Controllable growth of organic nanostructures from 0D to 1D with different optical properties

Abstract

Two donor–π–donor (D–π–D) type compounds containing carbazole as electron donors around the perylene core were synthesized and fabricated into controllable nano/microstructures from 0D to 1D by adjusting the growth rate. The difference in symmetry between the two molecules results in distinct self-assembly behaviours. Fluorescence microscopy images and fluorescence spectra of these self-assembled structures displayed different optical properties. It is indicated that the photophysical properties of these two compounds in the solid state are determined not only by their chemical structures but also by the mode of molecular packing.

---

Introduction

Nanomaterials have undergone rapid development by virtue of their excellent properties arising from their individual components.^1,2 To date, a great number of inorganic and organic materials have been used to prepare nano-sized structures with desired size and shape through “top-down” and “bottom-up” methods.^3–6 Among them, organic nanomaterials with well-defined structures and novel functions have aroused much attention because of their potential applications in photonic and electronic fields.^7–9 Utilizing intermolecular forces to construct such nanostructures through self-assembly of low-weight organic molecules is a facile and valid method to realize expected functions. This bottom-up technique offers the possibility to tune the functions of solid materials at the molecular level.

Apart from the common intermolecular forces of hydrogen bonding,¹⁰ π–π stacking¹¹ and electrostatic interaction,¹² dipole–dipole interactions are also used to drive self-assembly, especially in the highly polarized intramolecular charge-transfer (ICT) compounds.¹³ This directional supramolecular interaction has recently been used to induce the growth of low-dimensional, such as 1D or semi-1D, organic nanostructures.^14,15 Nanomaterials composed of ICT compounds are of great importance due to their fascinating applications in dye-sensitized solar cells,¹⁶ sensors,¹⁷ information storage materials,¹⁸ and nonlinear optical materials.¹⁹ Reasonably designed ICT compounds with controlled self-assembly processes are hot topics for novel applications in the future.

The design of ICT compounds based on carbazole as the donor with various aggregate nanostructures and properties has been widely investigated.^20–22 However, controlling the self-assembly processes to adjust the dimensionality of nanostructures and clarifying the structure–property relationships are still challenging. Herein, we report on design strategies and structure–property studies for self-assembly of conjugated compounds based on carbazoles. Asymmetric and symmetric compounds were synthesized through changing the number of carbazole units around the perylene core. The difference in molecular symmetry can lead to the polarization of the molecules. Supramolecular aggregates of 0D and 1D structures were obtained from these compounds by tuning the growth rate. Finally, the optical properties of these nanostructures were also investigated to clarify the structure–property relationships.

Results and discussion

The synthesis of target molecules TBC and TCP (Fig. 1) is shown in Scheme S1.† Details of the experiments are described in the Experimental section. In brief, we obtained two molecules with different symmetry by adjusting the number of carbazole units. It is important to note that the asymmetric TBC has additional dipole–dipole interaction between molecules apart from the evident π–π interaction, which dominates in TCP.

Fig. 1 Molecular structures of TBC and TCP.

Electrochemical properties

The electrochemical properties of TBC and TCP were studied by cyclic voltammetry (CV) and differential pulse voltammetry (DPV) at room temperature in dichloromethane solution. Both CV curves showed only two pairs of irreversible oxidation waves, which indicated the two compounds are more like the donor–π–donor (D–π–D) type rather than donor–π–acceptor (D–π–A). The irreversible processes are related to the oxidation steps of carbazole²⁰ and perylene.²³ Due to the irreversible electrochemical processes, DPVs were tested and ferrocene was used as internal standard substance to calculate the HOMO energies^24,25 (Fig. S1b and d† and Table 1). Moreover, for TBC, the first oxidation potential (E^ox₁) is 0.57 V and the second oxidation potential (E^ox₂) is 0.84 V. Compared with TBC, the E^ox values of TCP showed a slight positive shift of about 0.02 V (E^ox₁ = 0.59 V, E^ox₂ = 0.86 V), which might result from the additional electron-donating carbazole group²⁶ in TCP.

Table 1 Photophysical properties and calculated data of TBC and TCP

Compd	λ^absmax (nm) (log ε)^a	λem^b (nm)	Φfl^c	HOMO^d/LUMO^e (eV)	Eg^f (eV)
a Measured in 1 × 10^−5 M CH2Cl2 solution.b TBC and TCP were excited at 398 nm and 405 nm, respectively.c In CH2Cl2, quinine sulfate dihydrate, Φfl = 0.53 in 0.1 M H2SO4 as the standard.d HOMO energies were calculated by the onset oxidation potential. HOMO = −e(E^onsetox + 4.40) eV.e LUMO energies were calculated through the formula: Eg = LUMO − HOMO.f Eg was obtained from the onset of absorption spectra according to Eg = 1240/λ.
TBC	398(4.45), 421(4.79), 449(4.93)	461 488 522	0.90	−4.97, −2.27	2.70
TCP	405(4.06), 430(4.37), 457(4.48)	462 492 528	0.85	−4.99, −2.40	2.59
TBCfilm	406, 432, 460	467 537	—	—	—
TCPfilm	438, 473	465 524	—	—	—

---

Photophysical properties

UV-Vis and fluorescence spectra are very sensitive to the intermolecular distance and orientation, therefore the interaction between molecules could be inferred from them to some extent. Fig. 2a shows the normalized UV-Vis absorption spectra of TBC and TCP in CH₂Cl₂ solution (1 × 10⁻⁵ M) and cast films. The spectral parameters are summarized in Table 1. Four distinct UV-Vis absorption bands were observed in both, and the bands at around 290 nm would be associated with the π–π* transition of the carbazole units²⁷ in two molecules. Three bands in the visible region could be assigned to the π–π* transition in the perylene core.²⁸ From Fig. 2a, it is revealed that the absorption bands would experience red shift as the π-conjugated systems become longer and the electron donors become stronger.^29,30 Accordingly the maximum absorption band had a red shift of 8 nm as TCP has a longer conjugated system and stronger electron donors. Furthermore, the absorption band of TBC film presented an clear red shift of 11 nm compared with its dilute solution, which indicates the formation of J-type aggregates and π–π interactions among neighbouring molecules in solid states.^11,31 With regard to compound TCP, the absorption spectrum of film became structureless and with a red shift compared to that in solution. It means that compound TCP would also form the J-type aggregates in solid states. The long tail absorption of TCP solid might result from the scattering of some microparticles in the film, which at the same time could affect the light transmittance and thus give rise to the structure-less absorption band and also low absorbance obtained (Fig. S2†). From the onset of absorption spectra, E^opt_g of 2.70 eV and 2.59 eV were obtained for TBC and TCP, respectively. Thus LUMO energy levels of −2.27 eV and −2.40 eV were derived for TBC and TCP.

Fig. 2 (a) Normalized UV-Vis absorption spectra of the ICT compounds in CH₂Cl₂ solution (1 × 10⁻⁵ M) and solid states. (b) Normalized PL emission spectra both in solution and solid states. The excitation wavelengths used were 398 nm and 405 nm for TBC and TCP, respectively.

Fluorescence spectra of TBC and TCP were also obtained in the two states and both molecules gave intense fluorescence in solution and films (Fig. 2b). The emission spectra in solution are almost mirror images with respect to their absorption bands. The maximum emission band of TBC monomers was located at 461 nm with two shoulders (488 nm and 522 nm) when excited at 398 nm. The solution of TCP gives a band centred at 462 nm with two shoulder peaks at 492 and 528 nm when excited at 405 nm. The Stokes shifts of TBC and TCP were very small (Δλ = 12 nm and 5 nm, respectively) and their fluorescence quantum yields are given in Table 1. It is interesting to note that the emission spectra in solid states displayed different trends. For example, TBC film showed a broad red-shifted band having a maximum at 537 nm with a minor monomer emission around 467 nm. The minor band could be assigned to the shortest 0-0 transition observed in the fluorescence spectrum in solution and the decrease in intensity was thought to originate from two factors: one is as a result of self-absorption, which according to the literature, also has a small Stokes shift,³² and the other is due to the number of individual molecules existing in the film decreased.³³ In addition, the broad red-shifted band was due to the perylene excimer, which was usually observed in other studies,^31,34,35 whereas the TCP film displayed a very prominent monomer emission located at 465 nm and a relatively weak, structure-less emission band in the longer wavelength range. This observation indicates that in solid states, the perylene core in TCP compounds would not interact or there were a very small number of perylene molecules interacting with neighbouring perylene in the excited states. This might be derived from its intrinsic molecular structure, that is there are four carbazoles around the perylene core, which could cause less planarity for the whole molecule. Together with the steric repulsion, carbazoles mainly interact with each other in solid states rather than with perylene molecules. Thus, the perylene excimer would not appear under this condition. The abovementioned results suggested that although both molecules formed a J-type stacking in solid states, there would be a difference in their microstructure. As a consequence, we could obtain various aggregate nanostructures with different optical properties through tuning parameters such as solvent and growth rate during the self-assembly processes. Detailed data are shown in Table 1.

Self-assembly and optical properties

Self-assembly is a valid method to construct organic nanostructures with high efficiency without the need for templates. It is important to note that the growth rate plays a key role in self-assembly to obtain nanostructures with desirable shape and size. In the following study, we adopted two pathways, namely, the solvent vapour technique³⁶ and the phase transfer methodology,³⁷ to fabricate the ICT compounds into various nano/microstructures from 0D to 1D by adjusting the growth rate. The initial concentration (1 × 10⁻³ M) of both molecules was kept the same.

The solubility of TBC and TCP is “good” in dichloromethane and tetrahydrofuran but “poor” in hexane, methanol and acetone. By the solvent vapour technique, compound TBC could self-assemble into sphere-like architectures. On injection of the same volume of hexane into a solution of TBC in CH₂Cl₂ with sufficient stirring, a certain structure of aggregates was obtained after solvent evaporation. As shown by scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images (Fig. 3a and b), nanospheres with an average diameter of 500 nm was formed and TEM image further confirmed that the nanospheres had hollow morphology. It is expected that during the solvent evaporation process, TBC molecules segregated from CH₂Cl₂ should have been organized by intermolecular interactions to form definite structures in hexane after the better solvent evaporated first. The hollow nanospheres, called vesicles, are also obtained in other systems such as the amphiphilic perylene bisimide dyes.^33,38,39 The self-assembled nanostructures would undergo various shapes as a result of the different solvent polarity, which can affect the shape and size of the formed micelles. However, in our study, the hollow-shaped nanostructures can form even if adjusting the solvent polarity (THF/CH₃OH). It is suggested that though the carbazole units have a certain hydrophilic property, it is not enough to be affected by the solvent polarity. Thus, the mechanism of this process would be attributed to the certain directional molecular stacking and its relevant growth rate (Fig. 3g and 4g).

Fig. 3 SEM, TEM and corresponding fluorescence microscopy images (excited by a UV band (330–380 nm) light source) of self-assembled nanospheres (a–c) and nanotubes (d–f) of TBC. The inset of (a) and (b) are SEM and TEM imagine of the single nanosphere respectively. Note: for the different techniques, the same poor and good solvents were used. The difference in morphologies mainly results from the growth rate. (g) The proposed mechanism for the self-assembly of TBC.

Fig. 4 Large-area SEM and TEM images of self-assembled TCP: nanospheres (a and b) and nanotubes (d and e). The corresponding fluorescence microscopy images (c and f) show the similar emission property of both nanostructures. (g) The proposed mechanism for the self-assembly of TCP.

With the same mixture of solvents, CH₂Cl₂/hexane, slowing down the growth rate of TBC molecules during self-assembly, one-dimensional nanotubes were formed. The operating process is described as follows: adding the poor solvent hexane slowly to the CH₂Cl₂ solution with the volume ratio of 4 : 1, 1D nanotubes were obtained as yellow precipitates after the mixture was placed for five days without disturbance. SEM and TEM images (Fig. 3d and e) showed that the nanotubes were well monodispersed with an average width of 100 nm and length up to several micrometers. Under this condition, nucleation first happened at the interface between the two solvents followed by the slow growth process of TBC molecules and finally well-defined nanotubes were formed by the synergistic effect of different intermolecular forces (Fig. 3g). The possible mechanism for the morphology transition was proposed as a “curvature strain releasing” process driven by donor–acceptor dipole–dipole interactions.⁴⁰

Fluorescence microscopy images (Fig. 3c and f) of TBC in 0D and 1D supramolecular architectures were totally different. Both structures were intense emitters and for the nanospheres, they radiated strong yellow fluorescence but for the nanotubes, green fluorescence was observed. It is interesting to note that for the same compound, we could control the dimensionality of nanostructures from 0D to 1D just by adjusting their growth rate and these nanostructures would possess disparate optical properties.

For compound TCP, large-scale spheres measuring several hundred nanometers (800–1000 nm) were observed from the solvent vapor technique (Fig. 4a). TEM images (Fig. 4b) also indicated these nano/microspheres were hollow structures. Morphology transition was also observed when slowing down the growth rate by the phase transfer methodology. Fig. 4d shows typical SEM image of nanorods with width of about 760 nm and TEM image (Fig. 4e) further indicated they were solid structures with length up to several micrometers (Fig. 4g). The parameters of the self-assembly processes, namely, the operation method, solvent, initial concentration of ICT compounds, and volume ratio, were the same as that of TBC described above. Fluorescence microscopy images (Fig. 4c and f) were taken to investigate the morphologies and fluorescence emission of nano/microspheres and nanorods. Interestingly, in contrast with TBC molecules, the different supramolecular structures of TCP showed similar emission behaviour; both displayed yellow fluorescence.

To compare the optical properties of various nanostructures directly, their corresponding fluorescence spectra were obtained (Fig. 5). For TBC and its relevant self-assembled structures, the maximum emission bands of nanospheres and nanotubes both experienced red-shifts compared with molecules in solution but differing in degree (Fig. 5a). In accordance with the fluorescence microscopy images, these two types of morphologies had different illuminant properties. The λ_em of nanotubes red-shifted from 488 nm in solution to 519 nm, whereas the λ_em of nanospheres red-shifted much more to 545 nm. As mentioned before that the broad red-shifted band was attributed to the excimer emission. This emission band of nanospheres is more red-shifted and broader than that of nanotubes, which means better exciton migration originating from stronger π–π interaction among perylene cores but less well-organized molecular aggregation into spherical structures.^33,35 Thus, the microstructures within nanospheres and nanotubes are different, which would in turn affect the optical properties of solid materials.

Fig. 5 Normalized PL emission spectra of (a) TBC and (b) TCP molecules both in solution and in the self-assembled aggregates states. They were excited at 398 nm and 405 nm, respectively.

Moreover, the fluorescence spectra of nanostructures of TCP showed the same results as displayed in the fluorescence microscopy images. The shape of the bands and λ_em of nano/microspheres and nanorods were nearly the same (∼464 nm) only with a little difference in the region between 530 nm and 580 nm (Fig. 5b). According to the previous study (Fig. 2b), it is difficult for perylene units to interact with each other in excited states, which is in accordance with the inconspicuous excimer emission bands observed here. However, the π–π interactions among perylenes and thus excitonic coupling is stronger in nanorods, which can be deduced from the emission spectra between 530 nm and 580 nm in Fig. 5b and also in Fig. 4g. To form the rod-shaped nanostructures, TCP molecules should be packed more tightly than those in hollow nanospheres, therefore the probability of interaction among perylenes increases. The results demonstrated that the optical properties of organic supramolecular aggregates are not only related to the molecular structure itself (comparing TBC and TCP), but also has a relationship to the molecular packing in solid states (when comparing the same compound in different shapes).

At this stage, we acquired self-assembled nanostructures of TBC and TCP with controllable shapes and dimensionality, namely, nano/microspheres, nanotubes and nanorods, which have high morphological purity without other types of structures. It has been discussed that for the same compounds, 0D to 1D nanostructures could be formed by tuning the growth rate. As can be observed from the abovementioned results, the properties (electrochemistry, absorption and fluorescence spectra) of TBC and TCP in solution have very subtle differences. However, the optical properties have considerable differences once they were fabricated into self-assembled nanostructures, which could be attributed to different molecular packing in solid states.

It is interesting to note that the difference in symmetry of the chemical structures would result in rather distinct self-assembly behaviour and distinct optical properties under the same conditions. TCP has very symmetric structure and dipole moments from all directions are offset. The major driving force for formation of nanostructures is the π–π interaction. However, TBC is asymmetric and the resultant dipole moment from several directions would be the driving force. Thus, there are mainly two types of intermolecular forces, π–π interactions and dipole–dipole interactions, which would induce molecules to form certain structures under different environments. In the solvent vapor technique, although spherical structures were obtained for both compounds, the molecular packing was largely different. This can be reflected from the emission spectra (Fig. 5a and b) that both nanospheres had two emission bands at around 465 nm (monomer emission) and 545 nm (excimer emission) but differed in relative intensity. The different molecular packing during self-assembly resulted in different size structures (Fig. 3g and 4g). The difference between TBC and TCP compounds became more obvious in the phase transfer methodology, which has a slower growth process, that is, TBC could be fabricated into tube-like structures with green fluorescence, whereas TCP could self-assemble into rod-like structures with yellow fluorescence.

In addition, it is found that in the solvent vapor technique, acetone as a poor solvent could induce totally different self-assembly results between TBC and TCP. The symmetric TCP would produce 2D flower-like structures when a little amount of acetone was added; however, the asymmetric TBC could not form similar structures no matter how the amount of acetone was tuned (see ESI, Fig. S3†). The growth of TCP was carried out in CH₂Cl₂/acetone (v/v, 40 : 3) by the solvent vapor technique and large scale microflowers bearing a series of nanowires were obtained (Fig. 6a, arrows). It is clearly observed that some microflowers were only composed of a few nanowires, whereas others displayed hierarchical structures with a large number of nanowires stacked together (Fig. 6a and b). The inset in Fig. 6b confirmed that the branches of microflowers were not smooth but were pod-like with many spheres fusing together. The microflowers also displayed strong yellow emission as do other nanostructures of TCP. TEM images (Fig. 6d–f) further confirmed that these microflowers were composed of nanowires, which might be formed by fusing of primary nanospheres in certain directions.

Fig. 6 (a and b) SEM and (d–f) TEM images of TCP prepared in CH₂Cl₂/acetone (v/v 40 : 3) by solvent vapor technique at room temperature. Inset of (b) represents the details of the branches of microflowers. ​(e) One microflower with simple structures and (f) one typical microflower with hierarchical structures. (c) The corresponding fluorescence microscopy image obtained to examine the morphologies and fluorescence emission of flower-like structures.

To clarify the role of acetone during this process, different amounts of acetone were added to direct the self-assembly of TCP molecules. By directly evaporating CH₂Cl₂ without acetone, quasi-spheres were formed with the diameter of nearly 500–1000 nm (Fig. 7a). When increasing the amount of acetone to the volume ratio 40 : 10, microflowers appeared but more unordered (Fig. 7b). It was shown in Fig. 6 that there were some nanospheres (about 500–1000 nm) dispersed around these microflowers with the same size as the width of the branches. We proposed that in the first stage, small nanospheres appeared. Then, the existing small amount of acetone induces these nanospheres to fuse together forming the nanowires, which would further connect together to form the final microflowers.

Fig. 7 SEM images of TCP prepared in CH₂Cl₂/acetone with different volume ratios (v/v) by solvent vapor technique. (a) v/v = 40 : 0, (b) v/v = 40 : 10, (c) v/v = 40 : 15 and (d) v/v = 40 : 40. Inset of (d) indicates the open holes on the surface of nanospheres.

However, further increasing the amount of acetone to reach the volume ratio of 40 : 15, as shown in Fig. 7c, nanorods combined with nanospheres were obtained in nearly the same proportions. When the volume ratio reached 40 : 40, plenty of nanospheres were formed with some open holes on their surface (inset in Fig. 7d) and these nanospheres were connected together (Fig. 7d), due to the fact that the large amount of acetone functions, as the “poor” solvent in these cases. The results indicated that acetone could induce molecules of TCP to form 2D flower-like hierarchical structures when a little amount was added. Further increasing the acetone would not produce such microstructures but result in nanospheres.

Conclusions

In summary, two donor–π–donor (D–π–D) type compounds containing carbazole as the electron donor around the perylene core were synthesized and fabricated into well-defined and controllable nanostructures. It is found that the dimensionality of the self-assembled nanostructures could be tuned from 0D to 1D by changing the growth rate. It is reasonable to conclude that the optical properties of organic nanomaterials are not only related to their molecular structures, but also have relationships with modes of molecular packing in solid states.

Acknowledgements

This study was supported by the National Natural Science Foundation of China (21290190, 21322301, 20803029, 51072064) and the National Basic Research 973 Program of China (2011CB935800, 2012CB932900), and the “Strategic Priority Research Program” of the Chinese Academy of Sciences (XDA09020302).

Notes and references

1. Y. Guo, L. Xu, H. Liu, Y. Li, C. Che and Y. Li, Adv. Mater., 2015, 27, 985–1013.
2. M. Safont-Sempere, G. Fernández and F. Würthner, Chem. Rev., 2011, 111, 5784–5814.
3. Y. Fernandez, T. Gschneidtner, C. Wadell, L. Fornander, S. Avila, C. Langhammer, F. Westerlund and K. Moth-Poulsen, Nanoscale, 2014, 6, 14605–14616.
4. H. Yu, M. Regulacio, E. Ye and M. Han, Chem. Soc. Rev., 2013, 42, 6006–6018.
5. X. Li, Y. Ma, L. Qin, Z. Zhang, Z. Zhang, Y. Zhang and Y. Qu, J. Mater. Chem. A, 2015, 3, 2158–2165.
6. R. Huang, Y. Wang, W. Qi, R. Su and Z. He, Mater. Lett., 2014, 128, 216–219.
7. Y. Li, T. Liu, H. Liu, M. Tian and Y. Li, Acc. Chem. Res., 2014, 47, 1186–1198.
8. Y. Yan and Y. Zhao, Chem. Soc. Rev., 2014, 43, 4325–4340.
9. L. Zang, Y. Che and J. Moore, Acc. Chem. Res., 2008, 41, 1596–1608.
10. L. Zhang, Y. Long, Z. Chen and M. Wan, Adv. Funct. Mater., 2004, 14, 693–698.
11. C. Huang, L. Wen, H. Liu, Y. Li, X. Liu, M. Yuan, J. Zhai, L. Jiang and D. Zhu, Adv. Mater., 2009, 21, 1721–1725.
12. T. Tsai, G. Heckert, L. Neves, Y. Tan, D. Kao, R. Harrison, D. Resasco and D. Schmidtke, Anal. Chem., 2009, 81, 7917–7925.
13. J. Xu, L. Wen, W. Zhou, J. Lv, Y. Guo, M. Zhu, H. Liu, Y. Li and L. Jiang, J. Phys. Chem. C, 2009, 113, 5924–5932.
14. X. Zhang, X. Zhang, K. Zou, C. Lee and S. Lee, J. Am. Chem. Soc., 2007, 129, 3527–3532.
15. T. Liu, Y. Yu, S. Chen, Y. Li and H. Liu, RSC Adv., 2013, 3, 9973–9977.
16. S. Ito, H. Miura, S. Uchida, M. Takata, K. Sumioka, P. Liska, P. Comte, P. Pechyb and M. Gratzel, Chem. Commun., 2008, 41, 5194–5196.
17. Q. Li, M. Peng, H. Li, C. Zhong, L. Zhang, X. Cheng, X. Peng, Q. Wang, J. Qin and Z. Li, Org. Lett., 2012, 14, 2094–2097.
18. Y. Shang, Y. Wen, S. Li, S. Du, X. He, L. Cai, Y. Li, L. Yang, H. Gao and Y. Song, J. Am. Chem. Soc., 2007, 129, 11674–11675.
19. J. Xu, S. Semin, D. Niedzialek, D. Niedzialek, P. Kouwer, E. Fron, E. Coutino, M. Savoini, Y. Li, J. Hofkens, H. Uji-I, D. Beljonne, T. Rasing and A. Rowan, Adv. Mater., 2013, 25, 2084–2089.
20. S. Chen, Z. Qin, T. Liu, X. Wu, Y. Li, H. Liu, Y. Song and Y. Li, Phys. Chem. Chem. Phys., 2013, 15, 12660–12666.
21. X. Sun, Y. Liu, X. Xu, C. Yang, G. Yu, S. Chen, Z. Zhao, W. Qiu, Y. Li and D. Zhu, J. Phys. Chem. B, 2005, 109, 10786–10792.
22. D. Jiang, Z. Xue, Y. Li, H. Liu and W. Yang, J. Mater. Chem. C, 2013, 1, 5694–5700.
23. A. Aleshinloye, J. Bodapati and H. Icil, J. Photochem. Photobiol., A, 2015, 300, 27–37.
24. T. Johansson, W. Mammo, M. Svensson, M. Andersson and O. Ingana, J. Mater. Chem., 2003, 13, 1316–1323.
25. Y. Li, Y. Cao, J. Gao, D. Wang, G. Yu and A. Heeger, Synth. Met., 1999, 99, 243–248.
26. S. Chen, Y. Li, W. Yang, N. Chen, H. Liu and Y. Li, J. Phys. Chem. C, 2010, 114, 15109–15115.
27. H. Zhang, X. Wan, X. Xue, Y. Li, A. Yu and Y. Chen, Eur. J. Org. Chem., 2010, 1681–1687.
28. L. Zhao, W. Yang, Y. Ma, J. Yao, Y. Li and H. Liu, Chem. Commun., 2003, 19, 2442–2443.
29. Y. Zhu, A. Kulkarni, P. Wu and S. Jenekhe, Chem. Mater., 2008, 20, 4200–4211.
30. K. Chen, Y. Hsu, Q. Wu, M. Yeh and S. Sun, Org. Lett., 2009, 11, 377–380.
31. H. Wu, L. Xue, Y. Shi, Y. Chen and X. Li, Langmuir, 2011, 27, 3074–3082.
32. A. Iida and S. Yamaguchi, Chem. Commun., 2009, 21, 3002–3004.
33. B. Jiang, D. Guo and Y. Liu, J. Org. Chem., 2010, 75, 7258–7264.
34. D. Das, G. Sahoo, P. Mazumdar, A. Maity, D. Chattopadhyay, G. Morán and A. Misra, J. Mol. Liq., 2015, 206, 47–55.
35. Y. Che, X. Yang, S. Loser and L. Zang, Nano Lett., 2008, 8, 2219–2223.
36. T. Liu, Y. Li, Y. Yan, Y. Li, Y. Yu, N. Chen, S. Chen, C. Liu, Y. Zhao and H. Liu, J. Phys. Chem. C, 2012, 116, 14134–14138.
37. S. Chen, N. Chen, Y. Yan, T. Liu, Y. Yu, Y. Li, H. Liu, Y. Zhao and Y. Li, Chem. Commun., 2012, 48, 9011–9013.
38. X. Zhang, S. Rehm, M. Safont-Sempere and F. Würthner, Nat. Chem., 2009, 1, 623–629.
39. X. Yang, X. Xu and H. Ji, J. Phys. Chem. B, 2008, 112, 7196–7202.
40. J. Xu, X. Liu, J. Lv, M. Zhu, C. Huang, W. Zhou, X. Yin, H. Liu, Y. Li and J. Ye, Langmuir, 2008, 24, 4231–4237.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra17516j

---