β-cyclodextrin mediated construction of porous helical nanoribbons from oligoaniline derivatives

β-CD mediated construction of PHNRs

According to our previous report, helical nanoribbons can be prepared from oligoaniline derivatives through the chemical oxidative polymerization of aniline in an i-propanol/water mixture, where right- or left-handed helical nanoribbons of oligoaniline derivatives can be obtained by controlling the i-propanol content of the mixed solution^[11]. Inspired by the controllable self-assembly performance of oligoaniline, and taking into account that oligoaniline can form a host-guest inclusion complex with β-CD, β-CD was employed to further manipulate the supramolecular assembly process of oligoaniline. Specifically, aniline was oxidized using ammonium peroxodisulfate (APS) as an oxidant in an i-propanol/water mixture, where the i-propanol content at 40% was chosen as a typical example for the fabrication of helical nanoribbons of oligoaniline^[12]. At the point of 4 min of polymerization, β-CD, of which the molar amount was kept the same as aniline, was added, and the reaction mixture was left for 12 h at 25 °C. After that, the product was centrifugated and washed with deionized water. Finally, the resulting solid was dried to afford the target black powder.

The morphology of the synthesized product was investigated using field emission scanning electron microscope (FESEM) and transmission electron microscope (TEM). As shown in Figure 1A-C, in the absence of β-CD, highly uniform and dense 1D helical nanoribbons with several microns in length were observed in both FESEM and TEM images, which is in high agreement with our previous report^[11]. By contrast, surprisingly, when mediated by β-CD, the helical nanoribbons obtained were porous [Figure 1D-F and Supplementary Figure 1], with an almost unchanged size. These results indicate that β-CD participated in the assembly process of oligoaniline and played a dominating role in the formation of porous helical nanoribbons (PHNRs).

Figure 1. (A-E) FESEM and (C and F) TEM images of helical nanostructures synthesized (A-C) in the absence of β-CD and (D-F) in the presence of β-CD. FESEM: Field emission scanning electron microscope; TEM: transmission electron microscope; β-CD: β-cyclodextrin.

The time point for the addition of β-CD during the polymerization of aniline was considered to be important for the fabrication of PHNRs. To verify this assumption, synthesis experiments were carried out on the modulation of the addition time of β-CD. The morphologies of the products obtained under different conditions were investigated by TEM. Interestingly, helical nanoribbons bearing large and relative uniform pore channels were observed upon the addition of β-CD at 4 min of polymerization [Figure 2B]. By contrast, the introduction of β-CD at shorter (1 min) [Figure 2A] and longer polymerization times (9 min) [Figure 2C] showed helical nanoribbons featuring small pore size and nonporous morphologies, respectively. According to the above results, it is logical to speculate that the dominating product at 1, 4, and 9 min of polymerization was different, and β-CD exhibited stronger binding capability towards the product formed at 4 min of polymerization of aniline. This deduction was supported by the UV-vis experiments. As shown in Supplementary Figure 2, the addition of 1 equiv. of β-CD at 1 min polymerization induced the observation of a singlet peak at 281 nm, which was ascribed to the monomer aniline. Remarkably, the absorption peak at 281 nm experienced an obvious decrease along with the appearance of a new peak at 374 nm after the addition of 1 equiv. of β-CD at 4 min polymerization. We assigned this new peak as oligoaniline^[12]. The addition of β-CD at a longer polymerization time (9 min) led to the slight diminishment of the aniline absorption and gradual enhancement of oligoaniline. This result not only illustrates the binding events between β-CD and oligoaniline formed at different polymerization times but also substantiates the importance of the additional time point of β-CD to the construction of porous nanostructures. Thus, the colors of the solutions obtained at different polymerization stages were also observed. The color changed from pale yellow to dark brown over time, as shown in Figure 2A-C (inset), which clearly confirmed the different polymerization degrees.

Figure 2. (A-C) (Left) Photographs of the polymerization mixture at the polymerization times of: (A) 1 min; (B) 4 min; and (C) 9 min. (Right) TEM images of β-CD mediated nanostructures assembled from oligoaniline for 12 h with the addition of β-CD at different polymerization times: (A) 1 min; (B) 4 min; and (C) 9 min. (A and B) (Inset) Magnified TEM images. (D-F) High-resolution mass spectra (HRMS with ESI mode) of the polymerization mixture at the polymerization times of: (D) 1 min; (E) 4 min; and (F) 9 min. Note that the aniline monomer and its oligomers (oligoaniline derivatives), such as dimer, trimer, tetramer, pentamer, hexamer, heptamer, and octamer, are carefully labeled in (D-F). TEM: Transmission electron microscope; β-CD: β-cyclodextrin; HRMS: high-resolution mass spectrometry.

To further verify the possible structure of the oligoaniline obtained at different polymerization times, the mixtures from different polymerization times were investigated using high-resolution mass spectrometry (HRMS with ESI mode; Figure 2D-F and Supplementary Figure 3, Supplementary Table 1). The mass spectrum showed multiple ion peaks consistent with various oligoaniline derivatives for the sample that was polymerized for 1 min [Figure 2D]. Typically, an intense ion peak at m/z = 290.1283, corresponding to the mono-charged species of [tri-aniline+H]⁺, was detected in addition to the free aniline ion peak at m/z = 94.0669. The ion peak of [tri-aniline+H]⁺ evolved into the most intense one after 4 min of polymerization, indicating that the tri-aniline derivative C₁₈H₁₆N₄ was the main component at this polymerization stage [Figure 2E]. However, when the polymerization time reached 9 min, an ion peak at m/z = 778.0974, which could be assigned to the octa-aniline derivative, appeared [Figure 2F]. Based on the above results, we assume that at the early polymerization time (1 min), a lower amount of tri-aniline oligomer was produced. When β-CD was added at this time point, a lower degree of β-CD/tri-aniline oligomer complex was formed, leading to less involvement of β-CD in the polymerization process.

At the polymerization time of 4 min, a higher amount of tri-aniline oligomer was obtained, which led to the relative concentrated formation of β-CD/tri-aniline oligomer complex and, thus, a higher engagement of β-CD during the polymerization. However, at a longer polymerization time (9 min), the product with a higher polymerization degree, namely the octa-aniline derivative, was too bulky to interact with β-CD, which led to a lower engagement of β-CD as well. In addition, the molar ratio of β-CD/aniline was modulated. As shown in Supplementary Figure 4, increasing the amount of β-CD led to a greater number of pores or even disruption of the helical nanoribbons. Above all, the involvement of β-CD in the initial polymerization process was considered to be the key factor for obtaining porous helical ribbons.

Morphological evolution of PHNRs

As discussed above, PHNRs with large pores were observed upon the addition of β-CD after 4 min polymerization of aniline; therefore, to clarify the role that β-CD played in the fabrication of PHNRs, this optimal condition was selected to monitor the formation process of PHNRs [Figure 3 and Supplementary Figure 5]. Figure 3 shows the nanostructures assembled from β-CD mediated oligoaniline derivatives at different time intervals. Spindle nanoflakes could be obtained after 1 h of polymerization. It is worth noting that the nanoflakes started to split and afford the nanofiber structures after 3 h. When the polymerization time was extended to 5 or 9 h, clusters of twisted nanoribbons, composed of entangled nanoribbons, were observed. Surprisingly, helical nanoribbons bearing multiple pore channels, namely PHNRs 100-150 nm in width and 200-250 nm in average pitch, were formed after 12 h of polymerization. As evidenced by the SEM and TEM images, we measured the pore sizes of the porous helical nanoribbons in the range of 15-35 nm. Based on these observations, we hypothesize that β-CD was involved in the early polymerization process to a certain extent and directed the formation of PHNRs through the abscission from the helical nanoribbons.

Figure 3. FESEM images of the nanostructures assembled from β-CD mediated oligoaniline derivatives at different polymerization times: (A) 1 h; (B) 3 h; (C) 5 h; (D) 9 h; and (E) 12 h. (F) A cartoon image of PHNR. FESEM: Field emission scanning electron microscope; β-CD: β-cyclodextrin; PHNR: porous helical nanoribbon.

To investigate the size effect of the β-CD on the formation of the porous helical nanoribbons, we selected its analogs, α-CD and γ-CD, to mediate the nanostructure formation under the same conditions. As shown in Supplementary Figure 6, the addition of 1 equiv. of α-CD at the 4 min polymerization time point along with polymerization for 12 h induced the observation of porous helical nanoribbons with smaller pore sizes compared to those obtained by β-CD. In contrast, neither porous character nor helical nanostructures were detected in the presence of 1 equiv. of γ-CD. Taken together, these observations demonstrated that α-CD and β-CD displayed similar binding affinity towards the oligoaniline, while negligible binding existed in the case of γ-CD due to its larger size. However, γ-CD also damaged the polymerization of aniline, which eventually led to the disappearance of helical nanoribbons.

To investigate the helical sense discrepancy of the helical nanostructures, we performed circular dichroism experiments in an i-propanol/water solvent mixture [Supplementary Figure 7]. The helical nanoribbons formed in the absence of β-CD showed a positive Cotton effect at 202 nm, which is in agreement with the literature^[11]. By contrast, the porous helical nanoribbons mediated by β-CD exhibited a positive Cotton signal with a blue-shift at 196 nm. This finding demonstrates that the emergence of the porous features did not affect the helical sense of the porous nanoribbons but enhanced the intermolecular π-π interaction to a certain extent.

Formation mechanism of PHNRs

To test the above formation hypothesis, fourier-transform infrared spectroscopy (FTIR) measurements were conducted. As shown in Figure 4, the spectrum of β-CD showed a typical -OH band at 3421 cm^-1 (Figure 4F, left), and the spectrum of helical nanoribbons assembled from oligoanilines, obtained with 12 h polymerization in the absence of β-CD, showed two characteristic bands at 1572 and 1509 cm^-1, corresponding to C=C stretching vibration in quinonoid and benzenoid rings, respectively. In addition, the peak located around 1352 cm^-1 could be ascribed to the C-N stretching mode of benzoquinone imine (Figure 4A, left). It is worth noting that, after the introduction of β-CD into the oligoaniline solution for 1 h, the three mentioned FTIR adsorption peaks of oligoaniline were slightly shifted to lower wavenumbers with reduced intensity compared to free oligoaniline (Figure 4B, left), indicating the establishment of a supramolecular complex between oligoaniline and β-CD. However, with the extension of the polymerization time, the specific bound bands of both oligoaniline and β-CD started to diminish (Figure 4C and D, left) and completely disappeared after 12 h (Figure 4E, left). Moreover, the characteristic bands of free oligoaniline reappeared simultaneously, which indicated the disassociation of the β-CD/oligoaniline complex.

Figure 4. (Left) FTIR spectra and (Right) selected ¹H NMR spectra of (A) oligoanilines obtained with 12 h polymerization in the absence of β-CD. β-CD mediated polymerization product at various polymerization times: (B) 1 h; (C) 3 h; (D) 5 h; and (E) 12 h. (F) β-CD. FTIR: Fourier-transform infrared spectroscopy; NMR: nuclear magnetic resonance; β-CD: β-cyclodextrin.

To further disclose the underlying mechanism of the β-CD mediated formation of PHNRs, the ¹H NMR spectra of β-CD, helical nanoribbons assembled from oligoanilines obtained with 12 h polymerization in the absence of β-CD, and the β-CD mediated polymerization product at various polymerization times in deuterated DMSO solution were obtained. Note that the samples taken from the polymerization system were centrifuged and washed with water to remove free aniline, APS, and β-CD. As shown in Figure 4 (right), when β-CD was added into the polymerization system for 1 h, the hydroxyl protons of β-CD, located in the range of 4-5 ppm, from the resulting product experienced a slight downfield shift in comparison with that of free β-CD, an indication of the formation of the β-CD/oligoaniline complex. However, with the increase of the polymerization time, the signals arising from β-CD were largely attenuated for the β-CD mediated polymerization product and completely disappeared after 12 h. The above results further verify the abscission of β-CD from the oligoaniline assemblies with the extension of polymerization time, which was considered the crucial reason for the formation of PHNRs.

Taken together, the formation mechanism of the PHNRs was proposed as follows: A supramolecular complex was formed between β-CD and tri-aniline at the early stage of polymerization and involved in the assembly of spindle nanoflakes [Figure 3A]. However, with the reorganization of the oligoaniline assemblies over time, the abscission of β-CD from the helical nanoribbons was observed, which eventually induced the formation of PHNRs.