A novel graphene oxide-based fluorescent nanosensor for selective detection of Fe3+ with a wide linear concentration and its application in logic gate

Biosensors and Bioelectronics 70 (2015) 69–73

Contents lists available at ScienceDirect

Biosensors and Bioelectronics journal homepage: www.elsevier.com/locate/bios

Short communication

A novel graphene oxide-based fluorescent nanosensor for selective detection of Fe3 þ with a wide linear concentration and its application in logic gate Liang He a, Jianna Li b, John H. Xin a,n a b

Institute of Textiles and Clothing, The Hong Kong Polytechnic University, Hung Hom, Hong Kong, P.R. China Shenzhen Key Laboratory of Translational Medicine of Tumor, Health Science Center, Shenzhen University, Shenzhen 518060, P.R. China

art ic l e i nf o

a b s t r a c t

Article history: Received 24 November 2014 Received in revised form 14 January 2015 Accepted 18 January 2015 Available online 3 March 2015

A graphene oxide-based fluorescent nanosensor AGO has been designed and synthesized by covalent grafting allylamine onto GO surface. In aqueous media, AGO displays a highly selective and sensitive discrimination of Fe3 þ from Fe2 þ and other metal ions through electron transfer-induced fluorescence quenching. The quenching of AGO fluorescence is linearly proportional to Fe3 þ concentration in a wide range of 0–120 μM (correlation coefficient R2 ¼0.9994). Moreover, AGO can be used to construct a combinational three-input logic gate to discriminate Fe3 þ and Fe2 þ . The logic gate works well in intracellular fluorescence imaging, which shows a potential as a promising platform for biosensing analysis. & 2015 Published by Elsevier B.V.

Keywords: Graphene oxide Nanosensor Electron buffer Logic gate Cell imaging

1. Introduction Biological species play important roles in various physiological and pathological activities (Bush, 2000). Among these, Fe3 þ is an indispensable ion for most organisms and performs significant roles in various biochemical processes at the cellular level including structure and activity retention for proteins and enzymes, oxygen transportation, photosynthesis, DNA synthesis and repair, and nitrogen fixation (Que et al., 2008; Lee and Helmann, 2006). Fe3 þ concentration in organisms is usually a health index and its overload or deficiency will result in various disorders. Excess levels of Fe3 þ will not only cause organ dysfunction and even cancers, but also easily initiate serious diseases, such as Parkinson's, Alzheimer's and Huntington's diseases (Galaris et al., 2008; Crichton et al., 2008). On the other hand, the deficiency of Fe3 þ will result in low oxygen delivery to cells, which will deprive oxygen in organs and cause diseases of anemia, hemochromatosis and even cancers, etc. (Narayanaswamy and Govindaraju, 2012). Therefore, it is important to develop efficient and convenient methods to detect intracellular Fe3 þ . Compared to other methods, fluorescent detection shows great superiority in its high selectivity, sensitivity, fast response and simplicity (Chen et al., 2012). Therefore, n

Corresponding author. Fax: þ 852 2766 1432. E-mail addresses: [email protected] (L. He), [email protected] (J.H. Xin). http://dx.doi.org/10.1016/j.bios.2015.01.075 0956-5663/& 2015 Published by Elsevier B.V.

considerable efforts have been devoted to develop fluorescent sensors for Fe3 þ detection (Yang et al., 2012; Hu et al., 2011a, 2011b, 2014; Choi et al., 2014; Kim et al., 2013; Zheng et al., 2013; Wang et al., 2013; Qu et al., 2013). However, many reported sensors are not suitable to be used in 100% aqueous media and only very few sensors are reported to conduct logic gates for the discrimination of Fe3 þ and Fe2 þ in living cells (Mei et al., 2012; Murale et al., 2013). Graphene, as a novel single-atom-thick carbon nanomaterial, has recently attracted more attention due to its superior optical and electrical properties (Geim and Novoselov, 2007; Kim et al., 2009; Geim, 2009). Its water-soluble derivative of graphene oxide (GO) has numerous carboxyl, hydroxyl and epoxy groups, which usually induce the non-radiative recombination of localized electron–hole pairs and thus result in the non-emission of GO. After modification to these oxygen-containing groups, the non-radiative recombination sites can be removed and the modified GO will show photoluminescent properties (Mei et al., 2010, 2012; Dreyer et al., 2010; Cheng et al., 2012; Jeon et al., 2013). Therefore, GO is an emerging nanomaterial for the construction of sensors with unique properties (Cheng et al., 2012; Lu et al., 2009; Dong et al., 2010; Zhang et al., 2011, 2013; Li et al., 2012; Wang et al., 2014; He et al., 2013). Saturated hydrocarbon-modified GO was reported as a fluorescent sensor for Fe3 þ detection through a mechanism of GO-Fe3 þ electron transfer (ET) (Mei et al., 2012). This process led to a quick quenching of the fluorescence emission and followed

70

L. He et al. / Biosensors and Bioelectronics 70 (2015) 69–73

Fig. 1. Schematic illustration of fluorescent GO logic gates with/without electron buffer in living cells.

the Fe3 þ detection concentration without a wide enough linear range. However, a wide linear range of detection concentration is important for practical application in analytical fields and it is worthy to be considered for developing sensors. To the best of our knowledge, the approach to enlarge linear range of detection concentration through conjugating an electron buffer to a sensor has not been documented. Herein, we designed and synthesized a novel fluorescent graphene oxide (AGO) through covalent grafting allylamine onto GO surface as an electron buffer. AGO shows a stable fluorescence emission, which can be selectively quenched by Fe3 þ even in the presence of biological species. Due to the unsaturated nature of the grafted allylamine, the linear range of Fe3 þ detection concentration is enlarged in this quenching process. Moreover, AGO can be used to construct a combinational logic gate for the discrimination of Fe3 þ and Fe2 þ through intracellular fluorescence imaging (Fig. 1).

wavelength of 350 nm with the slit set of 10 nm. 2.3. AGO synthesis GO was prepared using natural graphite flakes as raw material via the typical Hummer's method through acid-oxidation exfoliation, followed by filtration, washing and dryness under vacuum (Hummers and Offeman, 1958). Subsequently, 12 mg of dried GO was dissolved in 10 mL thionyl chloride and was then refluxed overnight after the addition of several drops of DMF. The resultant mixture was separated by centrifugation and washed with dried THF. Then, the residual was mixed with 1 mL allylamine in THF and stirred overnight at room temperature. The final mixture was dried out by rotary evaporation under vacuum. 10 mL deionized water was added to extract the target. The AGO powder was obtained after the centrifugation and dried under vacuum. 2.4. Measurement procedures

2. Experimental 2.1. Materials Natural graphite flakes and allylamine (analytical grade) were purchased from Sigma-Aldrich Chem. Co. All other regents (analytical grade) and solvents (spectroscopic grade) were commercially available and used as received. 2.2. Instruments UV–visible absorption spectra were measured on a HP 8453 spectrometer with the scanning range of 200–700 nm. Fourier transform infrared spectra (FTIR) were recorded on a Perkin-Elmer Paragon 1000 infrared spectrometer in the range of 4000–650 cm  1. Field emission transmission electron microscopy (FETEM) images were obtained from a JEOL Model JEM-2100F microscope. The chemical compositions were determined by X-ray photoelectron spectroscopy (XPS) on a SKL-12 X-ray photoelectron spectrometer (Shenyang, China) equipped with a VG CLAM 4MCD electron energy analyzer. XPS is configured with a dual anode source from VG (type XR3E2) and non-monochromatic Mg Kα radiation (1253.6 eV) at a current of 15 mA with an ultrahigh vacuum ( o8  10  10 Torr). To compensate for surface charging effects, all binding energies were referenced to the C 1s hydrocarbon peak at 284.6 eV. All fluorescence spectra were recorded on a Perkin-Elmer LS55 fluorescence spectrometer at the excitation

The metal ion stock solutions (1.0  10  2 mol L  1) were prepared in deionized water using their chlorides (Fe3 þ , Cr3 þ , Ni2 þ , Fe2 þ , Mn2 þ , Cu2 þ , Hg2 þ , Na þ , K þ ) or nitrates (Al3 þ , Pb2 þ , Mg2 þ , Zn2 þ ). H2O2 was diluted to 2.0  10  2 mol L  1 using deionized water. The AGO stock solution (0.6 mg mL  1) was prepared in deionized water. To a quartz cell containing AGO solution in deionized water, proper amounts of metal ion solutions were directly added with a micropipette. The fluorescence emission spectra were then measured instantly. Distilled water was used throughout the experiments. For each result, the average of three repeated measurements was used. 2.5. Cell incubation and imaging The living human lung cancer cells A549 were purchased from Kunming Cell Bank of Type Culture Collection, Chinese Academy of Sciences. The A549 cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum at 37 °C and 5% CO2 in 1-well plates with a density of 2.0  105 cells per plate. Before fluorescent imaging experiment, the cells were washed with phosphate buffered saline (PBS) for 3 times, followed by incubating with 100 mL AGO solution in 2 mL medium for 24 h at 37 °C. After removal of the extracellular AGO by washing with PBS for 3 times, experiments to recognize Fe3 þ in living cells were then performed by incubating the fluorescent cells with 100 mL Fe3 þ for 1 h. Experiments to recognize Fe2 þ were performed by incubating

L. He et al. / Biosensors and Bioelectronics 70 (2015) 69–73

3.1. Structural characterization The structures of the allylamine-grafted AGO and GO were first characterized by electron microscopy (Fig. S1). In their typical FETEM images, AGO and GO were transparent nanosheets. Namely, the grafting reaction did not change the nanostructure morphology of GO. Interestingly, the wrinkles were clearly observed in AGO FETEM image and its raised edges had smooth arc shapes. These indicated that the grafted AGO nanosheet displayed good flexibility. In absorption spectrum, GO solution showed two characteristic peaks at ca. 230 and 300 nm, which are assigned to the π–π* transition of CQC and n–π* transition of CQO, respectively (Fig. S2a). After the covalent grafting allylamine, these two peaks disappeared and a new absorption band at ca. 325 nm concomitantly occurred. These clearly exhibited the different absorption property of GO solution before and after the grafting. The covalent grafting allylamine onto GO surface was also confirmed by its FTIR analysis (Fig. S2b). The FTIR spectrum of GO showed the characteristic CQO stretching at 1717 cm  1 and C (O)–O stretching at 1263 cm  1. In the spectrum of AGO, these two bands disappeared and new vibration bands appeared at 1645 cm  1 and 1578 cm  1, which were assigned to the CQO stretching and C(O)–NH bending, respectively (Cheng et al., 2012; Mei et al., 2010; Compton et al., 2010). These results confirmed the formation of amide bonds between GO and allylamine in AGO. At the same time, the band at 1068 cm  1 for GO epoxides completely disappeared in AGO, while a new band accordingly appeared at 1128 cm  1 due to the C–N–C asymmetric stretching in the grafted allylamine (Mei et al., 2010). Obviously, the epoxy groups in GO were removed through a ring-opening amination reaction with allylamine. XPS analyses further confirmed the covalent grafting allylamine onto GO surface. In the survey spectra (Fig. S3a), GO only showed the C 1s and O 1s peaks at ca. 285 and 530 eV, respectively; while AGO exhibited an additional N 1s peak at ca. 400 eV, which was from the grafted allylamine (Cheng et al., 2012; Park et al., 2009; Yang et al., 2009). The C 1s core-level spectrum of GO could be curve-fitted into five peak components with the binding energy at ca. 283.9, 284.6, 286.3, 287.3, 288.4 eV (Fig. S3b), which were attributed to the sp2-hybridized carbon, sp3-hybridized carbon, C–O, CQO, and O–CQO species, respectively (Xu et al., 2010; Moulder et al., 1992; Liu et al., 2010). The C 1s core-level spectrum of AGO also could be curve-fitted into five peak components with the binding energy at ca. 284.6, 285.2, 286, 287.2, 288.8 eV (Fig. S3c), which were assigned to the C–C, C–N, C–O, CQO, and N–CQO species, respectively (Cheng et al., 2012; Yang et al., 2009; Moulder et al., 1992). The disappearance of O–CQO peak component and the concomitant appearances of C–N and N–CQO peak components in AGO C 1s spectrum clearly verified the covalent grafting allylamine onto GO surface. In addition, the N 1s peak in AGO could be curve-fitted two peak components at the binding energy of 399.5 and 401.3 eV (Fig. S3d), attributable to N–C and N–CQO peak components of the grafted allylamine, respectively (Cheng

3.2. Fluorescent discriminations The synthesized AGO showed highly stable fluorescence emission in aqueous media. Interestingly, its fluorescence could be selectively quenched by Fe3 þ through ET mechanism. In the selectivity profiles of AGO towards various metal ions, only Fe3 þ induced a significant quenching to AGO fluorescence (Fig. S5). On the contrary, other tested metal ions including Fe2 þ , Al3 þ , Cr3 þ , Cu2 þ , Pb2 þ , Zn2 þ , Mg2 þ , Hg2 þ , Mn2 þ , Ni2 þ , Na þ and K þ , induced relatively insignificant changes under the same conditions. The selectivity to Fe3 þ was further investigated through the competition experiments in the presence of 165 mM Fe3 þ and other metal ions (Fig. S5). It was found that the coexisted metal ions had no obvious interference to Fe3 þ discrimination in aqueous media. Compared to other metal ions, Fe3 þ has stronger electron-accepting ability and can more easily capture electrons from the negative ζ potential surface of AGO through the electrostatic attractions (Bricks et al., 2005; Mei et al., 2010). Therefore, AGO exhibited a high selectivity towards Fe3 þ among other metal ions, indicating a possibility to be as an efficient nanosensor. The fluorescence titration of AGO solution was carried out using Fe3 þ in aqueous media. As shown in Fig. 2, the fluorescence intensity of AGO gradually decreased with the increase of Fe3 þ concentration. About 10% of the fluorescence intensity was quenched when 20 mM Fe3 þ was added into the solution. The most fluorescence was quenched when Fe3 þ concentration was increased up to 500 mM. But, the maximum wavelength and shape of the fluorescence spectra kept unchanged even in the presence of high Fe3 þ concentration, which indicated that Fe3 þ did not

120 100

1.0

0

0.8 0.7

80 60

R = 0.9994

0.9

I/I

3. Results and discussion

et al., 2012; Mei et al., 2010; Compton et al., 2010). These XPS results clearly confirmed the success in covalent grafting allylamine onto GO surface. Under visible light, the color of GO solution changed from light brown to light blue after grafting allylamine (Fig. S4a). Under UV light, GO solution showed no color appearance; while AGO solution exhibited naked-eye-visible navy blue color (Fig. S4a). In their excitation spectra, GO solution only had a band centered at 380 nm, while a new excitation band appeared at 350 nm in AGO solution (Fig. S4b). When excited at 350 nm, AGO solution exhibited strong fluorescence at ca. 445 nm, but there was only very weak fluorescence in GO solution (Fig. S4c). These results indicated the increase in new fluorescence centers on GO surface after grafting allylamine (Li et al., 2008; Luo et al., 2009).

Fluorescence intensity

the fluorescent cells with 100 mL Fe2 þ for 1 h in the presence of ascorbic acid and glutathione. To observe the conversion from Fe2 þ to Fe3 þ , the fluorescent cells were incubated with 100 mL Fe2 þ and 100 mL H2O2 for 1 h. Experiments to H2O2 were performed by incubating the fluorescent cells with 30 mL H2O2 for 1 h. After the incubation and washing with PBS for 3 times, the cell suspensions were used for fluorescent imaging on a LEICA TCS SP5II laser scanning confocal fluorescence microscope.

71

0.6

500

0

20

40

60

80 100 120

[Fe ] (10 M)

40 20 0

350

400

450 500 550 Wavelength (nm)

600

650

Fig. 2. Fluorescence titration of AGO with Fe3 þ (0–500 mM) in aqueous media. Inset: relative fluorescence intensity of AGO (I/I0) as a linear function of Fe3 þ concentration.

72

L. He et al. / Biosensors and Bioelectronics 70 (2015) 69–73

change the structure of AGO. As shown in the inset of Fig. 2, the relative fluorescence intensity (I/I0) was linearly proportional to Fe3 þ concentration in a wide range of 0–120 mM, which was over two times to the reported result from the saturated hydrocarbonmodified GO (Mei et al., 2012). According to this linear relationship, the detection limit to Fe3 þ was calculated as 4.6 mM (3δ/slope). This wide linear range of Fe3 þ detection concentration was most possible due to the covalent grafting allylamine. Fe3 þ can quickly adsorb onto the negative ζ potential surface of AGO through electrostatic attractions. Thus, the electrons in the conductive and defect bands of AGO would transfer to the outer halffilled d orbits of Fe3 þ , which led to the quenching of AGO fluorescence (Fig. 1; Mei et al., 2012). On the other hand, the grafted allylamine was an unsaturated electron-rich group and the π bond in CH2 ¼CH group could form π–π interaction with the big π bond in AGO. This π–π interaction weakened the AGO-Fe3 þ ET process, thus delayed the ET-induced quenching of AGO fluorescence. Therefore, the linearly dependent range of I/I0 on Fe3 þ concentration obviously increased compared to that from saturated hydrocarbon-modified GO (Mei et al., 2012). The kinetic responses of AGO to Fe3 þ , Fe2 þ and H2O2 were further investigated (Fig. S6). AGO solution exhibited a stable fluorescence and its relative intensity kept unchanged during the tested time. Its relative intensity was also unchanged in the presence of 165 mM Fe2 þ or H2O2, respectively. Upon the addition of Fe3 þ , however, its fluorescence was quenched by about 45% in 4 min under the same conditions. Then the quenching reached equilibrium and remained unchanged even further increasing the interaction time. Similarly, when Fe2 þ and H2O2 were added into AGO solution at the same time, the fluorescence was quenched by about 48% in 4 min. Obviously, the Fe2 þ was oxidized to Fe3 þ by H2O2 and induced a similar fluorescence quenching by Fe3 þ . Through the nonlinear fitting analysis, their quenching were consistent with the first order exponential decay curves with the rate constants of 0.457 and 0.503 for the quenching induced by Fe3 þ and Fe2 þ /H2O2, respectively (Fig. S7). 3.3. Molecular logic gate interpretations From the above studies, a three-input logic gate signifying AND and OR signals has been interpreted based on the output changes of AGO fluorescence induced by different chemical inputs. In this combinational logic gate, Fe3 þ , Fe2 þ and H2O2 were employed as three inputs, whose presence and absence were defined as “1” and “0”, respectively. Concerning the outputs, the fluorescence

quenching and the unchanged fluorescence were defined as outputs “1” and “0”, respectively. When Fe2 þ and H2O2 were used as inputs together, it formed an AND logic gate. On the other hand, an OR logic gate formed when Fe3 þ or the output of the abovementioned AND logic gate was used as an input, respectively. Fig. 3a lists the fluorescence responses of AGO logic gate under eight possible inputs. There was no fluorescence quenching (output 0) in the absence of inputs (000) or in the presence of only Fe2 þ (010) or only H2O2 (001), respectively, while other five inputs caused an obvious fluorescence quenching (output 1). Due to the dual quenching by Fe3 þ and Fe2 þ /H2O2, the output in the presence of three inputs (111) was expectedly larger than the outputs under other inputs (110, 101, 011, 100). This combinational logic gate worked well in intracellular fluorescence imaging, as exampled in Figs. 3b–f and S8. After A549 cells were incubated with AGO solution, the center part of the suspended cells showed stable blue fluorescence without decay under the UV lamp illumination for over 1 h (input 000, output 0, Fig. 3b). This indicated that AGO was successfully transfected into the cells. However, the blue fluorescence in the cells obviously weakened after the cells were followed by a subsequent incubation in Fe3 þ medium (input 100, output 1, Fig. 3c). On the other hand, the blue fluorescence showed no obvious changes after incubating the fluorescent cells with Fe2 þ (input 010, output 0, Fig. 3d) or H2O2 (input 001, output 0, Fig. 3e), respectively. As expected, the blue fluorescence significantly weakened after incubating the fluorescent cells with Fe2 þ and H2O2 at the same time (input 011, output 1, Fig. 3f). Obviously, intracellular Fe2 þ transferred to Fe3 þ in the presence of H2O2, resulting in the similar quenching by Fe3 þ (input 100, output 1, Fig. 3b).

4. Conclusions In summary, we synthesized a novel fluorescent nanosensor AGO through the covalent grafting unsaturated allylamine onto GO surface. The surface grafting was fully characterized. The blue fluorescence emission of AGO could be selectively quenched by Fe3 þ with a low limit of detection. And the grafted allylamine provided Fe3 þ a wide linear range of detection concentration. Moreover, AGO can be used to construct a combinational logic gate through intracellular fluorescence imaging. The concept demonstrated in this study provides a promising alternative for the construction of fluorescent sensing platform.

Fig. 3. The outputs (a) and fluorescent cell images (b–f) of AGO logic gate under various inputs. Scale bar¼ 25 mm.

L. He et al. / Biosensors and Bioelectronics 70 (2015) 69–73

Acknowledgements The authors acknowledge the funding of the GRF project (No. PolyU 5316/10E) by the Research Grants Council of the Hong Kong SAR Government, the National Natural Science Foundation of China (No. 21376197), Shenzhen Basic Research Project (JCYJ20130326112522124) and Key Laboratory Project of Shenzhen (No. ZDSY20130329101130496). And we also acknowledge the use of the facilities and engineering support by Dr. Wei Lu at the Hong Kong Polytechnic University Research Facility in Materials Characterization and Device Fabrication (UMF) – Centre for Electron Microscopy.

Appendix A. Supplementary Information Supplementary data associated with this article can be found in the online version at http://dx.doi.org/10.1016/j.bios.2015.01.075.

References Bricks, J.L., Kovalchuk, A., Trieflinger, C., Nofz, M., Büschel, M., Tolmachev, A.I., Daub, J., Rurack, K., 2005. J. Am. Chem. Soc. 127, 13522–13529. Bush, A.I., 2000. Curr. Opin. Chem. Biol. 4, 184–191. Chen, X.Q., Pradhan, T., Wang, F., Kim, J.S., Yoon, J., 2012. Chem. Rev. 112, 1910–1956. Cheng, R.M., Liu, Y., Qu, S.J., Pan, Y.Q., Zhang, S., Chen, H., Dai, L.M., Qu, J., 2012. Anal. Chem. 84, 5641–5644. Choi, Y.W., Park, G.J., Na, Y.J., Jo, H.Y., Lee, S.A., You, G.R., Kim, C., 2014. Sens. Actuators B 194, 343–352. Compton, O.C., Dikin, D.A., Putz, K.W., Brinson, L.C., Nguyen, S.T., 2010. Adv. Mater. 22, 892–896. Crichton, R.R., Dexter, D.T., Ward, R.J., 2008. Coord. Chem. Rev. 252, 1189–1199. Dong, H.F., Gao, W.C., Yan, F., Ji, H.X., Ju, H.X., 2010. Anal. Chem. 82, 5511–5517. Dreyer, D.R., Park, S., Bielawski, C.W., Ruoff, R.S., 2010. Chem. Soc. Rev. 39, 228–240. Galaris, D., Skiada, V., Barbouti, A., 2008. Cancer Lett. 266, 21–29. Geim, A.K., 2009. Science 324, 1530–1534. Geim, A.K., Novoselov, K.S., 2007. Nat. Mater. 6, 183–191. He, Y., Xing, X.J., Tang, H.W., Pang, D.W., 2013. Small 9, 2097–2101. Hu, Z.-Q., Du, M., Zhang, L.-F., Guo, F.-Y., Liu, M.-D., Li, M., 2014. Sens. Actuators B 192, 439–443.

73

Hu, Z.-Q., Feng, Y.-C., Huang, H.-Q., Ding, L., Wang, X.-M., Lin, C.-S., Li, M., Ma, C.P., 2011a. Sens. Actuators B 156, 428–432. Hu, Z.-Q., Wang, X.-M., Feng, Y.-C., Ding, L., Li, M., Lin, C.-S., 2011b. Chem. Commun. 47, 1622–1624. Hummers Jr., W.S., Offeman, R.E., 1958. J. Am. Chem. Soc. 80, 1339. Jeon, E.K., Seo, E., Lee, E., Lee, W., Um, M.-K., Kim, B.-S., 2013. Chem. Commun. 49, 3392–3394. Kim, H., Kim, K.B., Song, E.J., Hwang, I.H., Noh, J.Y., Kim, P.-G., Jeong, K.-D., Kim, C., 2013. Inorg. Chem. Commun. 36, 72–76. Kim, K.S., Zhao, Y., Jang, H., Lee, S.Y., Kim, J.M., Kim, K.S., Ahn, J.-H., Kim, P., Choi, J.-Y., Hong, B.H., 2009. Nature 457, 706–710. Lee, J.-W., Helmann, J.D., 2006. Nature 440, 363–367. Li, D., Müller, M.B., Gilje, S., Kaner, R.B., Wallace, G.G., 2008. Nat. Nanotechnol. 3, 101–105. Li, F., Feng, Y., Zhao, C., Li, P., Tang, B., 2012. Chem. Commun. 48, 127–129. Liu, J.B., Fu, S.H., Yuan, B., Li, Y.L., Deng, Z.X., 2010. J. Am. Chem. Soc. 132, 7279–7281. Lu, C.-H., Yang, H.-H., Zhu, C.-L., Chen, X., Chen, G.-N., 2009. Angew. Chem. Int. Ed. 48, 4785–4787. Luo, Z.T., Lu, Y., Somers, L.A., Johnson, A.T.C., 2009. J. Am. Chem. Soc. 131, 898–899. Mei, Q.S., Guan, G.J., Liu, B.H., Wang, S.H., Zhang, Z.P., 2010. Chem. Commun. 46, 7319–7321. Mei, Q.S., Jiang, C.L., Guan, G.J., Zhang, K., Liu, B.H., Liu, R.Y., Zhang, Z.P., 2012. Chem. Commun. 48, 7468–7470. Moulder, J.F., Stickle, W.F., Sobol, P.E., Bomben, K.D., 1992. Handbook of X-ray Photoelectron Spectroscopy, second ed. Perkin-Elmer Corp., Eden Prairie, Minnesota. Murale, D.P., Singh, A.P., Lavoie, J., Liew, H., Cho, J., Lee, H.-I., Suh, Y.-H., Churchill, D. G., 2013. Sens. Actuators B 185, 755–761. Narayanaswamy, N., Govindaraju, T., 2012. Sens. Actuators B 161, 304–310. Park, S., An, J., Jung, I., Piner, R.D., An, S.J., Li, X., Velamakanni, A., Ruoff, R.S., 2009. Nano Lett. 9, 1593–1597. Qu, K.G., Wang, J.S., Ren, J.S., Qu, X.G., 2013. Chem. Eur. J 19, 7243–7249. Que, E.L., Domaille, D.W., Chang, C.J., 2008. Chem. Rev. 108, 1517–1549. Wang, C.Y., Yang, S., Yi, M., Liu, C.H., Wang, Y.J., Li, J.S., Li, Y.H., Yang, R.H., 2014. ACS Appl. Mater. Interfaces 6, 9768–9775. Wang, J.S., Xu, X.J., Shi, L.L., Li, L.D., 2013. ACS Appl. Mater. Interfaces 5, 3392–3400. Xu, L.Q., Yang, W.Y., Neoh, K.-G., Kang, E.-T., Fu, G.D., 2010. Macromolecules 43, 8336–8339. Yang, D.X., Velamakanni, A., Bozoklu, G., Park, S., Stoller, M., Piner, R.D., Stankovich, S., Jung, I., Field, D.A., Ventrice Jr., C.A., Ruoff, R.S., 2009. Carbon 47, 145–152. Yang, Z., She, M.Y., Yin, B., Cui, J.H., Zhang, Y.Z., Sun, W., Li, J.L., Shi, Z., 2012. J. Org. Chem. 77, 1143–1147. Zhang, M., Le, H.-N., Ye, B.-C., 2013. ACS Appl. Mater. Interfaces 5, 8278–8282. Zhang, M., Yin, B.-C., Tan, W.H., Ye, B.-C., 2011. Biosens. Bioelectron. 26, 3260–3265. Zheng, M., Tan, H.Q., Xie, Z.G., Zhang, L.G., Jing, X.B., Sun, Z.C., 2013. ACS Appl. Mater. Interfaces 5, 1078–1083.