Kinetically controlled Ag⁺-coordinated chiral supramolecular polymerization accompanying a helical inversion

Abstract

We report kinetically controlled chiral supramolecular polymerization based on ligand–metal complex with a 3 : 2 (L : Ag⁺) stoichiometry accompanying a helical inversion in water. A new family of bipyridine-based ligands (D-L¹, L-L¹, D-L², and D-L³) possessing hydrazine and D- or L-alanine moieties at the alkyl chain groups has been designed and synthesized. Interestingly, upon addition of AgNO₃ (0.5–1.3 equiv.) to the D-L¹ solution, it generated the aggregate I composed of the D-L¹AgNO₃ complex (D-L¹ : Ag⁺ = 1 : 1) as the kinetic product with a spherical structure. Then, aggregate I (nanoparticle) was transformed into the aggregate II (supramolecular polymer) based on the (D-L¹)₃Ag₂(NO₃)₂ complex as the thermodynamic product with a fiber structure, which led to the helical inversion from the left-handed (M-type) to the right-handed (P-type) helicity accompanying CD amplification. In contrast, the spherical aggregate I (nanoparticle) composed of the D-L¹AgNO₃ complex with the left-handed (M-type) helicity formed in the presence of 2.0 equiv. of AgNO₃ and was not additionally changed, which indicated that it was the thermodynamic product. The chiral supramolecular polymer based on (D-L¹)₃Ag₂(NO₃)₂ was produced via a nucleation–elongation mechanism with a cooperative pathway. In thermodynamic study, the standard ΔG° and ΔH_e values for the aggregates I and II were calculated using the van't Hoff plot. The enhanced ΔG° value of the aggregate II compared to that of the formation of aggregate I confirms that aggregate II was thermodynamically more stable. In the kinetic study, the influence of concentration of AgNO₃ confirmed the initial formation of the aggregate I (nanoparticle), which then evolved to the aggregate II (supramolecular polymer). Thus, the concentration of the (D-L¹)₃Ag₂(NO₃)₂ complex in the initial state plays a critical role in generating aggregate II (supramolecular polymer). In particular, NO₃⁻ acts as a critical linker and accelerator in the transformation from the aggregate I to the aggregate II. This is the first example of a system for a kinetically controlled chiral supramolecular polymer that is formed via multiple steps with coordination structural change.

---

Introduction

Control of helicity in dynamic supramolecular assemblies based on metal–ligand coordinated architectures such as capsules, cages, and helices has rapidly developed into one of the most successful and versatile strategies in supramolecular chemistry,^1–5 largely due to its essential role in biomolecules such as DNA and proteins. In addition, numerous artificial helical systems, ranging from oligonucleotides and peptide-mimetics to oligomers and polymers, have been developed.

Metallosupramolecular polymers based on metal–ligand coordination have attracted much attention as a significant class of organic/inorganic hybrid supramolecular materials.^6–30 Because of the incorporation of metal complexes in metallosupramolecular polymers, these metal-containing systems exhibit redox, optical, electrochromic, catalytic, and magnetic properties as well as the usual properties of organic polymers.^31–34 Chiral metallosupramolecular polymers are particularly attractive because they are adaptable to chiral recognition and sensing, and asymmetric catalysis.^35,36

In particular, recent advances in kinetic control over the dynamic structural and functional state of the metastable state (i.e., the out-of-equilibrium state) in supramolecular systems has allowed the development of supramolecular polymers with unprecedented structural complexity and sophisticated functionality.^{15–17,37–43} Such metastable states, in which spontaneous self-assembly is kinetically suppressed owing to competing aggregation and intramolecular hydrogen bonding, have been utilized to control the self-assembly behavior of supramolecular polymers.^16,40,41 Furthermore, understanding artificial supramolecular systems from kinetic perspectives has improved in terms of pathway complexity.^{39,41,43–45} Moreover, the growth mechanisms of supramolecular polymerization can now be interpreted by the thermodynamic (cooperative or isodesmic) models proposed by Meijer et al.^42,43,46 Such studies have demonstrated that self-assembly pathways play a significant role in determining thermodynamic or kinetic products. Thus, these previous works demonstrate that achieving kinetically programmable out-of-equilibrium states in supramolecular systems is a significant challenge related to achieving desired characteristics for dynamic self-assembly processes in supramolecular polymerization. For example, Besenius and co-workers demonstrated the kinetically controlled supramolecular polymerization of an Au(I)-metallopeptide in water.³⁷ This metal complex kinetically assembles into metastable nanorods at room temperature but is converted into a thermodynamic equilibrium fibril state upon heating. Similarly, Fernández and co-workers recently demonstrated packing polymorphism in a supramolecular polymer achieved from an amphiphilic Pt(II) complex through the kinetically controlled pathway.¹⁵ However, kinetically controlled chiral supramolecular polymerization accompanying helical inversion, based on metal-coordinated building blocks by controlling the concentration of the unique complex structure has not been successful in water. Thus, kinetically controlled metal-coordinated chiral supramolecular polymerization that exhibits pathway complexity with unique metal–ligand building blocks remains highly challenging.

Based on these considerations, we designed a new family of chiral bipyridine-based ligands (D-L¹, L-L¹, D-L², and D-L³) that have hydrazine and/or D- or L-alanine moieties at the alkyl chain groups as a chiral segment and a binding site for intermolecular hydrogen-bonding interactions in supramolecular polymerization (Fig. 1). The Ag⁺ ion was selected as a guest ion because it forms a variety of coordination geometries with bipyridine-based ligands, such as linear, triangular, square planar, and tetrahedral structures with different coordination numbers.^47–49 In particular, there is a rarely observed Ag–Ag interaction in complexes obtained from Ag⁺ ions when three ligand molecules are coordinated with two Ag⁺ ions. Furthermore, the Ag–Ag interaction also acts as an additional driving force and a metal-coordinated chiral seed in chiral supramolecular polymerization.

Fig. 1 Chemical structures of chiral bipyridine derivatives with hydrazinoethanol (D-L¹ and L-L¹), hydrazine (D-L²) and 3-aminopropanol (D-L³).

Surprisingly, we found that the left-handed (M-type) chiral bipyridine complexes (D-L¹)₂AgNO₃ and D-L¹AgNO₃ are formed at less than 0.5 equiv. and at over 1.4 equiv. of AgNO₃, respectively. More interestingly, the addition of 0.5–1.3 equiv. of AgNO₃ to a D-L¹ solution generates the D-L¹AgNO₃ complex, which forms aggregate I with a spherical structure. However, this eventually forms aggregate II with a fiber structure based on the (D-L¹)₃Ag₂NO₃ complex, involving left-handed (M-type) to right-handed (P-type) helicity conversion. The supramolecular polymerization that forms the fiber structure occurs via a cooperative pathway by a nucleation–elongation mechanism that is kinetically controlled by the concentrations of Ag⁺ and NO₃⁻. Thus, we herein report kinetically controlled chiral supramolecular polymerization in water based on a unique complex building block.

Results and discussion

Stoichiometric ratios of self-assembled complexes with Ag⁺ by ESI-MS

Because bipyridine-based ligands form a variety of complexes with Ag⁺, we first investigated the complexation behavior of ligand D-L¹, using high-resolution electrospray ionization mass spectrometry (HR ESI-MS) (Fig. S1–S4†). We confirmed that D-L¹ formed mainly (D-L¹)₂AgNO₃, (D-L¹)₃Ag₂(NO₃)₂, and D-L¹AgNO₃ complexes with 0.4, 1.0, and 2.0 equiv. of Ag⁺, respectively (see ESI: explained in detail†).

Helicity inversion in the formation of supramolecular polymers in the presence of Ag⁺ ion

To investigate the helicity of self-assembled complexes comprising D-L¹ (with D-alanine moieties) and L-L¹ (with L-alanine moieties) with Ag⁺, time-dependent circular dichroism (CD) and UV-Vis analysis of D-L¹ (8 mM) was performed for three different concentrations (0.4, 1.0, and 2.0 equiv.) of AgNO₃ at 25 °C (Fig. 2 and S5†). The negative CD signals immediately increase upon addition of 0.4 and 2.0 equiv. of AgNO₃ but were not inverted to a positive CD signal (Fig. 2C: (a and c)). Conversely, the negative CD signal at 334 nm was inverted to a positive CD signal with 1.0 equiv. of AgNO₃ after 70 min and was enhanced drastically (Fig. 2A and C: (b)). The transparent solution prepared with D-L¹ (8 mM) and AgNO₃ (1.0 equiv.) also changed to a colloidal suspension during the CD measurement (as observed by ESI/TOF-MS measurement). The CD inversion was due to the formation of the (D-L¹)₃Ag₂(NO₃)₂ complex, which has right-handed (P-type) helicity. Furthermore, the weak CD signal intensity of D-L¹ with 1.0 equiv. of AgNO₃ is due to the presence of three complexes with different helicities in the initial state (Fig. 2C). The plot of the CD intensity and UV-Vis spectral changes showed a non-linear sigmoidal transition characterized by a nucleation–elongation mechanism (Fig. 2C and D). The long lag time in the CD and UV-Vis observations implies the existence of the kinetic aggregate.

Fig. 2 Time-dependent (A) CD and (B) UV spectra changes of D-L¹ (8 mM) with AgNO₃ (1.0 equiv.) in water. (C) Plot for CD spectra changes of D-L¹vs. time in the presence of AgNO₃ (a: 0.4, b: 1.0, and c: 2.0 equiv.) at 334 nm in water. (D) Plot for absorbance changes of D-L¹vs. time in the presence of AgNO₃ (1.0 equiv.) at 334 nm in water.

We also observed the CD spectra of the supramolecular polymeric D-L¹ complexes at various concentrations (0–2.0 equiv.) of Ag⁺ after aging for 3 h to eliminate the time effect (Fig. 3). The CD signal of complex D-L¹ (8 mM) was observed at a mole ratio below 0.5 as a negative signal at 315 nm, which we ascribe to the bipyridine derivative D-L¹ chromophores. These negative signals in the CD analysis indicate that complex (D-L¹)₂AgNO₃ with four coordination was orientated to the left-handed direction (Schemes 1 and S1A†). Interestingly, the positive CD signal at 334 nm for supramolecular polymer D-L¹ in cases where the Ag⁺-to-D-L¹ mole ratio was greater than 0.4 was gradually enhanced (Fig. 3A and C), reaching a maximum at a Ag⁺-to-D-L¹ mole ratio of 1.0 (Fig. 3C). This CD inversion was due to the transformation of (D-L¹)₂AgNO₃ (anisotropic factor, |g| = 1.0 × 10⁻³ at 334 nm) and D-L¹AgNO₃ complexes into the (D-L¹)₃Ag₂(NO₃)₂ complex, as shown in Scheme S1B.† Thus, the addition of 0.5–1.3 equiv. of AgNO₃ generates aggregate I composed of the D-L¹AgNO₃ complex (|g| = 8.0 × 10⁻³ at 334 nm) with left-handed helicity as the kinetic product, and then finally aggregate II is formed based on the (D-L¹)₃Ag₂(NO₃)₂ complex (|g| = 1.45 × 10⁻² at 334 nm), which has right-handed (P-type) helicity (Schemes 1 and S1B†) as the thermodynamic product.

Fig. 3 CD spectra of (A) D-L¹ or (B) L-L¹ upon addition of AgNO₃ (0–2.0 equiv.) after aging for 3 h in water. (C) Plot for CD spectral changes of D-L¹ and L-L¹vs. equivalent of AgNO₃ at 334 nm. (D) Complex structures existed in various concentration of AgNO₃; (a) (D-L¹)₂AgNO₃, (b) (D-L¹)₂AgNO₃, (D-L¹)₃Ag₂(NO₃)₂, and D-L¹AgNO₃, and (c) D-L¹AgNO₃.

Scheme 1 Proposed chiral supramolecular polymerization for D-L¹ with AgNO₃ (0.5–1.3 equiv.) showing the assembly by nucleation–elongation mechanism; (a) complex formation (D-L¹ : Ag⁺ = 2 : 1, 1 : 1 and 3 : 2), (b) 1 : 1 complex formation, (c) the formation of aggregate I based on D-L¹AgNO₃ complex with M-helicity, (d) 3 : 2 complex formation, and (e) formation of the aggregate II based on (D-L¹)₃Ag₂(NO₃)₂ complex with P-helicity.

Furthermore, the transparent solution became a colloidal suspension at 0.5–1.3 equiv. of Ag⁺. After separation, we observed the CD spectra of both the colloidal particles and the filtered solution. The colloidal particles dispersed in water exhibited an intense positive CD signal (Fig. S6a†), which was almost the same as that for the solution before filtration. However, the filtrate exhibited an extremely weak negative CD signal (Fig. S6b†). These findings indicate that complexes (D-L¹)₂AgNO₃ and D-L¹AgNO₃ present before CD inversion change to (D-L¹)₃Ag₂(NO₃)₂ as the colloidal aggregate II upon achieving chemical equilibrium, as shown in Schemes 1 and S1B.† These results provide clear evidence that the (D-L¹)₃Ag₂(NO₃)₂ complex plays an essential role as a nucleus (seed) in the formation of aggregate II. In addition, it is noteworthy that AgNO₃ acted as a key component in forming aggregates I and II. More importantly, the positive CD signal for the supramolecular polymer obtained with 1.0 equiv. of Ag⁺ did not change upon further addition of AgNO₃ (Fig. S7†), indicating that aggregate II is the thermodynamic product. When the Ag⁺-to-D-L¹ mole ratio increases beyond 1.0, the intensity of the CD signal for the supramolecular polymer gradually decreases again (Fig. 3A and C). This decrease in the positive CD signal was due to the coexistence of aggregate II and the monomeric D-L¹AgNO₃ complex species. At 1.3 equiv. of AgNO₃, time-dependent CD spectra of D-L¹ (8 mM) did not change after 180 min, indicating that the kinetic product was not existed (Fig. S8†). However, a small amount of monomeric 1 : 1 complex species, which does not form the spherical structure, exists with the thermodynamic product. Finally, the negative CD signal for D-L¹ at over 1.4 equiv. of AgNO₃ is attributed to the formation of the D-L¹AgNO₃ complex as the main product (Scheme S1C†), which indicates that the D-L¹AgNO₃ complex has M-helicity (left-handed helicity). In addition, the negative CD signal obtained at 1.5 equiv. of Ag⁺ was inverted to the positive signal when 0.5 equiv. of free D-L¹ was added to a mixture of D-L¹ and Ag⁺ (1.5 equiv.) (Fig. S9†). This CD inversion was due to the (D-L¹)₃Ag₂(NO₃)₂ complex, which was generated from the D-L¹AgNO₃ complex. We conclude that the aggregate II as the thermodynamic product was only produced for 0.5–1.3 equiv. of Ag⁺ (Fig. 3D and Scheme 1).

In contrast, the CD spectra of complex L-L¹ (8 mM) upon addition of AgNO₃ (0–0.4 equiv.) exhibited a positive signal at 315 nm (Fig. 3B and C), which is exactly opposite to the behavior of complex D-L¹ with Ag⁺. This positive CD signal indicates that the bipyridine moiety of L-L¹ is orientated in the right-handed (P-type) direction. The positive CD signal of L-L¹ was inverted into an intensely negative CD signal when 0.5–1.3 equiv. of Ag⁺ was added to L-L¹. This chiral inversion is also due to the transformation from (L-L¹)₂AgNO₃ and L-L¹AgNO₃ complexes into complex (L-L¹)₃Ag₂(NO₃)₂, which led to the formation of the left-handed aggregate II. However, the positive CD signal of L-L¹ was also observed when 1.4–2.0 equiv. of AgNO₃ was added to L-L¹, which was attributed to the formation of aggregate I (nanoparticles) based on the L-L¹AgNO₃ complex by path A in Scheme 1.

In order to investigate their linear dichroism (LD) effects, the LD spectra of D-L¹ (8 mM) and L-L¹ (8 mM) with AgNO₃ (1.0 and 1.4 equiv.) were observed under the same conditions as those used for CD observations (Fig. S10†). Extremely weak LD signals were obtained for D-L¹ and L-L¹ in the presence of AgNO₃ (1.0 and 1.4 equiv.). These results imply that the LD effect is only marginally affected by the CD response of supramolecular polymers.

To determine how anions effect the production of supramolecular polymers, we further observed the handedness of the supramolecular polymer upon addition of AgBF₄, AgClO₄, AgPF₆, or AgOTf instead of NO₃⁻. As shown in Fig. S11,† the negative CD signals for complex D-L¹ formed with 0–2.0 equiv. of AgBF₄, AgClO₄, AgPF₆, or AgOTf showed a slightly enhanced intensity, but they did not invert to the positive signal even though 3 : 2 complex (D-L¹ : Ag⁺) formation was confirmed by HR ESI-MS (Fig. S12–S15†). These findings imply that the NO₃⁻ anion plays a critical role in the formation of aggregate II, as shown in Schemes 1 and S1B.†

Role of NO₃⁻ anion in the supramolecular polymerization

To demonstrate the essential role of NO₃⁻ in the formation of the supramolecular polymer, we also observed the CD spectral changes for D-L¹ in the presence of 1.0 equiv. of AgOTf, AgClO₄, AgBF₄, or AgPF₆ upon the addition of 2.0 equiv. of NaNO₃ (Fig. S16†). The weak negative CD signal became a positive CD signal when NaNO₃ (0.4 equiv.) was added to a mixed D-L¹ and AgOTf (1 : 1 equiv.) solution. The continuous addition of NaNO₃ gradually enhanced the positive CD signals, and the signal reached a maximum at 2.0 equiv. of NaNO₃ (Fig. 4), which was due to the generation of the aggregate II based on (D-L¹)₃Ag₂(NO₃)₂. The peak for the [(D-L¹)₃Ag₂NO₃]⁺ fragmented from the (D-L¹)₃Ag₂(NO₃)₂ complex was clearly observed in HR ESI/TOF-MS measurement (Fig. S17†). This led not only to the helical inversion but also to the formation of supramolecular polymers. These results indicate that NO₃⁻ plays a key role in not only the formation of the supramolecular polymer via the bridge effect but also helical inversion by path B, as shown in Scheme 1.

Fig. 4 (A) CD spectra of D-L¹ upon addition of NaNO₃ in the presence of AgOTf (1.0 equiv.) in water. (B) Plot for CD intensity of D-L¹ + AgOTf against added NaNO₃ concentration.

As a reference experiment, HR ESI-MS and CD spectra of D-L², and D-L³ without ethyl alcohol or the hydrazine moiety at both ends were observed upon addition of AgNO₃. The formation of the (D-L²)₂AgNO₃, (D-L²)₃Ag₂(NO₃)₂, or D-L²AgNO₃ complexes obtained at different concentrations of Ag⁺ (0.4, 1.0, and 2.0 equiv. of Ag⁺) was confirmed via HR ESI-MS (Fig. S18–S20†). As expected, the D-L² complex structures with different concentrations of Ag⁺ showed a similar tendency as that D-L¹. The CD inversion of D-L² (8 mM) with 0.8–1.1 equiv. of AgNO₃ occurred due to the formation of aggregate II based on the (D-L²)₃Ag₂(NO₃)₂ complex (Fig. S21†). However, the CD intensity for the supramolecular polymer D-L² was ca. 4.3-fold lower than that of supramolecular polymer D-L¹ under a same condition, which was attributed to the weak complex formation between D-L² and Ag⁺. However, the negative CD signal did not invert to a positive CD signal when 1.0 equiv. of AgNO₃ was added to D-L³ (8 mM), which does not have the hydrazine moiety (Fig. S22†), because of no formation of 3 : 2 complex (D-L³ : Ag⁺) as shown in Fig. S23.† These findings indicate that the amine in the hydrazine group plays a critical role in 3 : 2 complex formation (D-L³ : Ag⁺).

Interestingly, upon addition of NaCl (0.9 equiv.) to aggregate I, a positive CD signal resulted, and the maximum positive CD signal was reached with 1.3 equiv. of NaCl (Fig. S24†). The (D-L¹)₂AgNO₃ complex and free D-L¹ were generated by decomplexation of D-L¹AgNO₃ in the presence of NaCl, and this eventually produced (D-L¹)₃Ag₂(NO₃)₂, which enables the formation of aggregate II, as shown in Scheme S1C.† The largest positive CD signal induced via 1.3 equiv. of NaCl was ascribed to the formation of the largest amount of the (D-L¹)₃Ag₂(NO₃)₂ complex. Thus, this CD inversion was due to transformation from the D-L¹Ag to the (D-L¹)₃Ag₂(NO₃)₂ complex, which led to further formation of aggregate II.

NMR and IR studies of aggregates I and II

To determine the exact binding site of D-L¹ for NO₃⁻ in the formation of the supramolecular polymer, we obtained the ¹H NMR spectrum of D-L¹ in the presence and the absence of AgNO₃ in D₂O and H₂O (1 : 1 v/v) (Fig. S25–S27†). These results suggest that the supramolecular polymer constructed from π–π stacking interactions, and the intermolecular hydrogen-bonding interactions between D-L¹ and D-L¹ were weakened due to the linkage with NO₃⁻ between the hydrazine moiety of D-L¹ and D-L¹ (see ESI: explained in detail†). The FT-IR results provide evidence for the intermolecular hydrogen bonds between the hydrazine NH and the NO₃⁻ in the chiral supramolecular polymers (Fig. S28 and S29, see ESI: explained in detail†). In addition, a sharp powder X-ray diffraction (XRD) peak for aggregate II was observed at 2θ = 32 (Fig. S30†), which corresponds to an Ag–Ag interaction with a distance of 2.8 Å.⁴⁸ This suggests that NO₃⁻ did not coordinate with Ag⁺. The Ag–Ag interaction observed likely plays an important role in the chiral supramolecular polymerization. Thus, based on the FT-IR, powder XRD, and ¹H NMR results, we determined that NO₃⁻ acted as a bridge to produce the helical supramolecular polymeric structure. This is in agreement with results previously reported by the Miyake group.⁵⁰

Morphology observation of aggregates I and II by AFM

To confirm the morphology changes of the supramolecular polymers formed at different concentrations of AgNO₃, we performed time-dependent atomic force microscopy (AFM) experiments (Fig. 5). For this purpose, freshly prepared solutions of D-L¹ with and without AgNO₃ (0.4, 1.0, 1.3, and 2.0 equiv.) were quickly dropped onto silicon wafers at 298 K after a constant aging time. This fixed the sample on the silicon wafer surface because it acts like a mode and prevents structural changes in the material during drying.

Fig. 5 AFM images of complex D-L¹ with 1.0 equiv. of AgNO₃ in water after aging of (A) 10 min, (B) 90 min, (C) 120 min and (D) 240 min.

In the absence of AgNO₃, the AFM image of pristine D-L¹ showed fiber structures with widths of ca. 30 nm and lengths of several-hundred micrometers (Fig. S31†), indicating nanostructural formation of D-L¹ by π–π stacking. This fiber structure did not change with time, which indicative that the fiber structure was the thermodynamic product. In contrast, no distinct morphology for the D-L¹ solution with 0.4 equiv. of AgNO₃ was observed regardless of aging time, indicating that the (D-L¹)₂AgNO₃ complex did not elongate to form aggregates I or II. However, time-dependent AFM images of the D-L¹ solution containing AgNO₃ (2.0 equiv.), which formed complex D-L¹AgNO₃, were observed at different aging times. After aging for 10 min, the AFM image showed spherical structures with a diameter of ca. 50 nm (Fig. S32†). The spherical structures with dimensions close to 5 nm, as previously reported by Meijer's group,⁵¹ were similar to that of the micelle type containing a chiral molecular arrangement. This sphere would consist of the bilayer structure, because the full extended molecule length of D-L¹ is ca. 3 nm.

No significant morphological changes were observed after aging for two weeks, indicating that the spherical structure was the thermodynamic product. The morphology changes were further supported by dynamic light scattering (DLS) study (Fig. S33b†). Note that a growth pattern in the scattering intensity was not observed, which means that the morphology did not change. This finding suggests that the nanoparticles formed in the presence of 2.0 equiv. of AgNO₃ were a thermodynamic aggregate, as shown in Scheme S1C.†

Interestingly, the D-L¹ solution containing AgNO₃ (1.0 equiv.) had only small spherical structures with ca. 50 nm diameters at an aging time of 10 min (Fig. 5A), which originated from aggregate I (Scheme 1). After 90 min, the AFM image showed one-dimensional nanorods with 200 to 300 nm length and nanoparticles with ca. 50 nm diameter (Fig. 5B), which were due to aggregate II and the aggregate I, respectively. Upon allowing the solution to age for 120 min at 298 K, the nanorods grew into fibrous structures (Fig. 5C and D). This was clear evidence that the (D-L¹)₃Ag₂(NO₃)₂ complex grew into one-dimensional fibers with intermolecular interactions as shown in Scheme 1. These fibers had an average thickness of ∼3 nm by AFM observation (Fig. S34†), which is consistent with the fully extended molecular length of ligand D-L¹. Based on density functional theory (DFT) calculations, one fiber should consist of ca. 20 units of (D-L¹)₃Ag₂(NO₃)₂ complex in the width direction (Fig. S34†). As shown by the DLS study (Fig. S33a†), the growth pattern in scattering intensity was similar to the time-dependent change in the CD signal at 334 nm. This clearly indicates that the spherical structure as the kinetic product was transformed into the fiber structure as the thermodynamic product. Furthermore, only fibrous structures were observed in the presence of 1.3 equiv. of AgNO₃ after aging for 180 min (Fig. S35†), indicating that the thermodynamic product but not the kinetic product existed.

To confirm the anion effect in the formation of the supramolecular fibers, we obtained AFM images of the D-L¹ solution containing AgOTf (1.0 equiv.). No significant morphological changes were observed, regardless of aging time. However, a narrow size distribution for the nanoparticles was observed during the lag time when NaNO₃ (1.0 equiv.) was added to a D-L¹ solution containing AgOTf (1.0 equiv.) (Fig. S36A†). These nanoparticles were transformed into fibrous structures several micrometers in length upon 90 min of aging (Fig. S36B and C†). This is indirect evidence that NO₃⁻ acted as a bridging ligand between the hydrazine groups of complex D-L¹ to generate the chiral supramolecular polymer.

For D-L² with AgNO₃ (1.0 equiv.), the AFM image showed nanoparticles with ca. 50 nm diameter during the initial stage (Fig. S37A†). Then, helical fibrous structures with 100 μm length and 50 nm diameter was observed after 90 min of aging (Fig. S37B†). These results suggest that the supramolecular polymerization mechanism for D-L² was the same as that for D-L¹ and L-L¹. We also confirmed the morphology of D-L³ (8 mM) with and without AgNO₃ (1.0 equiv.) by AFM. The AFM image did not show any distinct morphologies for solutions of D-L³ in the presence and the absence of AgNO₃ (1.0 equiv.) (Fig. S38†), because D-L³ could not form intermolecular hydrogen-bonding interactions between the –NH of the hydrazine moiety and NO₃⁻.

DFT calculation for complexes formation with Ag⁺ ion

To gain insight into the influence of NO₃⁻ on formation of the D-L¹ and Ag⁺ complex, we used DFT to optimize the structures of the complexes as confirmed by HR ESI-MS. During optimization, we considered the solvation effect of water by employing a conductor-like polarizable continuum model.⁵² Here, we optimized three complexes, i.e., D-L¹AgNO₃, (D-L¹),₂AgNO₃, and (D-L¹)₃Ag₂(NO₃)₂, depending on the number of NO₃⁻ ions and D-L¹. As shown in Fig. 6, S39 and S40† (see ESI: explained in detail†), we found that the oxygen atoms of the NO₃⁻ moiety bonded to the hydrogen of the hydrazine group at the alkyl chain in the (D-L¹)₃Ag₂(NO₃)₂ complex. The distances between the NO₃⁻ oxygens and the –NH hydrogen were in the range 1.93–1.95 Å, which is typical of hydrogen bonds (1.5–2.5 Å) between an oxygen acceptor and a nitrogen or oxygen donor.^53,54 These results indicate that the NO₃⁻ group plays an essential role as an oxygen acceptor for hydrogen bonding in the complex formation of D-L¹ and Ag⁺.

Fig. 6 Optimized structure of (A) (D-L¹)₃Ag₂(NO₃)₂ with P-helicity. (B) Proposed structure of the aggregate II consisting of four building block units of (D-L¹)₃Ag₂(NO₃)₂.

Thermodynamic study of aggregates I and II

To determine the thermodynamic parameters for supramolecular polymerization, temperature dependent spectra were observed at different concentrations (5–8 mM) of the supramolecular polymer formed with 1.0 equiv. of AgNO₃ were observed from 293 to 373 K at a heating rate of 1 K min⁻¹ (Fig. 7 and S41A†) because a long time is required for complex formation at room temperature. Further, the supramolecular polymerization (nucleation–elongation process) was observed only at concentrations >5 mM, due to the weak 3 : 2 complex formation. To determine the thermodynamic parameters, we calculated the results by referring to a previous study,⁵⁵ which obtained thermodynamic parameters at high concentrations. The plot of CD intensity showed a non-sigmoidal curve that was characterized by the critical elongation temperature (T_e) (Fig. 7 and S41B†). The onset temperatures for the elongation process were in the range 334.2–339.7 K, representing the critical temperatures between nucleation and elongation. The CD intensities of the supramolecular polymers enhanced with increasing concentration. The heating curves could be fitted using a nucleation–elongation model that has the characteristics of a cooperative assembly.^56–58 We calculated the thermodynamic parameters using the van't Hoff plot.⁵⁹

Fig. 7 (A) Degree of aggregation of different concentrations of D-L¹ (8 mM) complex with 1.0 equiv. of AgNO₃ as a function of temperature at 334 nm in water. (B) Natural logarithm of the reciprocal C_T as a function of the reciprocal T_e showing the linear relationship.

As shown in Fig. 7B, a linear correlation was evident between the concentration and T_e in the van't Hoff plot. The standard ΔG° and ΔH_e values for the supramolecular polymer (aggregate II) were calculated to be −19.64 kJ mol⁻¹ and −63.90 kJ mol⁻¹, respectively. The thermodynamic parameters of aggregate I at 2.0 equiv. of AgNO₃ were also determined (Fig. S42†), even though the formation conditions of aggregate I differ slightly from those in Scheme S1B.† The standard ΔG° and ΔH_e values for aggregate I were −15.82 kJ mol⁻¹ and −33.08 kJ mol⁻¹, respectively. The reduced ΔG° value compared to that for the formation of aggregate II (ΔG° = −19.64 kJ mol⁻¹) confirms that aggregate II was thermodynamic product, as shown in Scheme S1B.† Accordingly, aggregate I was the kinetically favored state at 0.5–1.3 equiv. of AgNO₃, i.e., aggregate II was the thermodynamically favored state in the energy landscape as related to pathway complexity (Scheme 1 and Fig. S43†). These results indicate that the pathway complexity in the self-assembly of aggregates I and II can be rationally modulated by controlling the conformation of the metal–ligand complex depending on the AgNO₃ concentration. On the other hand, the standard ΔG° and ΔH_e values for pristine D-L¹ were calculated to be −15.23 and −25.26 kJ mol⁻¹, respectively (Fig. S44†), which are lower than those for aggregates I and II.

Kinetic study of aggregate II at various concentrations of Ag⁺ and NO₃⁻

To investigate kinetic control of the supramolecular polymer formation, we performed time-dependent CD experiments (Fig. 8). The kinetic profiles of supramolecular polymer D-L¹ (8 mM) that formed with different concentrations of AgNO₃ (0.6–1.0 equiv.) in aqueous solution at 25 °C showed non-sigmoidal curves with different lag times (Fig. 8). The t₅₀ value, at which the transformation is 50% complete, became short as the AgNO₃ concentration increased (Fig. 8B). The influence of AgNO₃ concentration on the kinetics confirmed the initial formation of aggregate I consisting of D-L¹AgNO₃, which developed into aggregate II consisting of (D-L¹)₃Ag₂(NO₃)₂. The largest amount of the (D-L¹)₃Ag₂(NO₃)₂ complex produced during the initial stage in the presence of 1.0 equiv. of AgNO₃, kinetically accelerated the transformation of aggregate I into the aggregate II. Thus, the concentration of the (D-L¹)₃Ag₂(NO₃)₂ complex during the initial state played a critical role in producing aggregate II, which exhibited a helical fiber structure.

Fig. 8 (A) Time course evolution of the aggregate I of D-L¹ into the kinetically controlled the aggregate II with different concentrations of AgNO₃; (a) 0.6, (b) 0.7, (c) 0.8, (d) 0.9 and (e) 1.0 equiv. at 334 nm in water. (B) Representation for t₅₀ values vs. AgNO₃ concentration.

Furthermore, the kinetic profiles of the supramolecular polymer were also observed upon addition of NaNO₃ (1.0, 1.4, 1.7, and 2.0 equiv.) in the presence of AgOTf (1.0 equiv.) (Fig. S45†). The highest concentration of NaNO₃ showed the largest positive CD intensity (Fig. S45d†) owing to formation of the largest amount of the (D-L¹)₃Ag₂(NO₃)₂ complex under these conditions. These results imply that the kinetics of aggregate II strongly influenced by NO₃⁻ anions. Furthermore, the lag time decreased with increasing NaNO₃ concentration, indicating that the NO₃⁻ anion plays an essential role in the helical inversion from aggregate I to aggregate II. To our knowledge, this is the first example of a system in which the NO₃⁻ anion acts as an accelerator and linker in the transformation of aggregates. In addition, the kinetic profile of the supramolecular polymer D-L² that does not have an ethyl alcohol moiety showed the same tendency as that of compound D-L¹ (Fig. S46†). However, the supramolecular polymerization of D-L² was observed in a narrower Ag⁺ concentration range compared to that for D-L¹ because D-L² forms the (D-L²)₂AgNO₃ complex as a main product, as shown by HR ESI-MS measurements.

Conclusions

We have demonstrated that supramolecular polymerization involving a distinct metal–ligand complex was controlled kinetically via the concentration of Ag⁺ and anions in water. This also enabled the controlled, reversible transformation between M- and P-type helices, with a change in complex structure as a function of Ag⁺ concentration. A change in the complex structure also accompanied the helical inversion, where the M-type helices of (D-L¹)₂AgNO₃ and D-L¹AgNO₃ were inverted to the P-type helix of (D-L¹)₃Ag₂(NO₃)₂. In the kinetic study, the (D-L¹)₃Ag₂(NO₃)₂ complex was found to play an essential role in chiral inversion. The distinct (D-L¹)₃Ag₂(NO₃)₂ complex produced the chiral supramolecular polymer (aggregate II) via a nucleation–elongation mechanism with a cooperative pathway in the range of 0.5–1.3 equiv. of AgNO₃, and this process showed an unusually long lag time. In addition, D-L¹, and L-L¹ in the presence of 0.5–1.3 equiv. of AgNO₃ formed intermediate aggregate I (nanoparticles) as the kinetic product that evolved to afford a helical supramolecular polymer (aggregate II). This transformation can be accelerated according to the Ag⁺ and NO₃⁻ concentration. Also, AgNO₃ acted as a key component in forming aggregates I and II. In particular, NO₃⁻ bridged between hydrazine–NH groups of the complex via intermolecular hydrogen bonds in the supramolecular polymer. In contrast, aggregate I formed at over 1.4 equiv. of AgNO₃ was the thermodynamic product.

This system is an extremely rare example of chiral supramolecular polymerization based on metal–organic building blocks via control of coordination number. We believe that this concept will greatly contribute to the development of metal-coordinated supramolecular systems with optimized functional properties and may provide new insights into the understanding of the chiral assembly process in complex biological systems.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the NRF (2017R1A4A1014595 and 2018R1A2B2003637) from the Ministry of Science, ICT and Future Planning, Korea and the Korea Basic Science Institute (KBSI) National Research Facilities & Equipment Center (NFEC) grant funded by the Korea government (Ministry of Education) (No. 2019R1A6C1010042). In addition, this work was partially supported by a grant from the Next-Generation BioGreen 21 Program (SSAC, grant#: PJ013186052019), Rural development Administration, Korea.

Notes and references

1. N. Ousaka, Y. Takeyama, H. Iida and E. Yashima, Nat. Chem., 2011, 3, 856–861.
2. J. Rebek, Jr, Angew. Chem., Int. Ed., 2005, 44, 2068–2078.
3. P. J. Stang, B. Olenyuk, D. C. Muddiman, D. S. Wunschel and R. D. Smith, Organometallics, 1997, 16, 3094–3096.
4. E. Yashima, N. Ousaka, D. Taura, K. Shimomura, T. Ikai and K. Maeda, Chem. Rev., 2016, 116, 13752–13990.
5. A. R. A. Palmans and E. W. Meijer, Angew. Chem., Int. Ed., 2007, 46, 8948–8968.
6. M. Albrecht, Chem. Rev., 2001, 101, 3457–3498.
7. Q. Wan, X.-S. Xiao, W.-P. To, W. Lu, Y. Chen, K.-H. Low and C.-M. Che, Angew. Chem., Int. Ed., 2018, 57, 17189–17193.
8. Q. Wan, W.-P. To, C. Yang and C.-M. Che, Angew. Chem., Int. Ed., 2018, 57, 3089–3093.
9. J. L.-L. Tsai, T. Zou, J. Liu, T. Chen, A. O.-Y. Chan, C. Yang, C.-N. Lok and C.-M. Che, Chem. Sci., 2015, 6, 3823–3830.
10. A. Mishra, D. B. Korlepara, M. Kumar, A. Jain, N. Jonnalagadda, K. K. Bejagam, S. Balasubramanian and S. J. George, Nat. Commun., 2018, 9, 1295.
11. K. K. Kartha, N. K. Allampally, A. T. Politi, D. D. Prabhu, H. Ouchi, R. Q. Albuquerque, S. Yagai and G. Fernández, Chem. Sci., 2019, 10, 752–760.
12. F. Mamiya, N. Ousaka and E. Yashima, Angew. Chem., Int. Ed., 2015, 54, 14442–14446.
13. N. Ousaka, K. Shimizu, Y. Suzuki, T. Iwata, M. Itakura, D. Taura, H. Iida, Y. Furusho, T. Mori and E. Yashima, J. Am. Chem. Soc., 2018, 140, 17027–17039.
14. M. E. Robinson, A. Nazemi, D. J. Lunn, D. W. Hayward, C. E. Boott, M.-S. Hsiao, R. L. Harniman, S. A. Davis, G. R. Whittell, R. M. Richardson, L. De Cola and I. Manners, ACS Nano, 2017, 11, 9162–9175.
15. A. Langenstroer, K. K. Kartha, Y. Dorca, J. Droste, V. Stepanenko, R. Q. Albuquerque, M. R. Hansen, L. Sánchez and G. Fernández, J. Am. Chem. Soc., 2019, 141, 5192–5200.
16. S. Ogi, C. Grzeszkiewicz and F. Würthner, Chem. Sci., 2018, 9, 2768–2773.
17. L. Herkert, J. Droste, K. K. Kartha, P. A. Korevaar, T. F. A. de Greef, M. R. Hansen and G. Fernández, Angew. Chem., Int. Ed., 2019, 58, 11344–11349.
18. A. Langenstroer, Y. Dorca, K. K. Kartha, M. J. Mayoral, V. Stepanenko, G. Fernández and L. Sánchez, Macromol. Rapid Commun., 2018, 39, 1800191.
19. N. A. M. S. Caturello, Z. Csók, G. Fernández and R. Q. Albuquerque, Chem.–Eur. J., 2016, 22, 17681–17689.
20. K. Y. Kim, J. Kim, C. J. Moon, J. Liu, S. S. Lee, M. Y. Choi, C. Feng and J. H. Jung, Angew. Chem., Int. Ed., 2019, 58, 11709–11714.
21. H. Park, K. Y. Kim, S. H. Jung, Y. Choi, H. Sato and J. H. Jung, Chem. Mater., 2018, 30, 2074–2083.
22. K. Y. Kim, J. Kim, H. Park, Y. Choi, K.-Y. Kwon and J. H. Jung, Bull. Korean Chem. Soc., 2018, 39, 988–994.
23. V. C.-H. Wong, C. Po, S. Y.-L. Leung, A. K.-W. Chan, S. Yang, B. Zhu, X. Cui and V. W.-W. Yam, J. Am. Chem. Soc., 2018, 140, 657–666.
24. S. Y.-L. Leung, K. M.-C. Wong and V. W.-W. Yam, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 2845–2850.
25. M. H.-Y. Chan, S. Y.-L. Leung and V. W.-W. Yam, J. Am. Chem. Soc., 2018, 140, 7637–7646.
26. Y. Sakata, S. Chiba, M. Miyashita, T. Nabeshima and S. Akine, Chem.–Eur. J., 2019, 25, 2962–2966.
27. T. Nakamura, H. Kimura, T. Okuhara, M. Yamamura and T. Nabeshima, J. Am. Chem. Soc., 2016, 138, 794–797.
28. R. Matsuoka and T. Nabeshima, Front. Chem., 2018, 6, 349.
29. B. Kemper, M. von Gröning, V. Lewe, D. Spitzer, T. Otremba, N. Stergiou, D. Schollmeyer, E. Schmitt, B. J. Ravoo and P. Besenius, Chem.–Eur. J., 2017, 23, 6048–6055.
30. B. Kemper, Y. R. Hristova, S. Tacke, L. Stegemann, L. S. van Bezouwen, M. C. A. Stuart, J. Klingauf, C. A. Strassert and P. Besenius, Chem. Commun., 2015, 51, 5253–5256.
31. A. Winter and U. S. Schubert, Chem. Soc. Rev., 2016, 45, 5311–5357.
32. L. Yu, Z. Wang, J. Wu, S. Tu and K. Ding, Angew. Chem., Int. Ed., 2010, 49, 3627–3630.
33. G. R. Whittell, M. D. Hager, U. S. Schubert and I. Manners, Nat. Mater., 2011, 10, 176–188.
34. M. Higuchi, J. Mater. Chem. C, 2014, 2, 9331–9341.
35. D. Taura, S. Hioki, J. Tanabe, N. Ousaka and E. Yashima, ACS Catal., 2016, 6, 4685–4689.
36. L.-J. Chen, H.-B. Yang and M. Shionoya, Chem. Soc. Rev., 2017, 46, 2555–2576.
37. B. Kemper, L. Zengerling, D. Spitzer, R. Otter, T. Bauer and P. Besenius, J. Am. Chem. Soc., 2018, 140, 534–537.
38. A. Sorrenti, J. Leira-Iglesias, A. J. Markvoort, T. F. A. de Greef and T. M. Hermans, Chem. Soc. Rev., 2017, 46, 5476–5490.
39. T. Fukui, S. Kawai, S. Fujinuma, Y. Matsushita, T. Yasuda, T. Sakurai, S. Seki, M. Takeuchi and K. Sugiyasu, Nat. Chem., 2017, 9, 493–499.
40. W. Wagner, M. Wehner, V. Stepanenko, S. Ogi and F. Würthner, Angew. Chem., Int. Ed., 2017, 56, 16008–16012.
41. S. Ogi, V. Stepanenko, J. Thein and F. Würthner, J. Am. Chem. Soc., 2016, 138, 670–678.
42. M. F. J. Mabesoone, A. J. Markvoort, M. Banno, T. Yamaguchi, F. Helmich, Y. Naito, E. Yashima, A. R. A. Palmans and E. W. Meijer, J. Am. Chem. Soc., 2018, 140, 7810–7819.
43. P. A. Korevaar, S. J. George, A. J. Markvoort, M. M. J. Smulders, P. A. J. Hilbers, A. P. H. J. Schenning, T. F. A. De Greef and E. W. Meijer, Nature, 2012, 481, 492–496.
44. S. Ogi, K. Sugiyasu, S. Manna, S. Samitsu and M. Takeuchi, Nat. Chem., 2014, 6, 188–195.
45. J. S. Valera, R. Gomez and L. Sánchez, Small, 2018, 14, 1702437.
46. M. A. J. Gillissen, M. M. E. Koenigs, J. J. H. Spiering, J. A. J. M. Vekemans, A. R. A. Palmans, I. K. Voets and E. W. Meijer, J. Am. Chem. Soc., 2014, 136, 336–343.
47. T. Nakamura, H. Ube and M. Shionoya, Angew. Chem., Int. Ed., 2013, 52, 12096–12100.
48. A. Bellusci, M. Ghedini, L. Giorgini, F. Gozzo, E. I. Szerb, A. Crispini and D. Pucci, Dalton Trans., 2009, 7381–7389.
49. A. Bellusci, A. Crispini, D. Pucci, E. I. Szerb and M. Ghedini, Cryst. Growth Des., 2008, 8, 3114–3122.
50. H. Miyake, K. Yoshida, H. Sugimoto and H. Tsukube, J. Am. Chem. Soc., 2004, 126, 6524–6525.
51. P. Besenius, G. Portale, P. H. H. Bomans, H. M. Janssen, A. R. A. Palmans and E. W. Meijer, Proc. Natl. Acad. Sci. U. S. A., 2010, 107, 17888–17893.
52. J. Tomasi and M. Persico, Chem. Rev., 1994, 94, 2027–2094.
53. M. Shen, Y. Xie, H. F. Schaefer, III and C. A. Deakyne, J. Chem. Phys., 1990, 93, 3379–3388.
54. T. Vijayakumar, I. H. Joe, C. P. R. Nair and V. S. Jayakumar, J. Mol. Struct., 2008, 877, 20–35.
55. M. H.-Y. Chan, M. Ng, S. Y.-L. Leung, W. H. Lam and V. W.-W. Yam, J. Am. Chem. Soc., 2017, 139, 8639–8645.
56. T. F. A. De Greef, M. M. J. Smulders, M. Wolffs, A. P. H. J. Schenning, R. P. Sijbesma and E. W. Meijer, Chem. Rev., 2009, 109, 5687–5754.
57. M. Wehner, M. I. S. Röhr, M. Bühler, V. Stepanenko, W. Wagner and F. Würthner, J. Am. Chem. Soc., 2019, 141, 6092–6107.
58. S. Dhiman, A. Sarkar and S. J. George, RSC Adv., 2018, 8, 18913–18925.
59. M. M. J. Smulders, M. M. L. Nieuwenhuizen, T. F. A. de Greef, P. van der Schoot, A. P. H. J. Schenning and E. W. Meijer, Chem.–Eur. J., 2010, 16, 362–367.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c9sc04958d

---