Pathway-dependent supramolecular polymerization by planarity breaking

In controlled supramolecular polymerization, planar π-conjugated scaffolds are commonly used to predictably regulate stacking interactions, with various assembly pathways arising from competing interactions involving side groups. However, the extent to which the nature of the chromophore itself (planar vs. non-planar) affects pathway complexity requires clarification. To address this question, we herein designed a new BOPHY dye 2, where two oppositely oriented BF2 groups induce a disruption of planarity, and compared its supramolecular polymerization in non-polar media with that of a previously reported planar BODIPY 1 bearing identical substituents. The slightly non-planar structure of the BOPHY dye 2, as evident in previously reported X-ray structures, together with the additional out-of-plane BF2 group, allow for more diverse stacking possibilities leading to two fiber-like assemblies (kinetic 2A and thermodynamic 2B), in contrast to the single assembly previously observed for BODIPY 1. The impact of the less rigid, preorganized BOPHY core compared to the planar BODIPY counterpart is also reflected in the stronger tendency of the former to form anisotropic assemblies as a result of more favorable hydrogen bonding arrays. The structural versatility of the BOPHY core ultimately enables two stable packing arrangements: a kinetically controlled antiparallel face-to-face stacking (2A), and a thermodynamically controlled parallel slipped packing (2B) stabilized by (BF2) F⋯H (meso) interactions. Our findings underscore the significance of planarity breaking and out-of-plane substituents on chromophores as design elements in controlled supramolecular polymerization.


Materials and Methods
Chemicals and Reagents: All chemicals were purchased from Sigma Aldrich (St. Louis, MO, USA), TCI Europe N.V. (Tokyo, JP) and BLD pharm (Senefelder ring, Reinbeck, DE) and used without further purification methods unless otherwise mentioned.Silica gel was used for column chromatography unless otherwise mentioned.
Column chromatography: Preparative column chromatography was performed in self-packed glass columns of different sizes with silica gel (particle size: 40-60 µm, Merck).Solvents were distilled before usage.
NMR spectroscopy: 1 H and 13 C NMR spectra were recorded at 298 K on Avance II 300 and Avance II 400 from Bruker for routine experiments using tetramethylsilane (TMS) as internal standard.Additional 1 H as well as 2D 1 H-19 F HOESY spectra were recorded on an Agilent DD2 500 ( 1 H: 500 MHz) and an Agilent DD2 600 ( 1 H: 600 MHz) at a standard temperature of 298 K in deuterated solvents.Multiplicities for proton signals are abbreviated as s, d, t, q and m for singlet, doublet, triplet, quadruplet and multiplet, respectively.

Mass spectrometry (MS):
MALDI mass spectra were recorded on a Bruker Daltonics Ultraflex ToF/ToF or a Bruker Daltonics Autoflex Speed with a SmartBeamTM NdYAF-Laser with a wavelength of 335 nm.ESI mass spectra were measured on a Bruker MicrOToF system.The signals are described by their mass/charge ratio (m/z) in u.

UV-Vis spectroscopy:
UV-Vis absorption spectra were recorded on a JASCO V-770 or a JASCO V-750 with a spectral bandwidth of 1.0 nm and a scan rate of 400 nm min -1 .Glass cuvettes with an optical length of 1 cm, 1 mm and 0.1 mm were used.All measurements were conducted in commercially available solvents of spectroscopic grade.
Fluorescence spectroscopy: Fluorescence and excitation spectra were recorded on a JASCO Spectrofluorometer FP-8500 in quartz cuvettes (SUPRASIL®, Hellma) with an optical length of 1 cm.

FT-IR spectroscopy:
Solution and solid-state measurements were carried out using a JASCO-FT-IR-6800 equipped with a CaF 2 cell with a path length of 0.1 mm.

Atomic force microscopy (AFM):
The AFM images were recorded on a Multimode®8 SPM System manufactured by Bruker AXS.The used cantilevers were AC200TS by Oxford Instruments with an average spring constant of 9 N m -1 , an average frequency of 150 kHz, an average length of 200 µm, an average width of 40 µm and an average tip radius of 7 nm.All samples were drop-casted from freshly prepared solutions onto an HOPG surface.

Gel permeation chromatography (GPC):
Gel permeation chromatography was performed on a Shimadzu prominence GPC system equipped with two Tosoh TSKgel columns (G2500H XL; 7.8 mm I.D. x 30 cm, 5 µm; Part.No. 0016135) using CH 2 Cl 2 as eluent.The solvent flow was set to be 1 mL/min.Detection was carried out via a Shimadzu prominence SPD-M20A diode array detector (DAD).

Scanning electron microscopy (SEM):
SEM images of self-assembled species were recorded on a Thermo Fisher Scientific Phenom ProX Desktop SEM.All samples were drop-casted on a silicon wafer surface.

Transmission electron microscopy (TEM):
The TEM images were recorded on a FEI TITAN Themis G3 60-300 transmission electron microscope manufactured by Thermo Fischer Scientific with an operation voltage of 60 kV and 300 kV.The X-FEG field emission gun gives a bright and highly stable electron source for the measurements for high resolution images.This device is also equipped with monochromator, Cs image corrector, quadruple EDX-system, Fischione model 3000 HAADF detector, a fast CMOS camera to capture high resolution images with very fast frame rates and a high-resolution EEL spectrometer (GATAN Quantum 965) for detailed analysis of the structures.The samples were prepared on carbon coated mesh copper grid by drop casting the sample and the excess liquid was drained using a filter paper that was placed under the grid.Compound 1, B, C, D and 3,4,5-tris(dodecyloxy)-N-(4-ethynylphenyl)benzamide (E) were prepared by following the reported synthetic procedures and showed similar spectroscopic properties to those reported therein. [1,2]nthesis of linear BOPHY derivative (2):

Nucleation-Elongation model for cooperative supramolecular polymerization
The equilibrium between the monomeric and supramolecular polymer species can be described in a cooperative process with the Nucleation-Elongation model which was developed by Ten Eikelder, Markvoort and Meijer. [3,4] is model is used to describe the aggregation of 2, which exhibits a non-sigmoidal cooling curve as shown in temperature-dependent UV-Vis experiments.The model extends nucleation-elongation based equilibrium models for growth of supramolecular homopolymers to the case of two monomer and aggregate types and can be applied to symmetric supramolecular copolymerizations, as well as to the more general case of nonsymmetric supramolecular copolymerizations.In a cooperative process, the polymerization occurs via two steps: in a first step (nucleation), a nucleus, which is assumed to have a size of 2 molecules, is formed.In a subsequent step, the elongation of the nuclei into one-dimensional supramolecular polymers occurs.The values T e , ΔH°n ucl , ΔH° and ΔS° can be determined by a non-linear least-square analysis of the experimental melting curves.The equilibrium constants associated with the nucleation and elongation phases can be calculated using the following equations: (1) And the cooperative factor (σ) is given by:

Denaturation Model for Supramolecular Polymerization
The denaturation model [5] is based on the concentration-dependent supramolecular polymerization equilibrium model by Goldstein, [6] where the polymerization is described as a sequence of monomer addition equilibria. [ For the cooperative model, K n < K e and for the isodesmic process K n = K e .The concentration for each species P i is given by The dimensionless mass balance is obtained by inserting the dimensionless concentration   =   [  ], the monomer concentration x= K e [X] and the concentration of each species P i (for i≤ n):   =  −1   and for  > n :   =  −1   ): Both sums are evaluated by using standard expressions for converging series: With   =     and   = total monomer concentration The sum solved by standard numerical methods (Matlabfzerosolver) yields the dimensionless monomer concentration .Considering that every species with is defined as aggregate, the degree of  > 1 aggregation results in: Via   = ⅇ (−  0  ) the denaturation curves can be obtained with f defined as volume fraction of good solvent: It is assumed that the cooperativity factor  is independent of the volume fraction and the m value for the elongation regime equals the m value for nucleation.The denaturation data needs to be transformed into the normalized degree of aggregation, if fitted to the supramolecular polymerization equilibrium model: The optimization of the four needed parameters (∆G 0 , m, σ and p) to fit the equilibrium model to the experimental data (normalized degree vs. f) is done by the non-linear least-squares analysis using Matlab (lsqnonlinsolver).The data is then fitted with the non-linear least squared regression (Levenberg Marquardt algorithm).

Fluorescence quantum yield
Absolute luminescence quantum yields were measured on a JASCO spectrofluorometer FP-8500 (equipped with an ILF-835 integrating sphere) with a band width of 5 nm and a scan rate of 1000 nm/min.Quartz cuvettes with an optical path of 5 mm were employed.The measurements were carried out with a specific excitation wavelength for each sample, as shown in Table S4.

Supplementary Figures
Minor shifts in emission in solvents such as toluene arise from solvatochromism, which takes place without any aggregation process.TEM measurements further confirm that the aggregates 2A and 2B also exhibit similar morphologies to those observed in AFM and SEM.

Theoretical Calculations
The DFT B3LYP/6-31g(d,p) basis set [7,8] was used to perform the geometry optimization of the different supramolecular species (monomer, dimers and trimers).To reduce the computational cost of theoretical calculations, the long alkoxy chains were replaced by methoxy groups.The corresponding absorption spectra for the monomer and trimers were calculated by using the rcam-B3LYP/6-31g(d,p) method.All computations were carried out using Gaussian-16 (G16RevC.01). [11]The time-dependent density functional theory (TD-DFT) [9] was selected for the geometry optimization (monomer, dimers and trimers), employing the CAM-B3LYP density functional [10] together with the 6-31G(d,p) basis set. [7,8] he corresponding absorption spectra for the monomer and trimers were calculated by TD-DFT using the rcam-B3LYP/6-31g(d,p) method including 80 excitation energies.PyMOL was used as molecular visualization program.The unusual emission features of the face-to-face stacked aggregate 2A probably arise from the defects in the packing as also evident from the above-mentioned theoretical calculations.

Figure S8 :
Figure S8: Variable Temperature (VT) cooling (a) and heating (b) UV-Vis spectra of a 10 µM MCH solution of 2 with a cooling/heating rate of 1 K min -1 .c) α agg vs. T at wavelength of 505 nm.

Figure S11 :
Figure S11: Variable Temperature (VT) UV-Vis spectra obtained upon heating a sonicated solution of 2A of compound 2 at different concentrations: a) 5 µM b) 10 µM c) 15 µM and d) 30 µM with a heating rate of 1 K min -1 in MCH.

Figure S19 :
Figure S19: SEM images of aggregate 2B prepared by drop-casting 10 µL of 2 (c= 20 µM in MCH) on a silicon wafer substrate.The images reveal elongated fibre-like structures with a more significant bundling than those of 2A.

Figure S20 :Figure S21 :Figure S22 :
Figure S20: AFM height (a,c) and corresponding phase (b,d) images of aggregate 2A prepared by cooling a 10 µM solution in MCH from 363 K to 298 K with a cooling rate of 1 K min -1 followed by drop-casting the sample on HOPG surface.
13 C

Table S5 :
Different H-bond distances in aggregates 2A and 2B.