SERS in Plain Sight: A Polarization Modulation Method for Signal

Subscriber access provided by EKU Libraries

Article

SERS in Plain Sight: A Polarization Modulation Method for Signal Extraction Pietro Strobbia, Tyjair Sadler, Ren Abelard Odion, and Tuan Vo-Dinh Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b04360 • Publication Date (Web): 24 Jan 2019 Downloaded from http://pubs.acs.org on January 25, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

SERS in Plain Sight: A Polarization Modulation Method for Signal Extraction Pietro Strobbia a,b, Tyjair Sadler a,c, Ren A. Odion a,b and Tuan Vo-Dinh a,b,c* a. Fitzpatrick Institute for Photonics, Duke University, Durham, NC 27708, USA b. Department of Biomedical Engineering, Duke University, Durham, NC 27708, USA c. Department of Chemistry, Duke University, Durham, NC 27708, USA * corresponding author: [email protected] Abstract Surface-enhanced Raman spectroscopy (SERS) is a powerful

analytical

spectroscopy

offering

advantages ranging from “vibrational fingerprints” to multiplexed detection. However, the use of this technique in real-world applications has been limited due to difficulties in detecting inherently weak Raman signals often embedded in strong interfering background signals. A variety of plasmonics-active platforms have been developed to increase Raman signals but are not sufficient to extract weak SERS signals from intense interfering background signals. Herein, we describe a practical method, referred to as Polarization Modulation SERS (PM-SERS), which utilizes the polarization dependence of anisotropic SERS-active nanostructures to modulate the plasmonic effect to extract SERS signals and remove background. The modulation is obtained by switching the polarization of the excitation source at a specific frequency involving addition of only few optical components such as liquid crystal polarizers to a typical Raman setup. In this work, we characterized the polarization-dependent response of the SERS substrates fabricated using the oblique angle evaporation (OAV) technique and their response under laser excitation using a polarization modulated source. We demonstrated that the PM-SERS method can extract the analyte weak SERS signals from the strong interfering background signal in different situations, involving a fluorescent sample and a strong background light, and we show the possibility of using PM-SERS at a quasi-real time rate (0.5 Hz). We believe that the PM-SERS method will help expand the translation of applications that utilize SERS-substrates to real-world settings.

Keywords: Surface-enhanced Raman scattering (SERS), nanoposts, polarization modulation, background subtraction

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 14

Introduction After its discovery in the 1970s, surface-enhanced Raman scattering (SERS) has become a powerful analytical technique thanks to the practical application and further understanding of this phenomenon unveiled in the following decades.1-5 SERS offers many advantages, including specific chemical information and multiplex detection (i.e., simultaneous detection of multiple analytes).5, 6 These attributes make SERS an exceptional tool for a wide variety of bio-analytical applications ranging from biomedicine to defense.7-17 However, due to the inherently low cross-section of Raman scattering, SERS applications require sensing platforms with large enhancements (i.e., the Raman signal amplification) to be useful in real-world scenarios. furthermore, the low signal-to-background ratio of Raman limits field applications of SERS.

With the goal of achieving high enhancement factors (EF), many different SERS substrates have been developed in this and other laboratories, as reported in numerous reviews.18-22 The EF of a substrate depends on the structure/shape of the nanostructures that produce the SERS phenomenon. More than three decades ago, our laboratory reported the first practical use of SERS for trace analysis using a special platform called the “nanowave”, a SERS substrate based on a silver metal film coated on nanosphere arrays.5 Other metal film on nanoparticles included substrates base on quartz nanoposts,23 and silica nanoparticles.24 These structures take advantage of structural motifs as sharp protrusions and interstitial gaps to achieve a strong SERS enhancement, due to “lightning rod effect” (i.e. charge density gradients) and coupling between closely spaced localized surface plasmons, respectively.4, 25-27 Additionally, multi-layered materials have also been observed to improve the enhancement factors in substrates with respect to single-layered analogous.28-31 Since reproducible nanostructures that include the nano-gaps motifs are hard to fabricate in a scalable fashion, substrates based on sharp protrusions, such as nanostars, nanorods and nanoposts arrays, have attracted interest as highly enhancing SERS substrates for analytical applications.24, 32-35

Oblique angle evaporation (OAV), also known as oblique angle deposition (OAD), was first used to deposit nanoparticles onto stochastic quartz posts to produced effective SERS-active substrates, which exhibited stronger SERS enhancement compared to island films and crossed-grating structures.36, 37 The OAV method was later used with thermal evaporation of silver on nanoposts substrates at angles close to 90º with respect to the normal of the surface.38 The glancing angle at which the deposited material encounter the surface produces an array of posts all oriented towards the evaporation source. These nanoposts were demonstrated to be efficient SERS substrates for analytical applications, showing EF of 108.38, 39 In addition to generate a large SERS enhancement, the anisotropic and ordered nature of nanorod structures make them sensitive to the polarization of the incident light.40, 41 Other reports have demonstrated the polarization response of SERS signals in various ACS Paragon Plus Environment

Page 3 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


structures, for nano-gaps (e.g., nanoparticles dimers) and anisotropic nanoparticles.42-44 Previous work by Brolo’s group has used the SERS polarization response of plasmonic micro-arrays to subtract unwanted background from Raman spectra.45-47

Although many effective substrates have been developed over the years, an obstacle for the widespread use of SERS in realworld applications is the difficulty to observe the Raman signal over the background in complex media or field conditions. Shifted-excitation Raman difference spectroscopy (SERDS) and modulated Raman spectroscopy are techniques developed to fix extract Raman signal from a complex background. Both these techniques employ the modulation of the wavelength of the Raman signal to remove the background.48-50 While wavelength modulation techniques are effective for the removal of background signal, the instrumentation needed for their realization is not readily accessible (mostly lab-built lasers) and require to substitute the excitation source in the Raman setup. Here, we describe a method that utilizes the polarization dependence of anisotropic plasmonics-active nanostructures to modulate the SERS response for signal extraction and background removal. The modulation is obtained by switching the polarization of the excitation source at a known frequency, producing a modulated SERS response from the substrate. Firstly, we characterized the polarization-dependent response of the SERS substrates fabricated with the OAV technique, to optimize the polarization-dependence of the substrate. We then characterize the frequency dependent response of the substrates under laser excitation with a polarization modulated source. Finally, we illustrate the usefulness of this method to extract signal under different scenarios involving a Raman analyte within a strongly fluorescing sample and illumination light background. The method we developed will benefit the translation of SERS-based applications to real-world settings.

Experimental methods Materials Silver pellets (99.999% pure) and and Ti shots used in the deposition were obtained from Kurt J. Lesker. Gold(III) chloride trihydrate (HAuCl4·3H2O), L(+)-ascorbic acid (AA), trisodium citrate dihydrate, sodium borohydride (NaBH4), 1 N hydrochloric acid solution (HCl), mercaptobenzoic acid (MBA) were purchased from Sigma-Aldrich at the highest purity grade available. Silver nitrate (AgNO3, 99.995%) was supplied by Alfa Aesar. Ammonium hydroxide (NH4OH, 29.5%) and absolute ethanol were obtained through VWR. All glassware and stir bars were thoroughly cleaned with aqua regia and dried prior to use. Ultrapure water (18 MΩ·cm) was used in all preparations.

Nanopost substrate fabrication ACS Paragon Plus Environment


Page 4 of 14

The posts were fabricated on silicon wafers (50 mm; University Wafer) in a 2 steps process modified from a previously reported OAD nanopost fabrication.33 Initially, a wafer was coated with 5 nm of Ti and 200 nm of Ag using e-beam evaporation in a PVD system (PVD75; Kurt J. Lesker). The coated wafer was then cleaved in rectangular pieces with sides of of approximately 10×20 mm. One of the pieces was then placed at 90o with respect to the horizontal axis and displaced of 2 mm on the horizontal axis with respect to the center of the source position, generating an angle between the surface and the source of approximately 86o.38 Using this geometry, the square was coated with different thicknesses of silver forming the nanoposts array. The silver deposition thickness was measured by mass with a quartz crystal microbalance.

Nanostars synthesis Gold Nanostars (AuNS) were synthesized with a previously described procedure.32 Briefly, 12nm gold seed solution was first prepared using a modified Turkevich method. AuNS were then synthesized by the simultaneous addition of 50 μL of 2 mM AgNO3 and 50 μL of 0.1 M ascorbic acid to a solution containing 10 mL of 0.25 mM HAuCl4, 10 μL of 1 N HCl, and 100 μL of the 12 nm gold seed solution under gently stirring at room temperature. The process was completed in less than a minute along with color change from a light orange to dark blue within 10 seconds, indicating formation of AuNS. The silver-coated gold nanostars (AuNS@Ag) were prepared as previously described.51 For synthesis of AuNS@Ag, unfunctionalized AuNS were kept stirring and 50 µL of AgNO3 0.1 M and 10 µL of NH4OH were added to the solution. The color of the solution changed from blue to dark brown. The obtained solution was used for further functionalization without purification. The AuNS@Ag were functionalized 2 h after the synthesis.

Optical setup A backscattered Raman setup was used to measure the SERS signal. A 632.8 nm HeNe laser (Melles-Griot) was

Figure 1. Schematic diagrams of the optical setups used in the experiments. (a) Setup used for the substrate charactrizetion (b) Optical setup used for the polarization modulation experiments.

passed through a laser line filter (Thorlabs) and reflected ACS Paragon Plus Environment

Page 5 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


on a notch filter (Thorlabs). The light was then focused on a sample using a 10× objective. The light scattered from the sample was collected though the same optics, passed through an additional notch filter and collected into the spectrometer using a lens (f/4; Thorlabs). The spectrometer used was an Acton 2300 (Princeton Instruments) coupled to a PiMAX CCD camera (Princeton Instruments). The spectra acquisisition was controlled using Winspec 32 (Roper Scientific). Figure 1a shows the optical setup used in the characterization of the nanopost substrates. The laser source is passed through a polarizer (Thorlabs) kept in a labeled rotational mount to permit the control of the laser polarization axis. The setup used in the polarization modulation studies is shown in Figure 1b. In this case, the laser source was passed though the polarizer and then through a half-wave liquid crystal variable retarder (LCC1111-A; Thorlabs). The liquid crystal controller (LCC25; Thorlabs) produced a square-wave at a set frequency and optimized amplitude. By orienting the polarizer at 45º from the slow axis of the halfwave retarder, the polarization is shifted of 90º between the two voltages of the square-wave. To observe the Raman signal, both the nanostars and the nanoposts were incubated in 2 mM solution of MBA in ethanol.

Results and Discussion Substrate polarization dependence characterization To characterize the polarization-dependent response of the substrates fabricated through oblique angle deposition, we functionalized the surface of the substrates with a Raman reporter (i.e., MBA) and detected the SERS signal from the reporter using an incident excitation laser of varying polarization. These studies were performed using the optical setup with a rotatable polarizer (Figure 1a). In the results, the polarization angle is calculated with respect to the long axis of the substrate. The nanoposts on the substrate are oriented at 90º with respect to this axis due to the substrates orientation during the deposition, as confirmed in SEM micrographs. The polarization response of the substrates was quantified as polarization ratio defined as the peak height of the Raman reporter at 0º polarization incidence over that of 90º. Figure 2 shows one of the Raman peaks characteristic of MBA obtained from the nanopost substrate

Figure 2. SERS spectra from an aligned nanopost substrate measured with incident laser light polarized at different angles with respect to the post orientation. The blue spectra are relative to the polarization oriented along the axis of the posts (0 and 180o), the green spectra are relative to a 45 o shift respect to the post axis (45, 135, 225 and 315 o) and the red spectra are relative to a 90 o shift (90 and 270 o).

(silver thickness ≈ 1500 nm). As it can be observed, the ACS Paragon Plus Environment


Page 6 of 14

spectra obtained with polarization parallel to the long axis of the substrate (0, 180 and 360º) show a stronger SERS signal with respect to those having perpendicular (90 and 270º) and oblique (45, 135º) polarizations. These results are consistent with previous results reported on the polarization dependence of this type of substrates.41

To optimize the substrates for their application in PM-SERS, we investigated the polarization response as a function of the thickness of silver deposited to fabricate the nanoposts. To this end, we deposited silver in increasing thicknesses (measured by mass) and characterized the resulting substrates via SEM and by measuring the polarization ratio. Figure 3 shows the SEM micrographs of the posts for different deposition thicknesses. The nanopost diameter and height increased with an increase in deposition thickness, in agreement with what was previously reported for this type of substrates.39 The polarization as a function of the deposition thickness is reported in Figure 3b. The polarization ratio has a non-monotonic behavior and reaches a maximum between 900 and 1500 nm of deposition thickness, showing that the nanopost height is not the only factor determining the polarization response of a substrate. We believe that this behavior could be explained by the different

Figure 3. Scanning electron micrographs of nanoposts for deposition thicknesses of 532nm (a), 750nm (b), 1094nm (c) and 1800nm (d). e. Plot of the polarization ratio of the nanopost substrates as a function of deposited silver thickness. mechanisms by which the polarization response acts in these structures. The polarization response arises from interaction of surface plasmons excited on adjacent posts.39 With an increase in the size of the posts, the anisotropy in the substrate increases generating a sharper difference between the interactions among posts on the parallel axis with respect to the normal axis. However, while anisotropy increases, the simultaneous increase in diameter results in reducing the points in which adjacent posts interact in the area illuminated by the laser. The combination of these competing factors produces the behavior observed in the polarization ratio data (Figure 3b).


Page 7 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The concept of background subtraction via polarization modulation We exploited the polarization properties of the substrates described in the previous section to remove unwanted background from SERS spectra through PM-SERS, a signal modulation and frequency filtering. The SERS signal extraction is achieved by illuminating the substrate with a laser source that had its polarization modulated to periodically switch from parallel to perpendicular with respect to nanopost orientation. The laser is first passed through a polarizer and then is passed through a

Figure 4. Simulated data showing the concept behind the background subtraction mechanism used on the polarization dependent nanoposts. a. Original SERS data with background fluorescence as a function of Raman shift. b. Signal as a function of time for the two wavelengths (on- and off-peak) highlighted in the spectrum (a). c. Signal reconstructed after the background removal through filtering in the frequency domain. d. Signal as a function of frequency for wavelength 1 e. Signal as a function of frequency for wavelength 2. half-wave liquid crystal variable retarder, which was setup to slow the polarized beam to achieve a phase shift of 90º (Figure 1b). Figure 4 illustrates the basic concept of polarization modulation for interfering background removal. Considering a spectrum with modulated SERS signal and a constant background (Figure 4a), the SERS signal should display a modulation intensity in the time domain unlike the background signal (Figure 4b). In the frequency domain, the constant background and high frequency noise are distinguishable from the modulated SERS signal (Figure 4d,e). Since the SERS modulation frequency is known, it is possible to filter out the unwanted background from the modulated signal by selecting the correct frequency in the Fourier-transform of each wavelength of the spectrum, generating a background-free Raman spectrum



Page 8 of 14

(Figure 4c). While modulated Raman spectroscopy methods have previously been exploited for background removal, these methods have usually involved a modified laser source to produce the wanted signal modulation. The advantage of this method is to rely solely on optics (i.e., a polarizer and a variable half-wave retarder), which makes it transferable to possibly any optical setup.

Figure 5. Nanoposts (top). Intensity map of the SERS signal vs Raman shift (x axis) and vs time (y axis) (left), representative SERS spectrum for a single frame (center) and SERS signal as a function of time for an on-peak (II) and off-peak wavelgnth (III) (right), for a nanoposts substrate. Nanostars (bottom). Intensity maps of the SERS signal vs Raman shift (x axis) and vs time (y axis) (left), representative SERS spectrum for a single frame (center) and SERS signal as a function of time for an on-peak and off-peak wavelength (right), for nanostars in solution. The Intensity maps report the time frame at which the representative spectra (b, e) were taken (I), and the wavelengths for which the signal is reported as a function of time in c and f (II, III). The maps also show the time at which the polarization modulation was turned on (retarder on). To show that our system was exclusively modulating SERS through polarization modulation, we detected the SERS signal from MBA coated on the polarization-dependent nanoposts and on nanostars in solution (used as a polarization-independent reference). In both cases, we monitored the signal over time (500 spectra in 100 s) initially at constant polarization and after the activation of the half-wave retarder. Figure 5 shows the intensity map of the signal over time (left), a sample spectrum (center) and the signal over time from a SERS peak and from the background (right), for the two cases of a nanoposts substrate (top panel) and nanostars in solution (bottom panel). For the case of the nanoposts, after the activation of the polarization modulation the signal starts to vary following the polarization of the laser source. This result can be observed in the stripes present in the intensity map (left), as well as in the signal over time graph (right). We set the modulation to 0.5 Hz for


Page 9 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


visualization purposes and the signal was observed to change sharply every 2 s for analytes on the nanopost substrate, in agreement with the modulation frequency (Figure 5 top right). On the other hand, there is no modulated change in the SERS signals of analytes in the nanostars solution (Figure 5 bottom right). As it results clear from both the map and the signal over time graph, both the background and the SERS peak depends on the modulated polarization. This observation can be explained by the presence of scattering/SERS background in the spectrum. All the scattering phenomena are enhanced on the surface of the substrates and they follow the same enhancement rules. Thereby, while the SERS signal from MBA is enhanced with excitation sources perpendicular to the post axis, the same stands for light scattered from the substrate surface and SERS background signal.

Fluorescence background subtraction via polarization modulation We used the developed SERS-signal modulation method to extract SERS peaks from an interfering fluorescence background signal. To obtain a strong background signal, we placed over a functionalized nanoposts substrate a 1 µM solution of Cyanine 5 dye, which exhibits a strong fluorescence signal under 633 nm excitation. To avoid direct interaction between the dye and the substrate surface, Cyanine 5 is separated from the substrate by a cover slip. Then, we detected the signal from the substrate using the excitation laser with modulated polarization. Figure 6 shows the results from this experiment. The raw intensity map on the left of the Figure 6 (a) shows the signal as a

Figure 6. a. Intensity map of the SERS signal vs Raman shift (x axis) and vs time (y axis) for the substrate under a fluorescent solution. b. Intensity map of the SERS signal vs Raman shift (x axis) and vs time (y axis) after filtering in the frequency domain. c. Representative SERS spectrum for a single frame of the raw signal. d. Spectrum reconstructed from the amplitudes at the modulation frequency for each wavelength.

function of time and Raman shift. In the low Raman shift region of the spectrum, it can be distinguished the fluorescence broad peak, which has a relatively constant intensity and does not follow the modulation frequency. The MBA peaks can be noticed at 1075 and 1588 cm-1 with their intensity oscillating following the polarization modulation. A sample spectrum is also shown in Figure 6 (c), highlighting the features of the raw spectra (i.e., fluorescence background and SERS peaks). The series of raw spectra were filtered in the frequency domain (see Polarization modulation concept section) to isolate the modulated SERS signal. The resulting spectra are shown in the intensity map on the right (b), showing the filtered series of spectra as a function of time and Raman shift. It is noteworthy ACS Paragon Plus Environment


Page 10 of 14

that the amount of signal extracted is proportional to polarization ratio, because that is the part of the signal subject to polarization-modulation. Thus, structures with large polarization ratio are more efficient as substrates for PM-SERS. Thorough this process the constant fluorescence background is removed and the filtered series of spectra only shows the MBA peaks and the SERS background. A filtered spectrum is reconstructed from the signal amplitudes for modulation frequency at each wavelength of the spectrum and the resulting spectrum is shown in Figure 6 (d). As it can be observed the fluorescence background was completely removed, enhancing the signal-to-background ratio. Furthermore, while the peak around 1350 cm-1 relative to the COO stretch of MBA is not visible in the raw spectrum (c), this peak is easily observable in the reconstructed spectrum (d) due to the signal extraction capabilities of this technique. It is important to note that intensitybased modulation techniques cannot achieve the results achieved herein because the fluorescence background is also modulated by changes in excitation intensity.

Background illumination subtraction via polarization modulation To further test the background subtraction capabilities of the developed method in a different scenario, we detected the signal from a functionalized substrate in the presence of a strong background light, simulating the use of s SERS substrate in an illuminated room. A white LED light was directed towards the entrance slit of the spectrometer to generate the background signal. Figure 7 shows the results from this experiment. The raw intensity map is in Figure 7 (a), displaying the signal as a Figure 7. a. Intensity map of the SERS signal vs Raman shift (x axis) and vs time (y axis) for the substrate with LED background illumination. b. Intensity map of the SERS signal vs Raman shift (x axis) and vs time (y axis) after filtering in the frequency domain. c. Representative SERS spectrum for a single frame of the raw signal. d. Spectrum reconstructed from the amplitudes at the modulation frequency for each wavelength.

function of time and Raman shift. In the low Raman shift region of the spectrum, it can be observed the strong background from the LED light (Figure 7c). As observed in the previous study, the background is constant, while the MBA peaks at 1075 and 1588 cm-1 oscillates following the polarization modulation. The

filtered series of spectra is shown in the intensity map on the right of Figure 6 (b), plotting the filtered series of spectra as a function of time and Raman shift. The filtered series of spectra only shows the MBA peaks and the SERS background. The resulting background subtraction can be observed in the spectra shown in the bottom of Figure 7 (c,d). The raw spectrum (c)


Page 11 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


shows only one of the MBA peaks at 1588 cm-1, while the reconstructed spectrum (d) shows the characteristic MBA peaks at 1075, 1350, and 1588 cm-1, demonstrating the possibility of extracting the SERS signal in the presence of a background light.

For completeness, we also compared the average spectrum from the series used to produce the spectrum reconstructed after the filtering process with the reconstructed spectrum. Figure 8 shows the average spectrum (a) from the series of raw spectra shown in Figure 7 and the reconstructed spectrum (b) from the same series. As it can be observed,

Figure 8. a. Average SERS spectrum from the series of spectra in Figure 7 plotted in the intensity map. b. Spectrum reconstructed from the amplitudes at the modulation frequency for each wavelength.

the average spectrum shows 2 MBA peaks (i.e., 1075 and 1588 cm-1), however it also shows the strong background signal. In the reconstructed spectrum, the background signal is removed and the 1350 cm-1 peak from MBA is also visible. These results show how this method can effectively extract SERS signal and provide additional information even when compared to the average of the series of spectra used in the modulation.

Fast polarization modulation and signal reconstruction To demonstrate the possibility of using PM-SERS for online background subtraction, we repeated the background extraction at a higher frequency. This study used 10 Hz modulation frequency for the excitation and the spectra were collected at 100 Hz (i.e., 10 ms integration time). These parameters allowed for a series of 200 spectra, necessary to perform the background extraction, to be detected in 2 s. Figure 9 shows the signal reconstruction process for this fast PM-SERS study. As it can be observed, the original SERS signal (Figure 8a) can be reconstructed from 2 s of spectra collection (Figure 8b). While the signal is noisy due to the low integration time (10 ms per spectrum), the strong background from the LED emission was removed keeping the Raman peaks in the reconstructed spectrum. The noise of the reconstructed spectrum can be decreased

Figure 9. a. SERS spectrum taken on substrate in the presence of LED light at 10 ms integration time. b. Spectrum reconstructed from the amplitudes at the modulation frequency (10 Hz) for each wavelength of the 10 ms series of spectra. c. Spectrum from b smoothed with a Savitzky–Golay filter.



Page 12 of 14

by a simple smoothing procedure (Figure 8c). These results show how the PM-SERS technique can be used as an online background subtraction technique to remove unwanted background in a quasi-real time fashion.

Conclusion Many different applications for SERS substrates have been developed over the years; however, the use of SERS in real-world applications has been limited, due to the difficulties in observing weak Raman signals in strong interfering background. Several techniques have been developed extraction of Raman signal from different type of interfering background. Herein, we developed a method for background subtraction and SERS signal extraction based on polarization-responsive substrates and a polarization-modulated excitation source. We characterized the polarization response of the substrates and demonstrated the possibility of modulating the SERS signal by using the substrates in combination with a laser source modulated by a variable half-wave retarder. Finally, we show how this technique can be used to successfully remove the background in two different scenarios: a fluorescent sample and background from LED illumination. Furthermore, we show how the technique can be used at speeds compatible with real-time monitoring of the Raman signal (i.e., 10 Hz modulation with a 0.5 Hz reconstruction response). While background subtraction through signal or wavelength modulation was previously achieved in Raman spectroscopy, we developed a technique that can achieve similar results in SERS signal extraction without the need of complex and expensive hardware (e.g., multi-wavelength laser or gated ICCDs). This manuscript is the first proof-ofprinciple demonstration of PM-SERS, a technique to extract SERS signal from a strong background, that will benefit SERS field applications, as well as photochemistry experiments involving SERS.

Acknowledgements This work was supported by the U.S. Department of Energy Office of Science, under Award Number DE-SC0014077 and DE-SC0019393.

References 1. 2. 3. 4.

Jeanmaire, D. L.; Van Duyne, R. P., Surface raman spectroelectrochemistry: Part I. Heterocyclic, aromatic, and aliphatic amines adsorbed on the anodized silver electrode. Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 1977, 84, 1-20. Kelly, K. L.; Coronado, E.; Zhao, L. L.; Schatz, G. C., The Optical Properties of Metal Nanoparticles: The Influence of Size, Shape, and Dielectric Environment. J. Phys. Chem. B 2003, 107, 668-677. Moskovits, M., Surface roughness and the enhanced intensity of Raman scattering by molecules adsorbed on metals. J. Chem. Phys. 1978, 69, 4159-4161. Moskovits, M., Surface-enhanced Raman spectroscopy: a brief retrospective. J. Raman Spectrosc. 2005, 36, 485-496.


Page 13 of 14 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28.


Vo-Dinh, T.; Hiromoto, M. Y. K.; Begun, G. M.; Moody, R. L., Surface-enhanced Raman spectrometry for trace organic analysis. Anal. Chem. 1984, 56, 1667-1670. Rodriguez-Lorenzo, L.; Fabris, L.; Alvarez-Puebla, R. A., Multiplex optical sensing with surface-enhanced Raman scattering: A critical review. Anal. Chim. Acta 2012, 745, 10-23. Wang, H.-N.; Crawford, B. M.; Fales, A. M.; Bowie, M. L.; Seewaldt, V. L.; Vo-Dinh, T., Multiplexed Detection of MicroRNA Biomarkers Using SERS-Based Inverse Molecular Sentinel (iMS) Nanoprobes. J. Phys. Chem. C 2016, 120, 21047-21055. Ngo, H. T.; Wang, H.-N.; Fales, A. M.; Vo-Dinh, T., Plasmonic SERS biosensing nanochips for DNA detection. Anal. Bioanal. Chem. 2016, 408, 1773-1781. Zhang, X.; Zhao, J.; Whitney, A. V.; Elam, J. W.; Van Duyne, R. P., Ultrastable Substrates for Surface-Enhanced Raman Spectroscopy: Al2O3 Overlayers Fabricated by Atomic Layer Deposition Yield Improved Anthrax Biomarker Detection. J. Am. Chem. Soc. 2006, 128, 10304-10309. Yuen, J. M.; Shah, N. C.; Walsh, J. T.; Glucksberg, M. R.; Van Duyne, R. P., Transcutaneous Glucose Sensing by Surface-Enhanced Spatially Offset Raman Spectroscopy in a Rat Model. Anal. Chem. 2010, 82, 8382-8385. Farrell, M. E.; Holthoff, E., L.; Pellegrino, P., M. , Surface-Enhanced Raman Scattering Detection of Ammonium Nitrate Samples Fabricated Using Drop-on-Demand Inkjet Technology. Appl. Spectrosc. 2014, 68, 287-296. Penn, M. A.; Drake, D. M.; Driskell, J. D., Accelerated Surface-Enhanced Raman Spectroscopy (SERS)-Based Immunoassay on a Gold-Plated Membrane. Anal. Chem. 2013, 85, 8609-8617. Ngo, H. T.; Freedman, E.; Odion, R. A.; Strobbia, P.; Indrasekara, A. S. D. S.; Vohra, P.; Taylor, S. M.; Vo-Dinh, T., Direct Detection of Unamplified Pathogen RNA in Blood Lysate using an Integrated Lab-in-a-Stick Device and Ultrabright SERS Nanorattles. Sci. Rep. 2018, 8, 4075. Graham, D.; Thompson, D. G.; Smith, W. E.; Faulds, K., Control of enhanced Raman scattering using a DNA-based assembly process of dye-coded nanoparticles. Nat. Nanotechnol. 2008, 3, 548-551. Negri, P.; Choi, J. Y.; Jones, C.; Tompkins, S. M.; Tripp, R. A.; Dluhy, R. A., Identification of Virulence Determinants in Influenza Viruses. Anal. Chem. 2014, 86, 6911-6917. De Luca, A. C.; Reader-Harris, P.; Mazilu, M.; Mariggiò, S.; Corda, D.; Di Falco, A., Reproducible SurfaceEnhanced Raman Quantification of Biomarkers in Multicomponent Mixtures. ACS Nano 2014, 8, 2575-2583. Pallaoro, A.; Hoonejani, M. R.; Braun, G. B.; Meinhart, C. D.; Moskovits, M., Rapid Identification by SurfaceEnhanced Raman Spectroscopy of Cancer Cells at Low Concentrations Flowing in a Microfluidic Channel. ACS Nano 2015, 9, 4328-4336. Yuan, H.; Register, J.; Wang, H.-N.; Fales, A.; Liu, Y.; Vo-Dinh, T., Plasmonic nanoprobes for intracellular sensing and imaging. Anal Bioanal Chem 2013, 405, 6165-6180. Strobbia, P.; Languirand, E.; Cullum, B. M., Recent advances in plasmonic nanostructures for sensing: a review. Opt. Eng. 2015, 54, 100902-100902. Banholzer, M. J.; Millstone, J. E.; Qin, L.; Mirkin, C. A., Rationally designed nanostructures for surface-enhanced Raman spectroscopy. Chem. Soc. Rev. 2008, 37, 885-897. Vo-Dinh, T.; Dhawan, A.; Norton, S. J.; Khoury, C. G.; Wang, H.-N.; Misra, V.; Gerhold, M. D., Plasmonic Nanoparticles and Nanowires: Design, Fabrication and Application in Sensing. J. Phys. Chem. C 2010, 114, 74807488. Vo-Dinh, T.; Fales, A. M.; Griffin, G. D.; Khoury, C. G.; Liu, Y.; Ngo, H.; Norton, S. J.; Register, J. K.; Wang, H.-N.; Yuan, H., Plasmonic nanoprobes: from chemical sensing to medical diagnostics and therapy. Nanoscale 2013, 5, 10127-10140. Enlow, P. D.; Vo-Dinh, T., Detection of nitro-polynuclear aromatic compounds by surface-enhanced Raman spectrometry. Anal. Chem. 1986, 58, 1119-1123. Alak, A.; Tuan, V. D., Silver-coated fumed silica as a substrate material for surface-enhanced Raman scattering. Anal. Chem. 1989, 61, 656-660. Gersten, J.; Nitzan, A., Electromagnetic theory of enhanced Raman scattering by molecules adsorbed on rough surfaces. J. Chem. Phys. 1980, 73, 3023-3037. Dhawan, A.; Canva, M.; Vo-Dinh, T., Narrow groove plasmonic nano-gratings for surface plasmon resonance sensing. Opt. Express 2011, 19, 787-813. Dhawan, A.; Du, Y.; Batchelor, D.; Wang, H. N.; Leonard, D.; Misra, V.; Ozturk, M.; Gerhold, M. D.; Vo‐Dinh, T., Hybrid Top‐Down and Bottom‐Up Fabrication Approach for Wafer‐Scale Plasmonic Nanoplatforms. Small 2011, 7, 727-731. Leverette, C. L.; Shubert, V. A.; Wade, T. L.; Varazo, K.; Dluhy, R. A., Development of A Novel Dual-Layer Thick Ag Substrate for Surface-Enhanced Raman Scattering (SERS) of Self-Assembled Monolayers. J. Phys. Chem. B 2002, 106, 8747-8755.



29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51.

Page 14 of 14

Li, H.; Cullum, B. M., Dual Layer and Multilayer Enhancements from Silver Film over Nanostructured SurfaceEnhanced Raman Substrates. Appl. Spectrosc. 2005, 59, 410-417. Strobbia, P.; Henegar, A. J.; Gougousi, T.; Cullum, B. M., Layered gold and titanium dioxide substrates for improved surface enhanced Raman spectroscopic sensing. Appl. Spectrosc. 2016, 70, 1375-1383. Farrell, M. E.; Strobbia, P.; Pellegrino, P. M.; Cullum, B., Surface regeneration and signal increase in surfaceenhanced Raman scattering substrates. Appl. Opt. 2017, 56, B198-B213. Hsiangkuo, Y.; Christopher, G. K.; Hanjun, H.; Christy, M. W.; Gerald, A. G.; Tuan, V.-D., Gold nanostars: surfactant-free synthesis, 3D modelling, and two-photon photoluminescence imaging. Nanotechnology 2012, 23, 075102. Chaney, S. B.; Shanmukh, S.; Dluhy, R. A.; Zhao, Y. P., Aligned silver nanorod arrays produce high sensitivity surface-enhanced Raman spectroscopy substrates. Appl. Phys. Lett. 2005, 87, 031908/1-031908/3. Greeneltch, N. G.; Blaber, M. G.; Henry, A.-I.; Schatz, G. C.; Van Duyne, R. P., Immobilized Nanorod Assemblies: Fabrication and Understanding of Large Area Surface-Enhanced Raman Spectroscopy Substrates. Anal. Chem. 2013, 85, 2297-2303. Asiala, S. M.; Schultz, Z. D., Characterization of Hotspots in a Highly Enhancing SERS Substrate. The Analyst 2011, 136, 4472-4479. Meier, M.; Wokaun, A.; Vo-Dinh, T., Silver particles on stochastic quartz substrates providing tenfold increase in Raman enhancement. J. Phys. Chem. 1985, 89, 1843-1846. Vo-Dinh, T.; Meier, M.; Wokaun, A., Surface-enhanced Raman spectrometry with silver particles on stochastic-post substrates. Anal. Chim. Acta 1986, 181, 139-148. Driskell, J. D.; Shanmukh, S.; Liu, Y.; Chaney, S. B.; Tang, X. J.; Zhao, Y. P.; Dluhy, R. A., The Use of Aligned Silver Nanorod Arrays Prepared by Oblique Angle Deposition as Surface Enhanced Raman Scattering Substrates. J. Phys. Chem. C 2008, 112, 895-901. Liu, Y. J.; Zhang, Z. Y.; Dluhy, R. A.; Zhao, Y. P., The SERS response of semiordered Ag nanorod arrays fabricated by template oblique angle deposition. J. Raman Spectrosc. 2010, 41, 1112-1118. Tao, A. R.; Yang, P., Polarized Surface-Enhanced Raman Spectroscopy on Coupled Metallic Nanowires. J. Phys. Chem. B 2005, 109, 15687-15690. Zhao, Y. P.; Chaney, S. B.; Shanmukh, S.; Dluhy, R. A., Polarized Surface Enhanced Raman and Absorbance Spectra of Aligned Silver Nanorod Arrays. J. Phys. Chem. B 2006, 110, 3153-3157. Jiao, J.; Wang, X.; Wackenhut, F.; Horneber, A.; Chen, L.; Failla, A. V.; Meixner, A. J.; Zhang, D., Polarization‐Dependent SERS at Differently Oriented Single Gold Nanorods. ChemPhysChem 2012, 13, 952-958. Greco, S.; Dal Zilio, S.; Bek, A.; Lazzarino, M.; Naumenko, D., Frequency Modulated Raman Spectroscopy. ACS Photonics 2018, 5, 312-317. Liu, K.-K.; Tadepalli, S.; Kumari, G.; Banerjee, P.; Tian, L.; Jain, P. K.; Singamaneni, S., Polarization-Dependent Surface-Enhanced Raman Scattering Activity of Anisotropic Plasmonic Nanorattles. J. Phys. Chem. C 2016, 120, 16899-16906. Anema, J. R.; Brolo, A., The use of polarization-dependent SERS from scratched gold films to selectively eliminate solution-phase interference. Plasmonics 2007, 2, 157. Anema, J. R.; Brolo, A. G., Use of polarization-dependent SERS from scratched gold films to monitor the electrochemically-driven desorption and readsorption of cysteine. J. Electroanal. Chem. 2010, 649, 159-163. Hong, K. Y.; Brolo, A. G., Polarization-dependent surface-enhanced Raman scattering (SERS) from microarrays. Anal. Chim. Acta 2017, 972, 73-80. Maiwald, M.; Erbert, G.; Klehr, A.; Kronfeldt, H. D.; Schmidt, H.; Sumpf, B.; Tränkle, G., Rapid shifted excitation Raman difference spectroscopy with a distributed feedback diode laser emitting at 785 nm. Appl. Phys. B 2006, 85, 509-512. Zhao, J.; Carrabba, M. M.; Allen, F. S., Automated Fluorescence Rejection Using Shifted Excitation Raman Difference Spectroscopy. Appl. Spectrosc. 2002, 56, 834-845. De Luca, A. C.; Mazilu, M.; Riches, A.; Herrington, C. S.; Dholakia, K., Online Fluorescence Suppression in Modulated Raman Spectroscopy. Anal. Chem. 2010, 82, 738-745. Fales, A. M.; Yuan, H.; Vo-Dinh, T., Development of Hybrid Silver-Coated Gold Nanostars for Nonaggregated Surface-Enhanced Raman Scattering. J. Phys. Chem. C 2014, 118, 3708-3715.


SERS in Plain Sight: A Polarization Modulation Method for Signal

Recommend Documents