Simultaneous enhancement of Raman scattering and fluorescence emission on graphene quantum dot-spiky magnetoplasmonic supra-particle composite films

Abstract

Graphene quantum dots (GQDs) are a new type of quantum dot that have low cytotoxicity, excellent solubility, chemical inertia, and stable photoluminescence and Raman signal, and are promising materials for various applications. Due to their Raman scattering (RS)-fluorescence emission (FE) multi-mode optical property, GQDs have been considered as an important candidate for applications in photovoltaics, light-emitting devices, bioimaging and diagnostic techniques. However, GQD-related applications are severely restricted because of the high signal attenuation, low contrast, and low resolution. To overcome these problems, we have designed a reliable strategy for enhancing the RS-FE of GQDs to make them suitably efficient for their applications. In this study, we utilized spiky Au-coated Fe₃O₄ supraparticles (Fe₃O₄@Au SPs) to enhance the RS-FE of the GQDs by fabricating layer-by-layer (LbL) assembled GQD-spiky Fe₃O₄@Au SPs composite films. The RS and FE enhancements of 13-fold and 7.8-fold, respectively, were observed for the GQD-spiky Fe₃O₄@Au SPs composite films. It is also proved that external magnetic fields have an attenuation effect on the Raman activity of spiky Fe₃O₄@Au SPs. Moreover, the GQD-spiky Fe₃O₄@Au SPs composite films exhibited remarkable multi-mode imaging capabilities for Raman mapping and fluorescence imaging, which are expected to find application in photonic devices.

---

Introduction

Recently, multi-mode probes based on nanomaterials have been widely explored for enhanced cellular imaging and biomedical diagnostic techniques, such as nanoprobes integrating Raman scattering (RS) measurements with fluorescence emission (FE), X-ray computed tomography (CT), or magnetic resonance imaging (MRI) signals.^1–4 In particular, RS-FE duplex nanoprobes appear to be a promising route to improve imaging quality; the fluorescence signal is an immediate and intuitive indicator, while the RS signal is used to distinguish specific targets from the multiplex interactions.^5,6 Therefore, RS-FE duplex nanoprobes can be an extremely sensitive analytic tool for a variety of applications.

Graphene quantum dots (GQDs), a new type of multi-mode nanoprobes, have recently emerged and ignited tremendous research interest.^7,8 The pronounced quantum confinement and edge effects in GQDs result in numerous novel chemical and physical properties, such as low cytotoxicity, excellent solubility, chemical inertia, stable photoluminescence, and strong Raman scattering, that have been utilized for novel applications in optoelectronic devices, sensors, and biological imaging.⁹ Therefore, GQDs can be considered as ideal RS-FE duplex nanoprobes. Despite their potential success as optical nanoprobe, GQDs have intrinsic problems, such as signal attenuation, low contrast, and low resolution. These issues have prevented their utilization in the applications.

It has long been known that the optical properties of quantum dots can be dramatically altered when they are adsorbed onto or near a rough metallic surface. There have been many studies focused on enhancing the RS or FE from adsorbates on rough metallic surfaces. One effect that has been shown to enhance the RS or FE substantially is the electromagnetic interactions between the optical fields, quantum dots, and electronic plasma resonances localized on the roughness features of the metal surface. The presence of nearby metallic nanoparticles (NPs) can not only enhance the fluorescence intensity, but also stabilize adjacent quantum dots against photobleaching, further improving their practical use in bioimaging and biosensor applications.^10–14

The development of multifunctional magnetoplasmonic supraparticles (MPSPs) provides new materials for designing RS-FE multi-mode nanomaterials. The MPSPs integrate Au nanostructures with superparamagnetic Fe₃O₄ core to combine the plasmonic and magnetic properties in a confined cluster.^15–17 The MPSPs have the potential to lead the development of various applications, such as immunomagnetic separation under plasmonic imaging monitoring,¹⁸ biosensors,^11,19,20 dual mode imaging (e.g., MRI and CT imaging),^21,22 and the detection of surface-enhanced Raman scattering (SERS).^23,24 Previously, we have developed a synthetic route for the fabrication of self-assembled, spiky, Fe₃O₄@Au supraparticles (SPs) with diameters >50 nm, which were characterized as having high SERS activity, high colloidal stability, and good magnetic properties for biological sensing and imaging applications.^16,20

The surface roughness has long been considered one of the critical parameters for optimizing the metal-enhanced fluorescence and SERS. It has enabled the precise control over localized surface plasmon resonance (LSPR), as well as surface plasmon polaritons (SPPs).^12,25 Tuneable plasmonic properties can be achieved by controlling the thickness of the Au nanoshell or the morphology of the MPSPs.^16,20 The sensitivity of SERS is strongly dependent on the surface morphology of the Au NPs, and thus, it is expected that the spiky Fe₃O₄@Au SPs could act as highly active substrates for designing RS-FE multi-mode nanomaterials. Therefore, study of the RS-fluorescence enhancement of GQDs when using the spiky Fe₃O₄@Au SPs as the active substrate is essential.

Layer-by-layer (LbL) self-assembly, as a versatile, bottom-up nanofabrication technique, exhibits prominent advantages over conventional approaches on versatility and simplicity, and furnishes molecular-level control over the thickness, structure, and composition of the multilayered films with simple benchmark operations.^26–28 Over the past few decades, LbL assembly of thin films has been of considerable interest because of its ability to exert nanometers control over film thickness and its extensive choice of usable materials for coating planar and particulates substrates.²⁹ Therefore, the GQD-spiky Fe₃O₄@Au SPs composite films can be constructed through LbL self-assembly approach, in which GQDs and spiky Fe₃O₄@Au SPs were employed as nanobuilding blocks for the self-assembly of well-defined multilayered films. The poly-electrolytes exert a profound influence on RS-fluorescence enhancement of GQDs, which acts as a robust scaffold to control the distance between GQDs and spiky Fe₃O₄@Au SPs for LbL self-assembly.

In this study, we constructed GQD-spiky Fe₃O₄@Au SP composite films with the layer-by-layer (LbL) assembly method for the simultaneous enhancement of RS and FE (as shown in Fig. 1). This platform is composed of two components: GQDs acting as a duplex nanoprobe and spiky Fe₃O₄@Au SPs acting as the active substrate. With the spiky Fe₃O₄@Au SPs, the GQD nanoprobe generated strongly enhanced RS and FE signals. Due to the magnetic property of spiky Fe₃O₄@Au SPs, we also studied the effect of external magnetic field on the Raman enhancement. Furthermore, we evaluated the imaging capabilities of the GQD-spiky Fe₃O₄@Au SP films.

Fig. 1 Schematic representation of the layer-by-layer-assembled GQD-spiky Fe₃O₄@Au SP films.

Experimental section

Materials

Hydrogen tetrachloroaurate(III) trihydrate (HAuCl₄·3H₂O, purity = 99.9%), sodium citrate (Na₃C₆H₅O₇), FeCl₃·6H₂O, poly(diallyldimethylammonium chloride) (PDDA, molecular weight (M.W.) = 400 000–500 000), poly(acrylic acid) (PAA, M.W. ≈ 450 000) and hydroquinone, FeCl₂·4H₂O, 28% ammonia solution were obtained from Sigma-Aldrich. Pitch-based carbon fiber was purchased from Fibre Glast Development Corporation (United States). Deionized water (>18.2 mΩ cm⁻¹) was used throughout the experimental procedure. All chemicals were of analytical grade and used as received.

Synthesis of spiky Fe₃O₄@Au SPs

The spiky Au coating was achieved via the reduction of Au³⁺ on the surface of spherical, self-assembled, Fe₃O₄@Au seeds. Details of the self-assembly method can be found in our previous publication.¹⁶ For a typical synthesis of the spiky Fe₃O₄@Au SPs, 10 mL of aqueous HAuCl₄ (0.25 mM) was added with vigorous stirring. Subsequently, 500 μL of the Fe₃O₄@Au seed solution and 500 μL of the hydroquinone solution (30 mM) were added dropwise. The solution was then stirred at room temperature for 30 min.

Synthesis of GQDs

The GQDs were prepared via the chemical oxidation and cutting of micrometer-sized, pitch-based, carbon fiber according to the method reported by Peng et al.³⁰ The pitch-based carbon fiber (0.30 g) was added to a mixture of concentrated H₂SO₄ (60 mL) and HNO₃ (20 mL). The solution was sonicated for 2 h and stirred for 48 h at 100 °C. The mixture was cooled and diluted with deionized water (800 mL). The solution's pH was adjusted to 8 with Na₂CO₃. The solution containing the final product was further dialyzed in a dialysis bag (retained molecular weight = 2 kDa) for 3 days.

Preparation of GQD-spiky Fe₃O₄@Au SP nanocomposite films

Silica wafers (1 × 1.5 cm) were cleaned by immersion in piranha solution (3 : 1 mixture of H₂SO₄ : H₂O₂) for 1 h and then thoroughly rinsed with deionized water; the wafers were kept in deionized water prior to use. A 1 wt% solution of PDDA and a 1 mg mL⁻¹ solution of PAA were prepared. The glass substrates were consecutively immersed in the PDDA and PAA solutions for 10 min per solution. The washing process between the consecutive adsorptions was carried out with copious amounts of deionized water, and N₂ gas drying was performed to build the (PDDA/PAA)₂/PDDA base layers. These substrates were then individually immersed in the spiky Fe₃O₄@Au SP solution for 10 min. After repeating the washing and drying steps, the substrates were immersed in the PDDA/PAA/PDDA solutions to produce a spacer layer between the metallic surface and the GQDs. Finally, the substrates were immersed in the GQD solution for 5 min.

Optical and microscopic characterization

The absorbance of the spiky Fe₃O₄@Au SPs solution was measured with UV-Vis spectroscopy (SCINCO, S310, South Korea) and photoluminescence spectroscopy (Hitachi, F-7000, Japan). The morphology and size of the SPs were characterized with high-resolution transmittance electron microscopy (HR-TEM, JEOL, JEM-3010, Japan) and atomic force microscopy (AFM, Park System Inc., XE7, South Korea). The surface potentials and particle-size distribution of the SPs were monitored with a Zetasizer (Malvern Instruments, Nano ZS, UK) (see Fig. S1†). The change in PL of the samples was measured at an angle of 45° with respect to the incident light using a fluorescence spectrophotometer (Hitachi, F7000, Japan). Raman spectroscopy (Ramboss 500i, Dongwoo Optron Inc., South Korea) was performed with a 532 nm excitation laser (2.36 mW at the sample's surface) and Rayleigh-scattered light was removed with a holographic notch filter. The RS light was directed to an Andor Shamrock spectrograph equipped with a charge-coupled device (CCD).

Results and discussion

The sophisticated morphology of the nanostructures may alter the effective radii of the particle. Thus, the morphology of the spiky SPs can significantly increase the effective radii of the NPs compared to spherical NPs with the same geometric core volumes. According to the Mie theory, the spiky Fe₃O₄@Au SPs may possess much higher levels of surface plasmon resonance (SPR) than spherical NPs with similar geometric volumes. Fig. 2A shows typical UV-Vis spectra of spherical Au NPs and spiky Fe₃O₄@Au SPs. The spherical Au NPs show a maximum absorption centered on 534 nm, while the SPR peak of the spiky SPs is centered on 566 nm. Therefore, the optical response of the spiky SPs is dominated by the two different plasmon modes associated with a hollow spherical structure and many short spike structure.³¹ The plasmons of the spiky Fe₃O₄@Au SPs arise because of the hybridization of the plasmons supported by the inner radius (Au cavity), the outer radius of the shell layer (spiky Au surface), and low frequency plasmon oscillations at the spikes. Thus, the plasmon bands of the spiky SPs are red-shifted with respect to those for similarly sized spherical Au NPs.³²

Fig. 2 (A) UV-Vis spectra of the spherical Au NPs and spiky Fe₃O₄@Au SPs. (B) UV-Vis (a) and PL (b) spectra of the GQDs. TEM images of the (C) spherical Au NPs, (D) spiky Fe₃O₄@Au SPs, and (E) GQDs. The GQDs are outlined with white circles.

GQDs, which are edge-bound, nanometer-sized, pieces of graphene, have fascinating optical and electronic properties.⁸ Fig. 2B shows the UV-Vis absorption and PL spectra of the prepared GQDs. In the UV-Vis spectrum of the GQDs in aqueous solution, the typical absorption peak at 290 nm could be assigned to the π–π* transition of graphitic sp₂ domains. To explore the optical properties of the water soluble GQDs further, their PL spectrum in aqueous solution was measured with a 290 nm excitation wavelength. As shown in Fig. 2B, the PL spectrum of the GQDs shows the strongest peak at 406 nm. Fig. 2C–E shows TEM images of the spherical Au NPs, spiky Fe₃O₄@Au SPs, and GQDs, respectively. These TEM micrographs show that the Au NPs are spherical in shape and have a size of 53.91 ± 2.34 nm, while the diameter of the GQDs is 4.35 ± 0.77 nm. In addition, the as-synthesized spiky Fe₃O₄@Au SPs are uniform in size and shape, with an average diameter of 54.28 ± 2.19 nm. The mean diameters of all the NPs were calculated from TEM images (see the ESI† for more details).

Metal-enhanced fluorescence is an established phenomenon where the interaction between the quantum dots and metallic NPs results in enhanced fluorescence, increased photostability, and decreased lifetime.³³ On a rough metallic surface, the scattering of SPP modes can produce photons that can decrease the diffraction limit and resolve the sub-wavelength structure, thereby unlocking the prospect of utilizing metal–semiconductor nanocomposite films to enhance PL emission.¹² Fig. 3A shows the PL spectra of the GQDs, GQD-spherical Au NP, and GQD-spiky Fe₃O₄@Au SP LbL-assembled films (excitation wavelength = 290 nm). As can be seen, there is an emission enhancement greater than 13-fold when the GQDs are combined with the spiky Fe₃O₄@Au SPs, while the emission enhancement for the GQD-spherical Au NP film is only 9-fold. These results show that the spiky Fe₃O₄@Au SPs are better than the spherical Au NPs for fabricating fluorescence-active tags. To get a statistically meaningful result, the relative standard deviation (RSD) of the PL was calculated. Fig. 3B shows that RSD of the PL intensity for GQDs, GQD-spherical Au NP, and GQD-spiky Fe₃O₄@Au SP films at 406 nm, were 0.36, 0.28 and 0.56, respectively.

Fig. 3 (A) PL spectra and (B) relative PL intensities of the GQDs, GQD-spherical Au NP, and GQD-spiky Fe₃O₄@Au SP films at 406 nm. (C) Raman spectra and (D) the relative intensities of the GQDs (Control), GQD-spherical Au NP, and GQD-spiky Fe₃O₄@Au SP LbL-assembled films at 1579 cm⁻¹.

It is difficult to determine the precise contribution of each process responsible for the fluorescence enhancement of the GQDs related with absorption, scattering, and the radiative decay rate because these processes are interdependent. Nevertheless, for the experimental parameters and NP geometries discussed here, the scattering efficiency appears to provide the most important mechanism for the fluorescence enhancement of the GQDs. The spiky Fe₃O₄@Au SPs predominantly enhance FE of the GQDs through the enhanced absorption caused by the high-intensity local field resulting from the longitudinal plasmon resonance. There is a significant difference between the scattering cross sections of the spherical Au NPs and spiky SPs, and therefore, the spiky SPs greatly increase the far-field coupling efficiency of the fluorescence emission compared to that of the spherical Au NPs.^34,35 As shown in Fig. 2A, the spiky SPs show a stronger surface plasmon resonance than the spherical NPs. These properties explain why the fluorescence enhancement of the GQDs by the spiky SPs is more than 1.4-fold that by the spherical Au NPs.

The resulting electromagnetic field enhancement around metallic surfaces under laser excitation means some noble metallic nanostructures are very effective as RS-active materials. The Raman spectrum of the GQDs (Fig. 3C) shows the two well-known bands: the tangential stretching mode of the E2g phonon of sp₂ atoms (G band) at 1500–1600 cm⁻¹ and the breathing mode of K-point phonons with A1g symmetry (D band) at 1300–1400 cm⁻¹. Furthermore, the other peak at 1750 cm⁻¹ was clearly observed in Fig. 3C. Based on previous researches, it attribute to the M band owe to the intravalley double resonance scattering process of bilayer graphene and a few layers of graphene.^36,37 Similar to the G′ mode, the evolution of this M band coupled with the increasing number of graphene layers could be inflicted by that of electronic band with graphene film structure. As shown in Fig. 3C, without the metal-induced RS effect, the GQDs have low intensity Raman signals at 1351 cm⁻¹, 1579 cm⁻¹ and 1750 cm⁻¹, while both the films with spherical Au NPs and spiky SPs exhibit very intense Raman signals. Fig. 3D shows the Raman peak at 1579 cm⁻¹ for the GQD-spiky SP film is 7.8-fold more intense than that of the GQD-only film. Furthermore, the RSD of the Raman vibrations at 1579 cm⁻¹ were 0.16, 0.34 and 0.28, respectively, as shown in Fig. 3D. On the other hand, the GQD-spherical NP film shows a 3-fold increase in RS. These results demonstrate that the spiky Fe₃O₄@Au SPs are better than the spherical Au NPs for fabricating RS-active tags. There are a number of reasons for this: (1) the spikes of nanoscale bumps or tiny cavities on the surface of the spiky Fe₃O₄@Au SPs are potential “hot” spots for localized near-field enhancement effects; (2) the relatively large total surface area of the spiky Fe₃O₄@Au SPs increases the chances of interaction; and (3) the SPR peak of the spiky Fe₃O₄@Au SPs (566 nm) is nearer to the excitation wavelength (532 nm). The small difference between the excitation and SPR wavelengths incites stronger enhancement effects.

Moreover, the thickness of polyelectrolyte space layer significantly affect the enhancement factor of RS and PL. Usually, the thickness of the LbL films can be well controlled by adjusting the number of layers, deposition time, and environmental conditions, including temperature, ionic strength, and the pH value of the solutions. Herein, we try to measure the thickness of polyelectrolyte spacer layer (PDDA/PAA/PDDA) in sample film. The one layer of polyelectrolyte is too thin; it is too hard to measure the thickness of the space layer directly. Therefore, we prepared a LbL multilayer films which was composed of (PDDA/PAA)₁₀, and measured the cross section of this multilayer film by FE-SEM. As shown in Fig. S3† of ESI, the thickness of this multilayer film is 360 nm. After calculation, the average thickness of single polyelectrolyte layer is about 18 nm. Therefore, the thickness of the polyelectrolyte spacer layer (PDDA/PAA/PDDA) is 54 nm.

Magnetic field (MF) as surrounding environment can be considered for active-SERS substrates.³⁸ Fig. 4 shows the Raman spectra of GQDs obtained using the spiky SPs substrate under an external magnetic field. These Raman spectra, both in the presence and absence of the magnetic field were obtained at exactly the same position on the substrate. As shown in Fig. 4, three distinct peaks are observed at 1381, 1579 and 1750 cm⁻¹ which is correspondent on D, G and M band of GQDs, respectively. The Raman activity of spiky Fe₃O₄@Au SPs substrate under external MFs was significantly reduced from 0 to 426 mT. It is clearly seen that the magnetic field have a negative effect on the Raman activities for spiky Fe₃O₄@Au SPs substrate. The Raman intensity of GQDs decreased with increasing the strength of external magnetic field, and the attenuation degree of Raman activity of spiky Fe₃O₄@Au SPs by the MFs is more than 3.3-fold compare to that without MFs. Therefore, the weakening effect of MFs on Raman spectra of GQDs has been confirmed; yet the reasons of weakening effect still need to be investigated deeply. Superparamagnetic Fe₃O₄ cores can strengthen the local magnetic field at the site of plasmonic Au shells when an external magnetic field is applied, and thus, the degree of the broadening of the energy gap for charge-transfer and a decrease in the electrons on the gold surface would be enhanced.³⁹ This could lead to the higher degree of attenuation of Raman activity under the magnetic field in the case of spiky Fe₃O₄@Au SPs.

Fig. 4 (A) Raman spectra of GQD-spiky Fe₃O₄@Au SP composite films under external magnetic field from 0 to 426 mT.

Fig. 5A and B show optical fluorescence images of the GQD and GQD-spiky Fe₃O₄@Au SP films under irradiation by 532 nm laser, respectively. The image of the GQD-spiky Fe₃O₄@Au SP film is clear and has a strong red colour, while only a few faint red dots can be seen in the image of the GQD-only film. Moreover, Fig. 5C and D show Raman mapping images of the GQD and GQD-spiky Fe₃O₄@Au SP films at 1579 cm⁻¹, respectively. Note that the colour scale of blue to deep red represents the Raman intensity of the GQDs. As can be seen, the GQD-only film is blue without the metal-induced Raman scattering effect, while the GQD-spiky Fe₃O₄@Au SP film is a deep red. These results show that the GQD-spiky SP films have remarkable multi-mode imaging capabilities for Raman mapping and fluorescence imaging.

Fig. 5 Optical fluorescence images of the GQD-only film (A) and GQD-spiky SP film (B) under irradiation by 532 nm laser. Raman mapping images of the GQD-only film (C) and GQD-spiky SP film (D) at 1579 cm⁻¹.

Conclusions

By utilizing GQDs as duplex nanoprobes, the RS and fluorescence activity of LbL-assembled GQD-spiky Fe₃O₄@Au SPs films were investigated. RS and fluorescence enhancements of 13-fold and 7.8-fold, respectively, were measured for the GQD-spiky Fe₃O₄@Au SP films. These enhancements are due to the electromagnetic interactions between the optical fields, quantum dots, and electronic plasma resonances localized on the roughness features of the spiky metallic surfaces. Furthermore, under external MFs, the Raman activity of spiky Fe₃O₄@Au SPs was weakened as a function of applied MFs strength. The attenuation degree of Raman activity of spiky Fe₃O₄@Au SPs by the MFs is more than 3.3-fold compare to that without MFs, due to the broadening of the energy gap for charge-transfer and a decrease of the electrons on the metal surface under the MFs, and the superparamagnetic Fe₃O₄ cores also enhance the local magnetic field at the area of the Au shells. This work may open a wide range of application for GQDs composite-based photonic and light-emitting devices, bioimaging and diagnostic technology.

Acknowledgements

This study was supported by grants from the Korea Healthcare Technology R&D Project and National Research Foundation of Korea (NRF-2013R1A1A4A01004637, 2014R1A1A2007222) of the Ministry for Education, Republic of Korea; and also supported by grants from the Natural Science Foundation of China (Grant no. 51502296), P. R. China.

Notes and references

1. M. S. Jeong and D. R. Ahn, Analyst, 2015, 140, 1995.
2. S. W. Han, E. Jang and W. G. Koh, Sens. Actuators, B, 2015, 209, 242.
3. K. C. Han, E. G. Yang and D. R. Ahn, Chem. Commun., 2012, 48, 5895.
4. C. Zong, J. Wu, J. Xu, H. Ju and F. Yan, Biosens. Bioelectron., 2013, 43, 372.
5. Z. Wang, S. Zong, W. Li, C. Wang, S. Xu, H. Chen and Y. Cui, J. Am. Chem. Soc., 2012, 134, 2993.
6. X. Niu, H. Chen, Y. Wang, W. Wang, X. Sun and L. Chen, ACS Appl. Mater. Interfaces, 2014, 6, 5152.
7. L. Li, G. Wu, G. Yang, J. Peng, J. Zhao and J. J. Zhu, Nanoscale, 2013, 5, 4015.
8. M. Bacon, S. J. Bradley and T. Nann, Part. Part. Syst. Charact., 2014, 31, 415.
9. F. Liu, M. H. Jang, H. D. Ha, J. H. Kim, Y. H. Cho and T. S. Seo, Adv. Mater., 2013, 25, 3657.
10. J. R. Lakowicz, Anal. Biochem., 2005, 337, 171.
11. H. Zhou, J. Dong, V. K. Deo, E. Y. Park and J. Lee, Sens. Actuators, B, 2013, 178, 192.
12. S. R. Ahmed, M. A. Hossain, J. Y. Park, S. H. Kim, D. Lee, T. Suzuki, J. Lee and E. Y. Park, Biosens. Bioelectron., 2014, 58, 33.
13. D. Darvill, A. Centeno and F. Xie, Phys. Chem. Chem. Phys., 2013, 15, 15709.
14. T. Ming, L. Zhao, Z. Yang, H. Chen, L. Sun, J. Wang and C. Yan, Nano Lett., 2009, 9, 3896.
15. H. Zhou, F. Zou, K. Koh and J. Lee, J. Biomed. Nanotechnol., 2014, 10, 2921.
16. H. Zhou, J. P. Kim, J. H. Bahng, N. A. Kotov and J. Lee, Adv. Funct. Mater., 2014, 24, 1439.
17. F. Zou, Q. Ding, G. Wang, Y. Zhang, S. Kang, J. Lee and H. Zhou, RSC Adv., 2015, 5, 56653.
18. W. Chen, N. Xu, L. Xu, L. Wang, Z. Li, W. Ma, Y. Zhu, C. Xu and N. A. Kotov, Macromol. Rapid Commun., 2010, 31, 228.
19. H. Zhou, J. Kim, F. Zou, K. Koh, J. Y. Park and J. Lee, Sens. Actuators, B, 2014, 198, 77.
20. H. Zhou, J. Lee, T. J. Park, S. J. Lee, J. Y. Park and J. Lee, Sens. Actuators, B, 2012, 163, 224.
21. J. Kim, Y. Piao and T. Hyeon, Chem. Soc. Rev., 2009, 38, 372.
22. L. Moriggi, C. Cannizzo, E. Dumas, C. R. Mayer, A. Ulianov and L. Helm, J. Am. Chem. Soc., 2009, 131, 10828.
23. Y. Zhai, J. Zhai, Y. Wang, S. Guo, W. Ren and S. Dong, J. Phys. Chem. C, 2009, 113, 7009.
24. F. Bao, J. L. Yao and R. A. Gu, Langmuir, 2009, 25, 10782.
25. S. R. Ahmed, H. R. Cha, J. Y. Park, E. Y. Park, D. Lee and J. Lee, Nanoscale Res. Lett., 2012, 7, 1.
26. F. X. Xiao, J. Miao and B. Liu, J. Am. Chem. Soc., 2014, 136, 1559.
27. Y. Jin and X. Gao, Nat. Nanotechnol., 2009, 4, 571.
28. J. J. Richardson, M. Björnmalm and F. Caruso, Science, 2015, 348, aaa2491.
29. K. Ray, R. Badugu and J. R. Lakowicz, Chem. Mater., 2007, 19, 5902.
30. J. Peng, W. Gao, B. K. Gupta, Z. Liu, R. Romero-Aburto, L. Ge, L. Song, L. B. Alemany, X. Zhan and G. Gao, Nano Lett., 2012, 12, 844.
31. E. Prodan, C. Radloff, N. J. Halas and P. Nordlander, Science, 2003, 302, 419.
32. E. Hao, R. C. Bailey, G. C. Schatz, J. T. Hupp and S. Li, Nano Lett., 2004, 4, 327.
33. F. Tam, G. P. Goodrich, B. R. Johnson and N. J. Halas, Nano Lett., 2007, 7, 496.
34. N. Akbay, J. R. Lakowicz and K. Ray, J. Phys. Chem. C, 2012, 116, 10766.
35. R. Bardhan, N. K. Grady, J. R. Cole, A. Joshi and N. J. Halas, ACS Nano, 2009, 3, 744.
36. C. Cong, T. Yu, R. Saito, G. F. Dresselhaus and M. S. Dresselhaus, ACS Nano, 2011, 5, 1600.
37. F. Zou, H. Zhou, V. T. Tran, J. Kim, K. Koh and J. Lee, ACS Appl. Mater. Interfaces, 2015, 7, 12168.
38. R. Li, Q. W. Chen, H. Zhang, X. K. Kong, Y. B. Sun, H. Zhong, H. Wang and S. Zhou, J. Raman Spectrosc., 2013, 44, 525.
39. Q. Zhang, J. Ge, J. Goebl, Y. Hu, Y. Sun and Y. Yin, Adv. Mater., 2010, 22, 1905.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra14176a

---