Self-assembly and phase separation of amphiphilic dyads based on 4,7-bis(2-thienyl)benzothiodiazole and perylene diimide

Abstract

To get highly ordered organic structures at the nanoscale, a series of new electronic donor–acceptor (D–A) dyads were synthesized. The dyads bearing different side chains (lipophilic or amphiphilic) or linkers (Long or Short) showed variable self-assembly behaviour. Long-linker dyads can fold in dilute solution, but the folding of short-linker dyads was not observed. Intermolecular D–A interactions and acceptor–acceptor (A–A) aggregation were proved to co-occur in chloroform for all four dyads. In the bulk state, amphiphilic dyads could overcome the intrinsic D–A interactions and achieve better D–A phase separation than lipophilic ones due to the incompatibility of the side chains. The dyad with a short-linker and amphiphilic side chains, S_amphi, had the ability to form a gel, and both of the amphiphilic dyads could form nanofibres.

---

Introduction

In bulk heterojunction organic photovoltaic devices, bicontinuous arrays of electron donors and acceptors are a prerequisite.¹ As an inherent property of organic semiconductors, the exciton diffusion length is ca. 5–10 nm.² In traditional bulk heterojunction devices, donor and acceptor materials are blended with insufficient microphase separation, which results in exciton loss.³ To maximize the heterojunction interface, researchers paid much attention to small molecule dyads consisting of electron donor (i.e. p-type semiconductors) and electron acceptor (i.e. n-type semiconductors) units,^4,5 which are promising for optimizing charge separation and charge transportation processes.

To link electron donors and acceptors together, various linkages including soft alkyl chains,^5a rigid phenylene acetylene,^5b coordination bonding and hydrogen-bonding⁶ were used. Linkers are of key importance in dyads, since they affect the donor–acceptor (D–A) distance and D–A interface. However, the donors tend to stack with acceptors, forming charge transfer complexes that trap excitons.⁷ To induce D–A separation at the nanoscale and drive the formation of well-defined nano-morphology, incompatible wedges (e.g. dodecyl and triethylene glycol (TEG) chains^5a,b) were equipped at each end of the amphiphilic dyads.^4,5 The cooperativity of π–π stacking and incompatibility of the side chains enables the phase separation of both electron donors and acceptors.

In the present contribution, a series of new D–A dyads containing 4,7-bis(2-thienyl)-2,1,3-benzothiodiazole (T2BTZ) and perylene diimide (PDI) were synthesized (Fig. 1). A benzothiodiazole unit was incorporated to lower the LUMO level and red-shift the absorption onset,⁸ compared with the short-wavelength absorption (ca. 500 nm onset position) of commonly used oligothiophene.^5a,b Incompatible triethylene glycol and dodecyl chains were utilized to improve phase separation (S_amphi and L_amphi). The corresponding dyads with dodecyl chains on both sides (S_lipo and L_lipo) were synthesized as reference compounds. The independent donor (T2BTZ) and acceptor (PDI) were also prepared as reference compounds. Moreover, two amino acids were used to tune the distance between the donor and acceptor. The aliphatic linkers would avoid conjugated electron coupling between both parts.

Fig. 1 Chemical structures of four dyads and reference compounds. The donor and acceptor are 4,7-bis(2-thienyl)benzothiodiazole (T2BTZ) and perylene diimide (PDI), respectively.

The influence of side chains and linker length on assembly behaviour, phase separation and nano-morphology was systematically investigated. Intramolecular D–A interaction of long-linker dyads was first observed in dilute solution. Intermolecular D–A interactions and A–A aggregation were both observed in concentrated solution. The dyads could achieve D–A phase separation assisted by the amphiphilic side chains, and the assembly morphologies differed with side chains and linker length. To the best of our knowledge, the linker length influence on dyad assembly and intramolecular D–A interactions has rarely been studied.⁹

Results and discussion

Self-assembly in solution

UV-visible spectroscopy was carried out to study the ground state electronic properties of the dyads in solution. S_lipo and S_amphi showed identical absorption, suggesting that the electronic properties of the donor and the acceptor are not influenced by the side chains. The spectra of S_lipo and S_amphi totally overlapped with the spectra summation of the reference compounds PDI and T2BTZ (Fig. 2a), demonstrating no electronic interaction between donor and acceptor (Fig. 2c). However, after the 0–0 transition peaks belonging to the acceptor component were normalized, L_lipo and L_amphi had relatively higher 0–1 transition intensities than S_lipo and S_amphi, along with red-shifted onset wavelengths¹⁰ (Fig. 2b, black arrows). Based on this result, we proposed that in L_lipo and L_amphi, T2BTZ may fold back to interact with PDI intramolecularly due to the flexibility of the long linker (Fig. 2c). To rule out the possibility of intermolecular aggregation, concentration-dependent UV-vis spectra of L_lipo and L_amphi were recorded (Fig. S1†). The result confirmed that the spectral change indeed originated from intramolecular folding instead of intermolecular aggregation. Furthermore, the 0–1 transition intensity of L_lipo is slightly higher than that of L_amphi (Fig. 2b, red arrow), and this happens at several selected concentrations from 10⁻⁶ M–10⁻⁵ M (Fig. S1†). We tentatively explain it as follows: when folding happens, the side chains on both sides approach each other intramolecularly. For L_lipo, the affinity of the dodecyl chains induced folding of the D–A unit, while for L_amphi, the incompatibility of the dodecyl and triethylene glycol chains slightly inhibited folding. As far as we know, this is the first report on PDI–T2BTZ interactions and intramolecular side chain incompatibility, characterized by absorption spectroscopy in dilute solution.

Fig. 2 (a) UV-visible absorption spectra of short-linker dyads and reference compounds; (b) normalized absorption spectra of short-linker and long-linker dyads (chloroform, 5 × 10⁻⁶ M); (c) schematic diagram of intramolecular folding of short-linker and long-linker dyads.

To understand the assembly behaviour of the donor and acceptor segments in solution, all four dyads were studied by concentration-dependent ¹H NMR in CDCl₃ (Fig. 3). Taking S_amphi as an example and diluting its CDCl₃ solution, we found that the peaks of the PDI and T2BTZ protons moved downfield and the fine structure became clear (Fig. 3, black arrows), indicating the assembly of S_amphi. Moreover, detailed evidence for assembly was found from the upfield chemical shifts (Fig. 3, red arrow) of protons a and b attached to PDI. Through chemical structural analysis, we propose that proton a forms a weak hydrogen bond with the nearby carbonyl oxygen in the aggregates, which leads to a lower chemical shift of proton a. As the solution was diluted, the aggregates slowly dissociated, enabling free rotation of the phenyl ring. Therefore, the hydrogen bond was weakened and the chemical shift of proton a moved upfield. Meanwhile, proton b had a slight upfield shift due to the electronic effect from a. These phenomena demonstrate that self-assembly of S_amphi exists in chloroform at this concentration range (ca. 1 mM to 10 mM). The other dyads showed similar chemical shift changes when diluting their CDCl₃ solutions (Fig. S7–S10†), indicating the occurrence of self-assembly.

Fig. 3 Molecular structure of S_amphi and the concentration-dependent ¹H NMR spectra of S_amphi in CDCl₃.

To further investigate the assembly details, a ¹H NMR dilution experiment of the reference compounds PDI and T2BTZ and a mixture of both compounds was carried out in CDCl₃. The chemical shifts of the most downfield PDI proton (c) and T2BTZ proton (d) at four different concentrations (30 mM to 9 mM) are summarized in Fig. 4 (see the spectra in Fig. S7–S13†). The δ of proton c changed by a maximum value (Δδ) of 0.09 indicating that PDI can self-assemble. Proton d only had a Δδ value of 0.013, which indicates that the self-assembly tendency of T2BTZ is weak. However, the Δδ of proton d increased significantly (Δδ = 0.09) when PDI and T2BTZ were mixed together. It means that T2BTZ has a tendency to co-assemble with PDI. In the four dyads, the Δδ of proton d was apparently larger than 0.013 (Table S1†), so the D–A co-assembly existed. Therefore, it is necessary to introduce different side chains to overcome the D–A interactions in order to achieve phase separation.

Fig. 4 Selected proton chemical shift changes of PDI (blue triangle), T2BTZ (black square) and their equimolar mixture (purple inverted triangle for PDI and red circle for T2BTZ) studied by ¹H NMR experiments. (CDCl₃, 30 mM to 9 mM, 25 °C).

Morphology control and phase separation

The active layer morphology is of paramount importance for organic semiconductor devices (OSCs, OFET, etc.).¹¹ Herein, the morphology and phase separation, which were influenced by side chains, linkers and aromatic π–π interactions, were investigated.

Since all four dyads had assembly abilities as identified by ¹H NMR, we were interested in their bulk self-assembly behaviour. The differential scanning calorimetry (DSC) of L_lipo and L_amphi showed only one peak for the first-order phase transition, and polarized optical microscopy (POM) showed that no liquid crystal state existed. L_amphi had a slightly higher T_m (179 °C) than L_lipo (166 °C) (Fig. S4†). After thermal annealing by DSC, L_lipo and L_amphi were used to conduct small angle X-ray scattering (SAXS) experiments. For L_lipo, a strong scatter peak was observed at q₁ = 1.36 nm⁻¹, and weak signals of 2q₁ and 3q₁ could also be observed, which correspond to a repeating length of d = 4.63 nm (Fig. 5a). On the other hand, diffraction peaks were observed at a smaller angle (q₁ = 0.59 nm⁻¹, 2q₁ = 1.18 nm⁻¹) for L_amphi (Fig. 5b), indicating longer repeating units of d = 10.6 nm. In MM2 force field optimization, the molecular length of L_lipo and L_amphi were both 6.4 nm. We can deduce that a layered periodic structure exists in both compounds. The d spacing of L_lipo was close to the molecular length indicating a repeating unit of one single molecule (Fig. 6, left), while the d spacing of L_amphi was almost twice as long as the molecular length corresponding to the repeating unit of two molecules, which indicates that phase separation of TEG and alkyl chains occur in L_amphi (Fig. 6, right). The observed length was smaller than the real value because of the overlapping of the alkyl chains. These results demonstrate that incompatible side chains assist D–A dyads to achieve donor–acceptor phase separation. For S_lipo and S_amphi, the DSC showed no phase transition before they were decomposed, so thermal treatment is not ideal for the annealing process (see DSC curves in Fig. S5†). Thus, solvent annealing was adopted. The dyad S_amphi formed a gel during the slow solvent evaporation of its solution in a dichloromethane (DCM)–hexane (v/v = 1/1) mixture, and the S_lipo solution remained clear under the same conditions (Fig. S6†). After evaporating the residual solvent under vacuum, the solid (S_lipo) and dried gel (S_amphi) were used to conduct a SAXS experiment. One scattering peak was obtained for each dyad (q = 1.40 nm⁻¹ for S_lipo, q = 0.83 nm⁻¹ for S_amphi). The smaller q value for S_amphi indicated similar phase separation behaviour with long-linker dyads. However, the absence of high-order scattering demonstrated poor phase separation results or improper assembly conditions (Fig. 5c and d).

Fig. 5 Small angle X-ray scattering of L_lipo (a), L_amphi (b), S_lipo (c), S_amphi (d) samples. L_lipo and L_amphi were treated by DSC, and S_lipo and S_amphi were treated by solvent annealing.

Fig. 6 Schematic phase separation diagram of long-linker dyads; L_lipo has a multilamellar structure (left) and L_amphi has a bilayer structure (right).

According to scanning electron microscopy, S_lipo, S_amphi and L_amphi form fibres while L_lipo formed nanoparticles after a heating–cooling cycle in chloroform–MeOH (v/v = 1/1) at a concentration of 6 × 10⁻⁴ M (Fig. S14†).

To explore the assembly in detail, the concentration was lowered to 6 × 10⁻⁵ M. After heating and then cooling, the assemblies were observed by transmission electron microscopy (Fig. 7). In an alcohol-containing solvent, amphiphilic dyads can form a more regular nanofibre structure. S_amphi formed thin fibres with a width of around 10 nm, which is about twice the length of the rigid core (3.5 nm). It's in agreement with the model proposed in the SAXS experiment. In contrast, without the regulation of side chains, S_lipo could only form thick fibres. However, L_lipo, which differed only in linker length from the S_lipo dyad, formed nanoparticles with a diameter around 50 nm. For L_amphi with a longer molecular length than S_amphi, nanofibres narrower than 10 nm were observed. We speculate the reasons for different morphology formation are as follows. At least two factors contribute to the morphology differences: (1) the properties of the side chains (amphiphilic or lipophilic); (2) the linker length of the dyads (long-linker ones can fold as evidenced by UV-vis absorption spectra, thus are more flexible than short-linker ones). The incompatible side chains enable amphiphilic dyads to form thin fibres. The thinner fibres for L_amphi may be attributed to the flexible long linker. Due to the lipophilic affinity of side chains and relative rigid molecular structure, S_lipo formed thick fibres with a diameter larger than 50 nm. For L_lipo, the flexible long linker allows the D–A molecules to fold, and the side-by-side intermolecular lipophilic interactions were disrupted, resulting in discrete nanoparticles.

Fig. 7 Transmission electron microscopy images of four dyads.

Cyclic voltammetry

To verify whether the energy levels of the donor and acceptor match, cyclic voltammetry was carried out. The data are summarized in Table 1 (see the redox curve in Fig. S3†). The HOMO of T2BTZ and LUMO of PDI were determined by the onset of the CV curve relative to ferrocene. Their corresponding LUMO or HOMO was calculated from the optical bandgap obtained from the UV-vis absorption spectra. The short-linker dyads had HOMO and LUMO values nearly identical to T2BTZ and PDI, respectively, which also demonstrated the independent D–A electronic properties in the dyads. The LUMO difference between the donor and acceptor was 0.62 eV (Table 1), which was sufficient to provide a driving force for charge separation. The intramolecular quenching of T2BTZ fluorescence by PDI also indicates the matched energy levels (Fig. S2†). Therefore, reasonable energy levels enable the dyads to be promising materials in organic photovoltaics.

Table 1 Redox potentials and energy levels of short-linker dyads and reference compounds in DCM at 1 × 10⁻⁵ M

Compound	φox/V	φred/V	Optical bandgap^a/eV	HOMO/eV	LUMO/eV
a Calculated from the onset of absorption spectra (λonset).^12
Slipo	1.14	−0.49		−5.51	−3.38
Samphi	1.08	−0.50		−5.45	−3.87
PDI	1.53	−0.51	2.22	−6.08	−3.86
T2BTZ	1.04	−0.80	2.17	−5.41	−3.24
FeCp2	0.43	—

---

Conclusions

Lipophilic and amphiphilic dyads were synthesized and unambiguously characterized. Their self-assembly behaviour was studied from dilute solutions to the bulk state.

In chloroform, no obvious difference in assembly behaviour between amphiphilic and lipophilic dyads was observed, probably due to the good solubility of these dyads. In dilute solutions (10⁻⁶ to 10⁻⁵ M), only long-linker dyads showed intramolecular D–A interactions. However, in concentrated solutions (10⁻³ to 10⁻² M), all four dyads showed D–A interactions and A–A aggregation. Among them, short-linker dyads showed stronger intermolecular interactions, evidenced by ¹H NMR, maybe because the short-linker enhanced the intermolecular cooperativity of the D–A interaction and A–A aggregation.

Different from its assembly behaviour in solution, in the bulk state, with the assistance of amphiphilic chains, L_amphi can overcome D–A interactions and achieve better nanoscale D–A phase separation compared with L_lipo. In contrast, S_amphi has poor phase separation but good crystallinity. Therefore, L_amphi is more suitable for film formation processes. Under the same aggregate growth conditions, the assembly morphology of the four dyads is as follows: both of the amphiphilic dyads could form thin nanofibres (Φ ≤ 10 nm), while the lipophilic dyads form thick fibres or nanoparticles (Φ ≈ 50 nm). The phase separation and morphology results provide guidance for organic device fabrication. Photovoltaic device performance of the dyads is under investigation in our lab.

Experimental

Materials

All the chemicals were purchased commercially. The oxygen and moisture sensitive reactions were performed under a nitrogen atmosphere using standard Schlenk techniques. Tetrahydrofuran (THF) was freshly distilled with sodium under a nitrogen atmosphere prior to use. N,N-Dimethylformamide (DMF) was stirred overnight with CaH₂, distilled under vacuum and then dried over molecular sieves.

Instrumentation

¹H and ¹³C NMR spectra were collected on a Mercury plus 300 (300 MHz) or Bruker Avance 400 (400 MHz) spectrometer, using CDCl₃ or d₆-DMSO as solvents. Chemical shifts are reported in parts per million (ppm) and coupling constants are reported in Hertz (Hz). ¹H NMR chemical shifts were referenced to the residual solvent peak (7.26 ppm in CDCl₃ or 2.50 ppm in d₆-DMSO). ¹³C NMR chemical shifts were referenced to CDCl₃ (77.0 ppm). Electron ionization (EI) mass spectra were collected on a VG ZAB-HS mass spectrometer. Elemental analyses were performed using a German Vario EL III elemental analyzer. Electro-spray ionization (ESI) mass spectrometry was performed using a Bruker Apex IV FTMS instrument. Matrix-Assisted Laser Desorption/Ionization Time of Flight Mass Spectrometry (MALDI-TOF-MS) was performed using an ABI 4800 Plus MALDI TOF/TOF™ Analyzer. The UV-vis absorption spectra were recorded using a Hitachi U-4100 spectrophotometer with a 1 cm quartz cell. The photoluminescence spectra were recorded on a Horiba Jobin Yvon FluoroMax-4P spectrofluorometer with right-angle geometry, using a 1 cm quartz cuvette. Cyclic voltammetry was carried out by using BASI Epsilon workstation. The compounds were detected in DCM (freshly distilled over CaH₂ prior to use) containing 0.1 M n-Bu₄NPF₆ as an electrolyte. The working electrode was a glassy carbon electrode and the counter electrode was a platinum sheet electrode. The redox potentials were recorded versus a Ag/AgCl electrode (reference electrode). Differential scanning calorimetry was carried out on a METTLER TOLEDO DSC 822 with a programmed heating procedure under nitrogen. Small-angle X-ray scattering experiments were carried out on a SAXSess high-flux small-angle X-ray scattering instrument (Anton Paar), equipped with a Kratky block-collimation system, at room temperature. The X-ray was generated by a sealed-tube X-ray generator (Philips PW3830) with a Cu target (λ: 0.1542 nm). The power of the generator used for measurement was 40 kV and 40 mA. The X-ray intensities were recorded on an imaging-plate detection system with a pixel size of 42.3 × 42.3 μm². The distance from the sample to the detector was 264.5 mm and the exposure time was 600 s. The Scanning Electron Microscopy (SEM) experiment was carried out on a Hitachi S4800 Cold Field Emission Scanning Microscope. Transmission electron microscopy was carried out on a JEOL JEM-2100F field-emission high resolution transmission electron microscope.

Synthesis

The p-type (Scheme 1) and n-type (Scheme 2) segments were synthesized separately and then connected with amino acid linkers (Scheme 3). Synthetic details of all compounds were summarized in the ESI.†

Scheme 1 Reagents and conditions for the synthesis of the p-type segment: (a) n-C₁₂H₂₅Br, K₂CO₃, KI, acetone, reflux, 83%; (b) bis(pinacolato)diboron, PdCl₂(dppf), KOAc, DMF, 60 °C, 74%; (c) N-bromosuccinimide, CHCl₃–AcOH, 73%; (d) Pd(PPh₃)₄, K₂CO₃ (2 M), THF, 80 °C, 76%; (e) LiAlH₄, THF, 89%.

Scheme 2 Reagents and conditions for the synthesis of the n-type segments: R = R₁, (a) n-C₁₂H₂₅Br, K₂CO₃, DMF, 80 °C, 91%; (b) N₂H₄·H₂O, EtOH–THF, reflux, 82%; (c) n-4, 4-dimethylaminopyridine, imidazole, 130 °C, 79%; (d) p-toluenesulfonic acid monohydrate, 95 °C, toluene, 88%. R = R₂, (a) p-toluenesulfonyl triethylene glycol monomethyl ether, K₂CO₃, DMF, 90 °C, 47%; (b) N₂H₄·H₂O, EtOH, reflux, 91%; (c) n-4, 4-dimethylaminopyridine, imidazole, 130 °C, 75%; (d) p-toluenesulfonic acid monohydrate, 95 °C, toluene, 88%.

Scheme 3 Reagents and conditions for the synthesis of the four dyads: R = R₁, (a) imidazole, DMF, 95 °C, H₂N(CH₂)₅CO₂H, 89%; (b) p-7, diisopropyl azodicarboxylate, PPh₃, THF, rt, 30 min, 54%; (c) imidazole, DMF, 95 °C, glycine, 92%; (d) (COCl)₂, DMF, DCM, rt, 2 h; 4-nitrophenol, Et₃N, DCM, overnight, 91%; (e) p-7, 4-dimethylaminopyridine, DMF, 72 h, 53%. R = R₂, (a) imidazole, DMF, 95 °C, H₂N(CH₂)₅CO₂H, 89%; (b) p-7, Et₃N, DCM, 2-chloro-1-methylpyridinium iodide (Mukaiyama's salt), 34%. (c) Imidazole, DMF, 95 °C, glycine, 62%; (d) (COCl)₂, DMF, DCM, rt, 2 h; 4-nitrophenol, Et₃N, DCM, overnight, 47%; (e) p-7, DCM, DMF, 72 h, 69%.

p-7 was synthesized from p-1 and p-2 in 5 steps. The diphenol p-1 and aldehyde p-2 were prepared according to the literature,¹³ and then p-1 was treated through alkylation and Miyaura boration to afford p-4. p-6 was obtained via a Suzuki coupling of p-4 and p-5. The reduction of p-6 by LiAlH₄ gave the final p-7(Scheme 1).

n-7 and n-8 were afforded by the unsymmetrical functionalization of perylene diimide. Starting materials n-1 and n-2 were synthesized according to the literature.¹⁴ Alkylation of n-1 followed by Gabriel synthesis afforded the aniline derivative n-4, which can be condensed with anhydride n-2. The perylene monoimide diester was treated with p-toluenesulfonic acid to afford another anhydride, n-6. n-6 was condensed with 6-aminocaproic acid or glycine to obtain n-7 or n-8, respectively.

Various condensation methods were tested in order to link both segments. The Mitsunobu reaction and Mukaiyama condensation were finally applied for L_lipo and L_amphi, respectively (Scheme 3). Carboxylic acid n-8 has lower condensation reactivity than n-7. The possible reasons were proposed as follows: (1) n-8 bearing a glycine linker has extremely poor solubility in common solvents (THF, CHCl₃, DCM, toluene, etc.); (2) the hydroxyl group of n-8 may be deactivated by forming intramolecular hydrogen bonds with one of the carbonyl oxygens of a diimide unit. To overcome these disadvantages, we utilised transesterification.¹⁵ The active ester n-9, prepared from acid chloride, has an enhanced solubility, which allows a mild conversion to obtain the target compounds S_lipo and S_amphi (Scheme 3).

Acknowledgements

This research was financially supported by National Natural Science Foundation (no. 21074004, 91227202), the Ministry of Science and Technology (no. 2013CB933501), and the Ministry of Education (NCET-08-0005) of China.

Notes and references

1. (a) S. Günes, H. Neugebauer and N. S. Sariciftci, Chem. Rev., 2007, 107, 1324; (b) W. Ma, A. Gopinathan and A. J. Heeger, Adv. Mater., 2007, 19, 3656; (c) H. Hoppe and N. S. Sariciftci, J. Mater. Chem., 2006, 16, 45.
2. (a) D. R. Kozub, K. Vakhshouri, S. V. Kesava, C. Wang, A. Hexemerb and E. D. Gomez, Chem. Commun., 2012, 48, 5859; (b) P. E. Shaw, A. Ruseckas and I. D. W. Samuel, Adv. Mater., 2008, 20, 3516.
3. G. Yu, J. Gao, G. C. Hummelen, F. Wudl and A. J. Heeger, Science, 1995, 270, 1789.
4. (a) Y. Yamamoto, T. Fukushima, Y. Suna, N. Ishii, A. Saeki, S. Seki, S. Tagawa, M. Taniguchi, T. Kawai and T. Aida, Science, 2006, 314, 1761; (b) Y. Yamamoto, T. Fukushima, A. Saeki, S. Seki, S. Tagawa, N. Ishii and T. Aida, J. Am. Chem. Soc., 2007, 129, 9276; (c) P. Samorì, X. Yin, N. Tchebotareva, Z. Wang, T. Pakula, F. Jäckel, M. D. Watson, A. Venturini, K. Müllen and J. P. Rabe, J. Am. Chem. Soc., 2004, 126, 3567; (d) C.-L. Wang, W.-B. Zhang, C.-H. Hsu, H.-J. Sun, R. M. Van Horn, Y. Tu, D. V. Anokhin, D. A. Ivanov and S. Z. D. Cheng, Soft Matter, 2011, 7, 6135; (e) T. N. Y Hoang, D. Pociecha, M. Salamonczyk, E. Gorecka and R. Deschenaux, Soft Matter, 2011, 7, 4948.
5. (a) W.-S. Li, Y. Yamamoto, T. Fukushima, A. Saeki, S. Seki, S. Tagawa, H. Masunaga, S. Sasaki, M. Takata and T. Aida, J. Am. Chem. Soc., 2008, 130, 8886; (b) W.-S. Li, A. Saeki, Y. Yamamoto, T. Fukushima, S. Seki, N. Ishii, K. Kato, M. Takata and T. Aida, Chem. – Asian J., 2010, 5, 1566; (c) R. Charvet, Y. Yamamoto, T. Sasaki, J. Kim, K. Kato, M. Takata, A. Saeki, S. Seki and T. Aida, J. Am. Chem. Soc., 2012, 134, 2524; (d) F. Würthner, Z. Chen, F. J. M. Hoeben, P. Osswald, C.-C. You, P. Jonkheijm, J. v. Herrikhuyzen, A. P. H. J. Schenning, P. P. A. M. van der Schoot, E. W. Meijer, E. H. A. Beckers, S. C. J. Meskers and R. A. J. Janssen, J. Am. Chem. Soc., 2004, 126, 10611.
6. (a) M. E. El-Khouly, A. M. Gutiérrez, Á. Sastre-Santos, F. Fernández-Lázaro and S. Fukuzumi, Phys. Chem. Chem. Phys., 2012, 14, 3612; (b) T. K. Khan and M. Ravikanth, Tetrahedron, 2012, 68, 830; (c) P. Piotrowiak, Chem. Soc. Rev., 1999, 28, 143; (d) F. D'Souza and O. Ito, Chem. Commun., 2009, 4913; (e) G. Bottari, G. de la Torre, D. M. Guldi and T. Torres, Chem. Rev., 2010, 110, 6768.
7. (a) D. Markovitsi, H. Bengs and H. Ringsdorf, J. Chem. Soc., Faraday Trans., 1992, 88, 1275; (b) E. H. A. Beckers, S. C. J. Meskers, A. P. H. J. Schenning, Z. Chen, F. Würthner, P. Marsal, D. Beljonne, J. Cornil and R. A. J. Janssen, J. Am. Chem. Soc., 2006, 128, 649; (c) W. Pisula, M. Kastler, D. Wasserfallen, J. W. F. Robertson, F. Nolde, C. Kohl and K. Müllen, Angew. Chem., Int. Ed., 2006, 45, 819.
8. J.-L. Wang, Q. Xiao and J. Pei, Org. Lett., 2010, 12, 4164.
9. M. Ince, M. V. Martínez-Díaz, J. Barberá and T. Torres, J. Mater. Chem., 2011, 21, 1531.
10. W. Wang, L.-S. Li, G. Helms, H.-H. Zhou and A. D. Q. Li, J. Am. Chem. Soc., 2003, 125, 1120.
11. (a) A. M. Ballantyne, T. A. M. Ferenczi, M. Campoy-Quiles, T. M. Clarke, A. Maurano, K. H. Wong, W. M. Zhang, N. Stingelin-Stutzmann, J. S. Kim, D. D. C. Bradley, J. R. Durrant, I. McCulloch, M. Heeney, J. Nelson, S. Tierney, W. Duffy, C. Mueller and P. Smith, Macromolecules, 2010, 43, 1169; (b) M. M. Ling and Z. Bao, Chem. Mater., 2004, 16, 4824.
12. Y.-J. Cheng, S.-H. Yang and C.-S. Hsu, Chem. Rev., 2009, 109, 5868.
13. (a) W.-w. Mao, T.-t. Wang, H.-p. Zeng, Z.-y. Wang, J.-p. Chen and J.-g. Shen, Bioorg. Med. Chem. Lett., 2009, 19, 4570; (b) T.-W. Zeng, C.-C. Ho, Y.-C. Tu, G.-Y. Tu, L.-Y. Wang and W.-F. Su, Langmuir, 2011, 27, 15255.
14. (a) M. Peterca, M. R. Imam, C.-H. Ahn, V. S. K. Balagurusamy, D. A. Wilson, B. M. Rosen and V. Percec, J. Am. Chem. Soc., 2011, 133, 2311; (b) A. Wicklein, M.-A. Muth and M. Thelakkat, J. Mater. Chem., 2010, 20, 8646.
15. J. Hongrapipat, P. Kopečková, S. Prakongpan and J. Kopeček, Int. J. Pharm., 2008, 351, 259.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental procedures, details of spectroscopic and analytical data. See DOI: 10.1039/c3ra47633b

---