Determination of Ag⁺ ions by a graphene oxide based dual-output nanosensor with high selectivity

Abstract

A new highly selective chemosensor for Ag⁺ ions was designed and synthesized by covalently introducing well-known fluorophore 1,8-diaminonaphthalene (DAN) onto graphene oxide (GO) sheets. The resultant fluorescent sensor 1,8-diaminonaphthalene-graphene (GAP) was quenched due to the photoinduced electron transfer (PET) mechanism, and the intensity of second-order scattering (SOS) was enhanced upon addition of Ag⁺ ions. The fluorescent probe GAP exhibits no or a slight response to many other metal ions.

---

Silver is well known for its application as an effective antibiotic agent in various products.^1–3 Nanosilver was popularly used in many products such as cosmetics, textiles, food boxes, sprays, wound dressings, etc.^4–7 However, Ag⁺ as a heavy metal ion could bring adverse effects to aquatic ecosystems.⁸ It was found that silver ions could bind with various metabolites and inactivate sulfhydryl enzymes and produce dose-dependent cytopathogenic effects in distinct cell types, including human gingival fibroblasts, keratinocytes, human tissue mast cells and endothelial cells.^9–11 They entered living organisms through the alimentary chain and led to human internal organ edema, and even to death.¹² Accordingly, there is an urgent need to develop chemical sensors that are capable of detecting the presence of silver ions in environmental and biological samples under physiological relevant conditions.

A variety of classical analytical methods, including atomic absorption spectrometry and plasma atomic emission spectrometry, have been successfully developed to detect trace levels of silver ions.¹³ However, they have limited practical applications due to the high cost, time consuming and complex handling. Recently, great emphasis was placed on the development of new highly selective fluorescent sensors of biologically active metal ions.^14–17 In particular, the development of fluorescent molecular sensors directed specifically toward the detection and estimation of Ag⁺ either in vitro or in vivo has attracted considerable attention in different fields.^18,19

Considerable efforts have been devoted to developing fluorescent chemosensors for Ag⁺ over the last few decades. For example, a cysteamine capped CdS quantum dots (Cys–CdS QDs) was synthesized as a selective fluorescence probe to measure trace silver ions by enhanced fluorescence intensity.²⁰ The conductometric biosensor with three-enzyme system (invertase, mutarotase and glucose oxidase) was developed for determination of Ag⁺ ions based on inhibition of the cascade of enzymatic reactions resulting in changes of the medium conductivity.²¹ The graphene oxide (GO) and DNA were also mixed for determination of Ag⁺ by fluorescent method.²² However, complicated synthesis of sensors and delay of response restrict the development of silver ion sensors. Up to now, fluorescent sensors based on graphene oxide for detecting metal ions in aqueous system have been seldom found.

In this paper, we successfully synthesized a 1,8-diaminonaphthalene-GO (GAP) fluorescent sensor for detection of Ag⁺ with high sensitivity and selectivity. The Ag⁺ ion shows fluorescence (FL) quenching effect and enhancement of second-order scattering (SOS) of GAP. The SOS phenomenon gave rise many interests in recent years for sensors.^23,24 He et al. investigated the interaction between CdSe quantum dots and chitosan showing that the chitosan and CdSe quantum dots formed a network structure aggregates by electrostatic attraction and hydrophobic force. After the interaction of CdSe quantum dots with chitosan, the intensities of SOS enhanced and the enhancements were in proportion to the concentration of chitosan. The increase of the molecular volume and the hydrophobic force between CdSe QDs-chitosan and water were the reasons for the enhancement of nonfluorescence signals.²⁵ The interaction between emodin and ethyl violet was also studied by SOS and resonance Rayleigh scattering methods.²⁶ In our research, the dual-output signals make the GAP a highly sensitive and selective fluorescent sensor for Ag⁺.

Results and discussion

Synthesis of the GAP compound is simple and mild. As is schematically shown in Fig. 1a, GAP could be synthesized by coupling of 1,8-diaminonaphthalene (DAN) with graphene oxide using well-established EDC/NHS coupling chemistry. Fig. 1b exhibits graphene oxide sheets in the AFM image, the corresponding height profile shows an average thickness of ∼0.694 nm for GO, suggesting that the GO film consists of single layer graphene sheets. The height of GAP becomes almost 1.68 nm, which is much higher than that of GO (Fig. 1c). This is because DAN has been incorporated on the surface of GO.

Fig. 1 Synthetic pathway of GAP and AFM images of GO and GAP. (a) Synthetic pathway of GAP (b) GO and the corresponding height profile, (c) GAP and the corresponding height profile.

The characteristic absorption peaks of GAP appeared in the UV-vis spectrum at 262 nm and 347 nm, while the DAN at 232 and 335 nm (Fig. 2a). These peaks are associated with the π–π* electron transitions of aromatic rings.²⁷ The spectrum of GAP shows significant bathochromic-shift due to the covalently linking of GO and DAN. The FTIR spectrum of GAP (Fig. 2b) shows very strong O–H and N–H vibration at 3382 cm⁻¹. The C=O stretch at 1721 cm⁻¹, the amide band at 1632 and 1585 cm⁻¹, and aromatic C–C stretch (in ring) at 1455 and 1340 cm⁻¹ were also clearly observed. Signals at 850–1250 cm⁻¹ can be attributed to the =C–H bend, C–O and C–H stretch. Raman spectroscopy was used to study the ordered/disordered crystal structures of carbonaceous materials. In many cases, the Raman spectrum of carbon materials is characterized by two main features, the G band (∼1580 cm⁻¹) and the D band (∼1380 cm⁻¹). The D band is ascribed to edges, other defects, and disordered carbon, whereas the G band arises from the zone center E_2g mode, corresponding to ordered sp²-bonded carbon atoms.²⁸ The R-value (I_D/I_G) is a measure of the degree of disorder and average size of the sp² domain. The R value decreasing from 1.3 to 0.92 shows a significant decrease upon grafting DAN onto GO, indicating an increase in the average size of the sp² domains for GAP (Fig. 2c).²⁹ At the same time, we further characterized the crystal structure of GAP by XRD. Fig. 2d shows XRD spectra for graphite, GO, GAP and GAPAg. Compared with the pristine graphite, GO shows a broad band over low diffraction angles (2θ = 11.7°). The diffraction peak of modified GO down-shifted and broadened, indicated the covalently bonded organic molecule increased the interlayer space between graphene sheets in the GAP hybrid.³⁰ The stability of GO before and after reaction was researched by the thermal analysis method. From Fig. 2e, we can see that the complex GAP shows a lower decomposition rate than that of GO, indicating the difference of crystal structure of carbon materials. Meanwhile a complete weight loss was observed for DAN at ∼220 °C, presumably due to thermal evaporation.

Fig. 2 Characterization of GAP, GO and DAN. (a) UV-vis spectra of the GO, DAN, GAP; (b) FTIR spectra of the GO and GAP, (c) Raman spectra of the GO and GAP, (d) XRD of the graphite, GO, GAP, and GAPAg, (e) thermogravimetric analysis of the GO, DAN, GAP, (f) ¹H NMR spectra of the DAN and GAP in d₆-DMSO.

To further illustrate structure nature, the complex GAP was characterised by ¹H NMR and XPS. The ¹H NMR spectrum (Fig. 2f) of DAN shows four typical peaks at 5.43 ppm (signal of Ph–NH₂), 6.57 ppm, 6.97 ppm, and 7.05 ppm (Ph–H). After DAN incorporation onto GO, the signal of Ph–NH₂ became broad. Such signal is attributed to the unreacted amine. A new peak at about 8.19 ppm is the signal of amide groups (O=C–NH), it appears at downfield. There exist many peaks at 6.40–7.29 ppm. They are the peaks of phenyl of DAN and GO.³¹ The peaks of 7.46 and 7.57 ppm are the Ph–OH on the GO. Although the GAP can be well dispersed in aqueous and DMSO solutions, its solubility is low and the signal of GAP was not well constructed as the DAN. The peaks of GAP significantly broadened due to low solubility and hydrogen bonding. The above observation showed the DAN linked the GO with an amide group as shown in Fig. 1a.

The C 1s XPS spectrum (Fig. S1 and Table S1, ESI†) of GO clearly showed a considerable degree of oxidation with four components corresponding to C–C (285.0 eV), C–O (286.4 eV), C=O (287.5 eV), and C(O)O (289.1 eV), respectively.³² For the complex GAP, there is two additional components at 285.53 eV and 289.13 eV corresponding to C–N and N–C=O respectively, accompanied by the reduced O 1s peak with respect to the C 1s peak in the XPS survey spectrum of GAP. The corresponding N 1s peak from GAP shown in Fig. S1d† can be fitted to amine N at 399.42 eV and –NH–C=O at 400.92 eV. These XPS results confirmed the successful covalent binding between DAN and GO.

Titration experiments are carried out and demonstrate that Ag⁺ promotes remarkable response (Fig. 3). On addition of Ag⁺ up to 3.5 μM to the GAP solution, the absorbance at 232, 263 nm and 285 nm gradually rise confirming the interaction process, thereby induce a promising ultraviolet absorption.

Fig. 3 UV-vis absorption spectra of the GAP titration by Ag⁺ (the concentration of GAP is 2 mg L⁻¹).

The fluorescence emission spectra of the GAP and DAN were obtained by excitation at 293 nm and 337 nm, respectively. Compared with DAN, the emission peaks of complex GAP appeared at 392 nm, meanwhile showed a SOS peak at 589 nm (Fig. 4a). Fluorescence intensity at 392 nm decreased significantly after adding Ag⁺, however, the SOS intensity at 589 nm sharply increased, which was ascribed to the second-order scattering (Fig. 4b), the limit of detection is 0.2 mg L⁻¹. Titration of Ag⁺ to the DAN solution did not induce significant changes of fluorescent signal (Fig. S2, ESI†). Compared with GAP, the emission peaks of physical mixture of GO and 1,8-diaminonaphthalene only showed a SOS peak at about 589 nm excited by 293 nm (Fig. S3, ESI†), and no fluorescent peak was observed due to quenching with free GO. When adding Ag⁺ to the mixture solution of GO and DAN, the SOS peak did not significantly increase. These facts suggested the Ag⁺ coordinated to the amide moiety, and the electron transferred from the amide linkage to empty orbital of Ag⁺ center caused fluorescent quenching. Fig. S4 (ESI†) showed that the lifetime of GAP with Ag⁺ is lowed than that of GAP. Such mechanism is a photoinduced electron transfer (PET) signaling mechanism.³³ It is well-known that many transition metal ions binding to the organic sensors linked by amide induce electrons transferring from the fluorescent moiety to the unbound orbital of metals. Such courses are always accompanied by fluorescent quenching.³⁴ As shown in Fig. 4c, the fluorescent peak at 392 nm was quenched, and the quenching efficiency was linear related to the Ag⁺ concentration from 6 mg L⁻¹ to 12 mg L⁻¹, where the calibration equation was (I₀ − I)/I₀ = 8.7875C_Ag⁺ − 29.425 (R² = 0.9937). The relationship between the SOS response and the Ag⁺ concentration from 6 mg L⁻¹ to 12 mg L⁻¹ was also linear. The calibration equations at peak 589 nm (enhancement of SOS signal) was (I − I₀)/I₀ = 0.294C_Ag⁺ + 0.71142 (R² = 0.9932), as shown in Fig. 4d. Such phenomenon indicated the quantities of Ag⁺ can be accurately determined by the two calibrations. In aqueous solutions with different pHs (from 4.5 to 6.5), the GAP/Ag⁺ showed similar fluorescence spectra (Fig. 5). The peak at 392 nm decreased significantly at the pH of 5.5 and 6.5, while the peak at 589 nm increased dramatically. The solution pH 5.5 was chosen for the system.

Fig. 4 Fluorescence properties of GAP. (a) Fluorescence emission spectra of the DAN and GAP; (b) fluorescence response of GAP upon addition of Ag⁺ in a aqueous solution (with the concentration from 0.2 mg L⁻¹ to 16 mg L⁻¹); (c) calibration curve for Ag⁺ detection at peak of 392 nm with the concentration from 6 mg L⁻¹ to 12 mg L⁻¹; (d) calibration curve for Ag⁺ detection at peak of 589 nm with the concentration from 6 mg L⁻¹ to 12 mg L⁻¹ (the concentration of GAP and DAN are 2 mg L⁻¹, excitation at 293 nm).

Fig. 5 Fluorescence spectra of GAP upon addition of 5-fold Ag⁺ in a aqueous solution with various pHs (the concentration of GAP is 2 mg L⁻¹, excitation at 293 nm).

The fluorescence and SOS response of GAP to various metal ions and its selectivity for Ag⁺ are shown in Fig. 6. With the presence of alkali and alkaline earth metals, only slight changes of fluorescence intensity were observed due to their low affinity with the sensors. Upon the addition of Cu²⁺, the fluorescence of 392 nm was quenched significantly due to coordination interaction between Cu²⁺ and N of GAP. The behaviour is similar to Ag⁺ binding to GAP at 392 nm. However, the Cu²⁺ did not significantly enhance the intensity of GAP at 589 nm. Although the Cu²⁺ and Ag⁺ are in the IB group, the Cu²⁺ only induced the change of one signal at 392 nm. The fluorescent quenching at peak of 392 nm and enhancement of SOS signal at 589 nm were obviously shown in the presence of Ag⁺. As can be seen from Fig. 2d, the GAP after adsorbing the Ag⁺ produced the GAPAg complex. The XRD spectrum of GAPAg exhibited two new weak peaks at 38.7 and 44.6° assigning to the (111) and (200) lattice planes of Ag crystals.³⁵ Also, the peak of GO further reduced to a low position suggesting the increase of the interlayer space between graphene sheets by the reduced Ag particles. The coordination of Cu²⁺ or Ag⁺ induced the fluorescent signal at 392 nm to decrease, while the reduced Ag enhanced the SOS peak at 589 nm. The Cu²⁺ did not induce the enhancement of SOS peak probability due to non-reduction of Cu²⁺ ions. It has been reported that the Cu²⁺ is much stable when chelating to polyaminonaphthalene derivatives.³⁶ Ag⁺ can be reduced by GAP due to its higher oxidation property than other cations. In aqueous solution, when other metal cations combine with GAP to form binding products, they are still hydrophilic and the SOS intensities of GAP and cation-GAP complexes are all weak. But for the Ag–GAP complex, a reduction immediately occurs resulting in aggregation of Ag particles. Most of silver particles lose their charges in the course of reduction, and the Ag–GAP complex became hydrophobic. The liquid–solid interface between the hydrophobic complex and water may form, which produces surface enhanced scattering and enhances the scattering intensities.³⁷ The GAP responses to Ag⁺ ion by dual-output signals make it specifically recognize Ag⁺. The competition experiments indicated that the Ag-induced luminescence response was almost unaffected in the background of 5 equiv. of environmentally relevant alkali or alkaline-earth metals, such as Na⁺, K⁺, Mg²⁺, and Ca²⁺. In addition, the first-row transition-metal ions, including Mn²⁺, Fe²⁺, Co²⁺, Ni²⁺, and Cu²⁺, did not interfere with the Ag⁺-induced fluorescence increase, suggesting that compound GAP has a remarkable selectivity for Ag⁺.

Fig. 6 Metal ion selectivity of GAP in 1 : 1 ethanol–water solution with addition of a 5-fold excess concentration of various ions. (1) Blank; (2) Na(I); (3) K(I); (4) Ca(II); (5) Mg(II); (6) Cr(III); (7) Mn(II); (8) Fe(II); (9) Co(II); (10) Ni(II); (11) Cu(II); (12) Zn(II); (13) Fe(III); (14) Al(III); (15) Cd(II); (16) Ag(I), respectively. (a) Fluorescent quenching at peak of 392 nm; (b) enhancement of SOS signal at peak of 589 nm (red column: GAP + interferent; green column: GAP + interferent + Ag⁺).

The method was compared with other published techniques for Ag⁺ determination in Table 1. Although the limited of detection (LOD) of the other methods is lower, they require additional equipments and complicated sample preparation procedures. This methodology is a reproducible, simple and low cost technique, which is presented as a suitable alternative to more expensive instruments for Ag determination at trace levels.

Table 1 Comparison of analytical characteristics of the other reported techniques for determination of silver^a

Method	Linear range, mg L^−1	LOD, mg L^−1	Reference
a FAAS, flame atomic absorption spectrometry; ICP-OES, inductively coupled plasma optical emission spectrometry; IC-CL, ion chromatography-chemiluminescence.
FAAS	0.01–1	0.0039	38
IC-CL detection	0.001–0.1	0.0034	39
ICP-OES	0.1–0.9	0.02	40
Voltammetry	0.07–10	0.06	41
Spectrofluorometry	6–12	0.2	This work

---

Conclusions

To summarize, for the first time, a simple, highly selective and highly sensitive nanosensor was developed by covalently bonding DAN onto the GO for the detection and quantitative analysis of Ag⁺ based on the mechanism of PET and SOS. The coordination of Ag⁺ induced the fluorescent signal at 392 nm to decrease, while the reduced Ag enhanced the SOS peak at 589 nm. It is through this unique dual-output signal (in contrast to other metal cations) that the nanosensor can selectively recognize Ag⁺, with high sensitivity as well. The spectroscopic investigations indicated that the structure integrity of GO sheets was significantly enhanced by covalently bonding with DAN. The GAP exhibited an approximately linear response to the Ag⁺ in aqueous solution. The Ag⁺ efficiently chelated to the GAP and the reducing Ag enhanced the SOS signal. The results are not only of technical importance for Ag⁺ measurement but also provide a practical basis to other fluorescent chemosensors for detection of various other types of trace metal ions.

Acknowledgements

Financial supports for this work from the National Natural Science Foundation of China (21405115, 51433005), the National “Thousand Talents Program”, Technology project of Zhejiang Province Medical and Health Project (2015KYB254), and Ophthalmology & Optometry of Wenzhou Medical University (YNCX201408).

References

1. H. Qu, C. Knabe, S. Radin, J. Garino and P. Ducheyne, Biomaterials, 2015, 62, 95.
2. M. Norouzi, S. M. Boroujeni, N. Omidvarkordshouli and M. Soleimani, Adv. Healthcare Mater., 2015, 4, 1114.
3. K. Chaloupka, Y. Malam and A. M. Seifalian, Trends Biotechnol., 2010, 28, 580.
4. M. E. Quadros and L. C. Marr, Environ. Sci. Technol., 2011, 45, 10713.
5. C. Lorenz, L. Windler, N. von Goetz, R. P. Lehmann, M. Schuppler, K. Hungerbuehler, M. Heuberger and B. Nowack, Chemosphere, 2012, 89, 817.
6. N. von Goetz, L. Fabricius, R. Glaus, V. Weitbrecht, D. Gunther and K. Hungerbuhler, Food Addit. Contam., Part A, 2013, 30, 612.
7. S. W. P. Wijnhoven, W. J. G. M. Peijnenburg, C. A. Herberts, W. I. Hagens, A. G. Oomen, E. H. W. Heugens, B. Roszek, J. Bisschops, I. Gosens and D. Van de Meent, Nanotoxicology, 2009, 3, 109.
8. V. K. Bhardwaj, N. Singh, M. S. Hundal and G. Hundal, Tetrahedron, 2006, 62, 7878.
9. Z. Yuan, X. Yang, A. Hu and C. Yu, Chem. Eng. J., 2015, 276, 83.
10. E. Hidalgo and C. Dominguez, Toxicol. Lett., 1998, 98, 169.
11. V. K. M. Poon and A. Burd, Burns, 2004, 30, 140.
12. M. Li, H. Gou, I. Al-ogaidi and N. Wu, ACS Sustainable Chem. Eng., 2013, 1, 713.
13. H. Abdolmohammad-Zadeh and Z. Javan, Microchim. Acta, 2015, 182, 1447.
14. L. Zhou, Q. Wang, X. Zhang and W. Tan, Anal. Chem., 2015, 87, 4503.
15. E. L. Que, D. W. Domaille and C. J. Chang, Chem. Rev., 2008, 108, 1517.
16. Z. Hai, Y. Bao, Q. Miao, X. Yi and G. Liang, Anal. Chem., 2015, 87, 2678.
17. S. Gui, Y. Huang, F. Hu, Y. Jin, G. Zhang, L. Yan, D. Zhang and R. Zhao, Anal. Chem., 2015, 87, 1470.
18. L. Zeng, E. V. Mller, A. Pralle, E. Y. Isacoff and C. J. A. Chang, J. Am. Chem. Soc., 2006, 128, 10.
19. Y. Zhang, H. Jiang and X. Wang, Anal. Chim. Acta, 2015, 870, 1.
20. T. Khantaw, C. Boonmee, T. Tuntulani and W. Ngeontae, Talanta, 2013, 115, 849.
21. O. O. Soldatkin, I. S. Kucherenko, V. M. Pyeshkova, A. L. Kukla, N. Jaffrezic-Renault, A. V. El'skaya, S. V. Dzyadevych and A. P. Soldatkin, Bioelectrochemistry, 2012, 83, 25.
22. W. Y. Xie, W. T. Huang, N. B. Li and H. Q. Luo, Chem. Commun., 2012, 48, 82.
23. J. Yang, E. Wang, S. Zhou and Q. Yang, Talanta, 2015, 134, 754.
24. J. Li, X. Yang, J. Yang and L. Lai, Luminescence, 2015, 30, 643.
25. J. J. Peng, S. P. Liu, L. Wang, Z. W. Liu and Y. Q. He, J. Colloid Interface Sci., 2009, 338, 578.
26. E. Wang, J. Yang and D. Yu, RSC Adv., 2014, 4, 34204.
27. Y. S. Zou, A. J. Hou, G. F. Zhu, Y. F. Chen, H. D. Sun and Q. S. Zhao, Bioorg. Med. Chem., 2004, 12, 1947.
28. K. N. Kudin, B. Ozbas, H. C. Schniepp, R. K. Prudhomme, I. A. Aksay and R. Car, Nano Lett., 2008, 8, 36.
29. V. Chandra, J. Park, Y. Shun, J. W. Lee, I. C. Hwang and K. S. Kim, ACS Nano, 2010, 4, 3979.
30. J. Shen, Y. Hu, M. Shi, X. Lu, C. Qin, C. Li and M. Ye, Chem. Mater., 2009, 21, 3541.
31. H. Bai, W. Jiang, G. P. Kotchey, W. A. Saidi, B. J. Bythell, J. M. Jarvis, A. G. Marshall, R. A. S. Robinson and A. Star, J. Phys. Chem. C, 2014, 118, 10519.
32. S. Stankovich, R. D. Piner, X. Chen, N. Wu, S. T. Nguyen and R. S. Ruoff, J. Mater. Chem., 2006, 16, 155.
33. S. Ou, Z. Lin, C. Duan, H. Zhang and Z. Bai, Chem. Commun., 2006, 4392.
34. A. P. De Silva, D. B. Fox, A. J. M. Huxley and T. S. Moody, Coord. Chem. Rev., 2000, 205, 41.
35. P. Wang, B. Huang, X. Qin, X. Zhang, Y. Dai, J. Wei and M. H. Whangbo, Angew. Chem., Int. Ed., 2008, 47, 7931.
36. B. J. Palys, M. Skompska and K. Jackowska, J. Electroanal. Chem., 1997, 433, 41.
37. K. J. Huang, C. Y. Wei, Y. M. Shi, W. Z. Xie and W. Wang, Spectrochim. Acta, Part A, 2010, 75, 1031.
38. M. R. Fathi, N. Pourreza and S. Purweis, J. Chin. Chem. Soc., 2009, 56, 725.
39. P. Jones and H. G. Beere, Anal. Proc. incl. Anal. Commun., 1995, 32, 169.
40. A. Väisänen, R. Suontamo, J. Silvonen and J. Rintala, Anal. Bioanal. Chem., 2002, 373, 93.
41. H. Zejli, J. L. Hidalgo-Hidalgo de Crisneros, J. Naranjo-Rodriguez and K. R. Temsamani, Talanta, 2007, 71, 1594.

---

Footnotes

† Electronic supplementary information (ESI) available: Experimental section, fluorescent spectra, and lifetime curves. See DOI: 10.1039/c6ra00048g

‡ Lingli Li and Rumei Cheng contributed equally to this work.

---