Facile synthesis of Gd(III) metallosurfactant-functionalized carbon nanodots with high relaxivity as bimodal imaging probes

Abstract

Fluorescence and magnetic resonance (MR) dual-modal imaging contrast agents were prepared by the self-assembly of carbon nanodots (CDs) and gadolinium(III)-containing metallosurfactants (MS). This self-assembly process, driven by facile ionic interaction between MS and CDs, provides a new method for the preparation of dual-modal contrast agents. The obtained assemblies (MS–CDs) were investigated by various characterization techniques. Zeta potential measurement validated the driving force for the self-assembly. Transmission electron microscopy (TEM), dynamic light scattering (DLS) and Fourier transform infrared spectroscopy (FTIR) confirmed the formation of the aggregates, MS–CDs. The relation between zeta potential and particle size revealed the formation process of MS–CDs assemblies. It was noteworthy that the MS–CDs assemblies exhibited high relaxivity (r₁ = 19.73 mM⁻¹ s⁻¹, at 1.5 T) for MR imaging, which was more than fourfold that of commercial agents. The stability of the MS–CDs was evaluated by relaxation study and aggregation behaviour in phosphate buffer solution (PBS). Furthermore, good photoluminescence and low cytotoxicity were observed in model cell lines, suggesting that MS–CDs have great potential for bioimaging applications.

---

Introduction

In the development of fluorescent nanomaterials, the discovery of semiconductor quantum dots (QDs) is considered as a milestone. Although QDs exhibit high performance in optical imaging, most of them are composed of toxic heavy metal elements such as cadmium.¹ Tremendous research efforts have been devoted to the use of non-toxic materials as substitutes, like ZnS, to prevent the release of heavy metals.^2–4 Carbon-based quantum dots (CDs) with fascinating properties have gradually become a rising star as a new nanocarbon member due to their benign, abundant and inexpensive nature.^5,6 CDs are widely considered as candidates for optical bioimaging.^7–11 Superior to QDs, CDs show good solubility in aqueous solutions, low toxicity and biocompatibility. However, optical imaging has the capacity for single-cell sensitivity and subcellular resolution, but possesses poor spatial resolution and tissue penetration. Therefore, it is necessary to combine two or more complementary imaging techniques for validation of the in vivo imaging experiments. The combination of MRI and optical methods is one of those complementary techniques under current investigation and has become a hot topic in bioimaging.^12–15

MRI is a routine diagnostic tool in modern clinical practice.^16–19 It derives directly from the phenomenon of nuclear magnetic resonance and combines many advantages including non-ionizing, high spatial resolution, and deep tissue penetration. In order to enhance the imaging contrast, contrast agents (CAs) are commonly used in clinical application. In particular, gadolinium(III)-based paramagnetic contrast agents have be widely employed and attracted much research interest.^20–22 Although Gd(III) provides excellent contrast, free Gd(III) is toxic and CAs must be administered in the form of stable chelate complexes that prevent the release of the metal ion in vivo.

The observed relaxation rate of water molecules consists of diamagnetic and paramagnetic contributions, and the paramagnetic relaxation rate depends on the concentration of Gd(III) when these CAs present. The overall relaxivity is governed by a number of parameters, which characterize the structure and dynamics of the complex in solution, such as the number of inner-sphere water molecules directly coordinated to the Gd(III) centre (q), the residence time of the coordinated water molecule (τ_M) and the rotational correlation time representing the molecular tumbling rate of a complex in solution (τ_R). Shorter water residence time and longer rotational correlation time are generally considered essential for enhancing the relaxivity.^23–25

Growing attention has focused on the development of multimodal probes that can be visualized by MRI and by one or more other imaging modalities. As mentioned above, carbon nanodots can be the good candidate for combination of MRI CAs as bimodal CAs. The common routine for multimodal CAs is to bound to CDs and CAs covalently associated. A carbon nanodots functionalized with Gd(III) chelates was prepared by Ren et al.,²⁶ which was obtained through one-step pyrolysis of gadopentetate monomeglumine. This precursor provided simultaneously a carbon source and a Gd(III) source. Similar CDs functionalized bimodal CAs for T₁ (ref. 27 and 28) and T₂ (ref. 29) imaging have also been reported. Shi et al.³⁰ utilized amine-terminated CDs as fluorescent source, and cyclic DTPA dianhydride was covalently conjugated onto the outer surface of the CDs through carbodiimide chemistry to produce bimodal CAs. In addition, many trimodal imaging agents also have been developed for complementary information and synergistic advantages, combing magnetic, radionuclide, and fluorescent imaging capabilities.^31–33

Nevertheless, preparation of multimodal CAs through noncovalent bonding of CDs and MRI CAs was less exploited. Noncovalent bonding strategy is a facile and variable way to prepare advanced functional materials. Generally, this approach is to design of relatively simple structures using hydrogen bonding, π–π stacking, metal–ligand interactions, electrostatic forces, strong dipole–dipole association, hydrophobic forces, and steric repulsion.³⁴ It is important to note that these interaction should be sufficiently strong to provide sufficient stability. It is notable that ionic interaction possess rather high bonding strength and nonselective nature.³⁵ The excellent availability of the starting products (charged tectonic units) and the simplicity of synthesis, by neat addition, allow the recombinatorial synthesis of a whole range of functional materials and hybrids with interesting and versatile functions.³⁶ Recently, Wang et al.³⁷ prepared a bimodal MRI/NIR agents via a self-assembly approach of three types of amphiphilic structures in aqueous solution. The r₁ relaxivity was up to 11.87 mM⁻¹ s⁻¹. Meanwhile, excellent fluorescence and targeting effect were observed. However, the biocompatibilities of these agents were not good especially at high doses, probably owning to the organic dyes used.

Herein, we applied a facile approach to prepare CDs based bimodal CA by using a self-assembly technique as shown in Scheme 1. The MRI CA chosen was a metallosurfactant (MS) with a hydrophilic Gd(III)-complex head structurally similar to Omniscan® and two hydrophobic long alkyl chain tails.^38–40 The MS is positively charged due to the presence of two quaternary ammonium groups. The CDs used in this technique were passivated with strong oxidative acid, possessing abundant negatively charged carboxy groups on the surface. The MS formed micelles above critical micelle concentration (CMC) in solution, and CDs self-assembled with the MS through ionic interaction. The product of the self-assembly between MS and CDs (MS–CDs) were highly dispersible in solutions. Furthermore, high relaxivity for MR, good photoluminescence properties for optical imaging and good biocompatibility of these assemblies suggested the potential application in clinic.

Scheme 1 Schematic illustration of the synthetic route for CDs and self-assembly process of MS–CDs.

Experimental section

Materials

Citric acid, ethylenediamine, nitric acid and other materials were commercially available and used as received. Double chain cationic Gd(III)-containing metallosurfactant (MS) was synthesized according to our recently published work.⁴¹ Deionized (DI) water was used for all the experiments.

Preparation of bare CDs⁴²

Citric acid (1.0507 g) and ethylenediamine (335 μL) was dissolved in DI-water (10 mL). Then the solution was transferred to a poly(tetrafluoroethylene) (Teflon)-lined autoclave (30 mL) and heated at 150 °C for 5 h. The product was dialyzed (MWCO 500) against deionized water for 2 days.

Surface passivation of CDs

5 M nitric acid (6 mL) was added to the bare CDs, and the mixture was refluxed at 120 °C for 12 h. After neutralization with sodium bicarbonate, the CDs were desalted by dialysis (MWCO 500) against deionized water for 2 days.

Self-assembly of MS and CDs

MS was slowly dropped into the CDs solutions with continuous stirring. After 1 h, the mixture was dialyzed (MWCO 1000) against deionized water for 2 days.

Quantum yield (QY) measurements

Quinine sulphate in 0.1 M H₂SO₄ (QY = 0.54 at 360 nm) was chosen as a standard.⁴³ The quantum yield of the C-dots in water was calculated according to the following equation:
where ϕ is the quantum yield, I is the measured integrated emission intensity, n is the refractive index (1.33 for water), and A is the optical density. The subscript “r” refers to the reference fluorophore of known quantum yield.

Characterizations

Zeta potential and dynamic light scattering (DLS) measurements were carried out with a zetasizer analyser (Nano ZS, Malvern). Transmission electron microscopy (TEM) images were obtained on a JEOL JEM 1400 and JEOL JEM 2100 microscope operating at 100 kV and 200 kV, respectively. TEM samples were prepared by depositing a drop of the dilute solution on a carbon/copper grid, then frozen by liquid nitrogen and quickly vacuumed to remove the solvent. The FTIR spectra were analysed by infrared spectrometer (Nicolet 6700). The contents of gadolinium were measured through inductively coupled plasma atomic emission spectrometry (ICP-AES, IRIS 1000, Varian 710ES). The PL spectra were collected on a Fluorolog-3-P UV-vis-NIR fluorescence spectrophotometer (JobinYvon, France) and the UV-vis spectra were obtained using a UNICO UV-21-2 PCS spectrometer.

T₁ relaxivity measurements

T₁ relaxivity measurements were performed on a 1.5 T GE SIGNA EXCITE at ambient temperature. The samples were diluted to different concentrations in 1.5 mL centrifuge tubes. For T₁ measurement, the samples were imaged collectively with a high-resolution inversion recovery pulse sequence (repeat time (T_R) = 1600 ms, echo time (T_E) = 9 ms, inversion time (T_I) = 50, 100, 200, 300, 400, 600, 700, 900, 1200, 1500 ms. Field of view = 150 mm × 150 mm, matrix = 320 × 320). The resulting images were analysed on a pixel-by-pixel basis to a single exponential. These T₁ values were averaged over at least 45 pixels in the centre of each sample and plotted as 1/T₁ versus [Gd³⁺]. The relaxivity values for MS and MS–CD were calculated using the following equation:

The slope of the plot represents the relaxivity, r₁.

In vitro cytotoxicity

The cytotoxicity of MS–CDs were individually tested using the MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay with the human cervical carcinoma cell (HeLa) as model cell line. The cells were cultured in 25 mL flasks with Dulbecco's modified Eagle's medium (DMEM) containing 10% fetal bovine serum (FBS), antibiotics (100 U mL⁻¹ penicillin-G, and 100 μg mL⁻¹ streptomycin) at 37 °C under a humidified atmosphere of 5% CO₂/95% air.

To detect the cytotoxicity, the cells were seeded onto 96-well plates at a density of 4.0 × 10³ cells per well in 0.1 mL culture medium and allowed to attach for 24 h. Afterwards, the growth media were removed and the cells were washed with PBS (2×), and then each well was added DMEM medium with the MS–CDs at various concentrations of 20, 40, 60, 80, 100, 150, and 200 μg mL⁻¹. The MTT assay was then conducted following the standard protocol after incubating cells for 24 h, and 48 h. The purple formazan in the supernatant was quantified by measuring at 595 nm absorbance in a spectrophotometer (SPECTRAmax384, Molecular Devices, USA).

Permeability and photoluminescence of MS–CDs in cells

HeLa cells were pre-cultured on chamber slides in DMEM for 12 h and the medium was supplemented with 1.25 mg mL⁻¹ of MS–CDs. After incubation for 24 h, the medium was removed and was replaced by a pre-warmed (37 °C) probe-containing medium. After incubation for another 30 minutes under growth conditions, the medium was removed and the cells were washed three times with ice-cold PBS solution. Afterwards, the cells were fixed with 3.7% formaldehyde for 15 minutes, and then washed with PBS solution twice. Finally, the cell imaging was taken at a confocal laser scanning microscope (CLSM, Nikon A1R, Japan).

Results and discussion

The bare CDs prepared through the hydrothermal method were surrounded with amino and carboxyl groups on the surface.⁴² The amino groups were positively charged, and the carboxyl groups were negatively charged. The zeta potentials were measured through the process, as shown in Table 1. The zeta potential of the bare CDs is 2.75 mV, and it is not an appropriate acceptor for the self-assembly with the MS with positively charged quaternary ammonium groups (8.96 mV). After passivation treatment with strong oxidative acid, the surface charge of the CDs was changed to −22.5 mV. After the self-assembly of MS with CDs, the zeta potential was changed to −9.99 mV, indicating that the MS was successfully conjugated to the CDs through ionic interaction.

Table 1 Zeta potential of different complexes as CDs on the surface of aggregates

Sample	Zeta potential^a/mV
a Zeta potential describes the electrokinetic potential in colloidal systems.
Bare CDs	2.75
CDs	−22.5
MS	8.96
MS–CDs	−9.99

---

Fig. 1A and B show the TEM images of MS–CDs assemblies. The as-prepared MS–CDs possessed nearly spherical shape and uniformly dispersed without apparent aggregation. It may attribute to the intrinsic high surface energy, which has been widely seen in nanomaterial.⁴⁴ The size distribution is shown in Fig. 1C. The average size of the sphere-like particles counted in TEM images was 52 nm, which was consistent with the result in dynamic light scattering (DLS), as shown in Fig. 1D. The size of CDs was about 2 nm,⁴² and particle diameter of MS measured by DLS was 6 nm (see Fig. S1†), which were far smaller than the size of the MS–CDs assemblies. The great increase in size indicates that complexed aggregates have formed. Many black dots were observed from the TEM images, attributing to the CDs interacted with MS (Fig. S2†). These MS–CDs assemblies may consist of large amount of micelle complexes between MS and CDs, similar to many ionic micelle complexes reported in literature.^45,46

Fig. 1 TEM images (A and B), size distribution (C), and DLS curve (D) of the MS–CDs assemblies.

The relation of zeta potential and particle size was investigated to reveal the formation process of MS–CDs assemblies in Fig. 2A. The zeta potential would decrease with the growing content of CDs, due to the negatively charged nature. The particle size gradually increased, with the addition of CDs, which indicated that MS began to connect together under the influence of CDs. At zeta potential of −7 mV, the particle size reached a maximum of 60 nm. With the further addition of CDs, the particle size of MS–CDs assemblies decreased slightly, indicating that the binding of CDs to MS reached saturation.⁴⁷

Fig. 2 (A) The relation between zeta potential and particle size. (B) FTIR spectra of MS and MS–CDs.

The self-assembly process was also characterized by FTIR spectroscopy. As shown in Fig. 2B, the stretching vibrations of N–H, C=O, C–N in MS are at 3421, 1601 and 1405 cm⁻¹, respectively. After the formation of the assemblies, these peaks slightly shift to 3414, 1599 and 1403 cm⁻¹ in MS–CDs, respectively. The decrease in the wavenumber suggests an enhancement of hydrogen bonding strength and stronger hydrogen bonds are formed in the MS–CDs assemblies.⁴⁸ The antisymmetric and symmetric stretching vibrations of the CH₂ alkyl chains were observed at 2923 and 2853 cm⁻¹. Moreover, the intensity of carbonyl stretching was significantly enhanced in MS–CDs, owning to the complexation with the CDs.

The MS–CDs assemblies prepared are well dispersible in aqueous solutions. The fluorescence intensity of the MS–CDs assemblies stayed constant, and no precipitation was found even after a long store period. Meanwhile, fluorescence emission intensity keeps stable during a long-time continuous shinning with UV light, which was comparable to the photostability profile⁴² of the CDs used in this method (Fig. S7†), indicating high photostability and store-stability of the MS–CDs assemblies.

The UV-vis and PL spectra of the samples are shown in Fig. 3. In the UV-vis absorption spectrum of an aqueous solution of the passivated CDs (Fig. 3A), the peak is focused on 310 nm, which is slightly blue shifted than the bare CDs,⁴² indicating that the surface passivation had affected the band gap of these complex carbonaceous materials. Another absorbance peak was found at 220 nm, attributing to the carbonyl groups in CDs (see Fig. S3†). Excitation-dependent PL behaviour was found for these CDs, which is typical for CDs prepared from different kinds of methods.⁴⁹ With the increase of the excitation wavelength, the PL intensity reached a maximum at excitation of being 360 nm, and then gradually decreased with a simultaneously red shift of the emission. This behaviour is attributed to the surface state affecting the band gap of CDs. The surface state is analogous to a molecular state whereas the size effect is a result of quantum dimensions, both of which result in the complexity of the excited states of CDs.⁵⁰ Fig. 3B shows the UV-vis and PL spectra of the MS–CDs assemblies. There was no distinct peak observed in the UV-vis absorption. The absorbance peak of CDs at 310 nm is buried in the absorption of MS (Fig. S4†) and barely seen. When excited with a UV light centred at 365 nm, the aqueous solution of MS–CDs assemblies shows bright blue fluorescence (inset of Fig. 3B), and the quantum yield was measured to be 3.8%. As shown in Fig. 3C, a slightly blue shift of the strongest emission peak was found after the self-assembly. The combination of CDs and MS may reduce the conjugated electron area of the emissive centres, causing an increase in their band-gap as predicted by density functional theory.⁵¹ The blue shift in fluorescence emission also indicates the change of CDs' surface environment resulted from the ionic interaction between CDs and MS. In a previous report by Li et al.,⁵² similar blue shift behaviour was observed. More interestingly, when the excitation wavelength was above 460 nm, the green light intensity was enhanced with the increase of excitation wavelength till 500 nm, which suggests that MS–CDs may have another fluorescence emission band around 500 nm. This finding of dual flourescence emission could also be found in other CDs system.⁵³

Fig. 3 The UV-vis absorption and PL spectra of (A) CDs and (B) MS–CDs. Inset: photo taken under 365 nm UV light. (C) Comparison of emission peak of CDs and MS–CDs, excited at 360 nm and 380 nm.

The presence of the Gd(III) chelates in these MS–CDs assemblies endows them possibility for MR imaging. Fig. 4B shows the grey-scaled T₁-weighted MR images. Concentration-dependent positive contrast enhancement was revealed for these MS–CDs assemblies. Even at a very low Gd(III) concentration, the brightness contrast was clear. The plot of relaxation rate versus Gd(III) concentration is shown in Fig. 4A. The relaxivity of MS was concentration dependant due to its surfactant nature.⁴¹ Therefore, MS–CDs exhibited two stages relaxations with the concentration changing. Firstly, the relaxation rate increased slowly with the Gd(III) concentration at a relaxivity of 5.19 mM⁻¹ s⁻¹. Then the relaxation rate increased sharply above the critical aggregation concentration (0.042 mM), exhibiting a relaxivity of 19.73 mM⁻¹ s⁻¹, which was more than fourfold that of Omniscan® (4.26 mM⁻¹ s⁻¹).³⁸ The enhancement of the relaxivity, r₁, can be divided into two components, inner-sphere (IS) and outer-sphere (OS) contributions (eqn (1)).^16,17

r₁ = r^IS₁ + r^OS₁ (1)

Fig. 4 (A) Relaxivity curve of MS–CDs assemblies. (B) MR images of MS–CDs assemblies at different concentrations.

Relaxivity arising from the inner-sphere water can be described by the Solomon–Bloembergen–Morgan theory,¹⁷ as summarized in eqn (2)–(4).

where p_M is the molar fraction of water coordinated to the metal centre. T_1M is the longitudinal relaxation time of water in the inner-sphere, which can be further expressed by eqn (3).

Therefore, it is an effective way to maximize the hydration number (q), optimize residence time of coordinated water molecules (τ_M) and rotational correlation time (τ_R) for high relaxivity.

In particular, to a spherical molecule, the rotational correlation time can be estimated by eqn (4).¹⁷ In eqn (4), a is the radius and η is the viscosity.

The size of MS micelle was about 6 nm as measured with DLS (see Fig. S1†). New assemblies, MS–CDs, were formed after the self-assembly process. As shown in Fig. 1, nearly spherical particles were observed, and the particle size of the MS–CDs was boosted to 50 nm, approximately 8 times larger than MS. Based on eqn (4), the rotational correlation time would increase dramatically. However, as discussed earlier, the MS–CDs assemblies were composed of many small micelle complexes. Therefore, the tumbling rate of the Gd(III) complexes were dictated by the local motion of these small complexes rather than the global motion of the MS–CDs assemblies. Nevertheless, after the self-assembly of the MS with CDs, the local motion of the Gd(III) complexes were restricted, resulting in the enhancement of the rotational correlation time and the relaxivity.

The stability of the MS–CDs assemblies is crucial for bioimaging. As shown in Fig. 4A, the MS–CDs displayed two stages relaxation rates with concentration dependant. It indicated that MS–CDs would disassemble in the first lower relaxivity stage but still in a Gd(III) chelate form. In spite of this, the concentration was extremely dilute, when the MS–CDs occurred to disassemble. So the nature of concentration dependant won't handicap the application of MS–CDs. On the other hand, the stability of MS–CDs in phosphate buffer solution (PBS) was investigated. PBS is a commonly used buffer solution, matching the osmolarity and ion concentrations of the human body, and is appropriate to simulate the environment of biological fluids. The comparison of aggregation behaviour in water solution and PBS was observed by DLS in Fig. S6.† Size and size distribution in the two solutions did not differ much. There was no obvious aggregation or disassembly in PBS.

The cytotoxicity of the MS–CDs assemblies was detected at different concentrations in HeLa cell line after incubation time for 24 h and 48 h, respectively (Fig. 5A). It was found that the HeLa cells showed good viability even at high doses of MS–CDs (200 μg mL⁻¹) and long incubation time (48 h) with the MS–CDs assemblies. Hence, the cytotoxicity of these MS–CDs assemblies was very low, which is essential for in vivo applications.

Fig. 5 Cellular toxicity and cellular imaging of MS–CDs assemblies. Viability of HeLa cells after treatment by MS–CDs (A). The permeability and photoluminescence of MS–CDs in HeLa cells observed by confocal laser scanning microscope under bright field (B) and 488 nm excitation (C) and the merged image (D). Scale bar: 50 μm.

In order to assess the permeability and photoluminescence in cells, MS–CDs were cultured with HeLa cells for 24 h and then was observed with confocal laser scanning microscope. As shown in Fig. 5B–D, green flourescence were observed throughout the cells when excited with laser pulse (488 nm). Specifically, the green light from the MS–CDs assemblies penetrated into the cells and located in membrane and cytoplasm of HeLa cells. MS–CDs were internalized into the cells possibly through endocytosis.⁵⁴ Furthermore, photoluminescence intensity of the labelled cells exhibited no obvious decay with continuous excitation for 10 min. The cytocompatibility and cell imaging results suggest MS–CDs have great potential applications as MR and fluorescence bioimaging agents.

Conclusions

In summary, we have successfully prepared a CDs based bimodal contrast agent by a facile self-assembly technique. Zeta potential measurements validated that the ionic interaction was the driving force for the self-assembly. The size of the assemblies was around 50 nm as characterized with TEM and DLS. FTIR spectroscopy showed the presence of hydrogen bonding between MS and CDs, indicating the formation of new aggregates, MS–CDs assemblies. The relation between zeta potential and particle size of MS–CDs assemblies revealed the formation process, MS connected together under the influence of CDs. The optical property was investigated by UV-vis and PL spectroscopy, showing relatively strong fluorescence in the MS–CDs. Moreover, T₁ relaxivity measurements revealed two stages relaxation rates due to the concentration dependant nature, and a high relaxivity for MR imaging (r₁ = 19.73 mM⁻¹ s⁻¹, at 1.5 T). The stability of MS–CDs assemblies was sufficiently stable, detected by relaxation study and aggregation behaviour in PBS. The cytotoxicity of MS–CDs in vitro cell experiments showed that the MS–CDs were benign to cells and displayed strong photoluminescence. Overall, high relaxivity, good photoluminescence and nontoxicity suggest a great potential application of these assemblies for bioimaging.

Acknowledgements

The authors gratefully acknowledge the financial support from National Natural Science Foundation of China (21274042) and Shanghai Leading Academic Discipline Project (B502). AH thanks the “Eastern Scholar Professorship” support from Shanghai local government.

Notes and references

1. C. Ding, A. Zhu and Y. Tian, Acc. Chem. Res., 2013, 47, 20–30.
2. E. Atabey, S. Wei, X. Zhang, H. Gu, X. Yan, Y. Huang, L. Shao, Q. He, J. Zhu, L. Sun, A. S. Kucknoor, A. Wang and Z. Guo, J. Compos. Mater., 2013, 47, 3175–3185.
3. H. Wei, X. Yan, Y. Li, S. Wu, A. Wang, S. Wei and Z. Guo, J. Phys. Chem. C, 2012, 116, 4500–4510.
4. B. Suo, X. Su, J. Wu, D. Chen, A. Wang and Z. Guo, Mater. Chem. Phys., 2010, 119, 237–242.
5. S. N. Baker and G. A. Baker, Angew. Chem., Int. Ed., 2010, 49, 6726–6744.
6. J. Zhu, M. Chen, Q. He, L. Shao, S. Wei and Z. Guo, RSC Adv., 2013, 3, 22790–22824.
7. Y.-P. Sun, B. Zhou, Y. Lin, W. Wang, K. A. S. Fernando, P. Pathak, M. J. Meziani, B. A. Harruff, X. Wang, H. Wang, P. G. Luo, H. Yang, M. E. Kose, B. Chen, L. M. Veca and S.-Y. Xie, J. Am. Chem. Soc., 2006, 128, 7756–7757.
8. P. G. Luo, S. Sahu, S.-T. Yang, S. K. Sonkar, J. Wang, H. Wang, G. E. LeCroy, L. Cao and Y.-P. Sun, J. Mater. Chem. B, 2013, 1, 2116–2127.
9. Y. Wang and A. Hu, J. Mater. Chem. C, 2014, 2, 6921–6939.
10. B. Zhu, S. Sun, Y. Wang, S. Deng, G. Qian, M. Wang and A. Hu, J. Mater. Chem. C, 2013, 1, 580–586.
11. Y. Song, S. Zhu and B. Yang, RSC Adv., 2014, 4, 27184–27200.
12. A. Louie, Chem. Rev., 2010, 110, 3146–3195.
13. C. S. Bonnet and É. Tóth, C. R. Chim., 2010, 13, 700–714.
14. P. Verwilst, S. Park, B. Yoon and J. S. Kim, Chem. Soc. Rev., 2015, 44, 1791–1806.
15. M. Doubrovin, I. Serganova, P. Mayer-Kuckuk, V. Ponomarev and R. G. Blasberg, Bioconjugate Chem., 2004, 15, 1376–1388.
16. P. Caravan, Chem. Soc. Rev., 2006, 35, 512–523.
17. P. Caravan, J. J. Ellison, T. J. McMurry and R. B. Lauffer, Chem. Rev., 1999, 99, 2293–2352.
18. E. J. Werner, A. Datta, C. J. Jocher and K. N. Raymond, Angew. Chem., Int. Ed., 2008, 47, 8568–8580.
19. S. Viswanathan, Z. Kovacs, K. N. Green, S. J. Ratnakar and A. D. Sherry, Chem. Rev., 2010, 110, 2960–3018.
20. A. Datta and K. N. Raymond, Acc. Chem. Res., 2009, 42, 938–947.
21. K. N. Raymond and V. C. Pierre, Bioconjugate Chem., 2004, 16, 3–8.
22. M. Botta and L. Tei, Eur. J. Inorg. Chem., 2012, 2012, 1945–1960.
23. P. Hermann, J. Kotek, V. Kubicek and I. Lukes, Dalton Trans., 2008, 3027–3047.
24. R. B. Lauffer, Chem. Rev., 1987, 87, 901–927.
25. Q. Liu, S. Chen, J. Chen and J. Du, Macromolecules, 2015, 48, 739–749.
26. X. Ren, L. Liu, Y. Li, Q. Dai, M. Zhang and X. Jing, J. Mater. Chem. B, 2014, 2, 5541–5549.
27. H. Chen, G. D. Wang, W. Tang, T. Todd, Z. Zhen, C. Tsang, K. Hekmatyar, T. Cowger, R. B. Hubbard, W. Zhang, J. Stickney, B. Shen and J. Xie, Adv. Mater., 2014, 26, 6761–6766.
28. A. B. Bourlinos, A. Bakandritsos, A. Kouloumpis, D. Gournis, M. Krysmann, E. P. Giannelis, K. Polakova, K. Safarova, K. Hola and R. Zboril, J. Mater. Chem., 2012, 22, 23327–23330.
29. B. Li, X. Wang, Y. Guo, A. Iqbal, W. Liu, W. Qin, Y. Dong, W. Li, S. Chen, X. Zhou and Y. Yang, Dalton Trans., 2016 10.1039/C5DT04488J.
30. Y. P. Shi, Y. Pan, J. Zhong, J. Yang, J. H. Zheng, J. L. Cheng, R. Song and C. Q. Yi, Carbon, 2015, 93, 742–750.
31. K. Saatchi, P. Soema, N. Gelder, R. Misri, K. McPhee, J. H. E. Baker, S. A. Reinsberg, D. E. Brooks and U. O. Häfeli, Bioconjugate Chem., 2012, 23, 372–381.
32. X. Cui, D. Mathe, N. Kovács, I. Horváth, M. Jauregui-Osoro, R. Torres Martin de Rosales, G. E. D. Mullen, W. Wong, Y. Yan, D. Krüger, A. N. Khlobystov, M. Gimenez-Lopez, M. Semjeni, K. Szigeti, D. S. Veres, H. Lu, I. Hernández, W. P. Gillin, A. Protti, K. K. Petik, M. A. Green and P. J. Blower, Bioconjugate Chem., 2015, 27, 319–328.
33. J.-W. Shen, C.-X. Yang, L.-X. Dong, H.-R. Sun, K. Gao and X.-P. Yan, Anal. Chem., 2013, 85, 12166–12172.
34. S. I. Stupp and L. C. Palmer, Chem. Mater., 2014, 26, 507–518.
35. M. Antonietti and C. Göltner, Angew. Chem., Int. Ed., 1997, 36, 910–928.
36. C. F. J. Faul and M. Antonietti, Adv. Mater., 2003, 15, 673–683.
37. H. Wang, H. Wu, H. Shen, S. Geng, B. Wang, Y. Wang, X. Ma, G. Li and M. Tan, J. Mater. Chem. B, 2015, 3, 8832–8841.
38. Y. Chen, H. Yang, W. Tang, X. Cui, W. Wang, X. Chen, Y. Yuan and A. Hu, J. Mater. Chem. B, 2013, 1, 5443–5449.
39. Y. Chen, Q. Zhu, X. Cui, W. Tang, H. Yang, Y. Yuan and A. Hu, Chem.–Eur. J., 2014, 20, 12477–12482.
40. Y. Chen, Q. Zhu, Y. Tian, W. Tang, F. Pan, R. Xiong, Y. Yuan and A. Hu, Polym. Chem., 2015, 6, 1521–1526.
41. Y. Chen, H. Yang, W. Tang, X. Cui, W. Wang, X. Chen, Y. Yuan and A. Hu, J. Mater. Chem. B, 2013, 1, 5443–5449.
42. S. Zhu, Q. Meng, L. Wang, J. Zhang, Y. Song, H. Jin, K. Zhang, H. Sun, H. Wang and B. Yang, Angew. Chem., Int. Ed., 2013, 52, 3953–3957.
43. J. R. Lakowicz, Principles of Fluorescence Spectroscopy, Kluwer Academic/Plenum Publishers, New York, 2nd edn, 1999.
44. Z. Y. Zhou, N. Tian, J. T. Li, I. Broadwell and S. G. Sun, Chem. Soc. Rev., 2011, 40, 4167–4185.
45. Y. Liu, L. Yang and R. Guo, Soft Matter, 2013, 9, 3671–3680.
46. R. Hayes, G. G. Warr and R. Atkin, Chem. Rev., 2015, 115, 6357–6426.
47. G. Nizri, A. Makarsky, S. Magdassi and Y. Talmon, Langmuir, 2009, 25, 1980–1985.
48. N. Yamada, K. Ariga, M. Naito, K. Matsubara and E. Koyama, J. Am. Chem. Soc., 1998, 120, 12192–12199.
49. Y. Wang and A. Hu, J. Mater. Chem. C, 2014, 2, 6921–6939.
50. J. Shang, L. Ma, J. Li, W. Ai, T. Yu and G. G. Gurzadyan, Sci. Rep., 2012, 2, 792.
51. G. Eda, Y.-Y. Lin, C. Mattevi, H. Yamaguchi, H.-A. Chen, I. S. Chen, C.-W. Chen and M. Chhowalla, Adv. Mater., 2010, 22, 505–509.
52. D. N. Li, X. Hu, Q. Li, L. J. Dong, J. Huang, H. R. Li, H. A. Xie and C. X. Xiong, New J. Chem., 2013, 37, 3857–3860.
53. Q. S. Huo, Z. A. Qiao, Y. F. Wang, Y. Gao, H. W. Li, T. Y. Dai and Y. L. Liu, Chem. Commun., 2010, 46, 8812–8814.
54. Y. P. Sun, L. Cao, X. Wang, M. J. Meziani, F. S. Lu, H. F. Wang, P. J. G. Luo, Y. Lin, B. A. Harruff, L. M. Veca, D. Murray and S. Y. Xie, J. Am. Chem. Soc., 2007, 129, 11318–11319.

---

Footnote

† Electronic supplementary information (ESI) available: DLS, TEM, UV-vis spectra and photograph image of the assemblies. See DOI: 10.1039/c6ra02654k

---