolecular nanostructures.

Figure S16 .
Figure S16.Calculated CD spectra of modified trimers with P helical and M helical arrangement of central phenyl rings.(a) P helical and (b) M helical.All calculations were performed using ωB97X-D3 def2-

Figure S17 .
Figure S17.IR spectra of assembly in (a) DCE/MCH and (b) DCE/HE with different concentrations.

Figure S18 .
Figure S18.The magnified images of central nitrogen atoms in the modified trimers and the schematic representation of measured C-N-C bond angles in (a) P and (b) M helical structures.

Figure S19 .
Figure S19.Time-dependent (a) CD and (b) UV spectra of 1S assembly in DCE/HE after addition of MCH (equal amount of HE).The absorbance barely changed within 60 min.

Figure S20 .
Figure S20.CD and UV spectra of 1S assembly in DCE/MCH.(a) CD and (b) UV spectra of 1S assembly in DCE/MCH prepared with monomer solution irradiated for 30 min (365 nm irradiation wavelength, dark green solution).

Figure S24 .
Figure S24.The glum values and DC spectrum of 1S assemblies in DCE/MCH.(a) The processed glum values of measured CPL spectrum of 1S assemblies in DCE/MCH.(b) Related luminescence spectrum.The maximum glum value is 1.6×10 -2 .

Figure S25 .
Figure S25.The glum values and DC spectrum of 1S assemblies in DCE/HE.(a) The processed glum values of measured CPL spectrum of 1S assemblies in DCE/HE.(b) Related luminescence spectrum.The maximum glum value is 6.5×10 -3 .

Figure S27 .
Figure S27.(a) Photos of initial assembly solution (vial 1, right after the addition of poor solvent, c = 32 μM) and the assembly solution after overnight incubation (vial 2) in DCE/MCH.The Tyndall effect was observed in both solutions.(b) Time-dependent CD intensities of assembly (319 nm) in DCE/MCH with different water content in MCH.