View Article Online / Journal Homepage / Table of Contents for this issue

PCCP
Cite this: Phys. Chem. Chem. Phys., 2011, 13, 2111621120 www.rsc.org/pccp

Dynamic Article Links

PAPER

Competitive surface-enhanced Raman scattering eects in noble metal nanoparticle-decorated graphene sheets
Shengtong Sun and Peiyi Wu*
Downloaded by Indian Institute of Science on 08 December 2012 Published on 21 October 2011 on http://pubs.rsc.org | doi:10.1039/C1CP22727K

Received 8th August 2011, Accepted 26th September 2011 DOI: 10.1039/c1cp22727k Herein, noble metal nanoparticle (Au or Ag NP) decorated graphene sheets, fabricated according to a facile one-pot environmentally friendly method, are used as good substrates for the investigation of the combined surface-enhanced Raman scattering (SERS) eect, where both the electromagnetic mechanism and the chemical mechanism eects coexist among the Au or Ag NPs, graphene sheets and the absorbed analytes. Our results show that, in aqueous solution, the SERS eect of both the Au and Ag NPs on the absorbed probe molecules and on graphene is competitive, which varies dependent on the species and the concentration of the absorbed probe molecule. By a detailed comparison of three probe molecules (rhodamine 6G, nile blue A, and 4-aminobenzenethiol) with dierent coupling abilities to the graphene sheets, we nally attribute this phenomenon to the result of the strong suppressing eect of the macrocyclic probe molecules on the SERS of graphene, induced by charge transfer, as the probe molecules are coupled to the graphene sheets. This competitive eect is a non-ignorable phenomenon when graphene/Au or Ag nanocomposites are used as SERS substrates, and our study may deepen our understanding of the SERS mechanism.

Introduction
Since its inception in the late 1970s, surface-enhanced Raman scattering (SERS) has been proven to be a powerful and reliable analytical tool for the ultra-sensitive detection of analytes even at the single-molecule level.1 Due to the signicant increase in the cross-section of Raman scattering, SERS can enhance the Raman signal by a factor of 106, and even as much as 1014, when an analyte is placed in the vicinity of the surface of a plasmonic nanostructure.2 The origins of the SERS eect is still in dispute, but may be attributed to two possible sources: the electromagnetic mechanism (EM), which is based on surface plasmon resonance (a factor of 106108 enhancement) and the chemical mechanism (CM), which is based on charge transfer (usually a factor of 10100 enhancement).2 Traditionally, SERS substrates are based on the rough surface of a noble metal taking advantage of the EM, mostly Au and Ag, and involves depositing a thin lm using electrochemical or vacuum evaporation methods,3 nanosphere lithography,4 or nanocrystal colloids,5 etc. Recently, graphene has been shown to have a clear CM Raman enhancing eect, due to its unique exible and smooth 2D honeycomb structure with a one-atom thickness.6 Several studies suggest that

The Key Laboratory of Molecular Engineering of Polymers, Ministry of Education, Department of Macromolecular Science, and the Laboratory of Advanced Materials, Fudan University, Shanghai 200433, China. E-mail: peiyiwu@fudan.edu.cn

graphene is an excellent SERS substrate for absorbed species owing to it eciently quenching the interfering excited-state luminescence.6,7 On the other hand, either placing graphene on a Ag lm8 or depositing arrays of Au or Ag nanoparticles (NPs) on graphene9 induce a large enhancement in the Raman signal of graphene. Additionally, combination of Au or Ag and graphene sheets can also be used for the detection of low concentration molecules.10 By decorating Au or Ag NPs on graphene sheets, a coupled EM eect caused by the aggregation of the nanoparticles and strong electronic interactions between Au or Ag NPs and the graphene sheets are considered to be responsible for the signicantly enhanced Raman signal of the analytes.11 However, the acquisition of the enhanced Raman signal of the absorbed molecules on Au or Ag NPdecorated graphene sheets cannot always be as successful as nanocrystal colloids or graphene sheets alone.11b It may be reasonable to suggest that the coexistence of the EM and the CM eects among the Au or Ag NPs, graphene sheets and the absorbed analytes makes the case much more complicated. Herein, noble metal nanoparticle (Au or Ag NP) decorated graphene sheets were fabricated for use as good SERS substrates to distinguish the combined SERS eect when both the EM and the CM eects coexist among the Au or Ag NPs, graphene sheets and absorbed analytes. Our results show that, in aqueous solution, the SERS eect of both the Au and Ag NPs on the absorbed probe molecules and on graphene, as well as that of graphene on the probe molecules, are competitive and vary dependent on the species and the concentration of the
This journal is
c

21116

Phys. Chem. Chem. Phys., 2011, 13, 2111621120

the Owner Societies 2011

View Article Online

by centrifugation until the supernatant was colorless, the resulting graphene/Au or Ag nanocomposites were re-dispersed in water (graphene, 0.1 mg mL1) prior to use. Characterization The UVvis spectra of graphene/Au and graphene/Ag dispersions (graphene, 0.01 mg mL1) were measured on a Hitachi U-2910 spectrophotometer. X-Ray diraction (XRD) patterns were acquired by a D8 ADVANCE and DAVINCI.DESIGN (Bruker) Xpert diractometer with Cu Ka radiation. Highresolution transmission electron microscopy (HRTEM) images were taken with a JEOL JEM 2011 at 200 kV. The samples for Raman and SERS measurements were all prepared by placing the aqueous solutions in capillary tubes and the spectra recorded on a Renishaw inVia Reex micro-Raman spectrometer with He/Ne laser excitation at 632.8 nm. A 200 objective was used to focus the laser beam and to collect the Raman signals. To ensure the obtained spectra were comparable, the settings, including the lasers power and the exposure time, were all the same.

Downloaded by Indian Institute of Science on 08 December 2012 Published on 21 October 2011 on http://pubs.rsc.org | doi:10.1039/C1CP22727K

Results and discussion

Fig. 1 (a) UVvis spectra of graphene/Au and graphene/Ag nanocomposites, reduced by sodium citrate, in their aqueous dispersions. The inset is the photograph of the aqueous dispersions (graphene, 0.01 mg mL1) (left: graphene/Au; right: graphene/Ag). (b) XRD patterns of graphene/Au and graphene/Ag nanocomposites. (c) TEM images of graphene/Au (left) and graphene/Ag (right) nanocomposites. The insets are the corresponding HRTEM images. (d) Size distribution histograms of the graphene/Au and graphene/Ag nanocomposites.

absorbed probe molecules. A proposed mechanism is also presented to account for this phenomenon.

Experimental
Preparation of graphene oxide (GO) GO was prepared by a modied Hummers method.12 Briey, expandible graphite powders (5 g, 8000 mesh, Aladdin Co. Ltd.) and sodium nitrate (2.5 g) were mixed in concentrated H2SO4 (115 mL) at 0 1C, followed by the addition of potassium permanganate (15 g) under vigorous stirring. After increasing the temperature to 35 1C, excess deionized water (300 mL) was added to the mixture before stirring for 30 min and the temperature was then increased to 90 1C for 15 min before adding 30% H2O2 (100 mL). The resulting suspension was ltered, washed with 5% HCl and dialyzed for 7 days to remove the remaining metal species. The GO aqueous dispersion was diluted prior to use. Preparation of graphene/Au or Ag nanocomposites Graphene/Au or Ag nanocomposites were prepared according to a modied method detailed in a previous report.13 In brief, Au or Ag nanoparticle-decorated graphene sheets were prepared via a facile one-pot method. HAuCl4 or AgNO3 (5.0 mM), graphene oxide (0.1 mg mL1) and sodium citrate (10 mg mL1) were mixed with the assistance of sonication for 2 min and then placed in an oil bath at 100 1C for 24 h with stirring. After washing
This journal is
c

Au and Ag NP-decorated graphene sheets (referred to as graphene/Au and graphene/Ag) were prepared via a facile one-pot environmentally friendly method by heating a mixed aqueous solution of HAuCl4 (AgNO3), graphene oxide and sodium citrate at 100 1C for 24 h. After washing several times to remove the residual salts, the resulting graphene/Au and graphene/Ag nanocomposites were well dispersed in water, as shown in Fig. 1a. From the UVvis spectra, we can observe the characteristic absorption peak at 260 nm for the restored pp conjugated structure of the graphene sheets14 due to the reduction of graphene oxide (229 nm) via sodium citrate. Additionally, there is also another characteristic surface plasmon absorption peak for the Au NPs (539 nm) and for the Ag NPs (405 nm),15 which implies the formation of graphene/Au and graphene/Ag nanocomposites. The XRD patterns of graphene/ Au and graphene/Ag (Fig. 1b) both exhibit four obvious peaks corresponding to the (111), (200), (220) and (311) diractions, indicating that the Au and Ag NPs both exist in the crystalline state. TEM and HRTEM images shows that the Au NPs (1030 nm) and Ag NPs (1035 nm) are homogeneously attached to the graphene sheets, which can be used as a good substrate for investigating the combining SERS eect. Due to the complete reduction of graphene oxide and Au or Ag NPs by sodium citrate, the mass ratio of graphene : Au and graphene : Ag in the resulting nanocomposites can be estimated to be 1 : 10 and 1 : 5, respectively. In our study, the laser excitation wavelength was always xed at 632.8 nm as Au and Ag NPs have a rather large Raman signal enhancement relative to graphene at an excitation of 632.8 nm, compared to other excitation wavelengths.9b,c Based on this laser excitation, three commonly used Raman probe molecules were chosen for comparison (Fig. 3). Rhodamine 6G (R6G) is a cationic dye possessing a maximum absorption band in the visible domain around 530 nm (non-resonant excitation);16 nile blue A (NBA) is also a cationic dye with a maximum absorption band around 625 nm (resonant excitation), but with a larger macrocyclic conjugated structure than R6G to form
Phys. Chem. Chem. Phys., 2011, 13, 2111621120 21117

the Owner Societies 2011

View Article Online

Downloaded by Indian Institute of Science on 08 December 2012 Published on 21 October 2011 on http://pubs.rsc.org | doi:10.1039/C1CP22727K

Fig. 2 Raman spectra of graphene, graphene/Au and graphene/Ag nanocomposites in their aqueous dispersions (graphene, 0.1 mg mL1).

pp stacking, which couples to the graphene sheets;17 4-aminobenzenethiol (4-ABT) is well-known to preferentially bind Au and Ag NPs via AuS and AgS bonds and has the weakest conjugated structure among the probe molecules invesitgated.18 We rst examined the SERS eect of graphene by decorating it with Au and Ag NPs, as shown in Fig. 2. An apparent enhancing eect of the Raman signals of graphene can be observed after the decoration with the noble metal nanoparticles, which is in

accordance with previous reports.9 The enhancement factor can be estimated to be about 3.6 for graphene/Au and 13.1 for graphene/Ag, according to the G peak, which mainly arises from the EM eect based on the charge transfer between graphene and the Au or Ag NPs.9b Ag is known to have a larger SERS eect than Au. It is also notable that, after the decoration with the dierent nanoparticles, the D and G peaks shift in opposite directions as compared to those of graphene without any decoration. This can be attributed to the dierent types of doping induced by decoration of graphene with either Au or Ag NPs (Au, p-type doping; Ag, n-type doping).9d As the Raman probe molecules were added to either graphene/ Au or graphene/Ag aqueous dispersions, it is interesting to note that a competitive SERS eect can always be observed, as shown in Fig. 3. For all the three probe molecules, when the concentration is relatively high (1 104 M), a SERS eect of the Au or Ag NPs on the probe molecules can be identied, while the eect on graphene is suppressed. However, when the concentration decreases below 1 105 M, only the SERS spectra of graphene can be detected. It seems strange because, for either the CM of graphene or the EM of the Au or Ag NP colloids individually, the intensities of the enhanced Raman signals of the probe molecules should gradually decrease as the concentration decreases as has been commonly observed.5,6 Thus, we presumed

Fig. 3 Raman spectra of R6G, NBA and 4-ABT in graphene/Au (top) and graphene/Ag (bottom) aqueous dispersions (graphene, 0.1 mg mL1) with dierent concentrations varying from 1 104 to 1 106 M. For comparison, the Raman spectra of the pure components (R6G, 1 102 M aqueous solution; NBA and 4-ABT, as solids) as well as those of the graphene/Au and graphene/Ag nanocomposites without any analytes are also presented.

21118

Phys. Chem. Chem. Phys., 2011, 13, 2111621120

This journal is

the Owner Societies 2011

View Article Online

Downloaded by Indian Institute of Science on 08 December 2012 Published on 21 October 2011 on http://pubs.rsc.org | doi:10.1039/C1CP22727K

Fig. 4 Raman spectra of R6G, NBA and 4-ABT (1 104 M) in graphene aqueous dispersions (0.1 mg mL1).

that the SERS eect of the Au or Ag NPs on the probe molecules and on graphene may be competitive. To comprehend this phenomenon, it is necessary to distinguish all the possible SERS eects in the mixed solution of the probe molecules and the Au or Ag NP-decorated graphene sheets, which is also strongly dependent on the probe molecule species. For R6G under non-resonant excitation, there exists a strong SERS eect of the Au or Ag NPs on the R6G molecules absorbed on them based on the EM, which can be determined by the higher enhancement factor of 168 for graphene/Au and 691 for graphene/Ag according to the peak at 1507 cm1, which corresponds to the aromatic CC stretching vibration.19 However, the graphene sheets only show a very weak CM-based SERS eect on the R6G molecules, as shown in Fig. 4. For NBA under resonant excitation, the uorescence is too strong to get a discernable enhanced Raman signal for the Au or Ag NPs individually. However, the graphene sheets can eciently quench the interfering excited-state luminescence for the absorbed species, as has been previously reported.7 Thus, it is possible for us to obtain a strong enhanced Raman signal of NBA, based on the CM eect of graphene, by coupling via the pp stacking. This can also be conrmed by the strong SERS eect of graphene individually on the NBA molecules (Fig. 4). 4-ABT has no macrocyclic conjugated structure, thus, the coupling between graphene and 4-ABT is considered to be very weak, and the CM-based SERS eect of graphene on 4-ABT can be ignored, as shown in Fig. 4. However, 4-ABT can preferentially bind to Au and Ag NPs via AuS and AgS bonds. Therefore, for 4-ABT, the EM-based SERS eect of the Au or Ag NPs predominates. For all these three cases, the SERS eect of the Au or Ag NPs on the graphene sheets always

Fig. 5 Schematic illustration of the possible enhanced Raman signals under laser excitation (632.8 nm) in the mixed solutions of the probe molecules (R6G, NBA and 4-ABT) and Au or Ag NP-decorated graphene sheets. The green balls represent the Au or Ag nanoparticles and the stars represent the probe molecules.

Table 1 A summary of the SERS eects existing in the mixed solutions of the probe molecules (R6G, NBA and 4-ABT) and the Au or Ag NP-decorated graphene sheets SERS of Au or Ag NPs on graphene (EM) R6G NBA 4-ABT | | | SERS of Au or Ag NPs on the probe molecules (EM) | | SERS of graphene on the probe molecules (CM) | (weak) |

exists. Thus, we can summarize all the competitive SERS eects for these three probe molecules, respectively, as shown in Table 1. A schematic illustration of all the possible enhanced Raman signals existing in these three systems under Raman laser excitation has also been presented in Fig. 5. Now that we have determined all the possible enhanced Raman signals that exist, the mechanism of the competitive SERS eect can be investigated. It is notable that NBA has the strongest suppressing eect on the SERS of graphene, wherein the broad D and G bands of graphene cannot even be detected at a concentration of 1 104 M (Fig. 3). 4-ABT has the weakest suppressing eect on the SERS of graphene, while R6G has a moderate eect. Interestingly, this order is totally consistent with the coupling ability of the probe molecules. It is well accepted that the electronic properties of graphene are highly sensitive towards the attachment of molecules.20 Thus, it is presumed that the charge transfer induced by the coupling between graphene and the probe molecules accounts for this competitive SERS eect in the Au or Ag NP-decorated graphene sheets. When the concentration is high, large amounts of the NBA molecules are absorbed on the graphene sheets, and the
Phys. Chem. Chem. Phys., 2011, 13, 2111621120 21119

This journal is

the Owner Societies 2011

View Article Online

Downloaded by Indian Institute of Science on 08 December 2012 Published on 21 October 2011 on http://pubs.rsc.org | doi:10.1039/C1CP22727K

charge transfer between them strongly suppresses the SERS of graphene. Due to the relatively weak coupling ability of the R6G molecules to the graphene sheets, the SERS of graphene is only slightly suppressed. In addition, because the 4-ABT molecules cannot be absorbed on the graphene sheets, but only bind to Au or Ag NPs, the whole Raman spectrum can be considered to be the simple superimposition of the spectrum of 4-ABT and that of graphene enhanced by the Au or Ag NPs individually. When the concentration is low and there are not enough molecules to couple to the graphene sheets, the enhanced Raman spectra of graphene by Au or Ag NPs emerges. Due to its stronger coupling ability, NBA can be detected at a much lower concentration (1 105 M) than R6G, as shown in Fig. 3. Nevertheless, the precise mechanism of the strong suppressing eect of the macrocyclic probe molecules on graphene in an aqueous solution is still unclear, and may need further investigation. Finally, it is also notable that Si substrates are commonly used to support Au or Ag NPs for SERS measurements in the solid state, according to previous reports.6,10 However, graphene sheets tend to aggregate layer-by-layer when they are evaporated from the substrates, which is unfavorable for the highly sensitive and reproducible SERS measurements.6e In performing the SERS measurements of graphene in an aqueous solution, it is possible to overcome these problems because the graphene sheets are nearly all monolayers when they are dispersed in water. The measurements can be carried out in vitro, which can be particularly useful for biological applications. Whereas, as we reported in this paper, due to the strong solvation of the graphene sheets and the probe molecules in water, the SERS enhancement is relatively weak and almost not detected when the concentration of the probe molecules is low, which needs to be paid careful attention to by researchers.

Notes and references

1 (a) G. Haran, Acc. Chem. Res., 2010, 43, 11351143; (b) Z. Q. Tian, B. Ren and D. Y. Wu, J. Phys. Chem. B, 2002, 106, 94639483. 2 (a) J. R. Lombardi and R. L. Birke, Acc. Chem. Res., 2009, 42, 734742; (b) S. Schlucker, ChemPhysChem, 2009, 10, 13441354; (c) X. M. Qian and S. Nie, Chem. Soc. Rev., 2008, 37, 912920. 3 (a) M. Fleischm, P. Hendra and A. McQuilla, Chem. Phys. Lett., 1974, 26, 163166; (b) C. E. Taylor, J. E. Pemberton, G. G. Goodman and M. H. Schoensch, Appl. Spectrosc., 1999, 53, 12121221. 4 T. R. Jensen, M. D. Malinsky, C. L. Haynes and R. P. Van Duyne, J. Phys. Chem. B, 2000, 104, 1054910556. 5 (a) R. A. Alvarez-Puebla, R. Contreras-Caceres, I. Pastoriza-Santos, J. Perez-Juste and L. M. Liz-Marzan, Angew. Chem., Int. Ed., 2009, 48, 138143; (b) S. S. R. Dasary, A. K. Singh, D. Senapati, H. T. Yu and P. C. Ray, J. Am. Chem. Soc., 2009, 131, 1380613812. 6 (a) X. Ling, L. Xie, Y. Fang, H. Xu, H. Zhang, J. Kong, M. S. Dresselhaus, J. Zhang and Z. Liu, Nano Lett., 2010, 10, 553561; (b) N. Jung, A. C. Crowther, N. Kim, P. Kim and L. Brus, ACS Nano, 2010, 4, 70057013; (c) X. Ling and J. Zhang, Small, 2010, 6, 20202025; (d) X. Yu, H. Cai, W. Zhang, X. Li, N. Pan, Y. Luo, X. Wang and J. G. Hou, ACS Nano, 2011, 5, 952958; (e) C. Qiu, H. Zhou, H. Yang, M. Chen, Y. Guo and L. Sun, J. Phys. Chem. C, 2011, 115, 1001910025. 7 L. Xie, X. Ling, Y. Fang, J. Zhang and Z. Liu, J. Am. Chem. Soc., 2009, 131, 98909891. 8 (a) L. Gao, W. Ren, B. Liu, R. Saito, Z. S. Wu, S. Li, C. Jiang, F. Li and H. M. Cheng, ACS Nano, 2009, 3, 933939; (b) N. Kim, M. K. Oh, S. Park, S. K. Kim and B. H. Hong, Bull. Korean Chem. Soc., 2010, 31, 9991003; (c) Y. Ouyang and L. Chen, J. Mol. Struct., 2011, 992, 4851. 9 (a) C. Xu and X. Wang, Small, 2009, 5, 22122217; (b) F. Schedin, E. Lidorikis, A. Lombardo, V. G. Kravets, A. K. Geim, A. N. Grigorenko, K. S. Novoselov and A. C. Ferrari, ACS Nano, 2010, 4, 56175626; (c) X. Q. Fu, F. L. Bei, X. Wang, S. OBrien and J. R. Lombardi, Nanoscale, 2010, 2, 14611466; (d) J. Lee, K. S. Novoselov and H. S. Shin, ACS Nano, 2011, 5, 608612; (e) J. Lee, S. Shim, B. Kim and H. S. Shin, Chem. Eur. J., 2011, 17, 23812387. 10 (a) Y. Wang, Z. Ni, H. Hu, Y. Hao, C. P. Wong, T. Yu, J. T. L. Thong and Z. X. Shen, Appl. Phys. Lett., 2010, 97, 163111163113; (b) G. Lu, H. Li, C. Liusman, Z. Yin, S. Wu and H. Zhang, Chem. Sci., 2011, 2, 18171821. 11 (a) G. Goncalves, P. A. A. P. Marques, C. M. Granadeiro, H. I. S. Nogueira, M. Singh and J. Gracio, Chem. Mater., 2009, 21, 47964802; (b) J. Huang, L. M. Zhang, B. A. Chen, N. Ji, F. H. Chen, Y. Zhang and Z. J. Zhang, Nanoscale, 2010, 2, 27332738; (c) Z. Zhang, F. G. Xu, W. S. Yang, M. Y. Guo, X. D. Wang, B. L. Zhanga and J. L. Tang, Chem. Commun., 2011, 47, 64406442. 12 W. S. Hummers and R. E. Oeman, J. Am. Chem. Soc., 1958, 80, 13391339. 13 Z. Zhang, H. H. Chen, C. Y. Xing, M. Y. Guo, F. G. Xu, X. D. Wang, H. Gruber, B. L. Zhang and J. L. Tang, Nano Res., 2011, 4, 599611. 14 D. Li, M. B. Muller, S. Gilje, R. B. Kaner and G. G. Wallace, Nat. Nanotechnol., 2008, 3, 101105. 15 (a) J. Q. Hu, Q. Chen, Z. X. Xie, G. B. Han, R. H. Wang, B. Ren, Y. Zhang, Z. L. Yang and Z. Q. Tian, Adv. Funct. Mater., 2004, 14, 183189; (b) B. K. Pong, H. I. Elim, J. X. Chong, W. Ji, B. L. Trout and J. Y. Lee, J. Phys. Chem. C, 2007, 111, 62816287. 16 J. Guthmuller and B. Champagne, J. Phys. Chem. A, 2008, 112, 32153223. 17 H. Yamada, H. Nagata and K. Kishibe, J. Phys. Chem., 1986, 90, 818823. 18 K. Kim and J. K. Yoon, J. Phys. Chem. B, 2005, 109, 2073120736. 19 J. Zhang, X. Li, X. Sun and Y. Li, J. Phys. Chem. B, 2005, 109, 1254412548. 20 S. K. Saha, R. C. Chandrakanth, H. R. Krishnamurthy and U. V. Waghmare, Phys. Rev. B: Condens. Matter Mater. Phys., 2009, 80, 155414.

Conclusions
In this paper, we attempted to determine the combined SERS eect that exists in noble metal (Au and Ag) nanoparticledecorated graphene sheets in aqueous solution, and a competitive SERS eect between the absorbed probe molecules and the graphene sheets can be identied, which varies dependending on the species and concentration of the absorbed probe molecules. By a detailed comparison between the three probe molecules (R6G, NBA, and 4-ABT) with dierent coupling abilities to the graphene sheets, we can nally attribute this phenomenon to the strong suppressing eect of the macrocyclic probe molecules on the SERS of graphene, which is induced by charge transfer as the probe molecules are coupled to the graphene sheets. This competitive eect is a non-ignorable phenomenon when graphene/ Au or Ag nanocomposites are used as SERS substrates, and our study may deepen our understanding of the SERS mechanism.

Acknowledgements
This work was nancially supported by the National Science Foundation of China (NSFC) (No. 20934002, 51073043) and the National Basic Research Program of China (No. 2009CB930000).
21120 Phys. Chem. Chem. Phys., 2011, 13, 2111621120

This journal is

the Owner Societies 2011