Open Access
Add to BibSonomyAbstract
The ‘KlariteTM’ SERS sensor platform consisting of an array of gold coated inverted square pyramids patterned onto a silicon substrate has become the industry standard over the last decade, providing highly reproducible SERS signals. In this paper, we report successful transfer from silicon to plastic base platform of an optimized SERS substrate design which provides 8 times improvement in sensitivity for a Benzenethiol test molecule compared to standard production Klarite. Transfer is achieved using roll-to-roll and sheet-level nanoimprint fabrication techniques. The new generation plastic SERS sensors provide the added benefit of cheap low cost mass-manufacture, and easy disposal. The plastic replicated SERS sensors are shown to provide ~107 enhancement factor with good reproducibility (5%).
©2013 Optical Society of America
1. Introduction
Since Fleischmann et al observed SERS on electro-chemically roughened silver electrodes, SERS has been of interest in electrochemistry [1–4], polymer [5] and material science [6–8], biochemistry [9–12], catalysis [13–15] and SERS-based optical detection and nonlinear photonics [16–18]. Improved Raman scattering efficiency combined with increased Plasmon pump coupling provides benefits to detection and identification of chemical and biological substances applicable to: environmental monitoring, medical diagnostics, food safety, pharmaceutical research, security and defense. SERS has successfully detected bacteria on food, pesticide on fruits [19,20], total antioxidant capacity of commercial fruit juices and herbal teas [21], detection of biological samples of various cancers, Parkinson’s disease [22], Alzheimer’s disease [23] and blood glucose [24]. SERS is well known as a rapid, label-free, information rich detection mechanism providing high spatial resolution. However, SERS has challenges particularly in lack of interaction of test molecules with the substrate and lack of reproducibility due to the spatially non-uniform nature of Plasmonic field enhancement and sensitivity to metal’s surface roughness. Moreover, SERS substrates must be suitable for low cost, high throughput manufacturing to make the transition from research labs to ‘real world’ applications. In order to fulfill such challenges, a variety of SERS active substrates have been developed using various fabrication methods and architectural substrates including periodic nanoclusters of spheres, voids/holes, rods/wires, pillars/discs.
Previously the commercial ‘KlariteTM’ SERS sensor has been demonstrated to provide highly reproducible SERS signals. ‘KlariteTM’ consists of an array of gold coated inverted pyramidal pits with 2000nm pitch length (P) and 1500nm pit diameter (D). The pits have a square base structure. Recently, we demonstrated improvement in sensitivity by using rectangular based pyramidal pits with 1:1.2 (length: width) aspect ratio, 1250nm Pitch and 1000nm pit diameter. The optimized design was found to provide 800% average improvement in sensitivity whilst maintaining less than 8% variability for a self-assembled monolayer of Benezenethiol [25]. This optimized silicon based design was shown to achieve ~109 enhancement factor. This paper reports the successful transfer of the optimized SERS substrate design from silicon to plastic based platform which is manufactured by sheet-level and roll-to-roll (R2R) nanoimprint lithography (NIL) methods.
2. Device fabrication
Nanoimprint lithography (NIL) is a replication technique suitable for low cost high throughput production. NIL is an embossing method whereby a surface relief pattern is replicated from a mould or template onto another surface. The pattern transfer process involves embossing a soft polymer layer followed by immediate UV curing. Compared to thermal imprint methods, UV-NIL gives better large-area scalability, feature accuracy and surface quality, all of which are important for optical components. To be able to replicate polymer SERS sensors, the template surface must first be fabricated. We use electron-beam patterning followed by anisotropic wet etching of a silicon wafer to form the pyramidal pit structure. There are two steps to fabricate a replicated device. Firstly a master template is replicated onto a mould (template) forming a negative copy of master. This is then imprinted onto the final substrate to form the positive replica.
Sheet-level nanoimprint process is illustrated in Fig. 1. First the silicon master stamp is cleaned and treated with a fluorosilane anti-adhesion layer to ease mould separation after UV curing. UV-curable hybrid polymer (Ormocer®) material is droplet-deposited onto a cleaned and surface-treated plastic (PMMA) carrier film which forms an interim stamp. During the mould pressing step, the mould, which is at the first stage a master stamp, is pressed onto the polymer. After the polymer has spread and filled all the cavities on the mould (with the help of applied pressure, time and possible heat), the polymer is cured by UV-light exposure through the PMMA carrier film. Finally, separation of the substrate and mould releases a negative copy of master. The same procedure is used to fabricate a polymer replica onto the final polymer substrate using the negative plastic copy of master instead of the master.
In roll-to-roll (R2R) UV imprinting process, the following steps are done to replicate SERS structure from silicon master to polymer replica. First a Nickel-shim is electroformed from the silicon master and is laser welded to an embossing reel. Since a PMMA carrier web is used in the R2R process, the PMMA is corona treated to ensure good adhesion between UV lacquer and carrier web. In the second step, UV-curable lacquer is coated on top of the web using a reverse gravure coater. During the embossing phase the SERS structure is replicated onto the lacquer using the embossing reel while the lacquer is cured by UV light exposure through the PMMA carrier film. After embossing, the lacquer is post-cured and a protective foil laminated on top of SERS structure. The properties of UV-curable polymer, such as viscosity and adhesion, are essential for both imprinting processes to fabricate the qualified replicated structures.
In this paper, a plastic SERS sensor replicated by R2R method (denoted as ‘plastic-1’) as shown in Fig. 2(b) and a polymer replica imprinted by sheet-level method is (defined as ‘plastic-2’)shown in Fig. 2(c). Figure 2 shows SEM images of the silicon master Fig. 2(a) and plastic replicas after 300nm gold(Au) deposition. The SEM pictures of the silicon master and polymer replica devices show good fidelity of the features indicating the successful transfer of pattern by both imprint methods.
Fig. 2 SEM images of inverted rectangular based pyramidal design after 300nm gold deposition: (a) silicon master. (b) polymer replica from R2R method (plastic-1). (c) polymer replica from sheet-level method (plastic-2)
However, careful inspection of the SEM images for plastic substrates replicated by the two methods Au grain reveals that a slight difference in gold layer morphology. This is expected to slightly affect the electric-field localization along the side walls of the pit and hence result in different Raman amplification.
All SERS substrates presented in this paper were coated with 300nm thick gold layer by electron-beam evaporation. 300nm thickness was chosen to be optically thick, and comparable to the standard silicon Klarite. Due to length constraints of this paper we do not discuss effect of gold thickness and its morphology. This will be the topic of a subsequent paper.
Figures 3(a) and 3(b) show larger area SEM pictures of two gold coated imprinted devices with successful transfer with a uniform gold coating.
Fig. 3 Low Magnification SEM images to validate that the samples have high fabrication quality and uniform gold coating of 300nm thickness: (a) sheet-level replicated plastic substrate and (b) R2R replicated plastic substrate.
2. Analysis of SERS effect by Raman spectroscopy
SERS enhancement signals were measured using a Renishaw Invia Raman spectroscopy system. Polymer replica and silicon master substrates were excited by an incident 785nm laser. The Si master and sheet-level replicated sample (plastic-2) were coated with a benzenethiol (C6H5SH) test molecule, and R2R replicated sample (plastic-1) is coated with benzyl mercaptan (C6H5CH2SH) test molecule. Although samples are coated with different molecules, we can still make relative assessment of SERS enhancement by analyzing SERS counts of trigonal ring breathing (990 cm−1-1010 cm−1 Raman shift) since both molecules are from the same group of thiols. Measurement parameters were: 10% of power (20mW actual power), 10 second exposure time, 3 times accumulation and 50xmagnification in all cases.
As 50 times magnification (N.A = 0.75) gives an approximate beam spot size of 0.66µm (when carefully taken into tight focus), the spot size is actually of the order of the size of a single pit. Hence the measured SERS enhancement figures are predominantly due to coupling to localized surface Plasmons supported by an individual pit and should include very limited contributions from propagating surface Plasmons associated with coupling between neighboring pits (although sideways scattering is certainly possible for periodic structures supporting propagating Plasmons).
Hence we plot Raman intensity per pit for master and replicated polymer substrates. As expected, the trend for Raman (SERS) intensity for polymer substrates with various geometrical parameters (change in pitch length) is the same as for silicon SERS substrate as shown in Fig. 4(b). However, Raman intensity for the plastic samples are lower than the silicon master (as shown in Fig. 4(a)) probably due to the slight difference in the Au grain formation on the base substrate. Raman intensity of R2R fabricated sample (plastic-1) is lower than sheet-level imprinted sample (plastic-2) probably due to the different material properties of UV lacquer and Ormocer® respectively. From Raman intensity plot Fig. 4(a), an enhancement factor of plastic-2 is derived in order of 107 while R2R replicated substrate (plastic-1) provides only in order of 106.
Fig. 4 (a) Plots show individual Raman intensity per pit of different base substrates. (b) Plots show Raman intensity per pit as a function of pitch length for rectangular pyramidal pit at 1003cm−1 Raman shift for different base substrate.
When the objective lens is switched to 5 times magnification with a larger beam size (~4.15µm), measurements become averaged over the area of a few pits. Under these conditions for ‘plastic-1’ (R2R imprint method) average Raman intensity improves from 6771.56 to 8450.24. Increasing the measurement spot size therefore provides benefit in terms of improved reproducibility but doesn’t greatly increase overall sensitivity. This again confirms that enhancement is mostly due to localized surface Plasmon supported by the individual pits, rather than propagating Plasmons supported by group of pits. In order to assess the measurement to measurement reproducibility of polymer replica, few Raman measurements (10/12) are taken across the surface and plotted as shown in inset of Figs. 5(a) and 5(b) after base line correction. In comparing peak intensity value of trigonal ring breathing (990 cm−1-1010 cm−1 Raman shift), reproducibility of plastic-1 (R2R imprint method) demonstrates 5% (8450.24 ± 417.74) measurement to measurement reproducibility (averaged over 12 measurements) as shown in Fig. 5(a). We find that plastic-2 (sheet-level imprint) provides worse measurement to measurement reproducibility (9.41% instead of 5% for R2R imprint method) as shown in Fig. 5(b).
Fig. 5 (a) Plot shows the measurement to measurement reproducibility of 5% for 12 measurements on R2R replicated sample. (b) Plot shows 9.41% reproducibility for sheet-level replicated sample. Insets are measured Raman spectra over the surface (813.50 µm x 745.79 µm) showing the related vibrational structures of test molecules.
3. Optical analysis
From the Raman measurements, we see that the SERS count changes significantly with base substrate type. In the case of the optically thick gold layer (300nm) the direct contribution of the base substrate should be negligible in terms of influence on the Raman signal. To analysis the disparity of SERS effect 3D modeling and simulation were performed using RSOFT Diffractmod which employs Rigorous Coupled Wave Analysis (RCWA) method with transmission-line treatment of boundary condition. Figure 6(a) shows the designed model for inverted rectangular pyramidal pit and Fig. 6(b) defines the geometrical parameters of the design. The theoretical zero order reflection results were then compared to the experimental angle resolved broadband reflection results. The experimental technique is described fully in [26]. Both simulation and experiments were made for a 300nm Au rectangular based pyramidal pit with 1250nm pitch-1000nm diameter under TM-incident polarization. The reflection results are plotted as a color map of intensity as a function of incident angle over a wide wavelength range. Figure 7(a) and 7(b) are simulation results for the dedicated design coloring the low to high absorption order from white to dark, respectively. Figures 7(c) and 7(d) give the experimental results observing the color order from dark red to dark blue for the low to high absorption, respectively.
Fig. 6 (a) The cross-sectional and 3D view of the modeled inverted rectangular pyramidal pit with different base substrate for 300nm Au thickness. (b) Schematic diagram for geometrical parameters of inverted pyramidal pit: left side shows the pitch length and pit diameter and right side indicates the change in (width to length) aspect ratio of pit from top view.
Fig. 7 Dispersion maps for 1250nm ‘P’- 1000nm ‘D’ rectangular based pyrmidal pits with 300nm Au thickness under TM-incident polarization. (a) and (b) simulation results for silicon base substrate and UV lacquer base substrate, respectively. (c) experimental result of silicon base substrate functionalized with benzenethiol molecule. (d) experimental result of UV lacquer base substrate (plastic-1) coated with benzyl mercaptan.
The simulation results (Figs. 7(a) and 7(b)) show near identical plasmon dispersion properties for the rectangular pyramidal design for both silicon and plastic base substrate. In addition to the dispersion lines, the absorption region (dark black) which corresponds to the possibility of localized Plasmon occurrence did not change with base substrate type as expected. In order to experimentally validate these results, angle resolved reflectivity measurements have been performed on 300nm Au coated silicon and plastic-1 substrate as shown in Figs. 7(c) and 7(d) respectively. The silicon and plastic-1 substrates were functionalized with benzenethiol and benzyl mercaptan respectively. We assume that the change in molecule between two samples with additional CH2 bond should not make a difference in reflection. From the experimental results in Figs. 7(c) and 7(d) we observe that there is change in Plasmon absorption region.
Dispersion lines attributed to features of the pyramidal design match well between simulation and experimental results. These dispersion lines give information about quality of the replicated pyramid structure, such as sharpness of corners. The Plasmon absorption dips/region (dark blue to blue) gives the solution of the incident angle and wavelength range where Plasmon resonance occurs and hence Raman amplification for the SERS effect. The difference in Plasmon behavior between simulation and experimental result is attributed to the exclusion of grain structure and morphology of the gold surface in the simulation (inclusion of this information is currently impractical). The experimental results however suggest that the silicon base substrate with 300nm Au has the possibility to support highly localized Plasmons around 750nm-800nm wavelength which is the operating range of our 785nm Raman pump laser. Hence the experimental angular reflection measurement serves as a guideline for observing Plasmon resonances and determining the physical reason behind increased SERS counts. Using the same approach, and analyzing the reflectivity map for plastic-1 sample, the absorption intensity is lower hence giving rise to lower SERS counts. From the above discussions we can see that the Au thickness and its morphology are critically essential for the SERS sensor.
4. Conclusion
Recently silicon based KlariteTM has been optimised by adjusting the geometrical parameters in order to determine an optimized design, resulting in a new design giving 109 enhancement factor (as opposed to 10^7 previously). In this paper, the optimized substrate design was successfully transferred to a plastic platform by roll-to-roll and sheet-level nano-imprint replication methods. Comparison of Raman measurements on each type of substrate shows that R2R imprinted polymer (plastic-1) and sheet-level replica (plastic-2) give very good reproducibility 5% and 9.41%, respectively. In terms of enhancement factor, polymer replica device (plastic-2) gives the order of ~107 with slightly lower enhancement factor than silicon master. From SEM images and dispersion maps of different base platforms, the difference in Raman intensity from the Si master to polymer replica is mainly due to gold grain formation relative to the base substrate. We believe that with more development the level of SERS signal enhancement for the plastic substrate devices will increase to a similar level as is currently obtainable with the optimised silicon designs. The current level of sensitivity demonstrated in this paper for both imprint methods is already equivalent to the previous generation silicon Klarite devices, which has already been proven sufficient in the market place for real applications. The work presented in this paper shows the proof of principal for transfer of the Klarite technology from silicon to plastic. Disposable plastic SERS sensors are promising for real-world SERS sensing requirements with high SERS enhancement factor, low equipment cost and high throughput manufacturing.
Acknowledgments
The authors acknowledge the funds and support from the FP7 ‘PHOTOSENS’ consortium project (FP7-NMP-2010-LARGE-4-263382) and would like to thank Sumit Kalsi for his help with molecule coating.
References and links
1. K. B. Biggs, J. P. Camden, J. N. Anker, and R. P. Van Duyne, “Surface-Enhanced Raman Spectroscopy of Benzenethiol Adsorbed from the Gas Phase onto Silver Film over Nanosphere Surfaces: Determination of the Sticking Probability and Detection Limit Time,” J. Phys. Chem. A 113(16), 4581–4586 (2009). [CrossRef] [PubMed]
2. N. Liu, M. L. Tang, M. Hentschel, H. Giessen, and A. P. Alivisatos, “Nanoantenna-enhanced gas sensing in a single tailored nanofocus,” Nat. Mater. 10(8), 631–636 (2011). [CrossRef] [PubMed]
3. L. S. Koodlur, “Layer-by-layer self assembly of a water-soluble phthalocyanine on gold. Application to the electrochemical determination of hydrogen peroxide,” Bioelectrochemistry 91, 21–27 (2013). [CrossRef] [PubMed]
4. N.-N. Bu, A. Gao, X.-W. He, and X.-B. Yin, “Electrochemiluminescent biosensor of ATP using tetrahedron structured DNA and a functional oligonucleotide for Ru(phen)32+ intercalation and target identification,” Biosens. Bioelectron. 43, 200–204 (2013). [CrossRef] [PubMed]
5. M. Muniz-Miranda, E. Giorgetti, G. Margheri, T. Del Rosso, S. Sottini, A. Giusti, and M. Alloisio, “SERS Investigation on the Polymerization of Carbazolyl-diacetylene Monolayers on Gold Surfaces,” Macromol. Symp. 230(1), 67–70 (2005). [CrossRef]
6. M. G. Blaber, M. D. Arnold, and M. J. Ford, “A review of the optical properties of alloys and intermetallics for Plasmonics,” J. Phys. Condens. Matter 22(14), 143201 (2010). [CrossRef] [PubMed]
7. M. J. A. Hore, A. L. Frischknecht, and R. J. Composto, “Nanorod Assemblies in Polymer Films and Their Dispersion-Dependent Optical Properties,” ACS Macro Lett. 1(1), 115–121 (2012). [CrossRef]
8. M. G. Blaber, M. D. Arnold, and M. J. Ford, “Designing materials for Plasmonic systems: the alkali-noble intermetallics,” J. Phys. Condens. Matter 22(9), 095501 (2010). [CrossRef] [PubMed]
9. E. López-Tobar, B. Hernandez, M. Ghomi, and S. Sanchez-Cortes, “Stability of the Disulfide Bond in Cystine Adsorbed on Silver and Gold Nanoparticles As Evidenced by SERS Data,” J. Phys. Chem. C 117(3), 1531–1537 (2013). [CrossRef]
10. K. W. Kho, U. S. Dinish, A. Kumar, and M. Olivo, “Frequency Shifts in SERS for Biosensing,” ACS Nano 6(6), 4892–4902 (2012). [CrossRef] [PubMed]
11. B. L. Scott and K. T. Carron, “Dynamic Surface Enhanced Raman Spectroscopy (SERS): Extracting SERS from Normal Raman Scattering,” Anal. Chem. 84(20), 8448–8451 (2012). [CrossRef] [PubMed]
12. R. P. Johnson, J. A. Richardson, T. Brown, and P. N. Bartlett, “A Label-Free, Electrochemical SERS-Based Assay for Detection of DNA Hybridization and Discrimination of Mutations,” J. Am. Chem. Soc. 134(34), 14099–14107 (2012). [CrossRef] [PubMed]
13. A.-X. Yin, W.-C. Liu, J. Ke, W. Zhu, J. Gu, Y.-W. Zhang, and C.-H. Yan, “Ru Nanocrystals with Shape-Dependent Surface-Enhanced Raman Spectra and Catalytic Properties: Controlled Synthesis and DFT Calculations,” J. Am. Chem. Soc. 134(50), 20479–20489 (2012). [CrossRef] [PubMed]
14. W. Xie, C. Herrmann, K. Kömpe, M. Haase, and S. Schlücker, “Synthesis of Bifunctional Au/Pt/Au Core/shell Nanoraspberries for In Situ SERS Monitoring of Platinum-Catalyzed Reactions,” J. Am. Chem. Soc. 133(48), 19302–19305 (2011). [CrossRef] [PubMed]
15. T. Tachikawa and T. Majima, “Single-Molecule, Single-Particle Approaches for Exploring the Structure and Kinetics of Nanocatalysts,” Langmuir 28(24), 8933–8943 (2012). [CrossRef] [PubMed]
16. V. P. Zharov, “Ultrasharp nonlinear photothermal and photoacoustic resonances and holes beyond the spectral limit,” Nat. Photonics 5(2), 110–116 (2011). [CrossRef]
17. V. V. Temnov, “Ultrafast acousto-magneto-Plasmonics,” Nat. Photonics 6(11), 728–736 (2012). [CrossRef]
18. M. Kauranen and A. V. Zayats, “Nonlinear Plasmonics,” Nat. Photonics 6(11), 737–748 (2012). [CrossRef]
19. B. Liu, G. Han, Z. Zhang, R. Liu, C. Jiang, S. Wang, and M.-Y. Han, “Shell Thickness-Dependent Raman Enhancement for Rapid Identification and Detection of Pesticide Residues at Fruit Peels,” Anal. Chem. 84(1), 255–261 (2012). [CrossRef] [PubMed]
20. D. Volpati, P. H. B. Aoki, C. A. R. Dantas, F. V. Paulovich, M. C. de Oliveira, O. N. Oliveira Jr, A. Riul Jr, R. F. Aroca, and C. J. Constantino, “Toward the Optimization of an e-Tongue System Using Information Visualization: A Case Study with Perylene Tetracarboxylic Derivative Films in the Sensing Units,” Langmuir 28(1), 1029–1040 (2012). [CrossRef] [PubMed]
21. M. Özyürek, N. Güngör, S. Baki, K. Güçlü, and R. Apak, “Development of a Silver Nanoparticle-Based Method for the Antioxidant Capacity Measurement of Polyphenols,” Anal. Chem. 84(18), 8052–8059 (2012). [CrossRef] [PubMed]
22. X. Wang, C. Wang, L. Cheng, S.-T. Lee, and Z. Liu, “Noble Metal Coated Single-Walled Carbon Nanotubes for Applications in Surface-Enhanced Raman Scattering Imaging and Photothermal Therapy,” J. Am. Chem. Soc. 134(17), 7414–7422 (2012). [CrossRef] [PubMed]
23. K. Kaaki, K. Hervé-Aubert, M. Chiper, A. Shkilnyy, M. Soucé, R. Benoit, A. Paillard, P. Dubois, M.-L. Saboungi, and I. Chourpa, “Magnetic Nanocarriers of Doxorubicin Coated with Poly(ethylene glycol) and Folic Acid: Relation between Coating Structure, Surface Properties, Colloidal Stability, and Cancer Cell Targeting,” Langmuir 28(2), 1496–1505 (2012). [CrossRef] [PubMed]
24. M. Delcea, N. Sternberg, A. M. Yashchenok, R. Georgieva, H. Bäumler, H. Möhwald, and A. G. Skirtach, “NanoPlasmonics for Dual-Molecule Release through Nanopores in the Membrane of Red Blood Cells,” ACS Nano 6(5), 4169–4180 (2012). [CrossRef] [PubMed]
25. S. Z. Oo, M. D. B. Charlton, D. Eustace, R. Y. Chen, S. J. Pearce, and M. E. Pollard, “Optimization of SERS enhancement from nanostructured metallic substrate based on arrays of inverted rectangular pyramids and investigation of effect of lattice non-symmetry,” Proc. SPIE 8234, 823406, 823406-7 (2012). [CrossRef]
26. M. F. A. Muttalib, S. Z. Oo, and M. D. B. Charlton, “Experimental measurement of photonic/plasmonic crystal dispersion, applied to the investigation of surface plasmon dispersion for SERS sensing applications,” Proc. SPIE 8264, 82641C (2012).
