Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Ag nanoparticle ink coupled with graphene oxide cellulose paper: a flexible and tunable SERS sensing platform

Open Access Open Access

Abstract

Surface-enhanced Raman scattering (SERS) is highly promising for ultra-sensitive detection in a series of applications. Although extensive advances have been achieved in SERS technologies, the preparation of highly efficient SERS substrates still suffers from several limitations, including complex preparation procedures, high cost, and instability for long time storage. To address these problems, we report a novel, to the best of our knowledge, SERS platform that combines graphene oxide (GO) and cellulose composite paper with colloidal silver nanoparticle (Ag NP) ink. As an efficient substrate, the GO and cellulose composite paper that features hierarchical micro-nanostructures and improved interaction with target molecules can be fabricated on a large scale, and the Ag NP ink can be well stored, avoiding being oxidized in ambient conditions. In this way, our SERS platform not only reduces the cost, but also improved the stability. The sensitivity, reproducibility, and tunable SERS detection performance were evaluated using rhodamine 6G as probing molecules. To demonstrate the capability of our SERS platform in practical analysis, the SERS spectra of two monosodium salt solutions of different concentrations have been collected. The SERS platform has revealed great potential for practical application of SERS technologies.

© 2020 Optical Society of America

Surface-enhanced Raman scattering (SERS) has attracted extensive research interests due to its structural fingerprint detection of target analytes at ultra-low concentrations, thus revealing great potential in practical detection [1]. In recent years, with the exciting surge in the demand of ultra-sensitive detection, SERS has been introduced to biochemical analysis for the rapid and highly sensitive test, including pesticide residue detection on surfaces of fruits in customs inspection [2], onsite diagnostics of aquatic products [3], and noninvasive monitoring of uric acid in human body fluids [4,5]. In this regard, SERS stands out as an ideal method to achieve the detection of these biomarkers due to its good selectivity and extremely high sensitivity [68]. Currently, the development of SERS substrates has become increasingly important.

Up to now, tremendous efforts have been devoted to the preparation of effective SERS substrates [911]. The main strategies are coating noble metals on the substrates with hierarchical micro-nanostructures [1214]. For example, Mohammadi et al. successfully combined Ag nanoparticle (NP) dendrites with graphene-based coatings for the improvement of SERS [15]. Instead of using noble metal NPs, Hurst et al. prepared SERS substrates by modifying the surface of metal oxide (${\rm{Zr}}{{\rm{O}}_2}$, ${\rm{Ce}}{{\rm{O}}_2}$, and ${\rm{Ti}}{{\rm{O}}_2}$) with enediol ligands [16]. In spite of these representative researches, several problems that hamper the further applications of SERS substrates in daily life still exist. Since the preparation of SERS substrates usually needs precision instruments in ultra-clean rooms, and the processing procedures are complex [17,18], the costs of the SERS substrates are expensive [19]. Therefore, it is quite difficult to produce SERS substrates in this way on a large scale, which makes SERS detection an in-lab-only technology [20]. Another open problem is the stability of SERS substrates [21]. As we know, Ag is an ideal material for making SERS substrates due to the abundant hot-spot-induced plasmonic resonance. But Ag substrates are always prone to be oxidized after long-term storage [9,22]. Moreover, the commonly used rigid SERS substrates are not flexible and bendable, thereby being cumbersome to collect samples directly [23]. Therefore, to meet the requirement for practical applications, the development of SERS substrates with easy preparation, low cost, long-term preservation, high reproducibility, as well as convenient collection manners for samples is still highly desired.

In this Letter, we propose a flexible SERS sensing substrate by coupling Ag NP ink with graphene oxide (GO) cellulose papers. The GO cellulose paper was prepared by drying the mixed solution of GO and paper pulp in air. The droplet of the Ag NP solution was dipped on the resultant substrate. The detection capabilities involving the detection limits, sensitivity, reproducibility, and tunable intensities of SERS spectra were investigated by collecting SERS spectra of the rhodamine 6G (R6G) solution. In addition, the flexibility of the SERS substrates was also studied. Subsequently, the as-prepared SERS substrate was used to detect WST-8 molecules by directly collecting samples from the limited liquids in petri dish to distinguish the concentration. The as-prepared SERS substrates may hold great promise in efficient detection, which would greatly facilitate practical applications in daily life and medical diagnosis in the near future.

As shown in Fig. 1, the GO solution and pulp were mixed. The mixture solution was then spin-coated on a piece of glass and dried in air to obtain GO cellulose papers. Drops of AgNP ink were subsequently dipped on this paper. After the solvent evaporation, the flexible SERS platform could be obtained. In this way, the GO cellulose substrate and Ag NPs are stored separately. GO and cellulose composite film can be stored in ambient conditions for a long time considering its stability. The Ag NPs can be stored in aqueous solution to avoid oxidization. The coupling between Ag NP ink with GO cellulose papers could not only reduce the cost, but also greatly prolong its storage time. The resultant SERS platform is flexible and bendable, which is of great significance for direct sample collection.

 figure: Fig. 1.

Fig. 1. Schematic illustration for the preparation of SERS platforms.

Download Full Size | PDF

Figure 2(a) shows the photograph of the as-prepared SERS platform. After water evaporation, the marks of Ag NPs on the substrate provide ideal sites for the target analytes. The morphology of GO cellulose paper was first characterized by confocal laser scanning microscopy (CLSM). The paper fiber skeleton is completely nested in the layered GO, exhibiting good mechanical strength and toughness. The surface of the substrate was further investigated by a scanning electron microscope (SEM). Micro-scale wrinkles could be clearly observed [Figs. 2(c) and 2(d)]. These wrinkles are beneficial for the adhesion of NPs, which is meaningful to the enhancement of Raman scattering light. The combination of GO and paper would facilitate the settlement of AgNPs.

 figure: Fig. 2.

Fig. 2. Characterization of the substrate. (a) Optical photograph of the SERS platform. (b) CLSM image of the GO cellulose paper. SEM images at the scales of (c) 100 µm and (d) 1 µm, respectively.

Download Full Size | PDF

In addition to GO cellulose paper, Ag NPs attached on the surface of thes substrates were also characterized. As shown in Fig. 3(a), a large number of Ag NPs adhered on the substrate could be clearly observed. The enlarged view of SEM photographs revealed the detailed features of Ag NPs [Fig. 3(b)]. The size of Ag NPs is relatively uniform. The distribution of Ag NPs is not massively stacked. The transmission electron microscopy (TEM) image also confirms the homogeneity of Ag NP sizes [Fig. 3(c)]. It is worth noting that wrinkles of the substrate around the Ag NPs could provide enough sites for selectively absorbing the target analytes. The high resolution TEM image [Fig. 3(d)] revealed continuous identical directional lattice fringes, which indicates that the Ag NPs were single crystalline. In addition to the surface morphology, X-ray diffraction (XRD) of Ag NPs was also implemented [Fig. 3(e)].

 figure: Fig. 3.

Fig. 3. SEM images of Ag NPs on GO cellulose paper substrate at the scales of (a) 5 µm and (b) 100 nm, respectively. TEM images of Ag NPs at the scales of (c) 50 nm and (d) 2 nm, respectively. (e) XRD image of Ag NP. (f) Statistical chart of particle size distribution.

Download Full Size | PDF

The five distinct diffraction peaks confirm five crystal surfaces associated with the surface of 111, 200, 220, 311, and 222. This verifies that Ag NPs were present on the surface of the substrate. As we know, size homogeneity of the Ag NPs is crucial for the repeatability of SERS detection. The distribution of the diameters of 120 Ag NPs was shown in Fig. 3(f). The diameters of more than 70% of the NPs range from 40 to 60 nm. The relative uniformity guarantees the detective repeatability. Therefore, it can be concluded that the as-prepared SERS platform possesses homogeneous hot spots and surface-enhanced nanostructures by morphological observations.

To evaluate the detection capability of the SERS platform, R6G solutions with concentrations ranging from ${{1}}{{{0}}^{- 4}}$ to ${{1}}{{{0}}^{- 10}}\;{\rm{M}}$ were prepared. Figure 4(a) shows the SERS spectra of the R6G solution. The characteristic peaks (such as 611, 1358, and ${{1651}}\;{\rm{c}}{{\rm{m}}^{- 1}}$) of R6G in solutions with concentrations from ${{1}}{{{0}}^{- 4}}\;{\rm{M}}$ to ${{1}}{{{0}}^{- 7}}\;{\rm{M}}$ can be clearly observed. The bands at 1651,1507, and ${{1358}}\;{\rm{c}}{{\rm{m}}^{- 1}}$ are associated with C–C stretching. While the bands at 1570 and ${{610}}\;{\rm{c}}{{\rm{m}}^{- 1}}$ are attributed to ${\rm{C}} = {\rm{C}}$ stretching and aromatic bending, respectively. When the concentration of the R6G solution was lower than ${{1}}{{{0}}^{- 8}}\;{\rm{M}}$, the noise became obvious, and only several characteristic peaks (1570 and ${{1651}}\;{\rm{c}}{{\rm{m}}^{- 1}}$) could be observed. To evaluate the enhancement capability of this platform, the enhancement factor (EF) was calculated by the following formula [10]:

$${\rm{EF}} = {I_{{\rm{SERS}}}} \times {N_{{\rm{vol}}}}/{N_{{\rm{SERS}}}} \times {I_{{\rm{vol}}}},$$
where ${I_{{\rm{SERS}}}}$ and ${I_{{\rm{vol}}}}$ are the intensity of SERS and normal Raman spectrum, respectively. ${N_{{\rm{SERS}}}}$ and ${N_{{\rm{vol}}}}$ are the molecules number absorbed on this platform within the laser spot area and molecules number in Raman scattering volume, respectively. The EF at ${{1650}}\;{\rm{c}}{{\rm{m}}^{- 1}}$ for our SERS substrate was ${\sim}{1.49} \times {{1}}{{{0}}^7}$, which is about six times larger than that of the Ag NP loaded glass substrate (control experiment, ${2.51} \times {{1}}{{{0}}^6}$). The high SERS EF can be attributed to the coupling effect of electromagnetic (EM) field enhancement and chemical enhancement. The EM enhancement (EME, range from ${{1}}{{{0}}^5}$ to ${{1}}{{{0}}^{10}}$) was attributed to the oscillation of localized surface plasmon resonance (LSPR) [10]. When a specific EM field of incident light induces collective oscillation of Ag NP conduction electrons, LSPR would be formed. As a result, the optical energy of the incident light is coupled to the EM field of the Ag NP oscillating electrons, which excites the molecules adsorbed on the Ag NP surfaces to emit enhanced Raman signals [24]. The as-formed Ag NPs could provide rich hot spots resulting from oscillation of the LSPR, creating prominent EM field enhancement. The chemical enhancement is original from the aromatic rings’ interaction between R6G and GO [25]. The long axis of the R6G molecules would be approximately parallel to the layer of GO because of strong $\pi - \pi$ stacking, which induces charge transfer between the R6G and GO [26]. GO had been proven to be an attractive Raman-enhancing substrate due to the strong interaction of $\pi - \pi$ bonding between GO and R6G, which contributes to chemical enhancement. Compared to the detection limit, the stability and reproducibility of SERS spectra are more important for practical detection. Thus, the SERS spectra of the R6G solution (${{1}}{{{0}}^{- 6}}\;{\rm{M}}$) at eight randomly selected sites were acquired [Fig. 4(b)]. The Raman spectra exhibited high uniformity at each characteristic peak, demonstrating its remarkable reproducibility. The reproducible SERS spectra could be ascribed to the relative uniformity of the Ag NPs. The high repeatability and sensitivity would be a strong guarantee for its practical applications.
 figure: Fig. 4.

Fig. 4. (a) SERS spectra of R6G solution in different concentrations. (b) Raman spectra of R6G with a concentration of ${{1}}{{{0}}^{- 6}}\;{\rm{M}}$ at eight different sites. (c) Distribution of NPs in different positions of GO cellulose paper. (d) Comparison of Raman spectrum intensity at different positions. (e) Direct sample collection of WS-8. (f) SERS spectra of WST-8 solutions with different concentrations.

Download Full Size | PDF

In addition, this SERS platform also exhibits good capability of tunable intensities of Raman spectra. GO cellulose paper can be folded before dropping droplets of Ag NPs onto it [Fig. 4(c)]. The build-up structures of Ag NPs are different at different positions after complete evaporation of the solvent. At the top corner, the thickness of the Ag NP layer is thin, since the droplet of Ag NP ink tended to fall down along the slope. The droplet on the slope had similar situations. However, the Ag NPs at the bottom corner would pile up due to the aggregation. Thus, the bottom site is likely to provide more hot spots resulting from stronger oscillation of LSPR. The SERS intensity can be tuned by folding the substrates. To confirm this assumption, the Raman spectra of the R6G solution with the concentration of ${{1}}{{{0}}^{- 6}}\;{\rm{M}}$ were collected at these three different sites [Fig. 4(d)]. Obviously, the Raman intensity at the bottom corner is obviously stronger than that at the other site. The intensity difference confirms that the SERS sensitivity can be flexibly tuned by folding the substrate, which is another option for practical application.

To confirm the efficient detection of the SERS platform, 2-(2-Methoxy-4-nitrophenyl)-3-(4-nitrophenyl)-5-(2,4-disulfop-henyl)- 2 H-tetrazolium, monosodium salt (WST-8, a kind of cell activity indicator) and its diluted solutions were adopted for practical demonstration. Detecting the concentration variation of WST-8 is of great significance for evaluating cell activity, which is meaningful for parasitic infection examination [27]. In instant diagnosis, the small amount of the sample and limited size of the container usually does not allow direct sampling by a traditional rigid substrate. However, in this case, WST-8 sample collection was easily realized by folding the flexible SERS platform in an appropriate size, as demonstrated in Fig. 4(e). We prepared 0.1 M and 0.01 M WST-8 water solution, respectively, which makes it possible to distinguish the two different concentrations by using this SERS platform [Fig. 4(f)]. The characteristic peaks of WST-8 molecules with respect to 345, 465, 650, 1125, 1180, 1560, 1595, and ${{1615}}\;{\rm{c}}{{\rm{m}}^{- 1}}$ can be clearly observed. Moreover, the SERS signal intensity became larger when the concentration of WST-8 increased. The good sensitivity was mainly attributed to the coupling of EM field enhancement and chemical enhancement. As we know, the WST-8 molecule has three benzene rings with a $\pi - \pi$ bond [WST-8 formula is shown in inset of Fig. 4(f)]. The platform would absorb large numbers of WST-8 molecules due to the strong interaction between the $\pi - \pi$ bond in GO and WST-8, which contribute to the chemical enhancement. Moreover, the Ag NPs on the substrate also make contributions to the high sensitivity due to the EM field enhancement.

To investigate the flexibility, the variation of deformation rates with numbers of bending cycles was calculated. As for this SERS platform, it can be bended (the bending radius is ${{{R}}_0}$) and unfolded (${{L}}$ for long, ${{w}}$ for width) for many times. In this process, the transverse deformation rate (${\gamma _L}$) and longitudinal deformation rate (${\gamma _W}$) with the bending cycles has been gathered [Fig. 5(a)]. The deformation rate in these two different directions were both less than 2% after 120 bending cycles, which means the deformation is limited into 0.4 mm. Taking advantage of the flexibility, direct samples collection can be easily achieved. Furthermore, the Raman intensity after bending was further studied. For easier explanation, we calculated the relative standard deviation (RSD) of Raman intensity at R6G concentrations of ${{1}}{{{0}}^{- 6}}\;{\rm{M}}$ after every five bending cycles [Fig. 5(b)]. With increasing bending cycles, the RSD of Raman intensity also increases. The increase of RSD of Raman intensity suggested that signal fluctuations became serious. Fortunately, the RSD of Raman intensity was less than 10% when the bending cycles were less than 25. Generally, sample collection could be accomplished within 10 times of bending cycles. Therefore, it can be concluded that the SERS platform has good flexibility and satisfactory detection capacity after cyclic bending.

 figure: Fig. 5.

Fig. 5. (a) Variation of transverse and longitudinal deformation rates with bending cycles. (b) Relative standard deviation (RSD) variation of Raman intensity with bending cycles at R6G concentration of ${{1}}{{{0}}^{- 6}}\;{\rm{M}}$.

Download Full Size | PDF

In summary, a flexible SERS platform consisting of GO cellulose paper and Ag NPs was prepared. The detection performance was confirmed by collecting Raman signals of R6G with a series of different concentrations, and the detection limit of R6G was as low as ${{1}}{{{0}}^{- 8}}\;{\rm{M}}$. The Raman spectra were highly uniform, which indicated high sensitivity and good repeatability. The capability of tunable SERS intensity was also demonstrated by collecting Raman signals of R6G. The SERS platform also shows good flexibility. Finally, the SERS platform was adopted to distinguish two different concentrations of WST-8 solutions for practical application. Ag NPs as one of the components of this SERS platform is feasible to be replaced by gold NPs and gold nanorods for the consideration of biological compatibility. The as-prepared SERS platform exhibits fascinating properties including low cost, easy experimental operation, long preservation, high reproducibility and sensitivity, flexibility, as well as the capacity of tunable SERS intensity, which would greatly facilitate practical applications, such as instant diagnosis at a large scale.

Funding

National Natural Science Foundation of China (61775078, 61935008).

Disclosures

The authors declare no conflicts of interest.

REFERENCES

1. Y. Zeng, K. M. Koo, M. Trau, A. G. Shen, and J. M. Hu, Appl. Mater. Today 15, 431 (2019). [CrossRef]  

2. V. Sharma, N. N. Som, S. B. Pillai, and P. K. Jha, Spectrochim. Acta. A 224, 117352 (2020). [CrossRef]  

3. L. Ouyang, L. Yao, T. H. Zhou, and L. H. Zhu, Anal. Chim. Acta. 1027, 83 (2018). [CrossRef]  

4. S. S. Yang, Z. C. Zhang, J. Zhao, Z. W. Yu, and H. Y. Jiang, Mater. Lett. 131, 78 (2014). [CrossRef]  

5. T. T. B. Quyen, W. N. Su, K. J. Chen, C. J. Pan, J. Rick, C. C. Chang, and B. J. Hwang, J. Raman. Spectrosc. 44, 1671 (2013). [CrossRef]  

6. H. Zhang, C. P. Fu, S. T. Wu, Y. Q. Shen, C. H. Zhou, J. Neng, Y. Yi, Y. Jin, and Y. Zhu, Anal. Meth. 11, 783 (2019). [CrossRef]  

7. A. H. Qu, X. L. Wu, L. G. Xu, L. Q. Liu, W. Ma, H. Kuang, and C. Xu, Nanoscale 9, 3865 (2017). [CrossRef]  

8. K. V. Kong, W. K. Leong, Z. Lam, T. X. Gong, D. Goh, W. K. O. Lau, and M. Olivo, Small 10, 5030 (2014). [CrossRef]  

9. R. Zhang, B. B. Xu, X. Q. Liu, Y. L. Zhang, Y. Xu, Q. D. Chen, and H. B. Sun, Chem. Commun. 48, 5913 (2012). [CrossRef]  

10. J. A. Huang, Y. L. Zhang, Y. Zhao, X. L. Zhang, M. L. Sun, and W. Zhang, Nanoscale 8, 11487 (2016). [CrossRef]  

11. K. Xu, R. Zhou, K. Takei, and M. Hong, Adv. Sci. 6, 1900925 (2019). [CrossRef]  

12. S. Yan, R. An, Y. J. Zou, N. Yang, and Y. Zhang, Sens. Actuator B Chem. 302, 127107 (2020). [CrossRef]  

13. J. Zhang, M. ElKabbash, R. Wei, S. C. Singh, B. Lam, and C. Guo, Light: Sci. Appl. 8, 53 (2019). [CrossRef]  

14. D. X. Liu, Y. L. Sun, W. F. Dong, R. Z. Yang, Q. D. Chen, and H. B. Sun, Laser Photon. Rev. 8, 882 (2014). [CrossRef]  

15. A. Mohammadi, D. L. Nicholls, and A. Docoslis, Sensors 18, 3404 (2018). [CrossRef]  

16. S. J. Hurst, H. C. Fry, D. J. Gosztola, and T. Rajh, J. Phys. Chem. C 115, 620 (2011). [CrossRef]  

17. Z. Zhang, P. Ma, J. Y. Li, Y. Sun, H. Shi, N. Chen, X. Zhang, and H. Chen, Opt. Express 27, 29247 (2019). [CrossRef]  

18. M. Yue, F. Li, N. H. Lu, P. Yao, T. Xue, and P. Liu, Appl. Opt. 58, 8290 (2019). [CrossRef]  

19. Y. Tian, H. F. Wang, L. Q. Yan, X. F. Zhang, A. Falak, Y. J. Guo, P. Chen, F. Dong, L. Sun, and W. Chu, Adv. Sci. 6, 1900177 (2019). [CrossRef]  

20. K. C. Xu, R. Zhou, K. Takei, and M. H. Hong, Adv. Sci. 6, 1900925 (2019). [CrossRef]  

21. G. J. Weng, Y. Feng, J. Zhao, J. J. Li, J. Zhu, and J. W. Zhao, Appl. Surf. Sci. 478, 275 (2019). [CrossRef]  

22. K. Q. Wang, D. W. Sun, H. B. Pu, Q. Y. Wei, and L. J. Huang, ACS Appl. Mater. Interface 11, 29177 (2019). [CrossRef]  

23. J. Zhou, J. S. Zhang, H. T. Yang, Z. Wang, J. A. Shi, W. Zhou, N. Jiang, G. Xian, Q. Qi, Y. Weng, and C. Shen, Nanoscale 11, 11782 (2019). [CrossRef]  

24. S. Almohammed, F. Zhang, B. J. Rodriguez, and J. H. Rice, J. Phys. Chem. Lett. 10, 1878 (2019). [CrossRef]  

25. X. Liang, B. Liang, Z. Pan, X. Lang, Y. Zhang, G. Wang, P. Yin, and L. Guo, Nanoscale 7, 47 (2015). [CrossRef]  

26. W. H. Park and M. Jung, J. Phys. Chem. C 120, 42 (2016). [CrossRef]  

27. K. Chamchoy, D. Pakotiprapha, P. Pumirat, U. Leartsakulpanich, and U. Boonyuen, BMC Biochem. 20, 4 (2019). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (5)

Fig. 1.
Fig. 1. Schematic illustration for the preparation of SERS platforms.
Fig. 2.
Fig. 2. Characterization of the substrate. (a) Optical photograph of the SERS platform. (b) CLSM image of the GO cellulose paper. SEM images at the scales of (c) 100 µm and (d) 1 µm, respectively.
Fig. 3.
Fig. 3. SEM images of Ag NPs on GO cellulose paper substrate at the scales of (a) 5 µm and (b) 100 nm, respectively. TEM images of Ag NPs at the scales of (c) 50 nm and (d) 2 nm, respectively. (e) XRD image of Ag NP. (f) Statistical chart of particle size distribution.
Fig. 4.
Fig. 4. (a) SERS spectra of R6G solution in different concentrations. (b) Raman spectra of R6G with a concentration of ${{1}}{{{0}}^{- 6}}\;{\rm{M}}$ at eight different sites. (c) Distribution of NPs in different positions of GO cellulose paper. (d) Comparison of Raman spectrum intensity at different positions. (e) Direct sample collection of WS-8. (f) SERS spectra of WST-8 solutions with different concentrations.
Fig. 5.
Fig. 5. (a) Variation of transverse and longitudinal deformation rates with bending cycles. (b) Relative standard deviation (RSD) variation of Raman intensity with bending cycles at R6G concentration of ${{1}}{{{0}}^{- 6}}\;{\rm{M}}$.

Equations (1)

Equations on this page are rendered with MathJax. Learn more.

E F = I S E R S × N v o l / N S E R S × I v o l ,
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.