One-step synthesis of fluorescent silicon quantum dots (Si-QDs) and their application for cell imaging

Abstract

Novel fluorescent silicon quantum dots (Si-QDs) were synthesized by a one-step hydrothermal procedure using (3-aminopropyl)trimethoxysilane (APTES) as a silicon source and sodium ascorbate (SA) as a reducing agent. This synthetic strategy is straightforward, efficacious and low-cost. The as-synthesized Si-QDs featured narrow size distribution, intense purple-blue fluorescence with a photoluminescence quantum yield (PL QY) of 21%, complete water solubility, and favorable biocompatibility. Moreover, the resultant Si-QDs present robust stability in different circumstances, such as for long-term storage in air, in a wide pH range of 2–12 and under continuous UV irradiation, and they also do not respond to some metal ions. Their superior stability and safety render Si-QDs applicable as fluorescent probes in biomedical applications. The use of Si-QDs as an optical probe is illustrated by fluorescent imaging of IE8 cells.

---

Introduction

Bulk silicon is one of well-studied semiconductors with some advantages such as inertness, abundance, economy and nontoxicity. Nanoscale silicon has attracted considerable attention because of its unique properties such as single-electron tunneling, nonlinear optical properties, and photoluminescence (PL) emission tunable in a wide wavelength window from UV to near infrared (NIR).¹ Studies of fluorescent silicon nanostructures have been extensively carried out since the first discovery of luminescent porous silicon (p-Si) by Canham.² It has been well recognized that quantum confinement effect predominantly involves in the PL emission mechanism of silicon nanostructures with sizes less than silicon Bohr exciton radius of ∼5 nm, namely silicon quantum dots (Si-QDs).³ In addition, their surface characteristics plays a crucial role in tuning the optical properties of Si-QDs through offering radiative and/or non-radiative recombination centers.⁴ The fluorescent Si-QDs have shown potential for various applications such as optoelectronic devices,⁵ biological imaging,⁶ photocatalyst for diverse reactions including reduction, decomposition and selective oxidation,^7,8 light-emitting diodes (LEDs),⁹ solar cells,^10–12 and sensors,¹³ etc. Due to their low toxicity and good biocompatibility, Si-QDs may have benefits over conventional toxic heavy metal based semiconductor QDs such as CdSe and PbS especially for the biomedical applications.

Extensive efforts have been devoted to synthesis of silicon nanoparticles (Si NPs). The current synthetic methods mainly include electrochemical etching of bulk silicon,¹⁴ laser-driven pyrolysis of silane,¹⁵ gas phase synthesis,^16,17 microemulsion synthesis,^18,19 and wet chemistry techniques,²⁰ etc. However, these methods usually require high temperatures, special equipments, extreme conditions, and tedious procedures. Moreover, most resulting Si NPs are hydrophobic, which limits their broad applications. Complicated surface modification and functionalization of these silicon nanostructures are thus necessary for improvement of their water solubility.^21,22 Therefore, development of new synthetic routes is still a challenging toward versatile control of the fluorescent Si NPs.

Herein, we demonstrate a facile hydrothermal synthesis using mild reagents of (3-aminopropyl)trimethoxysilane (APTES) as a silicon source and sodium ascorbate (SA) as a reduction reagent to achieve water-dispersable and fluorescent Si-QDs. To explore the effects of the specific surface states of the synthesized colloidal Si-QDs on their PL properties, thorough instrumental analysis were measured using UV-vis absorption spectroscope, PL spectroscope, transmission electron microscope (TEM) and high resolution (HR) TEM coupled with electron diffraction compartment, X-ray photoelectron spectroscope (XPS) and Fourier transform infrared spectrometer (FTIR). The resulting Si-QDs possess stable and saturated surface states in comparison with those obtained by a solution synthesis at ambient temperature and pressure.

High quality Si-QDs were successfully obtained, possessing uniform sizes/shapes, strong PL emission, favorable biocompatibility, and superior stability. In an in vitro experiment, the synthesized Si-QDs showed acceptable cytotoxicity and safety for IE8 cells even at a high dose (1 g silicon per L) and for long incubation time (24 h), and were thus employed for fluorescent cell imaging. When appropriately excited, the cellular uptake of the fluorescent Si-QDs by IE8 cells was clearly observed and acquired by confocal fluorescent microscope (CFM), showing that Si-QDs are promising as fluorescent probes for biomedical imaging.

Experimental

Chemicals and apparatus

(3-Aminopropyl)trimethoxysilane (APTES, 98%) and sodium ascorbate (SA) were purchased from Aladdin.

UV-visible (UV-vis) absorption spectra were acquired with a Hitachi U-4100 UV/visible spectrophotometer (Japan). Photoluminescence (PL) spectra were recorded on a Gangdong F-280 fluorospectrophotometer (China). FTIR spectra over a range of 400–4000 cm⁻¹ were performed on a Nicolet Avatar-360 FTIR spectrometer (Japan). Transmission electron microscope (TEM) images were examined by Tecnai G2F30 transmission electron microscope (Hitachi Co., Japan). TEM samples were prepared by dropping aqueous Si-QDs solution onto carbon-coated copper grids (400 meshes) from Beijing Zhongjingkeyi Technology Co. (China) followed by evaporation of excess solvent. Electron diffraction spectroscopy (EDS) patterns were captured by the TEM microscope equipped with an energy dispersive X-ray spectrometer. The quantum yield (QY) of Si-QDs was measured according to literatures using quinine sulfate in 0.1 M H₂SO₄ (QY = 58%) as a reference standard.²³ X-ray photoelectron spectroscope (XPS) spectra were investigated at a PHI5000 Versa Probe X-ray photoelectron spectrometer (Japan) using Al K_α as source. All optical measurements were performed at room temperature under ambient conditions.

Synthetic procedure

Initially we prepared the Si-QDs via a room temperature solution method: 0.0594 g SA was dissolved in 10 ml deionized water in a flask with sustainable aeration of nitrogen for 20 min to remove oxygen. Then 1.25 ml the above SA solution and 1 ml APTES was added into 4 ml deionized water and then thoroughly stirred at room temperature and ambient pressure for 20 min. For surface passivation process, 0.1 g PVP was added into a fresh Si-QDs colloidal solution and stored for one month.

We chose hydrothermal route for the synthesis, which offers high temperature and pressure. A precursor solution at a concentration of 20 silicon g L⁻¹ was prepared following the above procedure, and then loaded into a Teflon-lined stainless steel autoclave, which was incubated at 200 °C for 24 h and cooled to room temperature. To remove byproducts, the resulting Si-QDs colloidal solutions were treated by dialysis (1 kDa). The purified Si-QDs with strong PL were used for cytotoxicity and immunofluorescent cellular imaging studies.

pH stability of Si-QDs

The pH of the Si-QDs colloidal solution was adjusted by dropwise addition of HCl or NaOH. The PL spectra of the resultant Si-QDs were then recorded.

MTT assay of cell viability

Human prostate cancer IE8 cells were selected for cytotoxicity and following cell imaging studies. IE8 cells were first cultured (37 °C, 5% CO₂) in a 96-well plate in DMEM with 10% heat-inactivated FBS and antibiotics (100 μg ml⁻¹ streptomycin and 100 U ml⁻¹ penicillin) overnight. Ten μL of the Si-QDs solution was then added to each well at a final concentration of 1 g silicon per L. Incubation was carried out for 0.5, 3, 6, 12 and 24 h. The cytotoxicity of the Si-QDs was evaluated by a well-established MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay, based on the accumulation of dark-blue formazan crystals created inside living cells after MTT exposure. The number of viable cells was proportional to the amount of the colorful formazan, which was quantified by measuring the absorbance at 570 nm (ELISA reader). A linear relationship between viable cell number and absorbance was established to assess the cell proliferation.

Cell imagination

Ten μL of the Si-QDs colloidal solution was added to each well containing 6000 cells of 96-well plate at a final concentration of 50 mg silicon per L for incubation of 0.5 h. Cellular uptake by IE8 cells was examined under a confocal laser microscope (Leica, TCS-SP5, Germany). Images were captured and processed with an image analysis software. The cellular uptake of Si-QDs by the cells were excited by diode laser (λ_ex = 405 nm) with 50 mW power. Diode laser at 30% power was used in our experiment. Detection window for Si-QDs was 450–510 nm.

Results and discussion

Among various routes for the preparation of silicon nanocrystals (Si-NCs),^14–22 a room temperature solution synthetic method is especially attractive due to its extremely simple conditions.²⁰ We at first followed a previous room temperature solution synthetic route after slight modification²⁰ to produce Si-QDs: simply stirring an aqueous mixture of APTES as a silicon source and SA as a reducing agent at ambient temperature and pressure for 20 min. The synthesis was monitored by observing visible fluorescence under a 365 nm UV lamp and measuring PL emission spectrum with a spectrometer.

Fig. 1 shows the PL spectra of the fresh and stored Si-QDs colloidal solutions. As can be seen, the Si-QDs showed changeable PL emission behavior during storage, corresponding to different visible colors from blue-green to purple-blue (Fig. 1 inset). Furthermore, the broad and unsymmetric PL peak shape reveals that multi emission mechanisms may involve. Based on the specific shapes, these PL peaks could be primarily fitted into two Gaussian peaks: one centered at 490 nm (2.53 eV) and another positioned at 430 nm (2.88 eV) (Fig. S1(a–d), see the ESI†). The former PL band at low energy probably originates from the surface defects,²⁴ while the latter one at high energy possibly arises from exciton radiative recombination at core-related levels and certain surface states within band gap.^7,24 The PL intensities of these two types of emission showed an opposite changing trend during storage in air. Under the current mild reaction conditions, there could be surface dangling bonds functioned as the surface defects. Since the surface dangling bonds are highly active, they prone to change during storage to form new surface states with different PL behavior, as evidenced by the decrease of the defects-related emission and increase of the core/surface states-related emission (Fig. S1(a–d), see the ESI†).

Fig. 1 The integrated PL spectra of the Si-QDs prepared freshly and stored for 24 h, 48 h, and 1 month. Inset shows the fluorescent photos of the Si-QDs prepared freshly (blue-green) and stored for 1 month (blue) (λ_ex = 365 nm).

In order to support the assumed defects-related and core/surface states-related emission, surface passivation of the Si-QDs using polyvinylpyrrolidine (PVP) was performed (Fig. S2, see the ESI†). As can be seen, except a PL band at 490 nm associated with the surface defects, the one at 530 nm occurred resulting from PVP-related surface states. The defects-related PL intensities of the Si-QDs stored for 24 h and passivated Si-QDs were comparable (Fig. S1(a) and S2†), suggesting that the PVP shell does effectively protect the surface defects.²⁵ These observations reveal that the diverse surface states play a determinant role in effective radiative recombination and consequential high PL QY.^26,27

It should be noted that there is still lack of consensus regarding the origin of PL of the nanosized silicon mainly due to high surface-to-volume ratio, broad particle size distribution, material defects, and complex PL recombination dynamics.²⁸ Currently, two major theories govern the discussion. One relies on quantum-confined core states²⁹ and another focuses on the surface states. Our observations reveal that the surfaces of the Si-QDs not only tailor the PL emission but also protect the inner cores from further inward variation.^30,31

Although the fluorescent Si-QDs were readily obtained via the room temperature solution route (Fig. 1), the poor quality in terms of weak PL intensity and broad PL peak may not satisfy the practical applications. The mild synthetic conditions can not provide enough energy to drive the reactions required for formation and surface construction of the qualified Si-QDs. Therefore, in this study, hydrothermal method offering higher temperature and pressure was selected to synthesize the Si-QDs with saturated surface composition and desired structure.

The Si-QDs synthesized via the one-step hydrothermal route were characterized by TEM, HRTEM, EDS, FTIR and XPS, respectively. TEM micrograph of the as-synthesized colloidal Si-QDs is shown in Fig. 2(a), in which they appear as quasi-spherical dots without obvious aggregation and agglomeration. It should be mentioned that low contrast in the TEM image is due to the low atomic weight of silicon and nanoscale dimensions of the silicon dots, resulting in poor visualization. Fig. 2(c) is the size distribution histogram of the Si-QDs, which well fits a Gaussian peak and demonstrates a size of 4.0 ± 0.7 nm. This small size (<5 nm) enables the QDs better biocompatibility since they facilitate the renal clearance in vivo.³² The calculated relative standard deviation (RSD) value of 17.5% suggests that the current synthesized Si-QDs are nearly monodisperse. HRTEM image (Fig. 2(b)) clearly displays the high crystallinity of the Si-QDs, as indicated by distinct lattice fringes with 0.30 nm interplanar spacing consistent with the (111) plane of diamond Si.^7,33 Elemental analysis of the Si-QDs using EDS is shown in Fig. 2(d), illustrating the presence of Si, N and O. The co-existence of Cu and C is unavoidable since carbon-coated copper grids were used as a sample supporter for the measurement. All the findings reveal that the Si-QDs are composed of a crystalline silicon core with the bulk lattice constant, capped with diverse functional groups.

Fig. 2 (a) TEM image of the Si-QDs. (b) HRTEM image of the Si-QDs. (c) The diameter distribution of the Si-QDs derived from several TEM images. (d) EDS pattern of the Si-QDs.

To further identify the main chemical bonds of the Si-QDs, FTIR spectrum was recorded (Fig. 3). There appeared pronounced bands located between 1000 and 1200 cm⁻¹ due to vibrational stretching of Si–OR/Si=O (ref. 34) and in a range of 1300 to 1600 cm⁻¹ arising from vibrational scissoring and symmetric bending of Si–CH₂.³⁵ There was an obvious absorbance at 2931 cm⁻¹ attributed to deformation and stretch vibration of O–H bond. The absorbance in the range of 600 to 800 cm⁻¹ as well as those at 1579 and 3356 cm⁻¹ are related to bending vibration and stretching vibration of N–H.^19,36 Besides, characteristic bands associated with Si–H/Si–H_x appeared, corresponding to bending and wagging modes at 572 cm⁻¹ and vibrational stretching mode at 2160 cm⁻¹.³⁷ Si–Si bonding corresponds to the absorbance around 520 cm⁻¹.³⁸ Despite the fact that residual byproducts of the reactions may interfere with the FTIR results, the existence of diverse surface terminations was evidenced.

Fig. 3 FTIR spectrum of the Si-QDs.

XPS was performed to determine the surface composition of the Si-QDs (Fig. 4). The survey XPS spectrum of the Si-QDs as shown in Fig. 4(a) presents four major peaks at 100.96, 283.43, 398.62 and 530.80 eV, which correspond to Si 2p, C 1s, N 1s and O 1s, respectively. The high resolution XPS spectrum of Si 2p (Fig. 4(b)) shows that the silicon exists in three different chemical environments, originating from Si–C (100.2 eV), Si–N (101.2 eV) and Si–O (102.3 eV), respectively.³⁹ The C 1s signal (Fig. 4(c)) can be described by six peaks at 282.9, 283.5, 284.3, 285.2, 286.3 and 287.7 eV, which are attributed to C–Si,³⁹ C–C/C=C, C–N, C–OH/C–O–C and C=O,⁹ respectively. Three Si–O peaks positioned at 530.3, 531.5 and 532.7 eV³⁹ were shown in the O 1s spectrum (Fig. 4(d)). The N 1s spectrum (Fig. 4(e)) exhibits four peaks at 397.7, 398.3, 399.6 and 401.3 eV, arising from Si–N–Si,⁴⁰ N–C,³⁹ Si–N–O⁴¹ and N–H,⁴² respectively. XPS results further confirm that the Si-QDs have diverse surface states.

Fig. 4 XPS spectra of the Si-QDs: (a) full scan, (b) Si 2p, (c) C 1s, (d) O 1s and (e) N 1s.

UV-vis absorption spectrum of the colloidal Si-QDs (Fig. 5) displays featured absorptions at approximate 270 nm (4.59 eV) and below. The former is likely associated with a direct band gap transition of L–L (4.4 eV),⁴³ which slightly blue-shifts 0.19 eV due to the quantum confinement of the Si-QDs. There also appears a typical absorption at 300 nm (4.13 eV), which is well consistent with another direct transition of Γ₂₅—Γ_2′ with the energy of 4.2 eV (295 nm).⁴⁴ Furthermore, an absorption around 360 nm (3.44 eV) may be attributed to a direct band gap transition of Γ₂₅–Γ₁₅ (3.4 eV) at the zone center of the Si-QDs,⁴³ which could be considered as the specific exciton absorption responsible for the PL since the maximum excitation wavelength of the current Si-QDs was 365 nm (Fig. S3, see the ESI†). The appearance of this characteristic absorption at 360 nm is an indicator reflecting the uniform size/shape of the Si-QDs. The PL spectrum (Fig. 5) of the colloidal Si-QDs under 365 nm excitation is nearly symmetric centered at 440 nm with a full width at half maximum height (FWHM) of ∼60 nm, further confirming that the size/shape distribution of the as-synthesized Si-QDs is homogeneous. The corresponding PL QY was ∼21% calculated using quinine sulfate (QY = 58%) as a reference.

Fig. 5 Absorbance and PL spectra of the Si-QDs synthesized by the hydrothermal process (λ_ex = 405 nm). Inset is the photos of the Si-QDs under normal light (left) and 365 nm UV light (right).

A excitation-independent emission behavior was observed as shown in Fig. S3 (see the ESI†), that is, within an excitation wavelength range of 300 to 400 nm, the emission position centered around 440 nm with slight shift as a result of size distribution, whereas the emission intensity obviously changes with the maximum value in response to 365 nm excitation.⁴⁵ Although the bulk silicon is indirect band gap material, single crystallite quantum confinement is still an important contributor to the emission mechanism along with the dominant surface states.

The stability of fluorescent materials is critically important for the practical applications. The current colloidal Si-QDs possess superior storage and chemical stability. They retained the original PL profile in terms of intensity, position and shape after storage for at least six months (data not shown). This is due to the presence of stable surfaces. Furthermore, in a wide pH range of 2–12, the PL of the colloidal Si-QDs remains stable (data not shown), indicating that the Si-QDs are pH insensitive. In addition, the PL of the Si-QDs exhibits stable after continuous UV irradiation (i.e., 365 nm for 2 h), preserving ∼90% of the initial intensity better than commonly used fluorescein isothiocyanate (FITC) and conventional CdSe QDs.²⁰ It has been reported that some fluorescent QDs such as carbon dots (CDs) respond remarkably PL quenching to certain metal ions such as Fe³⁺, Hg²⁺, and Cu²⁺, etc.^46–49 In this study, the fluorescent response of the Si-QDs to the selective metal ions of Li⁺, Ba²⁺, Hg²⁺, Mg²⁺, Na⁺, K⁺, Zn²⁺, Fe³⁺, Fe²⁺, Mn²⁺ and NH₄⁺ were investigated, showing no obvious PL quenching effects (Fig. S4†). Such remarkable stabilities are attributed to the intrinsic PL properties of the Si-QDs. Our findings suggest that the hydrothermal method effectively modify the surfaces of the Si-QDs and consequently improve their stability, solubility and surface states-related PL properties.

Various Si-QDs have been employed for bioimaging in different live cells because of their bright and stable PL, excellent water-solubility, good biocompatibility and nontoxicity.⁶ Before the fluorescent labeling experiment, in this study, MTT assay was performed to assess the cytotoxicity of the colloidal Si-QDs on human prostate cancer IE8 cells. It was found that almost all the experimental cells kept viability after incubation with the Si-QDs (1 g L⁻¹) for 24 h, revealing that the Si-QDs are of low toxicity (Fig. S5†). Cellular uptake of the Si-QDs by IE8 cells was monitored by CFM, showing bright and clearly resolved blue fluorescence (Fig. 6). All the findings suggest that the as-synthesized Si-QDs are a promising candidate for bioimaging living cells. It should be noted that the blue-green fluorescence in the region of 450–550 nm of native fluorophores (e.g., NADH and flavins) in IE8 cells did not cause obvious interference. In addition, due to the remarkable photostability of Si-QDs compared with photobleaching property of these native fluorophores, the Si-QDs are suitable for in vivo biological research.

Fig. 6 CFM images of IE8 cells incubated with the Si-QDs at a concentration of 1 g silicon per L for 0.5 h (λ_ex = 405 nm): (a) fluorescence image, (b) DIC image, and (c) merged image.

Conclusion

We present a simple synthetic strategy for the production of the Si-QDs with narrow size distribution, intense PL emission, excellent water solubility, favorable biocompatibility, and ultra stability in different circumstances such as long-term storage, photo irradiation and a wide pH. Instrumental characterization showed that the Si-QDs consist of crystalline silicon core terminated with diverse functional surface groups, which contributes to their unique optical properties. This synthetic route does not require expensive equipment and harsh conditions, and has some advantages over the previous methods involving tedious and complicated manipulation. The easy synthesis and remarkable PL emission enable the Si-QDs a potential candidate for the biomedical applications such as long-term and real-time cellular labeling, as indicated by the fluorescent imaging of IE8 cells.

Acknowledgements

This work was financially supported by the Fundamental Research Funds for the Central Universities (Grant No. HIT. IBRSEM. 201331) and the Fundamental Research Funds for the Central Universities and Program for Innovation Research of Science in Harbin Institute of Technology (Grant No. PIRS of HIT 201411).

References

1. M. Montalti, L. Prodi, E. Rampazzo and N. Zaccheroni, Chem. Soc. Rev., 2014, 43(12), 4243–4268.
2. A. G. Cullis and L. T. Canham, Nature, 1991, 353(6342), 335–338.
3. M. H. Mahdieh and A. Momeni, Laser Phys. Lett., 2015, 25(1), 015901.
4. Y. Kanemitsu, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 49(23), 16845–16848.
5. W. Liao, X. Zeng, X. Wen, W. Zheng, Y. Wen and W. Yao, J. Electron. Mater., 2015, 44(3), 1015–1020.
6. S. Chinnathambi, S. Chen, S. Ganesan and N. Hanagata, Adv. Healthcare Mater., 2014, 3, 10–29.
7. M. Dasog, Z. Yang, S. Regli, T. M. Atkin, A. Faramus and M. P. Singh, et al., ACS Nano, 2013, 7(3), 2676–2685.
8. Z. Kang, C. H. A. Tsang, N.-B. Wong, Z. Zhang and S.-T. Lee, J. Am. Chem. Soc., 2007, 129, 12090–12095.
9. J. Chen, W. Liu, L.-H. Mao, Y.-J. Yin, C.-F. Wang and S. Chen, J. Mater. Sci., 2014, 49(21), 7391–7398.
10. C. Y. Liu, Z. C. Holman and U. R. Kortshagen, Nano Lett., 2009, 9(1), 449–452.
11. M. Werwie, X. X. Xu, M. Haase, T. Basche and H. Paulsen, Langmuir, 2012, 28(13), 5810–5818.
12. D. V. Talapin, J. S. Lee, M. V. Kovalenko and E. V. Shevchenko, Chem. Rev., 2010, 110(1), 389–458.
13. P. D. Howes, R. Chandrawati and M. M. Stevens, Science, 2014, 346(6205), 1247390.
14. L. Wang, V. Reipa and J. Blasic, Bioconjugate Chem., 2004, 15(2), 409–412.
15. X. G. Li, Y. Q. He and M. T. Swihart, Langmuir, 2004, 20(11), 4720–4727.
16. J. Laube, S. Gutsch, D. Hiller, M. Bruns, C. Kübel and C. Weiss, et al., J. Appl. Phys., 2014, 116(22), 223501.
17. W. Yu, Y. Xu, H. Li, X. Zhan and W. Lu, Appl. Phys. A, 2013, 111(2), 501–507.
18. T.-H. Le and H.-D. B. Jeong, J. Korean Chem. Soc., 2014, 35(12), 3421–3422.
19. R. D. Tilley, J. H. Warner, K. Yamamoto, I. Matsui and H. Fujimori, Chem. Commun., 2005, 1833–1835.
20. Y. Zhong, F. Peng, F. Bao, S. Wang, X. Ji and L. Yang, et al., J. Am. Chem. Soc., 2013, 135(22), 8350–8356.
21. S. Hallmann, M. J. Fink and B. S. J. Mitchell, J. Exp. Nanosci., 2013, 10(8), 588–598.
22. Z. Yang, M. H. Wahl and J. G. C. Veinot, Can. J. Chem., 2014, 92(10), 951–957.
23. K. Zhang, L. S. Wang, G. H. Yue, Y. Z. Chen, D. L. Peng and Z. B. Qi, et al., Surf. Coat. Technol.Surf. Coat. Technol., 2011, 205(12), 3588–3595.
24. K. Dohnalová, A. N. Poddubny, A. A. Prokofiev, W. D. A. M. de Boer, C. P. Umesh and J. M. J. Paulusse, et al., Light: Sci. Appl., 2013, 2(1), 1–6.
25. Z. Kang, Y. Liu, C. H. A. Tsang, D. D. D. Ma, X. Fan and N.-B. Wong, et al., Adv. Mater., 2009, 21(6), 661–664.
26. D. C. Hannah, J. Yang, P. Podsiadlo, M. K. Chan, A. Demortiere and D. J. Gosztola, et al., Nano Lett., 2012, 12(8), 4200–4205.
27. D. C. Hannah, J. Yang, N. J. Kramer, G. C. Schatz, U. R. Kortshagen and R. D. Schaller, ACS Photonics, 2014, 1(10), 960–967.
28. A. Sa'ar, J. Nanophotonics, 2009, 3, 032501.
29. M. Green, Curr. Opin. Solid State Mater. Sci., 2002, 6, 355–363.
30. M. V. Wolkin, J. Jorne, P. M. Fauchet, G. Allan and C. Delerue, Phys. Rev. Lett., 1999, 82(1), 197–200.
31. S. Godefroo, M. Hayne, M. Jivanescu, A. Stesmans, M. Zacharias and O. I. Lebedev, et al., Nat. Nanotechnol., 2008, 3(3), 174–178.
32. Y. Y. Su, F. Peng, Z. Y. Jiang, Y. L. Zhong, Y. M. Lu and X. X. Jiang, et al., Biomaterials, 2011, 32, 5855–5862.
33. F. Huisken, H. Hofmeister, B. Kohn, M. A. Laguna and V. Paillard, Appl. Surf. Sci., 2000, 154, 305–313.
34. R. K. Baldwin, K. A. Pettigrew, E. Ratai, M. P. Augustine and S. M. Kauzlarich, Chem. Commun., 2002, 1822–1823.
35. J. H. Warner, A. Hoshino, K. Yamamoto and R. D. Tilley, Angew. Chem., 2005, 44(29), 4550–4554.
36. Y. Zhai, M. Dasog, R. B. Snitynsky, T. K. Purkait, M. Aghajamali and A. H. Hahn, et al., J. Mater. Chem. B, 2014, 2(47), 8427–8433.
37. J. Zou, R. K. Baldwin, K. A. Pettigrew and S. M. Kauzlarich, Nano Lett., 2004, 4(7), 1181–1186.
38. M. F. Omar, A. K. Ismail, I. Sumpono, E. A. Alim, M. N. Nawi and M. A.-R. Mukri, et al., Mal. J. Fund. Appl. Sci., 2012, 8(5), 259–261.
39. G. Wen, X. Zeng, X. Wen and W. J. Liao, Appl. Phys., 2014, 115(16), 164303.
40. J. J. Romero, M. J. Llansola-Portolés, M. L. Dell'Arciprete, H. B. Rodríguez, A. L. Moore and M. C. Gonzalez, Chem. Mater., 2013, 25(17), 3488–3498.
41. M. Dasog, G. B. De los Reyes, L. V. Titova, F. A. Hegmann and J. G. C. Veinot, ACS Nano, 2014, 8(9), 9636–9648.
42. D. Qu, M. Zheng, P. Du, Y. Zhou, L. Zhang and D. Li, et al., Nanoscale, 2013, 5(24), 12272–12277.
43. J. D. Holmes, K. J. Ziegler, R. C. Doty, L. E. Pell, K. P. Johnston and B. A. Korgel, J. Am. Chem. Soc., 2001, 123, 3743–3748.
44. J. P. Wilcoxon, G. A. Samara and P. N. Provencio, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 2704.
45. Y. Hu, J. Yang, J. Tian, L. Jia and J.-S. Yu, Carbon, 2014, 77, 775–782.
46. S. Li, Y. Li, J. Cao, J. Zhu, L. Fan and X. Li, Anal. Chem., 2014, 86(20), 10201–10207.
47. Y. Liu, C.-y Liu and Z.-y Zhang, Appl. Surf. Sci., 2012, 263, 481–485.
48. R. Zhang and W. Chen, Biosens. Bioelectron., 2014, 55, 83–90.
49. J. Zong, X. Yang, A. Trinchi, S. Hardin, I. Cole and Y. Zhu, et al., Biosens. Bioelectron., 2014, 51, 330–335.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra13119g

---