One-pot in situ photochemical synthesis of graphene oxide/gold nanorod nanocomposites for surface-enhanced Raman spectroscopy

Abstract

Nanoscale engineered plasmonic materials that can efficiently sustain surface enhanced Raman scattering (SERS) have been strongly pursued for high sensitivity molecular detection. In this work, we report the production of gold nanorod/graphene oxide (GNR/GO) nanocomposites by a simple, one-pot process whereby GNR are formed directly onto the GO flakes in solution by UV light irradiation. The proposed method is easily scalable and results in GNRs with low dispersion in size and aspect ratio of ∼3. The GNR/GO hybrids were deposited in glass by filtration yielding homogenous films that were systematically tested as SERS active substrates. Raman spectroscopy mapping revealed that the tested substrates present spatially homogenous and reproducible SERS responses. Analysis of Raman spectroscopy by using a model molecule (cresyl violet perchlorate) indicates that the produced substrates can provide very large SERS enhancement factors (∼10⁶) and very low molecular detection limits (10⁻¹¹ M).

---

1. Introduction

Plasmonic nanostructures have been a topic of intense research interest in recent years, mainly because of their excellent optical properties and many applications. This is due to the phenomenon of localized surface plasmon resonances (LSPRs) of noble metal nanoparticles which is associated with oscillations of electrons on the metal surfaces, inducing a resonant optical absorption. This SPR plasmon resonance phenomenon makes these metallic nanostructures behave as nanoantennas, amplifying the electromagnetic fields around the surface. This behavior leads to an important effect called surface-enhanced Raman spectroscopy (SERS) discovered by Richard Van Duyne in 1976.¹ This effect has been extensively studied since it opens the possibility of studying molecules at very low concentrations, which is interesting for applications such as biosensing.² The SERS effect originates from two principal mechanisms: the first is the charge exchange between the substrate and the molecule under study, and the other is the intense electromagnetic field originated by plasmon resonance around the metallic nanostructures.^3,4 The second one is most relevant in the gaps between two nanoparticles, called hot-spots, where the electric field is more intense and the highest enhancement is obtained.⁵ Homogeneous and reproducible fabrication of SERS active substrates is being widely studied.^6,7 Many substrates have been reported with different preparation methods^8–10 and metal nanostructures of several morphologies such as nanospheres,¹¹ nanorods,¹² nanostars¹³ and nanostarfruit¹⁴ have been used.

Gold nanorods^15,16 are excellent substrates for SERS due to strong electric field enhancements at their ends caused by their high curvature and because the easy tunability of the surface plasmon resonance in a wide spectral range in order to match the frequency of the excitation lasers. Many synthetic methods have been reported in recent years to obtain gold nanorods with high monodispersity, stability and homogeneity.¹⁷ The seed-mediated method has been the most traditional since it was discovered by El-Sayed et al.¹⁸ and Murphy et al.¹⁹ The synthesis of gold nanorods mediated by ultraviolet irradiation, described by Yang et al.,²⁰ has the advantage of being faster and more easily scalable. Graphene oxide (GO)²¹ and reduced graphene oxide (rGO)²² are normally obtained by acid/base treatments. GO has abundant surface functional groups such as hydroxyl, carbonyl, and carboxyl, which are responsible for the high solubility in various solvents. This solubility makes this material a great candidate for the formation of hybrids with metallic nanostructures as recently reported by Dong et al. (2013)²³ and Yin et al. (2013).²⁴ The formation of hybrids between graphene oxide and gold nanorods have been reported by several authors, and the employed methods usually make use of linkers to promote electrostatic interaction of GO with gold nanorods.^25–28 Chaofan Hu et al.²⁵ have report the fabrication of graphene oxide/gold nanorod hybrid materials by electrostatic self-assembly and their application in SERS. In that method, gold nanorods were synthesized separately and interacted with graphene oxide, through polymers. Recently, Jayabal et al.²⁷ reported a method for the preparation of reduced graphene oxide/gold nanorods embedded in an amine functionalized silicate sol–gel matrix. Dembereldorj et al.²⁸ have formed hybrids between gold nanorods and PEGylated-GO for photothermal cancer therapy. Other authors have used similar methods to form hybrids between gold nanorods and graphene oxide.^29–34

In this article we report a simple one-pot in situ photochemical formation process of graphene oxide and gold nanorods hybrids by UV light irradiation. This method has the advantage of being fast, simple and performed in just one step. Furthermore, the GO/GNR hybrids were used to produce highly sensitive SERS substrate.

2. Experimental section

2.1 Chemicals and reagents

Tetrachloroauric acid (HAuCl₄·3H₂O), silver nitrate (AgNO₃), ascorbic acid, cetyltrimethylammonium bromide (CTAB), acetone, cyclohexane, NaNO₃ (99% purity), H₂SO₄ (95–98%), KMnO₄ (99%), H₂O₂, and manganese oxide (Mn₂O₇) were purchased from Sigma Aldrich. Graphite powder 99% purity was supplied by National de Graphite Ltda, Brazil. All solvents were obtained from commercial suppliers and used without further purification. Milli-Q water with a resistivity of 18.2 MΩ cm was used. GO was synthesized using the modified Hummers method.³⁵

2.2 Synthesis of graphene oxide/gold nanorods hybrid

First a 15 mL solution of CTAB (0.1 M) was prepared and 100 μL of HAuCl₄ (0.1 M) were added. The solution was stirred for 15 minutes. Next, in sequence, 1 mL (1 mM) of silver nitrate, 400 μL cyclohexane, 330 μL of acetone, and 100 μL (0.1 M) of ascorbic acid were added. The solution changed from yellow to transparent, showing the reduction of Au⁺³ to Au⁺¹. The dispersion of GO (0.01 mg ml⁻¹) in water was assisted by sonication. Finally 5 ml of GO was added to the previous solution and the dispersion placed under UV-irradiation (256 nm, 30 W) for 25 minutes in a quartz tube. The final colloidal solution stayed stable for several months.

2.3 Characterization techniques

The optical characterization was carried out by UV-VIS-NIR using a Shimadzu UV 3600 spectrophotometer with 10 mm path length quartz cuvettes. TEM measurements were performed on a TEM TECNAI G2-20, using an accelerating voltage of 200 kV. Samples were prepared by drop-casting from the dispersion onto a TEM grid (200 mesh, holey carbon). Raman experiments were conducted using a Dilor XY triple spectrometer with different laser lines.

2.4 Substrate and SERS measurements

In order to prepare SERS substrates, excess CTAB was removed from the GNRs/GO solution through three rounds of centrifugation (5600 g for 10 min) and re-dispersion in deionized water (18.2 MΩ). After which the solution of hybrid material was filtered and washed with isopropanol several times using a 0.22 mm Millipore cellulose filter. The material collected in the filter was transferred to a glass substrate. The cellulose membrane/glass substrate was vacuum dried and then immersed in an acetone bath to remove the cellulose membrane. As a result, a thin film of the hybrid GNR/GO material was formed.

Cresyl violet perchlorate (CV) dye was used as model molecule for SERS measurements. CV was diluted in pure ethanol at different concentrations. The CV solution was either dropped on to the GNR/GO substrate film (for standard SERS measurement) or the substrate was dipped into this solution for 20 seconds and dried at room temperature (for SERS mapping). The surface enhanced Raman spectra were recorded with the 647 nm wavelength excitation with one 10 seconds accumulation. SERS measurements were also performed with wavelength lasers 488 nm, 514 nm and 568 nm. The laser power on the sample was kept low at 1 mW, focused in by an objective 10×. The reference Raman spectrum measured in a standard glass substrate was performed with 647 nm excitation wavelength with 60 seconds accumulation and focused in by an objective 100×. A SERS Mapping was performed on Raman Microscope alpha 30R (Witec), with 633 nm excitation wavelengths with 0.3 seconds accumulation. The laser power on the sample was kept low at 1 mW focused in by an objective 10×.

3. Results and discussion

The synthesis process of GNRs/GO hybrids was carried out by photochemical reduction of Au ions in presence of GO, under irradiation of ultraviolet light. The photochemical reactions promote the growth of gold nanorods onto graphene oxide sheets in situ. However, a specific environment has to be setup by a solution containing CTAB as surfactant, HAuCl₄ as precursor of gold, silver nitrate which acts in the induction of anisotropic growth of gold nanorods, similar to the one demonstrated by Placido et al.³⁶ Cyclohexane promotes the formation of micelle templates that assist the anisotropic growth, ascorbic acid promotes the reduction of Au⁺³ to Au⁺¹, a first reduction of the gold precursor, and acetone generates ketyl radicals that promote the reduction of gold Au⁺¹ to Au⁰, being considered as the radical initiator.^37,38 Silver nitrate and acetone are critical for the photochemical synthesis of GNR/GO hybrids. There is no growth of gold nanorods without their presence and, without irradiation by ultraviolet light, there is no formation of any type of nanoparticles. After this process, a solution of GNRs/GO hybrid material in CTAB is obtained. The transmission electron microscopy (TEM) images (Fig. 1 and S1 in ESI†) show the GNRs/GO hybrid material obtained by photochemical synthesis. In Fig. 1a–c, the gold nanorods were supported in graphene oxide sheets, as schematically shown in Fig. 1e. Fig. 1a–c shows clearly a large number of gold nanorods efficiently dispersed on graphene oxide sheets. The dimensions and aspect ratio of the gold nanorods were evaluated by statistical analysis, showing an average aspect ratio of 3 (30 nm/10 nm), as shown in Fig. 1d.

Fig. 1 (a–c) TEM images of hybrid GNR/GO at different magnifications. (d) Aspect ratio statistical distribution (e) schematic representation of the hybrid GNR/GO. (Note: TEM images were made using holey carbon microscopy grid).

The plasmon resonance of the GNRs/GO in aqueous solution was measured by optical absorption. Fig. 2 shows the optical absorption spectra of GO (red line), GNRs (blue line) and GNRs/GO (black line). The GNRs were synthesized by the same process described previously but without the presence of GO. The spectrum presents two absorption bands centered at 514 and 750 nm corresponding to the transversal and longitudinal plasmon modes respectively. When the synthesis was carried out in the presence of GO, an increase of the transverse peak was observed. The longitudinal resonance peak is shifted to 740 nm when the nanorods were allocated onto GO. The longitudinal plasmon damping and transverse peak increase are related to the strong interaction of the lateral surface of the gold nanorods with graphene oxide, the change of the local dielectric environment surrounding the surface of the nanorods, and mainly to an increase in polydispersity when the synthesis is performed in the presence of GO.

Fig. 2 Optical absorption spectra of GO (red line), pure gold nanorods (blue line) and hybrid of gold nanorods/graphene oxide (black line).

Considering using the GNR/GO hybrids as SERS substrate, the hybrid material was deposited onto a glass substrate as a thin film from solution. Fig. S2† presents optical and scanning electron microscopy images demonstrating that the obtained films are homogeneous in both macroscopic and microscopic scales. The advantage of this substrate is that it is simple to manufacture and can be applied to large surface areas. The optical absorption spectrum of the GNR/GO film is shown in Fig. 3 (black line), the spectrum of pure GO film is also shown for comparison (red line). The characteristic transverse and longitudinal peak can be observed in the GNR/GO film spectrum, however the longitudinal peak is strongly broadened and shifted in comparison with the peak in the GNR/GO in solution (Fig. 2). Such effect is expected and has been observed in spectra of pure GNR films.³⁹ It is attributed to the stronger coupling between plasmons in neighbor nanorods as the density of GNR is much larger in the films than in solution. It can also be due to the interaction of the GNR and the glass substrate.

Fig. 3 Absorption spectra of the GNR/GO film (black line) and pure GO film (red line).

SERS measurements

The dye cresyl violet perchlorate (CV)⁴⁰ was used for evaluation of the film as an active SERS substrate. The CV was deposited by drop casting onto the GNR/GO thin films. SERS measurements were carried out at various wavelengths to evaluate the best laser excitation resonance conditions of the GNR/GO film. Fig. S3† shows SERS measurements with a laser wavelengths of 488 nm, 514 nm, 568 nm and 647 nm. The 488 nm laser was out of resonance of the gold nanorods and a very weak Raman signal of CV was observed, clearly displaying the D (1330 cm⁻¹) and G (1600 cm⁻¹) Raman bands of the GO. The 514 nm laser was close to the resonance of the transverse plasmon mode of the gold nanorods and a significant increase in signal was observed. The use of the 568 nm and 647 nm lasers results in the largest increases in the Raman signal since they approach the longitudinal plasmon resonance of gold nanorods in the GNR/GO film. The 647 nm laser was chosen for SERS studies in this work because it is closest to the resonant plasmon for thin films of GNR/GO as shown in Fig. 3.

The CV was deposited by drop casting on two substrates; on the standard glass and onto GNR/GO film. In both substrates three concentrations were used (10⁻⁵, 10⁻⁸ and 10⁻¹¹ M). In the glass substrate, a very weak Raman signal was detected but only when higher concentrations with long acquisition times (60 seconds) were used (inset in Fig. 4). On the other hand, the SERS effect appears clearly on the GNR/GO substrate showing an excellent signal at all concentrations and a monotonically drop with the decreasing concentration (Fig. 4). The black line in the Fig. 4 shows the Raman spectrum of CV on glass without any SERS effect.

Fig. 4 SERS spectra of CV at different concentrations.

SERS mapping

SERS measurements of CV on the GNR/GO thin film substrates showed a high increase of Raman signal. However, the quality and uniformity of the substrates are fundamental for the development of commercial SERS substrates. In order to evaluate the GNR/GO substrate uniformity, SERS mapping was performed. The substrate was dipped for about 20 seconds in a solution of 10⁻⁷ M of CV. Then Raman mapping measurements were made using the 591 cm⁻¹ peak as a reference. Fig. 5a shows an optical image of GNR/GO film as substrate and the Fig. 5b shows the SERS mapping. This figure shows the variation of Raman signal depending on the position on the substrate; the yellow range is the intensity of the CV SERS Raman signal peak at 591 cm⁻¹. The variation of intensity is mainly due the random distribution of CV molecules on the substrate and by variation in the substrate surface topography, since the depth of focus is small. Fig. S4† shows the same mapping SERS measurements at two different focal distances.

Fig. 5 SERS mapping of the GNR/GO hybrid film. (A) Optical image of GNR/GO film as substrate (B) SERS mapping.

Substrate enhancement factors (SEFs) were calculated according to the methodology proposed by Le Ruet et al.⁴¹ We assumed that all CV molecules were uniformly adsorbed within the test samples, due to their dilute concentration (10⁻⁷ M). The results of our calculations indicate an enhancement factor of the order of 10⁶.

GNR/GO film features high SERS enhancement factors with detection limits near picomolar concentrations for the aromatic dye used. Other authors have reported SERS studies using hybrids between GO and metallic nanoparticles.⁴² However, gold nanorods are better for SERS than spherical nanoparticles as demonstrated by El-Sayed et al.⁴³ due to their high curvature and well-developed tunability of the surface plasmon resonance. Spherical nanoparticles only have a peak of plasmon resonance (∼520 nm). On the other hand, gold nanorods have two plasmon modes, one transverse (∼520 nm) and another longitudinal (∼640–1000 nm), and both of them are active, though the longitudinal mode is the most intense, as shown in Fig. 3 and S3.† Other studies published in the literature have achieved enhancement of about 10⁴–10⁵ and detection limits of nanomolar concentrations.^25,42,44,45 The detection limits and enhancement factors obtained in our experiment are comparable or higher to the ones recently reported in the literature.

4. Conclusions

We have developed a simple method for the synthesis of nanocomposite between graphene oxide and gold nanorods. The process was carried out through the photochemical synthesis of gold nanorods by irradiation of ultraviolet light and their growth on graphene oxide. The synthesized material shows excellent distribution of nanorods in aqueous solution as characterized by images of transmission electron microscopy (TEM). A thin film was formed with the hybrids on a glass substrate, showing high detection sensitivity, easy preparation and high enhancement in the Raman signal (magnification factor of 10⁶), while SERS mapping showed great uniformity.

Acknowledgements

This work was supported by CAPES, Fapemig, CNPq, and INCT/Nanomateriais de Carbono. The authors acknowledge MackGraphe for experiments involving SERS Mapping. The authors would like to acknowledge the Center of Microscopy at the Universidade Federal de Minas Gerais (http://www.microscopia.ufmg.br) for providing the equipment and technical support for experiments involving electron microscopy.

References

1. D. L. Jeanmaire and R. P. Van Duyne, J. Electroanal. Chem., 1977, 84, 1.
2. Y. Wang, B. Yan and L. Chen, Chem. Rev., 2013, 113(3), 1391.
3. N. Valley, N. Greeneltch, R. P. Van Duyne and G. C. Schatz, J. Phys. Chem. Lett., 2013, 4, 2599.
4. S. K. Saikin, Y. Chu, D. Rappoport, K. B. Crozier and A. Aspuru-Guzik, J. Phys. Chem. Lett., 2010, 1(18), 2740.
5. S. L. Kleinman, R. R. Frontiera, A. I. Henry, J. A. Dieringera and R. P. Van Duyne, Phys. Chem. Chem. Phys., 2013, 15, 21.
6. J. F. Betz, W. W. Yu, Y. Cheng, M. White and G. W. Rubloff, Phys. Chem. Chem. Phys., 2014, 16, 2224.
7. B. Sharma, M. F. Cardinal, S. L. Kleinman, N. G. Greeneltch, R. R. Frontiera, M. G. Blaber, G. C. Schatz and R. P. Van Duyne, MRS Bull., 2013, 38, 615.
8. W. J. Liao, P. K. Roya and S. Chattopadhyay, RSC Adv., 2014, 4, 40487.
9. Q. Jiwei, L. Yudong, Y. Ming, W. Qiang, C. Zongqiang, W. Wudeng, L. Wenqiang, Y. Xuanyi, X. Jingjun and S. Qian, Nanoscale Res. Lett., 2013, 8, 437.
10. W. Hasi, S. Lin, X. Lin, X. Lou, F. Yang, D. Lina and Z. Lu, Anal. Methods, 2014, 6, 9547.
11. R. W. Taylor, T. C. Lee, O. A. Scherman, R. Esteban, J. Aizpurua, F. M. Huang, J. J. Baumberg and S. Mahajan, ACS Nano, 2011, 5, 3878.
12. S. T. Sivapalan, B. M. De Vetter, T. K. Yang, T. van Dijk, M. V. Schulmerich, P. S. Carney, R. Bhargava and C. J. Murphy, ACS Nano, 2013, 7(3), 2099.
13. Q. Su, X. Ma, J. Dong, C. Jiang and W. Qian, ACS Appl. Mater. Interfaces, 2011, 3(6), 1873.
14. L. Vigderman and E. R. Zubarev, Langmuir, 2012, 28(24), 9034.
15. H. Chen, L. Shao, Q. Lia and J. Wang, Chem. Soc. Rev., 2013, 42, 2679.
16. X. Huang, S. Neretina and M. A. El-Sayed, Adv. Mater., 2009, 21, 4880.
17. X. Ye, L. Jin, H. Caglayan, J. Chen, G. Xing, C. Zheng, V. DoanNguyen, Y. Kang, N. Engheta, C. R. Kagan and C. B. Murray, ACS Nano, 2012, 6(3), 2804.
18. B. Nikoobakht and M. A. El-Sayed, Chem. Mater., 2003, 15(10), 1957.
19. N. R. Jana, L. Gearheart and C. J. Murphy, Adv. Mater., 2001, 13, 1389.
20. F. Kim, J. H. Song and P. Yang, J. Am. Chem. Soc., 2002, 124, 14316.
21. D. R. Dreyer, S. Park, C. W. Bielawski and R. S. Ruoff, Chem. Soc. Rev., 2010, 39, 228.
22. S. Pei and H. Cheng, Carbon, 2012, 50, 3210.
23. C. Zhua and S. Dong, Nanoscale, 2013, 5, 10765.
24. P. T. Yin, T. Kim, J. W. Choic and K. B. Lee, Phys. Chem. Chem. Phys., 2013, 15, 12785.
25. C. Hu, J. Rong, J. Cui, Y. Yang, L. Yang, Y. Wang and Y. Liu, Carbon, 2013, 51, 255.
26. Y. K. Kim, H. K. Na, Y. W. Lee, H. Jang, S. W. Hanb and D. H. Min, Chem. Commun., 2010, 46, 3185.
27. S. Jayabal, P. Viswanathana and R. Ramaraj, RSC Adv., 2014, 4, 33541.
28. U. Dembereldorj, S. Y. Choi, E. O. Ganbold, N. W. Song, D. Kim, J. Choo, S. Y. Lee, S. Kim and S. W. Joo, Photochem. Photobiol., 2014, 90, 659.
29. K. Yun, S. Seo, B. Kim, A. Joe, H. Han, J. Kim and E. Jang, Bull. Korean Chem. Soc., 2013, 34, 2795.
30. C. Xu, D. Yang, L. Mei, Q. Li, H. Zhu and T. Wang, ACS Appl. Mater. Interfaces, 2013, 5, 12911.
31. L. Li, H. Lu and L. Deng, Talanta, 2013, 113, 1.
32. W. Bai, H. Huanga, Y. Li, H. Zhang, B. Lianga, R. Guo, L. Dua and Z. Zhanga, Electrochim. Acta, 2014, 117, 322.
33. C. Xue, C. Kung, M. Gao, C. Liu, L. Dai, A. Urbas and Q. Li, Sensing and Bio-Sensing Research, 2015, 3, 7–11.
34. W. Zhu, W. Liu, T. Li, X. Yue, T. Liu, W. Zhang, S. Yu, D. Zhang and J. Wang, Electrochim. Acta, 2014, 146, 419–428.
35. W. S. Hummers Jr and R. E. Offeman, J. Am. Chem. Soc., 1958, 80(6), 1339.
36. T. Placido, R. Comparelli, F. Giannici, P. D. Cozzoli, G. Capitani, M. Striccoli, A. Agostiano and M. L. Curri, Chem. Mater., 2009, 21, 4192.
37. O. R. Miranda and T. S. Ahmadi, J. Phys. Chem. B, 2005, 109(33), 15724.
38. K. Nishioka, Y. Niidome and S. Yamada, Langmuir, 2007, 23(20), 10353.
39. S. Yun, M. Kyung Oh, S. Kyu Kim and S. Park, J. Phys. Chem. C, 2009, 113(31), 13551.
40. E. Vogel, A. Gbureck and W. Kiefer, J. Mol. Struct., 2000, 550–551, 177.
41. E. C. Le Ru, E. Blackie, M. Meyer and P. G. Etchegoin, J. Phys. Chem. C, 2007, 111(37), 13794–13803.
42. J. Huang, L. Zhang, B. Chen, N. Ji, F. Chen, Y. Zhanga and Z. Zhang, Nanoscale, 2010, 2, 2733–2738.
43. B. Nikoobakht, J. P. Wang and M. A. El-Sayed, Chem. Phys. Lett., 2002, 366(1), 17.
44. T. H. D. Nguyen, Z. Zhang, A. Mustapha, H. Li, M. Lin and J. Agric, Food Chem., 2014, 62(43), 10445–10451.
45. A. Saha, S. Palmal and N. R. Jana, Nanoscale, 2012, 4, 6649.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra17207h

---