Tuning of photoluminescence on different surface functionalized carbon quantum dots

Abstract

A simple microwave assisted route has been formulated to synthesise various surface functionalized carbogenic quantum dots from biodegradable polysaccharides. The photoluminescence (PL) properties of such surface functional carbon quantum dots (CQDs) have been tuned by the in situ addition of high boiling organic solvents during the synthesis of CQDs under microwave irradiation. Several divalent cations have also been added to investigate the variation of PL intensity after cationic modifications. Some straightforward mechanistic approaches have been predicted to rationalize the enhancement or quenching of the fluorescence by the introduction of organic and inorganic substrates.

---

Semiconductor quantum dots have attracted great attention over the past two decades owing to their numerous applications on electronics, electrochemistry, biotechnology, sensors and catalysis. Although ZnS, ZnSe, CdS, CdSe and silicon quantum dots have been extensively used as the fluorescent probes for cell labelling, the cytotoxicity of the quantum dots make them practically useless for in vivo biological applications.^1,2 Recently carbon quantum dots (CQDs) have emerged as the most alternative fluorescent nanoprobes due to their availability, thermal stability and relatively non-cytotoxic nature.^3,4 Versatile applications of CQDs include bioimaging,⁵ catalysis,⁶ and photoreduction of metals⁷ owing to the high electron donor–acceptor ability. Of late CQDs have shown promising applications in bio-sensing,^8,9 especially as they exhibit peroxidase like activity.¹⁰ CQDs or carbon nanoparticles (CNPs) can be prepared by several existing techniques, such as microwave irradiation of sucrose¹¹ or polyethylene glycol;⁶ combustion of carbon soot,¹² activated carbons¹³ or carbon xerogels¹⁴ by nitric acid; proton beam irradiation of nanodiamonds;¹⁵ laser ablation of graphite;¹⁶ solvothermal degradation of graphene oxide;¹⁷ electrochemical oxidation of graphite¹⁸ and using silica spheres as a template material.¹⁹ Although the most fascinating features of CQDs are the PL properties with λ_ex dependant emission intensity²⁰ which is quite similar to semiconductor quantum dots, but the origin of such a fascinating property is a matter of debate. It is speculated that quantum effect, emissive traps²⁰ and mostly radiative recombination of excitons²¹ contribute to the PL property of CQDs. Yang et al. have proposed that the introduction of different functionalities onto the surface of the CQDs can introduce defects on the surface which can lead to a PL property, but the actual mechanism requires further clarification.²² Meanwhile proper surface passivation is required to enhance the PL intensity of such carbon nanomaterials with poor quantum yields. Different amine terminated organic molecules have been used to achieve the above requirement.²³ Post passivated CQDs exhibit an abrupt enhancement in their PL intensity, which is significantly higher than the original intensity. Sun and co-workers have demonstrated that doping of carbon quantum dots with ZnO and ZnS followed by surface passivation with PEG-1500_N increased the PL intensity.²⁴ A rapid single step laser passivation of carbon dots in organic solvents has also been reported by Li et al.²⁵ Recently, Wang et al.²⁶ reported a simple method for the preparation of fluorescent carbon nanoparticles from carbohydrate without any surface passivation. In that article, a beneficial attempt has been made to illustrate the enhancement of the PL-properties for such CQDs by the effect of several cations and anions. Though surface passivation is the most useful technique for improving the quantum yield, the influence of metal ions or organic substances on the PL-intensity is still opaque in several cases. Therefore in this article we have taken into account all these unresolved explanations and tried to bridge the PL intensity with different functionalities along with the influence of inorganic salts and organic solvents in the course of the reactions.

Herein, we describe a simple route to achieve carbon dots with three different surface functionalities, starting from three individual biocompatible and biodegradable polysaccharides and the effect of various inorganic metal ions and organic solvents on their photoluminescence properties. Biocompatible polysaccharides or carbohydrates could be selected to yield nontoxic CQDs after reaction^22,27 according to other research articles. The functionality of the CQDs has been adapted by a selection of suitable precursors (chitosan, alginic acid and starch) under identical reaction conditions. In summary, the introduction of divalent metal ions, e.g., Sn²⁺, Cd²⁺ and Zn²⁺ effectively increases the PL-intensity of carbonaceous nano dots generated from starch, in which Sn²⁺ has the highest effect and Cd²⁺ the lowest, in this series. However, the most interesting phenomenon has been observed after the addition of a small amount of Cu²⁺, which can drastically quench the PL. Systematic enhancement and quenching of fluorescence was observed in the presence of organic solvents for all the above sets of quantum dots synthesized from chitosan, alginic acid and starch. Elaborate studies on the PL intensities with variations in pH and wavelength have been carried out and plausible mechanisms for the enhancement and quenching of the PL intensities have been speculated. The metal doped highest fluorescent quantum dots were applied as a bioimaging agent against S. aureus as the model organism.

The morphologies and sizes of the carbon quantum dots synthesized from biocompatible polysaccharides were analyzed by TEM analysis. The quantum dots obtained from chitosan, alginic acid and starch were designated CCQD, ACQD and SCQD respectively. The TEM image (Fig. 1) clearly revealed that SCQDs have the smallest particle size distribution and hence underwent some sort of aggregation,⁴ whereas CCQDs with the sizes between 2 to 10 nm were completely spherical and distinct in nature. Although the size distribution for the SCQDs was quite regular, ranging between 1–2 nm, the morphology of SCQDs showed less uniformity in comparison to CCQDs. Again the particle sizes of ACQDs were in between 2 to 4 nm and they were also distinct and spherical in comparison to SCQDs. As it had the smallest particle size, the fluorescence intensity (Fig. 2a) of SCQD was highest among the series, whereas CCQD had the lowest PL-intensity as it was comprised of relatively larger particles with sizes ranging from 2 to 10 nm.⁴ The PL-intensity of ACQD was in between the intensity of CCQD and SCQD, which was also expected from the TEM images represented in Fig. 1. The UV-Vis spectra for CCQD, ACQD and SCQD were shown in Fig. 2b, which demonstrated that the peak for SCQD was centered at 242 nm, whereas ACQD and CCQD both expressed their maxima at around 243 nm. However the nature of the absorbance peaks for ACQD and SCQD were very similar to each other. Fig. S1, ESI† described the variation of PL against the pH of the solution containing the quantum dots; which signified that the peak positions were unaltered for variations in pH for both ACQD and SCQD. A red shift phenomenon was observed for CCQD when raising the pH, which suggested that at alkaline pH there might be some sort of structural deformation. The highest PL-intensity for CCQD was observed at pH 3, while for ACQD and SCQD the maxima were noted at pH 9 and 7 respectively. The above result clearly indicated that the most stable particles for CCQD, SCQD and ACQD were formed at pH 3, 7 and 9 respectively. This was might be due to the surface charge and groups fabricated due the selection of different functionalized precursors. CCQD and ACQD most likely contained amine (chitosan) and carboxylic acid (alginic acid) respectively as the major constituent; while no or very little acidic or base specific groups are present in SCQDs as it was fabricated from a neutral species (starch). Since we were unable to separate CQDs from PEG-200, the functional groups on their surface could not be evaluated from their direct FTIR analysis. To elucidate the probable surface functionality, a parallel experiment was carried out; where crude carbons were obtained after treatment of each polysaccharide precursor with sulphuric acid.

Fig. 1 TEM image of (a) SCQD, (b) ACQD, (c) and (d) CCQD. The inset in Fig. 1a displays the SAED pattern of SCQD, whereas the inset in Fig. 1b illustrates the high resolution image of the corresponding ACQD represented at Fig. 1b.

Fig. 2 (a) Fluorescence and (b) UV-Visible spectra of CCQD, ACQD and SCQD.

The FTIR spectra (Fig. S2, ESI†) of the carbon material obtained from chitosan exhibited peaks at 3430 cm⁻¹, 2932 cm⁻¹, 1609 cm⁻¹, 1409 cm⁻¹, 1217 cm⁻¹, 1103 cm⁻¹ and 1046 cm⁻¹, indicating the presence of –O–H, C–H, C=O, –NH₂, C–O–C and C–O bonds, respectively. A very small peak appeared at 1716 cm⁻¹, representing the very small amount of carboxylic acid groups present on the surface. On the contrary, the material prepared using alginic acid gave three sharp peaks centred at 3438 cm⁻¹, 1631 cm⁻¹ and 1125 cm⁻¹, revealing the presence of –O–H, –COOH and C–O groups, respectively. The FTIR peaks obtained after pyrolysis of starch were almost identical to the particles obtained from chitosan, except the peak acquired at 1409 cm⁻¹ due to the absence of amine functionality. Therefore it was concluded that the nano carbons produced from chitosan (CCQD) contained –NH₂ groups, while the CQDs obtained from alginic acid (ACQD) were grafted with –COOH groups. Similarly, the surface of SCQD must be covered by carbonyl moieties and a small number of –COOH groups, but no amine functionality. These surface functionalities could be easily explained by the stability of CCQD, ACQD and SCQD at pH 3, 9 and 7 respectively (Fig. S1, ESI†); since the amine functionalized CCQDs, carboxylic acid functionalized ACQDs and almost neutral hydroxyl containing SCQDs should be most stable at acidic (pH 3), basic (pH 9) and neutral (pH 7) medium. Fig. S3, ESI† clarified how the PL-properties were varied with the variation in excitation wavelength. The maximum intensity for CCQD was observed at 380 nm, while both ACQD and SCQD showed the highest PL-intensity at 370 nm. A gradual red shift of maxima was noted for all of those quantum dots with excitation wavelength from 340 to 410 nm.

It is noteworthy to mention the most interesting part in this article was to focus on Fig. 3 which describes the change in PL-intensity of the QDs depending upon the presence of several organic solvents during the synthesis. Commencing from Fig. 3a, 3b and 3c, it could be easily understood that if a small amount of NMP was introduced during the synthesis of CCQD, ACQD or SCQD without altering the other reaction conditions; the PL-intensities for any of those carbogenic nano dots significantly increased, irrespective of their precursors. Even NMP exhibited a maximum PL intensity amongst all of the high boiling organic substances. On the contrary, the existence of aniline caused a dramatic quenching of fluorescence for each set of quantum dots and exhibited negligible PL intensity when excited at 340 nm. It was well recognized that the PL-property of a quantum dot depends on its stability, which was again consistent with the surrounding solvent molecules, making the particle more or less isolated and preventing non-radiative recombination by the surface modified organic layers.^28,29 Whenever a particle was transmuted into more isolated one, it could contribute more PL-intensity than its corresponding less acquirable or non-isolable particles. This was due to the fact that in a non-isolable state, the particles could transfer their excitation energy between themselves as they were in a close proximity to each other³⁰ and hence diminish their fluorescence properties. In our study, NMP followed a similar mechanistic route to passivate CQDs into their isolable forms; the carbonyl group of NMP strongly interacted with such CQDs by nanoparticle carbonyl co-ordination analogous to that previously reported by Shen et al.³¹. Again the mechanism of PL quenching in presence of aniline could be explained through the formation of an excited charge-transfer complex, attributed by transfer of an electron from aniline to the excited-state fluorophore or the carbogenic quantum dots.³² As a result, the PL-intensities of each and every particle were quenched by aniline. Though NMP and aniline showed the regular pattern for enhancement or quenching of PL intensities regardless of the precursor polysaccharides or their corresponding synthetic carbon nano materials, the introduction of γ-butyrolactone, triethanol amine and diethylene glycol displayed different phenomena depending upon the precursors. Fig. 3a and 3b revealed that a slight PL-enhancement was observed for both CCQD and ACQD in the presence of γ-butyrolactone, which was introduced before commencing the reaction. A red shifted peak was observed for CCQD, while a blue shifted phenomena was noticed for ACQD. Yet again, the use of triethanol amine and diethylene glycol played the opposite role for CCQD and ACQD. In CCQD, the PL-intensity was increased with the influence of the above substances, while for ACQD fluorescence quenching was observed with addition of both of those solvents. A blue shift was once again observed for CCQD with triethanol amine and a definite red shift was noticed for ACQD with diethylene glycol. These opposite phenomena could be explained by the inverse surface functionality of CCQD and ACQD. As CCQDs were surrounded with –NH₂ groups, the triethanol amine or diethylene glycol compelled the particles to move into a more isolated state due to the similar surface charge (positive) on the quantum dots and the introduced solvents. Therefore the particle experiences some repulsive interactions with those solvent molecules which isolated them from each other and increased their PL intensity. On the contrary, those compounds allow ACQDs into close proximity with each other as the particles contained –COOH groups on their surfaces. No other justification, such as pH or charge-transfer PL enhancement or quenching would be suitable, otherwise the reverse phenomena would be expected and CCQD and ACQD would give maximum PL-intensities at pH 3 and 9, respectively. If a charge-transfer complex was formed between the quantum dots (CCQD and ACQD) and the organic solvents then a sharp decrease in PL-intensity would be expected, analogous to the phenomenon observed for aniline.

Fig. 3 Photoluminescence spectra of (a) CCQD, (b) ACQD and (c) SCQD with introduction of different solvent before commencing the reaction.

The influence of anions and cations towards the PL-property were reported previously by Wang et al.,²⁶ which demonstrated that with the increasing valency of cations or anions or both, the PL-intensity increased gradually. However, no article has been published on the relationship between the PL-intensity of semiconductor carbogenic dots and the effect of different metal ions of same valance state. To clarify this effect, a 5 mL 0.001 M solution of Cu²⁺, Cd²⁺, Sn²⁺ and Zn²⁺ were separately and homogeneously mixed with a 10 mL 0.1 mg mL⁻¹ solution of starch before microwave irradiation. Unlike SCQDs, the TEM image (Fig. 4a) of Sn²⁺ modified SCQD evidently revealed that the particles were more discrete as they were large in comparison to the original SCQDs. The particle sizes were between 4 to 8 nm, suggesting that there was a strong interaction between Sn²⁺ and SCQDs, which could forcefully build a stable complex by means of enlarging their basal particle size. The inset image in Fig. 4a shows the corresponding SAED pattern for Sn²⁺ modified SCQDs, disclosing the crystallinity of the sample that confirms the doping of Sn²⁺ into the interior of the carbon dots, as the SAED pattern for SCQD (inset of Fig. 1a) demonstrated that it was completely amorphous in nature. The d-spacing calculated from SAED in Fig. 1a depicted the existence of [101] (d = 2.369 A°), [211] (d = 1.465 A°) and [202] (d = 1.221 A°) planes of Sn(0) (JCPDS card no. 19-1365) in Sn²⁺ modified SCQD. The presence of Sn atoms was further confirmed by the corresponding EDAX analysis (Fig. 4b). This fascinating response once more confirmed the formation of a perpetual complex between Sn²⁺ and SCQD. Similar to the Sn²⁺ modified SCQD, Cd²⁺ modified carbon dots were homogeneous and discrete in nature (Fig. 4c) with a particle size ranging from 2 to 5 nm. The inset image in Fig. 4c illustrated that it was polycrystalline in state. The existence of Cd atoms was again confirmed by the EDAX analysis given in Fig. 4d. Similarly, the TEM image and its corresponding SAED pattern shown in Fig. 4e signified that after modification with Zn²⁺, though the morphology and tendency to agglomerate remained unchanged, the nature of the sample moved from amorphous to polycrystalline. On the contrary, after the introduction of Cu²⁺, the product was pure crystalline (inset in Fig. 4f) with [220], [440] and [641] crystal planes indexing the d-spacing values of 2.160A°, 1.077A° and 0.822A° for cuprites respectively [JCPDS card no. 02-1067].

Fig. 4 TEM image of SCQD after modification with (a) Sn²⁺, (c) Cd²⁺, (e) Zn²⁺ and (d) Cu²⁺. The inset image illustrates the corresponding SAED patterns for the individual metal ion modified SCQD. Fig. 4b and 4d are the corresponding EDAX analysis for Fig. 4a and 4c respectively.

The UV-Vis absorbance for all of the compounds were shown in Fig. 5a, which revealed that except for Cu²⁺, all other metal ion modified CQDs gave the peaks centred at similar positions to CQD itself. Although the UV-Vis absorption spectra of SCQD and the passivated carbons by the introduction of Cd²⁺, Sn²⁺, Zn²⁺ were centred near 243 nm, the peak for Cu²⁺ coated CQDs was shifted towards a lower wavelength and centred at 221 nm. Apart from the UV-Vis absorbance spectra, Cu²⁺ played an opposing role in comparison to the other metal ions for the PL spectra. Fig. 5b shows that Cd²⁺, Sn²⁺ and Zn²⁺ enhanced the PL-intensity while Cu²⁺ was responsible for PL-quenching. This PL-quenching by Cu²⁺ might be attributed to the substantial interaction between Cu²⁺ and CQDs, inducing the donation of electrons from the fluorophore to the quencher.³³ Among all of those cations, Sn²⁺ imposed the highest PL intensity and the order of PL-intensities was as follows: SCQD with Sn²⁺ > SCQD with Zn²⁺ > SCQD with Cd²⁺ >SCQD > SCQD with Cu²⁺

Fig. 5 (a) UV-Visible absorption and the (b) fluorescence spectra of SCQD after modification with metal ions (c) fluorescence spectra of Sn²⁺ modified SCQD with variation of excitation wavelength (d) normalized PL pattern of Fig. 5c.

Due to the paramagnetic nature from the unfilled d⁹ electronic configuration of Cu²⁺, it could easily quench the emission of the fluorescent probes via photoinduced electron transfer (PET) or protoinduced-metal energy transfer mechanisms.³⁴ Therefore Cu²⁺ is converted to Cu⁺ after trapping the surface of SCQD. The existence of Cu⁺ rather than Cu²⁺ was confirmed by the SAED pattern discussed in the previous section (Fig. 4f, inset), where cuprite (Cu₂O) but not CuO was isolated on the surface of the quantum dots. The higher PL intensity from the Sn²⁺, Zn²⁺ and Cd²⁺ modified QDs was explained by the effect of surface states on the sensing property of the QDs as the modified CQDs showed greatly enhanced fluorescence intensity over the unmodified CQDs. The enhancement of PL-intensity by Cd²⁺ could be explained on the basis of photoinduced electron transfer (PET), whereas Zn²⁺ might follow the mechanism of internal charge transfer (ICT).³⁵ Although in some cases the mechanism of PL enhancement for Zn²⁺ adopted the pathway of ICT due to the closed shell electronic configuration (d¹⁰) for both Zn²⁺ and Cd²⁺; the PET mechanism was also very common for both of the metal ions.³⁴ Therefore, in this experiment the sensitivity of Zn²⁺ and Cd²⁺ was not only due to their charge transfer mechanism, but on the binding between the metal ions and the CQDs.²⁶ Most probably due to the small size of Zn²⁺ compared to Cd²⁺, the former had a strong affinity and specificity³⁶ to bind with CQDs in comparison to Cd²⁺. Therefore Zn²⁺ acted upon as a superior passivating agent. PL enhancement by Sn²⁺ is very uncommon in any practical situation as it can act like a partial fluorescent quencher^37,38 for several fluorophores. This phenomenon might be due to the deposition of metallic tin instead of metal ions into the interior of CQDs by the transformation of Sn²⁺ to Sn (0) during microwave irradiation. Therefore, the introduction of Sn²⁺ would afford crystallinity towards the quantum dots, as shown in Fig. 4a, and provide more nucleation sites for the CQDs³⁹ responsible for the enhancement in PL-intensity. Like SCQDs, Sn²⁺ modified carbon dots gave the highest PL-intensity at 370 nm excitation wavelength (Fig. 5c), which gradually decreased at either higher or lower excitation wavelengths. It was clearly observed that the compound also red shifted (Fig. 5d) when excited, from 330 to 410 nm.

The strong interactions of the metal ions, especially Sn²⁺and Cu²⁺, with the carbon dots was further clarified by determining the lifetime of the SCQDs and the metal ion modified SCQDs. The fluorescence lifetime was carried out at the emission maxima for unmodified and modified carbon dots. An interesting result came out for the Sn²⁺and Cu²⁺ modified SCQDs, in which the former decreased the lifetime for the quantum dots, whereas it was increased by the influence of the latter. The lifetime decays for SCQDs in the absence and presence of metal ions are shown in Fig. S4, ESI.† All the decays of the SCQDs with or without modification with metal ions were tri-exponential in nature. The average lifetime was determined by the equation given below:

τ_av = a₁τ₁ + a₂τ₂ + a₃τ₃

Where τ₁, τ₂, τ₃ were the first, second and third component of the decay time of the carbonaceous quantum dots and a₁, a₂, a₃ were the corresponding relative weightings of these components, respectively. The average lifetime for SCQDs at 480 nm was 1.63 ns, which increased to 1.76 ns with the introduction of Cu²⁺ and deceased to 1.56 ns after modification with Sn²⁺. The lifetime of the fluorophore was roughly unaltered by the addition of small amounts of Cd²⁺ and Zn²⁺. All of the values were summarized in Table S1 (ESI, †).

Finally the permeability of the particles into the cell membrane for bioimaging was examined by the simple incubation of Sn²⁺ modified SCQD with S. aureus as the model organism. The nano dots penetrated into the bacterial cell quite easily by simple incubation at 37 °C for 4 h. After entering into the intracellular part of the bacterial cells, they retained their high PL property and emitted bright blue and green fluorescent light (Fig. S5, ESI†) at 410 and 360 nm excitation respectively under a fluorescent microscope.

In conclusion, we have demonstrated the syntheses of carbon quantum dots with different functional groups on their surfaces. The surface decoration was controlled by the selection of suitable biodegradable polymeric precursors with amine, carboxylic acid or hydroxyl moieties. By this approach we were able to fabricate individual carbon quantum dots with different surface functionalities. A probable mechanism has been speculated to discuss the enhancement and quenching of fluorescence against the different organic solvents with high boiling points. The effect of surface functionality on optical excitation and relaxation has been investigated by variations in the pH and functional groups adjoined to the foreign substances. Finally the mechanism for varying the PL intensity by inorganic divalent cations has been evaluated using starch as the precursor, as it is able to produce the most stable particles at pH 7. Sn²⁺ modified SCQD possessed a maximum PL intensity amongst all the variations and was used as a bioimaging agent against S. aureus, the model organism in this study.

The authors are grateful to the TIFAC and the CSIR, Government of India for funding. Authors are grateful to Mr. Chiranjib Banerjee of IIT, Kharagpur (Department of Chemistry) for TCSPC analysis. Authors would also like to acknowledge the authority of Indian Institute of Technology, Kharagpur and Indian Statistical Institute, Kolkata for providing working facilities.

References

1. J. Kim, Y. Park, T. H. Yoon, C. S. Yoon and K. Choi, Aquat. Toxicol., 2010, 97, 116.
2. R. Hardman, Environ. Health Perspect., 2006, 114, 165.
3. S. T. Yang, X. Wang, H. Wang, F. Lu, P. G. Luo, L. Cao, M. J. Meziani, J. H. Liu, Y. Liu, M. Chen, Y. Huang and Y. P. Sun, J. Phys. Chem. C, 2009, 113, 18110.
4. S. C. Ray, A. Saha, N. R. Jana and R. Sarkar, J. Phys. Chem. C, 2009, 113, 18546.
5. C. Liu, P. Zhang, F. Tian, W. Li, F. Li and W. Liu, J. Mater. Chem., 2011, 21, 13163.
6. S. Mitra, S. Chandra, P. Patra, P. Pramanik and A. Goswami, J. Mater. Chem., 2011, 21, 17638.
7. X. Wang, L. Cao, F. S. Lu, M. J. Meziani, H. Li, G. Qi, B. Zhou, B. A. Harruff, F. Kermarrec and Y. P. Sun, Chem. Commun., 2009, 3774.
8. L. Zhou, Y. Lin, Z. Huang, J. Ren and X. Qu, Chem. Commun., 2012, 48, 1147.
9. W. Wei, C. Xu, J. Ren, B. Xu and X. Qu, Chem. Commun., 2012, 48, 1284.
10. X. Wang, K. Qu, B. Xu, J. Ren and X. Qu, Nano Res., 2011, 4, 908.
11. S. Chandra, P. Das, S. Bag, D. Laha and P. Pramanik, Nanoscale, 2011, 3, 1533.
12. H. Liu, T. Ye and C. Mao, Angew. Chem., Int. Ed., 2007, 46, 6473.
13. Y. Dong, N. Zhou, X. Lin, J. Lin, Y. Chi and G. Chen, Chem. Mater., 2010, 22, 5895.
14. S. Chandra, S. Mitra, D. Laha, S. Bag, P. Das, A. Goswami and P. Pramanik, Chem. Commun., 2011, 47, 8587.
15. S. J. Yu, M. W. Kang, H. C. Chang, K. M. Chen and Y. C. Yu, J. Am. Chem. Soc., 2005, 127, 17604.
16. L. Gao, X. Wang, M. J. Meziani, F. Lu, H. Wang, P. G. Luo, Y. Lin, B. A. Harruff, L. M. Veca, D. Murray, S. Y. Xie and Y. P. Sun, J. Am. Chem. Soc., 2007, 129, 11318.
17. S. Zhu, J. Zhang, C. Qiao, S. Tang, Y. Li, W. Yuan, B. Li, L. Tian, F. Liu, R. Hu, H. Gao, H. Wei, H. Zhang, H. Sun and B. Yang, Chem. Commun., 2011, 47, 6858.
18. Q. L. Zhao, Z. L. Zhang, B. H. Huang, J. Peng, M. Zhang and D. W. Pang, Chem. Commun., 2008, 5116.
19. T. W. Kim, P. W. Chung, I. I. Slowing, M. Tsunoda, E. S. Yeung and V. S. Y. Lin, Nano Lett., 2008, 8, 3724.
20. Y. P. Sun, B. Zhou, Y. Lin, W. Wang, K. A. S. Fernando, P. Pathak, M. J. Meziani, B. A. Harruff, X. Wang, H. F. Wang, P. J. G. Luo, H. Yang, M. E. Kose, B. L. Chen, L. M. Veca and S. Y. Xie, J. Am. Chem. Soc., 2006, 128, 7756.
21. S. N. Baker and G. A. Baker, Angew. Chem., Int. Ed., 2010, 49, 6726.
22. Y. Yang, J. Cui, M. Zheng, C. Hu, S. Tan, Y. Xiao, Q. Yang and Y. Liu, Chem. Commun., 2012, 48, 380.
23. H. Peng and J. Travas-Sejdic, Chem. Mater., 2009, 21, 5563–5565.
24. S. T. Yang, L. Cao, P. G. Luo, F. Lu, X. Wang, H. Wang, M. J. Meziani, Y. Liu, G. Qi and Y. P. Sun, J. Am. Chem. Soc., 2009, 131, 11308.
25. X. Li, H. Wang, Y. Shimizu, A. Pyatenko, K. Kawaguchi and N. Koshizaki, Chem. Commun., 2011, 47, 932.
26. X. Wang, K. Qu, B. Xu, J. Ren and X. Qu, J. Mater. Chem., 2011, 21, 2445.
27. B. R. Selvi, D. Jagadeesan, B. S. Suma, G. Nagashankar, M. Arif, K. Balasubramanyam, M. Eswaramoorthy and T. K. Kundu, Nano Lett., 2008, 8, 3182.
28. D. R. Jung, J. Kim and B. Park, Appl. Phys. Lett., 2010, 96, 211908.
29. S. A. Majetich and A. C. Carter, J. Phys. Chem., 1993, 97, 8727.
30. B. Ramachandram and A. Samanta, J. Phys. Chem. A, 1998, 102, 10579.
31. J. Shen, Y. Hu, M. Shi, N. Li, H. Ma and M. Ye, J. Phys. Chem. C, 2010, 114, 1498.
32. H. Knibbe, D. Rehm and A. Weller, Phys. Chem., 1968, 72, 257.
33. Y. Chen and Z. Rosenzweig, Anal. Chem., 2002, 74, 5132.
34. C. Lodeiro, J. L. Capelo, J. C. Mejuto, E. Oliveira, H. M. Santos, B. Pedras and C. Nunez, Chem. Soc. Rev., 2010, 39, 2948.
35. L. Xue, C. Liu and H. Jiang, Org. Lett., 2009, 11, 1655–1658.
36. G. K. Walkup and B. Imperiali, J. Am. Chem. Soc., 1996, 118, 3053.
37. S. W. Thomas, G. D. Joly and T. M. Swager, Chem. Rev., 2007, 107, 1339.
38. M. Boiocchi, L. Fabbrizzi, M. Licchelli, D. Sacchi, M. Vazquez and C. Zampa, Chem. Commun., 2003, 1812.
39. C. Ma, L. Bi and J. Y. Feng, J. Cryst. Growth, 2007, 305, 30.

---

Footnote

† Electronic Supplementary Information (ESI) available. See DOI: 10.1039/c2ra00030j/

---