Environment‐Adaptive Coassembly/Self‐Sorting and Stimulus‐Responsiveness Transfer Based on Cholesterol Building Blocks

Abstract Manipulating the property transfer in nanosystems is a challenging task since it requires switchable molecular packing such as separate aggregation (self‐sorting) or synergistic aggregation (coassembly). Herein, a unique manipulation of self‐sorting/coassembly aggregation and the observation of switchable stimulus‐responsiveness transfer in a two component self‐assembly system are reported. Two building blocks bearing the same cholesterol group give versatile topological structures in polar and nonpolar solvents. One building block (cholesterol conjugated cynanostilbene, CCS) consists of cholesterol conjugated with a cynanostilbene unit, and the other one (C10CN) is comprised of cholesterol connected with a naphthalimide group having a flexible long alkyl chain. Their assemblies including gel, crystalline plates, and vesicles are obtained. In gel and crystalline plate phases, the self‐sorting behavior dominates, while synergistic coassembly occurs in vesicle phase. Since CCS having the cyanostilbene group can respond to the light irradiation, it undergoes light‐induced chiral amplification. C10CN is thermally responsive, whereby its supramolecular chirality is inversed upon heating. In coassembled vesicles, it is interestingly observed that their responsiveness can be transferred by each other, i.e., the C10CN segment is sensitive to the light irradiation, while CCS is thermoresponsive. This unprecedented behavior of the property transfer may shine a light to the precise fabrication of smart materials.


Materials and methods
All starting chemicals and solvents were purchased from Aladdin Medicine Reagent Co. Ltd. and used without any further purifications. Cyanostilbene and C 10 CN precursor CN were synthesized according to our previous reports. S1,S2 Crystal CIF files for CCDC 1533821 (C 10 CN) and 1533820 (CCS).

Synthesis of CCS.
Cholesteryl chloroformate (450 mg, 10 mmol) was dissolved in DCM (20 mL) at 0 o C, which was dropped into a DCM (30 mL) solution of 3-bromopropylamine hydrochloride (218 mg, 10 mmol) with triethylamine (100 µL). After the reaction for ca. 30 min, the mixture was filtered. The transparent DCM solution was subjected to rotary evaporator to remove solvents, and the resultant solid was used for the next reaction directly.
Characterizations 1 H NMR spectra were measured on a Bruker-AC300 spectrometer. 13 C NMR spectra were measured on a Bruker BBFO-400 spectrometer. High-resolution mass spectrometry (HR-MS) was performed on a Waters Q-tof Premier MS spectrometer. Transmission electron microscopy (TEM) was recorded through a JEM-1400 (JEOL, 100kV). In the TEM sample preparation, ca. 30 µL self-assembled solution was dropped on TEM copper grids, followed by drying in air. HRTEM studies were carried out on a JEM-2010 (JEOL, 200 kV). Scanning electron microscopy (SEM) samples were tested by a field-emission JSM-6700F (JEOL).
Samples for SEM tests were prepared by dropping about 100 µL self-assembled solution on the polished silicon wafer, followed by sucking most of liquids and air drying. Powder X-ray diffraction patterns were obtained via a Bruker D8 powder X-ray diffractometer at 40 kV and 30 mA using Cu Kα radiation (λ = 1.5418 Å). Samples for Fourier transform infrared (FT-IR) measurements were recorded on an Avatar 370 FT-IR spectrometer. Circular dichroism (CD) spectra were measured with a Jasco J-810 CD spectrophotometer. For the rheological measurements, the gels were tested by a Thermo Scientific HAAKE Rheo-Stress 6000. The emission spectra were obtained via a Shimadzu RF-5301pc fluorescence spectrophotometer, and UV-vis absorption spectra were recorded using a Shimadzu UV-3600 spectrophotometer.

S3
Single crystals with appropriate sizes were chosen under an optical microscope, coated with vacuum grease to prevent deconstruction, and mounted on a glass fiber for data collection on a SuperNova X-ray diffraction system from Agilent Technologies at 293 K. Simulated molecular arrangements were obtained via the software Materials Studio 7.0 by Accelrys. Scheme S1 Synthetic route of CCS.

Scheme S2
Synthetic route of C 10 CN.     the aggregation-induced-emission (AIE) effect after the self-assembly as well as the solid-state bright emission. The self-assembly in polar solvents had profound effects on the spectroscopic spectra. As compared to the monomer state in THF, the increase in f w resulted in broadening, red-shifted, and decreased absorbance in absorption spectra ( Figure S9a,c).
Therefore, J-type π-π stacking was formed during the self-assembly. Supramolecular aggregation enabled red-shifted emission spectra of CCS from 470 nm to 550 nm with enhanced intensity ( Figure S9b). The emission intensity was elevated by a factor of 6, exhibiting the AIE effect that was also revealed by bright solid-state emission ( Figure S9e). In contrast, the fluorescence emission intensity of C 10 CN was greatly quenched when water was added, showing classic aggregation-caused-quenching (ACQ) effect (Figure 9d). Similarly, the absorbance at 445 nm (aggregation) and emission intensity at 555 nm (monomer) for C 10 CN as the function of concentration displayed mutant points at around 2 × S11 Figure S12 (a) TEM and (b,c) SEM proofs of "bundle" effect in C 10 CN gels.

Discussions regarding moisture sensitivity
We utilized fresh prepared anhydrous THF (10 vol%) with different amounts of pre-added water to study the moisture influence on the phase behavior. From the digital images of C 10 CN self-assembled systems in decane (with 10 vol% THF) upon increasing the water content ( Figure S15i), it could be observed that the gel phase started to collapse when 0.02 vol% water was added, and precipitates were obtained after adding more than 0.03 vol% water. The observations indicate that the gelation is ultra-sensitive to moisture. We then tried to evaluate if the C 10 CN self-assembly system could serve as a moisture sensor. A low concentration (10 -4 M) of C 10 CN without rapid precipitation or gelation was employed ( Figure S15a). With the increase in the moisture amount from 0 to 1 µL (1 mL of total volume), the turbidity became more and more obvious (inset of Figure S15a), yet the fluorescent emission variation was irregular. By summarizing the absorbance value as a function of water volume fraction (Figure S15b), the obtained curve fits Boltzmann model with an R 2 value of 0.993. Therefore, the system could serve as a supramolecular platform for moisture-sensing in water miscible solvents, and the limit of detection (LOD) that is defined as 10 % of absorbance change is determined to be 0.26 vol%. Next, we characterized morphologies of moisture-induced precipitates ( Figures S15c-h and S16). At the low concentration range of C 10 CN (such as 0.5 mM or 1 mM), a trace amount of water (0.1 vol%) induces the formation of flower-like microstructures, confirmed by TEM and CLSM. The relatively sharp edges demonstrate the presence of long-range ordered crystalline molecular packing. Under high-resolution TEM (HRTEM) analysis, the edges of micro-flowers ( Figure S15d,e) expose the crystal lattice with a mean distance about 0.5 nm. Unlike uniform crystal lattice of most crystalline materials, the orientation of individual units roughly exhibits a helical arrangement, schematically S15 illustrated in Figure S15e. It was noted that the morphologies of self-assembled flowers are basically independent of water contents, but dependent on the C 10 CN concentration. Scattered nanorods emerged instead of flower structures in the case of 2 mM concentration (Figures S15f and S19).

Figure S16
FT-IR spectra of gel sample and flower-like structure prepared from C 10 CN.
Peaks at 3355, 1638, and 1550 cm -1 are assigned to N−H stretch, amide-I stretch, and amide-II stretch bands respectively, which indicate the presence of inter-amide/imide hydrogen bonding interactions. Nevertheless, we noticed that the relative transmittance of amide-I stretch and amide-II stretch were reduced after the addition of water, elucidating that the inter-amide/imide hydrogen bonding was partially replaced by water. S16 Figure S17 Molar ratio-dependent emission spectra (a,b) and UV-vis absorption spectra (b) of CCS/C 10 CN system, of which total concentration was fixed at 10 -4 M. Inset of (a) shows digital images of co-assembly under 365 UV light, and the ratios of CCS and C 10 CN from left to right are 0:10, 1:9, 2:8, 3:7, 4:6, 5:5, 6:4, 7:3, 8:2, 9:1, and 10:0. UV-vis absorption of CCS (d) upon the titration of C 10 CN (concentration of CCS: 10 -4 M). Fluorescent emission quenching upon the addition of CCS into C 10 CN decane gel (e), and the inset of which shows the emission comparison images of gels with (40 mol%) and without CCS. Normalized absorption with the addition of CCS. (g) TEM image of the self-sorted system of CCS/C 10 CN in decane (molar ratio of C 10 CN to CCS was 1:0.4). S17 Figure S18 (a,b) Absorption and CD spectra by stepwise increasing the CCS component (from 0 mol% to 100 mol%) in C 10 CN self-assembled system (10 -4 M, THF/water, 1/9, v/v).