Role of novel silicon nanoparticles in luminescence detection of a family of antibiotics

Abstract

The design, synthesis, photophysical properties and practical application of fluorescent silicon nanoparticles (SiNPs) of a quantum-size (∼3 nm) have been studied. 3-Aminopropyltrimethoxysilane (APTMS) and 3-aminopropyltriethoxysilane (APTES) were employed as sources. The newly developed nanoparticles show excellent water solubility and luminescent features. Herein we report a novel chemical approach for selective tetracycline (TC) recognition and a new type of targetable fluorescent sensor has been prepared. The blue emission, peaking at 440 nm, was dramatically quenched in the presence of TCs. Moreover, we utilized this probe in a milk sample study for the determination of TCs, demonstrating its advantages of simplicity and practicality.

---

1. Introduction

The development of novel nanomaterials as chemical sensors has aroused considerable research interest in recent years due to their excellent optical, electronic and magnetic properties.^1,2 These nano sized sensors would be suitable for the simple, sensitive and on-site analysis of specific guests. Among them, fluorescent silicon nanoparticles (SiNPs), as important zero-dimensional silicon nanomaterials, have attracted growing attention for their unique properties, especially photo-stability, favorable biocompatibility and low toxicity.^3–5 As far as the synthetic approaches are concerned, researchers have contributed to the design of SiNPs dispersible in water^5,6 and they could be used in a variety of applications. Herein, some literature has reported that the modified SiNPs may be suitable candidates for biological imaging, but the detection of various analytes using label-free SiNPs was very limited.^7,8

Tetracyclines (TCs) is an important family of antibiotics. Because of their broad spectrum activity against pathogenic microorganisms, favorable absorption, relatively low toxicity and low cost, TCs have been widely used in the therapy of human and animal infections.⁹ But nowadays, the application of TCs seems to be excessive. The antibiotic residues were frequently found in animal products^10–12 such as meats, milk and honey, which may lead to allergic reactions to some hypersensitive individuals. Besides, the residues may also promote the development and distribution of bacterial resistance to antibiotics. Therefore, the maximal residues limits of TCs in animal products have been restricted by many countries.¹³ Meanwhile, numerous analytical methods including high performance liquid chromatography (HPLC), microbiological analysis (MA), chemiluminescence, capillary electrophoresis (CE) and dipstick colorimetric method have been successfully established as viable techniques for the determination of TCs.^14–18 Although these methods own high selectivity and adequate sensitivity, it should be mentioned that relative expensive and sophisticated instruments as well as complicated sample preparation procedures are usually indispensable, resulting in the fact that they are not suitable for on-site detection, especially in emergency cases. Therefore, the search for simple, efficient, sensitive and economical methods for determination of TCs is a real challenge.

Based on the advantages of fluorescence sensors, some researchers have made great efforts to fabricate novel luminescence probes for the detection of TCs based on nanoscale materials.^19–22 In this study, we proposed a simple and green method for the synthesis of silicon nanoparticles (SiNPs) employing APTES/APTMS and sodium citrate as precursors. The obtained SiNPs exhibited excellent water solubility and photo-stability. Meanwhile, such label-free SiNPs allow detection of TCs with high sensitivity and selectivity. The blue emission of SiNPs would be efficiently quenched by the addition of TCs. Moreover, as a fluorescence probe, the SiNPs was successfully utilized to determine TCs in milk sample with satisfactory detection limits and linear ranges. These interesting results indicated these novel photo-luminescent SiNPs have great potential applications in analytical detections.

2. Experimental

2.1. Materials

3-Aminopropyltrimethoxysilane (APTMS, ≥97%), 3-aminopropyltriethoxysilane (APTES, 98%), sodium citrate tribasic dihydrate (≥99.0%), tetracycline hydrochloride (98%), oxytetracycline hydrochloride (95%), chlortetracycline hydrochloride (USP) were purchased from J&K company. Glucose, sucrose, maltose, fructose and amino acids were purchased form aladdin company. Milli-Q water (18.25 MΩ cm at 25 °C) was used throughout the experiment. All the other reagents were provided by Guangzhou Chemical Reagent Factory and used without further purification.

2.2. Characterization

Fluorescence emission and excitation spectra were measured using an Edinburgh FLS920 spectrometer (Great Britain). Ultraviolet-visible absorption spectra were obtained with a Shimadzu UV-2550 spectrophotometer (Japan). High resolution transmission electron microscopy (HR-TEM) images were obtained by using a JEOL JEM-2100HR microscope (Japan). Zeta potentials were acquired by a ZetaPlus Zeta Potential Analyzer (Brookhaven instruments, USA).

2.3. Synthesis of SiNPs

The label-free SiNPs were synthesized via a one-pot hydrothermal process. 1.2 g of sodium citrate was dissolved in 25 mL N₂-saturated water. Subsequently, 6 mL of APTMS was added and magnetic stirred homogeneously (10 min) until the mixture became clear and transparent. The resultant solution was transferred to a 50 mL Teflon-lined stainless steel autoclave and filled with ultrapure water up to 80% filling capacity of the total volume. The autoclave was sealed and maintained at 180 °C for 20 h and then cooled to room temperature naturally. To exclude impurities, the resulting transparent solution was further dialyzed with a 1000 MWCO dialysis membrane. Finally, the purified SiNPs aqueous solution was stored at 4 °C for characterization. The SiNPs derived from APTES needs longer reaction time (30 min) during the magnetic stirring process in order to achieve homogenous solution.

2.4. Detection of TCs in aqueous solution

The stock solutions of tetracycline (TC), oxytetracycline (OTC) and chlortetracycline (CTC) were prepared by dissolving tetracycline hydrochloride, oxytetracycline hydrochloride and chlortetracycline hydrochloride with ultrapure water, respectively. For detections of TC, an aqueous solution of SiNPs with a volume of 50 μL was mixed with different volumes of PBS buffer (0.02 M, pH 7.4) and various concentrations of TCs solution were added. The total volume of the solution was set at 4 mL. The fluorescence intensity of the mixtures was measured upon being excited at 350 nm.

2.5. Detection of TCs in urine and milk samples

The pure milk sample was obtained from a supermarket and treated as follows before measurement.²³ Firstly, to remove the proteins in the milk sample, 1% (v/v) trichloroacetic acid was added and sonicated for about 20 min. Then, the lipids were removed by collecting the supernatant filtered through a 0.22 μm membrane. The concentration of TCs in real samples was detected using the standard addition method. Finally, a series of milk samples containing different concentrations of TCs were prepared by providing with stock TCs solutions of different volumes.

3. Results and discussion

3.1. Characterization of photoluminescent SiNPs

As illustrated in Fig. 1, the water-soluble and photo-luminescent SiNPs were prepared through a simple one-pot hydrothermal approach using APTMS/APTES and sodium citrate as precursors. It has been accepted that during the high temperature treatment, siloxane coupling reagents were readily reduced by trisodium citrate, forming silicon crystal nuclei.⁶ The reported work focused on the reduction reaction of 3-aminopropyltrimethoxysilane. Here we extended the choice of alkoxysilanes and 3-aminopropyltriethoxysilane was firstly used as the silicon source. In fact, we also carried out the experiment by changing the precursors to TEOS and SiO₂. However, the results demonstrated that that no significant emissions can be observed. It was estimated that no silicon nanoparticles (SiNPs) formed in this case. It was believed that the formation of the as-prepared SiNPs might be closely related to the “bottom-up” strategy⁶ through the microwave synthesis route. The water solubility of the silicon source would be very important for the successful synthesis of SiNPs. Both the APTES and APTMS were hydrophilic molecules and they would be easy to participate into the oxidation–reduction process. While for the inorganic SiO₂, we can not obtain clear and transparent solution after the magnetic stirring. Thus, the silicon source in the inhomogeneous solution could hardly react with sodium citrate. As a result, no SiNPs can be achieved after the hydrothermal treatment. As for the comparison between APTES (or APTMS) and TEOS, some researchers have found that the existence of –NH₂ group would play a significant role in the blue emission. It has been reported that the photoluminescence would be related to the lonely pair of electrons of the NH groups between the organic and inorganic phases.²⁴

Fig. 1 Schematic representation of the synthesis of SiNPs and possible mechanism of fluorescence quenching by TCs. Photo: SiNPs in PBS solution (pH 7.4) excited by UV light at 365 nm without (left) and with (right) 10 μM of TC.

The SiNPs prepared by APTMS and APTES were labeled as SiNPs-1 and SiNPs-2 in the below, respectively. As can be seen in Fig. 2, we can hardly observe absorption signals from organo alkoxysilanes in the wavelength range between 300–600 nm. In contrast, after hydrothermal treatment, a new and strong absorption peak (maximum around 350 nm) appeared, indicating SiNPs have been well-formed.^6,25 So far as the fluorescent property is concerned, the resulting SiNPs displayed strong blue luminescence under UV irradiation (inset in Fig. 2) compared to the non-emissive precursors.

Fig. 2 UV-Vis absorption spectra of solutions (precursor) and the purified solutions of SiNPs-1 and SiNPs-2.

Fig. 3 shows the typical transmission electron microscope images of the as-synthesized SiNPs. Silicon nano-particles prepared by APTMS or APTES as precursors were both mono-dispersed and homogeneous spherical dots with a diameter of 3–5 nm which were consistent with the result of dynamic light scattering (DLS) histogram in Fig. 2d. The SiNPs in this contribution were prepared by oxidation–reduction reaction between APTES/APTMS and trisodium citrate, and the surface of the SiNPs is attached with numerous hydrophilic amino groups that are closely related to their excellent water solubility. Besides, the HR-TEM image originated form the nanoparticle demonstrated a crystal lattice with spacing of ∼0.2 nm which is in accordance with the (220) lattice planes of crystalline Si.²⁶

Fig. 3 (a) TEM and (b and c) HRTEM images of the as-prepared SiNPs. ((a and b) SiNPs-1, (c) SiNPs-2, inset graphs in (b) and (c) gave the enlarged HRTEM image of a single SiNP); (d) DLS histogram of SiNPs.

The photoluminescence spectra gave the optimal emission bands at 440 nm under excitation at 350 nm (Fig. 4a), and their quantum yields were determined to be 20.5% and 17.8% respectively (relative to 0.05 mol L⁻¹ quinine sulfate). Both the excitation and emission curves emerged to be well-resolved and symmetrical peaks, suggesting the sizes of the SiNPs are uniform. The results were in agreement with TEM analysis. Under the excitation at 365 nm, striking blue luminescence can be observed from the SiNPs solution (Fig. 1). It has been found that the emission peak of the as-prepared SiNPs remains to be stable upon excitations at various wavelengths as shown in Fig. 4a, which is significantly different from the reported carbon nanoparticles.²⁷ The dominant excitation wavelength was determined to be 350 nm that is highly consistent with the absorption spectra of the obtained SiNPs (Fig. 2).

Fig. 4 (a) Excitation spectrum and emission spectra of the SiNPs at different excitation wavelength in PBS buffer solution (pH 7.4). (b) The fluorescence intensity of the solution of SiNPs at different pH conditions (pH 4–14).

To further confirm the stability of SiNPs, their PL intensities were investigated under different conditions. The as-prepared SiNPs possessed excellent solubility in water and there were no significant changes in both PL intensity and peak characteristics at different concentrations of NaCl solutions, indicating that SiNPs are stable under high ionic strength conditions. Besides, they exhibited superior photostability, the photoluminescence of the SiNPs showed no reduction in intensities after 3 h excitation at 365 nm UV light. For the pH stability, Fig. 4b shows the fluorescence intensities of SiNPs at different pH values (4–14). It is seen that the emission intensity gradually increases during pH from 4 to 9. Further increase of pH value to 14 would suppress the emission. This pH-influenced luminescence phenomenon can be assigned to the containing amino groups on the surfaces of SiNPs. These findings suggest that SiNPs have great potentials for biological labels and analytical detections under physiological conditions.

3.2. Detection of TCs in solution

To evaluate possible applications of the SiNPs in fluorescence sensor area, we explored the feasibility of using such label-free SiNPs for the detection of TCs based on the on–off fluorescence process. As demonstrated in Fig. 5, the presence of TC leads to an obvious decrease of fluorescence in intensity, indicating that TC can effectively quench the fluorescence of SiNPs. Significantly, this distinguished quenching process could be observed by the naked-eye under UV excitation at 365 nm. Fig. 6 presented the correlations between the fluorescence intensity and the concentration of TC. A good linear correlation (R² = 0.9997) was observed over the concentration range of 0–10 μM by the least-squares fitting method. F₀ or F is the luminescence intensity of SiNPs at 440 nm (excitation at 350 nm) without or with the addition of TC. Parallel experiments were carried out 5 times, and the standard error bars on the measured data were shown in Fig. 6. The limit of detection (LOD) for TC was calculated to 25.9 nM, which was obtained based on LOD = 3SD/slope. SD is the standard deviation of the blank sample (obtained by 5 consecutive scans of the blank sample).

Fig. 5 Emission spectra of SiNPs upon addition of 50 nM to 50 μM of TC in PBS buffer solution (pH 7.4) under excitation at 350 nm. Inset photo: SiNPs solution excited by UV light at 365 nm without (left) and with (right) 10 μM of TC.

Fig. 6 Correlation between fluorescence intensity ratios (F/F₀) and the concentration of TC.

Based on the results above, we also investigated the sensing behaviors of other kinds of tetracycline antibiotics including oxytetracycline and chlortetracycline (Fig. S1 and S2†). The results demonstrated that the fluorescence of SiNPs also exhibited a similar quenching process, indicating the as-prepared SiNPs were sensitive to TCs. In an analogous mode, the detection limit of OTC and CTC were calculated to be 20.4 nM and 28.3 nM respectively.

Selectivity is another important parameter to evaluate the performance of a sensor. To determine whether the SiNPs was specific for TCs, the influence of other kinds of antibiotics and common coexisting substances (anions, cations, saccharides and amino acids) were investigated. As shown in Fig. S3,† no obvious changes could be observed after the addition of other antibiotics such as cefalexin, sulfamonomethoxine, penicillin and trimethoprim, suggesting the sensing process might be derived from the specific structure of TCs. Fig. S4 and S5† also demonstrated that saccharides, amino acids and various biologically relevant ions both exhibited minor influence on tetracycline detection even at concentrations 10 times higher than that of TCs. Therefore, the detection of TCs by SiNPs as optical sensors presented a nice selectivity and sensitivity. Additionally, we have also carried out the measurements of zeta potentials of SiNPs in the absence or presence of tetracyclines (Table S1†). The zeta potentials of pure SiNPs-1 and SiNPs-2 with negative charge densities were −28.67 ± 2.17 mV and −32.14.67 ± 1.76 mV respectively. After the interactions with tetracyclines, the zeta potentials turned into −19.37 ± 1.56 mV and −21.37.67 ± 1.12 mV, indicating that tetracyclines could be easily adsorbed onto the surface of silicon surfaces.

To investigate the sensing mechanism, the lifetime measurement was carried out. As shown in Fig. S6,† the lifetime displayed a good fit using a single exponential decay. After addition of 10 μM TCs, no significant changes can be observed, indicating the lifetime of the SiNPs remained to be the same during the quenching process. Thus, the sensing system may be ascribed to the static quenching process.^19,20 It is estimated that the surface groups of SiNPs would interact with tetracyclines molecules to form ground-state complexes, which lead to dramatic decrease of the fluorescence intensity (as shown in Fig. 1). We could also clearly observe the strong absorption peaks of SiNPs in the presence of tetracyclines (Fig. S7†). Apart from the static quenching, we considered the inner filter effect might be also involved in the quenching process.^28,29 It was noticed that there was an overlap between the emission band of SiNPs and the absorption peak of TC (Fig. S8†). The emission of SiNPs could be partially re-absorbed by TC. In this way, the decay time of the ∼440 nm emission would also remain unchanged. However, the shared area did not cover the entire region. The static quenching process should play a dominant role in the sensing mechanism.

3.3. Detection of TCs in milk sample

To evaluate its applicability, the described method was applied to the analysis of TCs in milk. Although the utilization of milk in the test solution would slightly decrease the sensitivity, the regression curves remained to be linear. Analysis of tetracyclines in milk was performed with simple sample pretreatments such as the removal of proteins and lipids. The solution was added with appropriate amounts of tetracyclines (0.5, 1, 3 and 5 μM) and the final measured TCs contents in all samples (0.47, 1.11, 2.89 and 5.09 μM) were derived from the standard curves and regression equations (Table 1). Oxytetracycline hydrochloride and chlortetracycline hydrochloride could also be determined by this fluorescence titration method (Tables S2 and S3†). Significantly, the recovery of the supplemented TCs was above 90% and the RSD were generally satisfactory. These results demonstrate that the proposed assay strategy could be acceptable in real samples, indicating that it has potential application in the sensing field and analytical detections.

Table 1 Recoveries of TC in supplemented milk detected by SiNPs as optical sensor

Samples	TC supplemented (μM)	TC measured (μM)	Recovery (%)	RSD (%, n = 3)
1	0.5	0.47	94.0	2.8
2	1	1.11	111.0	3.1
3	3	2.89	96.3	2.9
4	5	5.09	101.8	3.3

---

4. Conclusions

In summary, we have reported a simple one-pot strategy to synthesize fluorescent SiNPs in water. The obtained SiNPs displayed narrow size distributions, excellent water miscibility and stable photo-luminescence features. Additionally, we have utilized such label-free SiNPs as optical sensor for quantitative analysis of TCs. The bright blue emission of SiNPs would be quenched in the presence of TCs. Experimental observations indicated that the fluorescence intensity of the incubated SiNPs is linearly dependent on the TCs concentration in the range of 50 nM to 10 μM with ideal detection limit. Furthermore, the result demonstrated that this novel proposed system can be used for detection TCs in milk sample, which may open new ways for detection of TCs in practical environment.

Acknowledgements

Q. M. appreciates National Natural Science Foundation of China (no. 21371063 and 21328503), Excellent University Young Scholar Fund of Guangdong Province (Yq2013053) and Science and Technology Project in Guangzhou (2014J4100054). This study was also supported by the Scientific Research Foundation of Graduate School of South China Normal University.

References

1. N. L. Rosi and C. A. Mirkin, Chem. Rev., 2005, 105, 1547–1562.
2. G. Aragay, J. Pons and A. Merkoc, Chem. Rev., 2011, 111, 3433–3458.
3. K. Sato, H. Tsuji, K. Hirakuri, N. Fukata and Y. Yamauchi, Chem. Commun., 2009, 3759–3761.
4. H. Mertens, J. S. Biteen, H. A. Atwater and A. Polman, Nano Lett., 2006, 6, 2622–2625.
5. Y. He, Y. L. Zhong, F. Peng, X. P. Wei, Y. Y. Su, Y. M. Lu, S. Su, W. Gu, L. S. Liao and S. T. Lee, J. Am. Chem. Soc., 2011, 133, 14192–14195.
6. Y. L. Zhong, F. Peng, F. Bao, S. Y. Wang, X. Y. Ji, L. Yang, Y. Y. Su, S. T. Lee and Y. He, J. Am. Chem. Soc., 2013, 135, 8350–8356.
7. Y. H. Yi, G. B. Zhu, C. Liu, Y. Huang, Y. Y. Zhang, H. T. Li, J. N. Zhao and S. Z. Yao, Anal. Chem., 2013, 85, 11464–11470.
8. J. N. Zhao, J. H. Deng, Y. H. Yi, H. T. Li, Y. Y. Zhang and S. Z. Yao, Talanta, 2014, 125, 372–377.
9. D. Schnappinger and W. Hillen, Arch. Microbiol., 1996, 165, 359–369.
10. K. D. Wasch, L. Okerman, H. D. Brabander, J. V. Hoof, S. Croubels and P. D. Backer, Analyst, 1998, 123, 2737–2741.
11. G. T. Peres, S. Rath and F. G. R. Reyes, Food Control, 2010, 21, 620–625.
12. S. Croubels, C. V. Peteghem and W. Baeyens, Analyst, 1994, 119, 2713–2716.
13. W. A. Moats, J. Agric. Food Chem., 2000, 48, 2244–2248.
14. M. F. Tavares and V. L. McGuffin, J. Chromatogr. A, 1994, 686, 129–142.
15. A. Pena, N. Pelantova, C. M. Lino, M. I. N. Silveira and P. Solich, J. Agric. Food Chem., 2005, 53, 3784–3788.
16. A. Townshend, W. Ruengsitagoon, C. Thongpoon and S. Liawruangrath, Anal. Chim. Acta, 2005, 541, 103–109.
17. A. L. Myllyniemi, L. Nuotio, E. Lindfors, R. Rannikko, A. Niemi and C. Backman, Analyst, 2001, 126, 641–646.
18. G. Alfredsson, C. Branzell, K. Granelli and Å. Lundström, Anal. Chim. Acta, 2005, 529, 47–51.
19. J. Hou, J. Yan, Q. Zhao, Y. Li, H. Ding and L. Ding, Nanoscale, 2013, 5, 9558–9561.
20. X. M. Yang, Y. W. Luo, S. S. Zhu, Y. J. Feng, Y. Zhuo and Y. Dou, Biosens. Bioelectron., 2014, 56, 6–11.
21. H. L. Tan and Y. Chen, Sens. Actuators, B, 2012, 173, 262–267.
22. X. M. Yang, S. S. Zhu, Y. Du, Y. Zhuo, Y. W. Luo and Y. J. Feng, Talanta, 2014, 122, 36–42.
23. H. L. Tan, C. J. Ma, Y. H. Song, F. G. Xu, S. H. Chen and L. Wang, Biosens. Bioelectron., 2013, 50, 447–452.
24. L. D. Carlos, R. A. S. Ferreira, V. D. Z. Bermudez and S. J. L. Ribeiro, Adv. Funct. Mater., 2001, 11, 111–115.
25. J. Zhang and S. H. Yu, Nanoscale, 2014, 6, 4096–4101.
26. Y. L. Zhong, F. Peng, X. P. Wei, Y. F. Zhou, J. Wang, X. X. Jiang, Y. Y. Su, S. Su, S. T. Lee and Y. He, Angew. Chem., Int. Ed., 2012, 51, 8485–8489.
27. S. N. Baker and G. A. Baker, Angew. Chem., Int. Ed., 2010, 49, 6726–6744.
28. N. Shao, Y. Zhang, S. M. Cheung, R. H. Yang, W. H. Chan, T. Mo, K. A. Li and F. Liu, Anal. Chem., 2005, 77, 7294–7303.
29. M. Kubista, R. Sjoback, S. Eriksson and B. Albinsson, Analyst, 1994, 119, 417–419.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra01769f

---