The Localized Surface Plasmonic Resonance and Sensing Properties

Subscriber access provided by UNIVERSITY OF CONNECTICUT

C: Plasmonics, Optical Materials, and Hard Matter

The Localized Surface Plasmonic Resonance and Sensing Properties of Ag-MgF Composite Nanotriangles 2

Steven Robert Larson, and Yiping Zhao J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b00122 • Publication Date (Web): 15 Mar 2018 Downloaded from http://pubs.acs.org on March 19, 2018

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 31 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 Journal of Physical Chemistry

The Localized Surface Plasmonic Resonance and Sensing Properties of Ag-MgF 2 Composite Nanotriangles Steven Larson* and Yiping Zhao

Department of Physics and Astronomy, University of Georgia, Athens, Georgia 30602 *Corresponding Author. E-mail: [email protected] Phone: 706/542-6230 Fax: 706/542-2492

1 ACS Paragon Plus Environment

The Journal of Physical 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 31

Abstract: Regular nanotriangular (NT) patterns and thin films composed of a mixture of Ag and MgF 2 with different composition ratios are prepared by combining nanosphere lithography and electron beam co-deposition. The plasmonic property of Ag-MgF 2 composites with a high Ag composition C Ag ( ≥ 90 at.%) are shown to be a function of C Ag as well as the size of the NTs, while for samples with low C Ag (< 90 at.%), a nearly constant localized surface plasmon resonance (LSPR) peak appears in all samples, regardless of C Ag , NT size, or thin film, which is confirmed to be due to Ag NPs formed during the deposition. Thus, the LSPR property of the composite NTs can be tuned by C Ag when C Ag ≥ 90 at.%. The resulting LSPR sensor at C Ag = 90 at.% with 500 nm-diameter polystyrene nanosphere monolayer can achieve a sensitivity of 696 RIU/nm, as compared to 312 RIU/nm for the same NTs with pure Ag. This significantly improved sensitivity is due to the modified dispersion relationship of the dielectric constant by the metal-dielectric composite, and will play an important role in future plasmonic material design and applications.


Page 3 of 31 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


Introduction Localized surface plasmon resonance (LSPR) based sensors, whose resonance absorbance wavelength, λ c , responds to the changes in the local dielectric environment, have attracted great attention and have been widely studied over the past decade.1-3 These sensors are compact, durable, repeatable, and more reliable than traditional sensors, offering real-time and label-free chemical and biological detection.2, 4-5 They are characterized by two important parameters, the refractive index sensitivity (RIS), i.e., the shift of the resonance wavelength (Δλ) with respect to the local index of refraction changes (ΔRI), RIS = Δλ/ΔRI, and the figure of merit (FOM), i.e., RIS divided by the full-width at half-maximum (FWHM) of the localized plasmon absorbance peak, which quantifies the ease of recognition of the spectral change. Research in the field of LSPR sensors has been focused on increasing these two parameters. Most studies of LSPR sensors tune the shape and size of plasmonic structures to improve RIS and FOM, including structures like thin films6, nanorods7, nanospheres8, nanopyramids9, nanohelix10, nanobranches/star11-12, nanotriangles13-14, and many others.5, 11, 15-17 These traditional approaches generally have RIS from 50 - 250 nm/RIU with FOMs in the single digits.18 However, with complex structures and fabrication techniques, some RISs have reached above 1000 nm/RIU.1, 1920

Tuning structure in this manner requires extremely complicated and low throughput methods

of fabrication, limiting the usefulness of the sensor. More recently, studies of composite materials have shown great potential in tuning plasmonic properties, in particular, the mixture of two noble metals, Ag-Au21, Ag-Cu22, and AuCu23-24. By varying their composition, these composites show promising tunability at nearly any wavelength between the two LSPR peaks of the individual materials. These mixtures offers a great possibility for improving the RIS and FOM of the corresponding LSPR sensor.25-26 One 3 ACS Paragon Plus Environment


Page 4 of 31

example of this tunability was demonstrated by Motl et al. who varied the composition of Au or Cu nanoparticles (NP) and changed their LSPR peak positions from 523 to 545 nm.24 Sharma and Gupta presented a theoretical study of Ag-Au alloy NP films and predicted that an alloy based sensor, whose structure was an alloy rather than NPs of one material suspended in a matrix of the other, could outperform all host-material only based LSPR sensors.27 Recently Jeong et al. showed that a Ti-Ag mixtures could significantly increase the RIS over that of a pure Ag structure. They demonstrated a RIS of ~1100 nm/RIU at a FOM of more than 2800 RIU-1 by using this Ti Ag mixture combined with a nano-helix morphology and a circular dichroism measurement strategy.10 However, most of the studies of the material systems have been limited to metal-metal mixtures. In fact, metal-dielectric composites could be another alternative to tune the LSPR of plasmonic material and improve sensing performance. Previously, intense studies of metaldielectric mixtures have involved metallic NPs (Ag or Au) suspended in a host matrix of dielectric rather than a true mixture, such as Ag or Au NPs embedded in matrixes of SiO 2 , MgF 2 , or polymers.28-29 While these mixtures with low metallic composition are interesting for some applications, they are ineffective for LSPR sensing for two reasons. First, only the metallic NPs that are exposed to air/environment interface can provide active sensing areas, and most NPs embedded inside the dielectric host do not contribute to the sensor. Therefore, the sensitivity of the NP-dielectric system is mainly determined by the exposed NPs, and is expected to be low. In fact, the reported RIS is around 160 nm/RIU.13, 29 Second, the LSPR peak wavelength depends only on the size and shape of the embedded NPs, as well as the host dielectrics. Therefore, they do not change appreciably with NP concentration, unless the NPs are so concentrated that the local electromagnetic field coupling between neighboring NPs becomes significant.29-31 4 ACS Paragon Plus Environment

Page 5 of 31 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


However, when the composition of Au or Ag increases and the metal becomes the host, it is expected that the LSPR properties of the nanostructures can be tuned by the guest composition. Taking a spherical nanoparticle as an example, its LSPR wavelength is determined by 𝑅𝑅𝑅𝑅(𝜀𝜀𝑚𝑚 ) = −2𝜀𝜀𝑑𝑑 , where 𝑅𝑅𝑅𝑅(𝜀𝜀𝑚𝑚 ) is the real part of the dielectric function of the metal and 𝜀𝜀𝑑𝑑 is the dielectric function of the local enviroment.32 Therefore the RIS is determined by the dispersion of 𝑅𝑅𝑅𝑅(𝜀𝜀𝑚𝑚 ) versus the wavelength. If

𝑑𝑑𝑑𝑑𝑑𝑑(𝜀𝜀𝑚𝑚 ) 𝑑𝑑𝑑𝑑

�

𝜆𝜆=𝜆𝜆𝑐𝑐

is small, the same change of

𝜀𝜀𝑑𝑑 would induce a larger change in 𝜆𝜆𝑐𝑐 . Thus, if one can tune the slope of the dispersion of

𝑅𝑅𝑅𝑅(𝜀𝜀𝑚𝑚 ) of the plasmonic material, one can increase the RIS. For most mixed material systems when one component dominates the other, a Maxwell-Garnett (MG) effective medium theory

(EMT) can be used to approximate the effective dielectric function,32 𝜀𝜀eff = 𝜀𝜀ℎ + 3𝑑𝑑𝑖𝑖 𝜀𝜀ℎ

𝜀𝜀𝑖𝑖 − 𝜀𝜀ℎ , 2𝜀𝜀ℎ + 𝜀𝜀𝑖𝑖 + 𝑑𝑑𝑖𝑖 (𝜀𝜀ℎ − 𝜀𝜀𝑖𝑖 )

(1)

where 𝜀𝜀ℎ and 𝜀𝜀𝑖𝑖 are the dielectric function of the host material and the inclusions, and 𝑑𝑑𝑖𝑖 is the

volume fraction of these inclusions. Taking Ag as the host material and MgF 2 as the inclusions, the representative 𝑅𝑅𝑅𝑅(𝜀𝜀𝑚𝑚 ) can be calculated for different Ag composition as shown in Fig 1.

Clearly, compared to pure Ag (100%), when small amount of MgF 2 is introduced into the material system, the slope

𝑑𝑑𝑑𝑑𝑑𝑑(𝜀𝜀𝑚𝑚 ) 𝑑𝑑𝑑𝑑

becomes smaller, which infers that if a LSPR sensor is made

of a Ag-MgF 2 mixture, it will have a larger RIS. Therefore, the RIS can be tuned by varying the MgF 2 composition. So far, no systematic studies have been carried out for the LSPR properties of metal-dielectric systems, especially at high metal content for which the shape and size of the fabricated nanostructure are expected to determine the optical properties, rather than the embedded NPs.



Page 6 of 31

Here we combined the nanosphere lithography and dual-source e-beam deposition process to fabricate composite Ag-MgF 2 nanotriangles with a wide range of composition. We demonstrated that these triangular nanopatterns possessed tunable plasmonic properties that could be easily controlled by their size and composition. We then showed that a mixed phase of Ag (90 at.%) and MgF 2 (10 at.%) could achieve a RIS of up to 696 nm/RIU, more than double the 312 nm/RIU of the same structure for pure Ag.

Experimental Materials Polystyrene nanospheres (PSNS) whose diameter D = 350 nm, 500 nm, and 750 nm (Polyscience, Lot # 660711, Lot # 679675, and Lot # 657567 respectively) were used to form the colloid monolayer onto clean glass slides (Gold Seal, Part# 301) and silicon wafers (University Wafer). Sulfuric acid (Fisher Scientific, 98%), ammonium hydroxide (Fisher Scientific, 98%), and hydrogen peroxide (Fisher Scientific, 30%) were acquired to clean the glass and silicon. Silver pellets (Plasmaterials, 99.99%) and magnesium fluoride pieces (Kurt J. Lesker, 99.9%) were purchased as the evaporation material. Ethanol (Sigma-Aldrich, 98%), toluene (Fisher Scientific, 99.8%), acetone (Fisher Scientific, 99.8%), and isopropanol (Fisher Scientific, 99.8%) were used for the colloid monolayer preparation and to remove residual PSNS from the substrates after Ag and MgF 2 deposition. Acetone, 1-hexanol (Tokyo Chemical Industry co., >98%), chloroform (J.T. Baker, 99%), carbon tetrachloride (Sigma-Aldrich, 99.9%), and toluene were obtained for LSPR sensing measurements. Deionized (DI) water (18 MΩ) was used throughout all the experiments. All chemicals and materials were used without further purification. 6 ACS Paragon Plus Environment

Page 7 of 31 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


Monolayer Fabrication Glass slides and silicon wafers were cut into small squares with dimensions of 2.5 cm × 2.5 cm and 1 cm × 1 cm, respectively. Glass substrates were washed in a heated piranha solution, 4:1 ratio of sulfuric acid to hydrogen peroxide, for 20 min. Silicon substrates were cleaned using the first step of the RCA method, heated in a 5:1:1 ratio of DI water, ammonium hydroxide, and hydrogen peroxide, for 20 min. All substrates were then thoroughly rinsed in DI water; substrates that were not coated with monolayers were then dried under a N 2 gas flow. D = 350, 500, and 750 nm PSNS monolayers were prepared using an air-water interface method as previously reported.22 Briefly, the desired PSNS solution was first diluted in DI water to a concentration of 0.01 w/v % and then washed several times via centrifugation. Next, the solution was further diluted to a 2:1 volume ratio of PSNS solution to ethanol. The resulting suspension was loaded into a syringe and droplets of PSNS solution were dispensed at a rate of 0.015 mL/min onto the surface of a tilted cleaned glass Petri dish (diameter of 10 cm) containing approximately 24 mL of DI water via a syringe pump. This process continued until a monolayer formed and covered the entire water surface. A Teflon ring was placed gently on the surface of the water to protect the monolayer film against adhering to the side wall of the glass Petri dish. The water level then was raised. Glass and silicon substrates were carefully slid below the monolayer film. Finally, the monolayer was lowered on the substrates by slowly pumping out the water with a peristaltic pump, followed by drying in the air overnight. The quality of the PS monolayers was very high, with < 0.8% defect percent as shown in large zoom-out SEM images of Fig. S1 of Supporting Information. Fabrication of the Ag-MgF 2 Composite Nanostructures and Thin Films



Page 8 of 31

The Ag-MgF 2 nanostructures and thin films were fabricated using a custom-built dual source electron deposition system (Pascal Technology). The Ag and MgF 2 crucibles were 17.8 cm apart from each other inside the chamber. The monolayer coated and uncoated substrates were mounted 53 cm above the crucibles and centered horizontally relative to the two crucibles. With respect to the substrates normal, the vapor incident angle of the Ag and MgF 2 were, 𝜃𝜃 = -

10° and 10°, respectively. After loading the samples, the deposition chamber was pumped down to a base pressure of < 1 × 10-6 Torr. The deposition rate and total thickness of each evaporation source were monitored by two quartz crystal microbalances (QCM) independently, such that the total deposition thickness, i.e., the sum of Ag and MgF 2 thickness, was fixed to be 60 nm, while keeping the total deposition rate R Ag + R MgF2 = 0.2 nm/s. The relative deposition rates of Ag 𝐶𝐶𝐶𝐶𝐶𝐶 ) of (R Ag ) and MgF 2 (R MgF2 ) were varied to achieve the estimated atomic composition of Ag (𝐶𝐶𝐴𝐴𝐴𝐴

0, 33, 50, 66, 70, 75, 80, 85, 90, 95, 97, and 100 at.%. During the deposition, the chamber

pressure was kept below 1 × 10-5 Torr, and the samples were rotated azimuthally at a speed of 15 rpm to ensure even mixing of Ag and MgF 2 . During the deposition, the substrate temperature increased to a maximum of 42oC. After the co-deposition, the substrates were allowed to cool to room temperature in vacuum before being taken out of the chamber. Then, the colloid template was removed from the substrate using Scotch tape, and any remaining PSNS residue was removed by rinsing in toluene, acetone, and isopropanol successively. Optical and Morphology Characterization The optical transmission spectra of the Ag-MgF 2 nanostructures were measured by an ultraviolet-visible spectrophotometer (UV-Vis, Jasco-750). The resulting transmission spectra were converted into extinction spectra. Scanning electron microscopy images and energy dispersive X-ray measurements of the nanostructures and thin films were taken by a field 8 ACS Paragon Plus Environment

Page 9 of 31 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


emission scanning electron microscope (SEM, FEI Inspect F). SEM images were analyzed by Image J software (NIH). The crystal structures of the thin films and nanostructures were characterized by a PANalytical X’Pert PRO MRD X-ray diffractometer (XRD) with a fixed incidence angle of 0.5°. The XRD scans of the thin films were recorded with a Cu Kα1 radiation (λ = 1.541 Å) in the 2α range from 20° - 80° with a step size of 0.010°. Ellipsometry measurements of the thin films were taken by a spectroscopic ellipsometer (M-2000, J.A Woollam Co., Inc.) at incident angles of 65°, 70°, 75°, and 80°, respectively, over a wavelength range of 370 - 1000 nm. Sheet resistance measurements of the thin film samples were taken by a Keithley 23700 multimeter and a custom four-point probe setup. LSPR Sensing Measurements LSPR sensing measurements were performed by immersing nanostructure samples in a quartz cuvette containing either acetone, 1-hexanol, chloroform, carbon tetrachloride, or toluene, then taking the transmission spectra over a wavelength range of 190 nm to 2500 nm. The resultant extinction spectra were then compared to those of the organic solvents, and any intrinsic peaks due to the solvent’s absorbance were removed. The LSPR peaks were then fit to determine the peak location and FWHM.

Results and Discussion Morphology and composition characterization Figs. 2 a-h show representative SEM micrographs of Ag-MgF 2 nanopatterns of different 𝑐𝑐𝑐𝑐𝑐𝑐 𝑐𝑐𝑐𝑐𝑐𝑐 compositions of Ag (𝐶𝐶𝐴𝐴𝐴𝐴 ) for D = 500 nm PSNS monolayer, where 𝐶𝐶𝐴𝐴𝐴𝐴 is the atomic



Page 10 of 31

compositions of Ag calculated from the deposition rate ratio. A complete set of SEM images are presented in the Supporting Information Fig. S2 for all PSNS templates and Ag compositions. As expected, all the samples consist of regular patterns of quasi-triangular shapes, nanotriangles (NTs), whose size is determined by the size of the PSNS and is consistent with previous reports.22 Some PS residue is also visible in the center of the hexagonal lattice; this residue is discussed further in the Supporting Information S3 and does not appreciably affect the optical properties of the sensor. The height of each NT, h p (defined in Fig. 2), is measured to be h p = 310 ± 10, 180 ± 10, and 145 ± 5 nm for PSNS with D = 750, 500, and 350 nm, respectively. The poor definition of the individual triangles shown in Fig. 2 is in fact due to the deposition configuration and the nature of Ag deposition. As stated in the experimental section, each source in the deposition system has a 10o incident angle with respect to surface normal of the substrate. During the deposition, while the substrate rotates continuously, the shadowing effect by this 10o incident angle as well as the small sticking coefficient of Ag vapor at room temperature causes the nanoparticle to grow around the triangles. This has been experimentally demonstrated by our previous publication and confirmed via Monte Carlo simulation.26 This demonstrates that all the nanopatterns produced with the same PSNS template are consistent in size and shape, 𝑐𝑐𝑐𝑐𝑐𝑐 independent of composition. However, for samples with 𝐶𝐶𝐴𝐴𝐴𝐴 between 33 and 85 at.%,

nanoparticles (NPs) were observed on the surfaces of the NTs with higher concentrations near the edge of each NT. Similarly, for thin films samples with the same Ag compositions, NPs were also observed as shown in Fig. S4 in the Supporting Information. These NPs are Ag NPs segregated from MgF 2 matrix during the deposition, as confirmed by Z-contrast SEM in our previous work,33 and has also been seen in SiO 2 -Ag codeposition.31 Figure 3(a) tracks the NP’s 𝑐𝑐𝑐𝑐𝑐𝑐 size d and density m as a function of the 𝐶𝐶𝐴𝐴𝐴𝐴 for D = 500 nm PSNS samples. Overall, m appears


Page 11 of 31 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


𝑐𝑐𝑐𝑐𝑐𝑐 to have a maximum when 𝐶𝐶𝐴𝐴𝐴𝐴 ≈ 50 at.% and decreases as the composition is dominated by one

of its constitute. However, the size of the Ag NPs in those samples is consistently between 15 and 35 nm, independent of the composition or template size. Apparently, the d of the NPs is

determined by the thermal dynamics of the Ag and MgF 2 system during the deposition while the m of the NPs is determined by the Ag composition. The NP size was further confirmed by XRD crystal domain analysis presented in the Supporting Information S5. The compositions of the nanostructures were experimentally determined by EDS measurements on the thin film samples deposited on Si substrates. Here we assume that the composition of the thin film samples and the NT samples is the same since they were prepared 𝑀𝑀 𝑀𝑀 simultaneously. Only the composition of Ag and Mg (𝐶𝐶𝐴𝐴𝐴𝐴 and 𝐶𝐶𝑀𝑀𝑀𝑀 ) were tracked. The

composition of fluorine was not considered due to its small Z value and inaccuracy of the 𝐸𝐸𝐸𝐸𝐸𝐸 determination of its content by EDS. 𝐶𝐶𝐴𝐴𝐴𝐴 was calculated by, 𝐸𝐸𝐸𝐸𝐸𝐸 𝐶𝐶𝐴𝐴𝐴𝐴

=

𝑀𝑀 𝐶𝐶𝐴𝐴𝐴𝐴

𝑀𝑀 𝑀𝑀 𝐶𝐶𝐴𝐴𝐴𝐴 + 𝐶𝐶𝑀𝑀𝑀𝑀

.

(2)

𝐸𝐸𝐸𝐸𝐸𝐸 𝑐𝑐𝑐𝑐𝑐𝑐 These 𝐶𝐶𝐴𝐴𝐴𝐴 were then compared to the 𝐶𝐶𝐴𝐴𝐴𝐴 , and the results are plotted in Fig. 3(b). A line

𝐸𝐸𝐸𝐸𝐸𝐸 𝑐𝑐𝑐𝑐𝑐𝑐 = 𝐶𝐶𝐴𝐴𝐴𝐴 is also drawn for reference. The measured compositions are consistent representing 𝐶𝐶𝐴𝐴𝐴𝐴 𝑐𝑐𝑐𝑐𝑐𝑐 with those calculated from the deposition rates, with slightly more offset towards the lower 𝐶𝐶𝐴𝐴𝐴𝐴 𝐸𝐸𝐸𝐸𝐸𝐸 𝑐𝑐𝑐𝑐𝑐𝑐 values. The reason that most 𝐶𝐶𝐴𝐴𝐴𝐴 < 𝐶𝐶𝐴𝐴𝐴𝐴 could be attributed to the low sticking coefficient of

Ag during the deposition as compared to MgF 2 .26 Low melting point metals generally have

significantly lower sticking coefficients and higher surface mobilities in thin film deposition than those of dielectrics. Due to the consistency between predicted composition and EDS cal measurements, and to avoid any confusion, all C Ag will now refer to 𝐶𝐶Ag .

ACS Paragon Plus Environment

11


Page 12 of 31

The sheet resistance of the thin film samples was measured to determine the metaldielectric percolation threshold of the material system. Figure 3(b) plots the resistivity, ρ, of the thin film samples as a function of the C Ag (The nanostructured samples were not measured as the gaps in-between the NTs do not allow significant conduction). Generally, as the C Ag decreases the resistivity increases until the Ag no longer percolates through the sample and the resistivity exceeds the measurement limit of our setup (~ 5×105 Ω·μm). The measured resistivity of the C Ag = 100 at.% thin film sample was 7 ± 4 × 10-2 Ω·μm which is consistent with the bulk value of Ag as well as other studies on Ag thin films.34 As the C Ag decreases, the resistivity increases to 1.0 ± 0.4 × 10-1 Ω·μm, then 3.5 ± 0.5 × 10-1 Ω·μm at C Ag = 97 and 95 at.% respectively. For the thin film at C Ag = 90 at.%, the resistivity continues to increase to 0.54 ± 0.02 Ω·μm, after which the resistance follows an exponential relationship, increasing drastically to 1.32 ± 0.02 Ω·μm at C Ag = 85 at.% until 27 ± 0.02 Ω·μm for C Ag = 66 at.% after which was not measurable by our equipment. This trend is consistent with other studies on percolations in composite materials.35 From these measurements, it is difficult to determine a threshold value. However, if we compare the SEM images to the resistivity results, we see that at C Ag = 90 at.% Ag NP formation starts to dominate the sample while the resistivity increases past 1 Ω∙μm, a similar threshold is also observed in the optical measurements. Optical properties of Ag-MgF 2 composite Figures 4(a)-(c) show the representative extinction spectra for the NT samples obtained from different PSNS templates for high C Ag . Spectra for all samples are presented in Figs. S6 and S7 of the Supporting Information. Overall, for the high C Ag samples, we observe a strong LSPR peak that redshifts as the C Ag decreases. This peak also redshifts as the template size increases. As the C Ag drops below the threshold observed in the SEM and resistance 12 ACS Paragon Plus Environment

Page 13 of 31 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


measurements, i.e., < 90 at.%, this strong peak vanishes and is replaced by a peak at ~ 500 nm for all the samples regardless of template size. This new peak slightly blue shifts as the C Ag continues to decrease in intensity until C Ag reaches 0 at.% where it vanishes. The polarization dependent spectra were also measured and are presented in Fig. S8 of the Supporting Information. Although we expected to see a C6 symmetry due to the hexagonal structure of the lattice, no significant polarization dependence was observed. This is due to the relatively large measurement spot size (~ 2×5 mm2) of the UV-Vis setup compared to the monolayer domain size (200 μm2), i.e., during the measurements, there are hundreds of randomly distributed domains were sampled, which washed out any C6 symmetry. Specifically, for the D = 750 nm PSNS template, shown in Fig 4(a), we observe that for the sample at C Ag = 100 at.% there is one main peak at λ o = 1033 nm which corresponds to a strong dipole resonance. This is consistent with other studies on nanosphere lithography and free NTs indicating that the NTs do not couple significantly with each other.14, 26 The additional broad shoulder on the visible side of this peak extends from 650 to 1000 nm, which can be explained by a weaker in-plane and out-of-plane quadrupole resonances.22, 26 There is also a small peak at λ o = 503 nm, which could be due to Ag NPs formed in between the NTs as shown in Fig. 2(a).36 The peak at λ = 265 nm is due to the Ag inter-band transitions and is present in all samples.37 When the C Ag decreases to 97, 95, and 90 at.%, the main dipole peak monotonically redshifts to λ o = 1204, 1327, and 1537 nm, respectively. The peak also significantly broadens as it redshifts. The peak intensity first increases for C Ag = 97 at.%, then decreases as C Ag decreases. The shoulder stays consistent for all the higher C Ag samples. When C Ag < 90 at.% there is an abrupt change and the main peak at λ o = 1537 nm vanishes along with its shoulder. The NP dependent peak at λ o ~ 500 nm first decreases in intensity from C Ag = 100 to 90 at.%, then at 13 ACS Paragon Plus Environment


Page 14 of 31

C Ag = 85% the peak shifts to λ o = 530 nm and increases in intensity. For samples with lower C Ag shown in S6(d) of the Supporting Information, this peak continues to decrease in intensity. The peak also blueshifts from λ o = 482 nm at C Ag = 80 at.%.to λ o = 396 nm when the C Ag = 20 at.%. This peak vanishes at C Ag = 0 at.%. For the samples made from the other template sizes, the general trends are consistent with that of the D = 750 nm template samples, with some slight differences. For D = 500 nm template samples (Fig 4(b)), the main dipole peaks are sharper than those of the D = 750 nm samples, and the shoulders are significantly smaller, while for D = 350 nm template samples (Fig 4(c)), very broad background absorption are observed, with the main dipole resonator on top of the broadband absorbance. Also, for the D = 350 nm template samples, instead of observing a single strong peak, a triplet is present. This new peak at λo = 520 nm could possibly be due to a Fano resonance since it was observed in other studies with nanosphere lithography where D < 500 nm.38 However further experiments and numerical calculations are needed to confirm the origin of this peak. When the C Ag is reduced to C Ag = 97 at.%, this Fano resonance peak strengthens significantly and slightly redshifts. The peak continues to redshift and then decreases in intensity for the C Ag = 95 at.% sample. The peak vanishes when the C Ag reaches 90 at.%. At C Ag < 90 at.% the main peak abruptly vanishes. The change in the optical properties when C Ag < 90 % is consistent throughout all the template sizes as described for the D = 750 nm template samples. To better understand the changes in the optical properties of the NT samples for different C Ag , we also measured the extinction spectra of the corresponding thin film samples. As shown in Fig. 4(d), for the high C Ag (= 100 and 97 at.%) samples, featureless broadband extinction is observed. When the C Ag decreases to 95 at.%, a peak starts to appear at λ = 555 nm. When C Ag = 90 at.%, this peak becomes more obvious and slightly blueshiftes to λ = 483 nm. It continues to 14 ACS Paragon Plus Environment

Page 15 of 31 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


slightly blueshift as the C Ag decreases to 85 at.%, and the broad extinction for λ > 600 nm also decreases. At C Ag < 85 at.% (see Fig S6 of the Supporting Information) the trend continues as the C Ag decreased, and the peak near λ = 500 nm slightly blueshifts and decreases in intensity. When C Ag = 0 at.%, the peak vanishes and the extinction becomes very small. Clearly, the peak near λ = 500 nm is present in all the samples with C Ag ≤ 90 at.%, regardless of template size, thin film, or NTs. As observed in SEM, the samples with C Ag ≤ 90 at.% have Ag NPs present with constant size but varying m, thus, it is expected that the λ ~ 500 nm peak is the LSPR wavelength of the AgNPs. Figure 5 summarizes the main extinction peak as a function of the C Ag for samples of all 𝑇𝑇 template sizes, where we observe a clear composition threshold at ~ 𝐶𝐶𝐴𝐴𝐴𝐴 = 90 at.%. When C Ag >

𝑇𝑇 𝐶𝐶𝐴𝐴𝐴𝐴 , the λ o of the main extinction peak of NTs depends closely on C Ag , i.e., λ o redshifts as C Ag

increases. When C Ag < 90 at.%, only one resonance peak is observed and λ o of this peak almost remains a constant when C Ag is changed. Such a peak location is also consistent with that of the thin films, indicating that this peak originates from the LSPR of the Ag NPs embedded in NTs and thin films. In fact, the optical properties of the low C Ag samples can be explained by a model developed by Garcia et al.,39 see Section S9 of the Supporting Information. Therefore, only when we consider C Ag ≥ 90 at.%, the plasmonic properties of the NTs can be tuned systematically by the C Ag . Thus, for LSPR sensor design, we only consider NT samples with C Ag ≥ 90 at.%.

Sensing performance The sensing performance of the NT samples for C Ag ≥ 90 at.% have been characterized by immersing the samples in 5 different organic solvents with index of refractions from 1.36 to 1.51 RIU. To ensure the accuracy of the sensitivity measurements, time dependent transmission 15 ACS Paragon Plus Environment


Page 16 of 31

spectra were taken and are presented in the Supporting Information Figure S10, no significant change in the samples was observed over a 90 minute period. A set of representative extinction spectra from C Ag = 90 at.% with a D = 500 nm PSNS are shown in Fig. 6(a). In general, as the index of refraction of the environment increases, the LSPR peak redshifts. The plots of the LSPR peak wavelength λ o versus the refractive index n for all the tested samples as well as the FOM as a function of C Ag are shown respectively in Fig. S11(a) & (b) of the Supporting Information. Unsurprisingly the λ o ~ n relationships all follow a linear function, and the slope of the linear fitting, the RIS, as a function of C Ag for different template sizes are summarized in Fig. 6(b). Two general trends were immediately observed. First, for the same C Ag , the larger the PSNS size, the larger the RIS. Second, for the same PSNS size, as the C Ag decreases the sensitivity increases, except for the samples at C Ag = 100 at.% which shows slightly higher RIS compared to that of samples at C Ag = 97 at.%. The sample with the highest RIS, ~ 696 RIU/nm, is the C Ag = 90 at.% D = 500 nm PSNS sample. While this is not the highest RIS of NT LSPR sensors reported in the literature, this sample showed more than twice the RIS of that from the pure Ag sample (RIS = 312 RIU/nm). The FOM of the NT samples are shown in Fig. S11 (b) of the Supporting Information. For all the NT samples the FOM is in the single digits, between 1 and 7; this is consistent with other studies on NTs.3, 14 The FOM could be further improved by incorporating the optimized Ag-MgF 2 composite to other plasmonic structures with higher FOM17 and/or by applying the differential sensing strategies to a linear40 or circular10 polarization dependence nanostructure. This increased RIS confirms our hypothesis that for a high Ag content Ag-dielectric mixture, the LSPR sensitivity can be tuned and improved, due to the change of the dielectric constant of the mixture as shown in Fig. 1. In fact, the complex dielectric constants of the Ag16 ACS Paragon Plus Environment

Page 17 of 31 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


MgF 2 mixture thin films were determined by the spectroscopy ellipsometry as shown in Fig S12 in the Supporting Information, and the real part ε eff is plotted in Fig. 6(c) for different C Ag . A table of n and k values for several common sensing wavelengths is also presented in the Supporting Information Table S1 for reference. As shown in Fig. 6(c), when C Ag decreases from 100 at.% to 80 at.%, the absolute value of the slope of the ε eff – λ curve become smaller and smaller. In fact, based on the measured ε of the pure Ag and MgF 2 thin films, one can apply the MG EMT to predict the ε eff , as shown by the dashed curves in Fig 6(c). For the sample at C Ag = 97 at.%, the experimental data agrees well with the EMT results. However, when C Ag decreases, more and more discrepancies appear between the measured data and the EMT. Especially when C Ag ≥ 90 at.%, the measured ε eff has a peak at λ ~ 500 nm while the EMT still gives a featureless curve. This is due to the appearance of Ag NPs in the sample as discussed before. In fact, regardless of C Ag or the PSNS size, there is a definite relationship between RIS and λ o , i.e., the LSPR resonance wavelength, as shown in Fig 6(d), where RIS is plotted as a function of λ o in air. Overall there is a clear trend that as the LSPR wavelength increases, the RIS increases. The dependence of RIS on LSPR wavelength has been studied both theoretically and through numerical calculations in similar wavelength regions.41-43 Saison-Francioso et.al.41 have given an analytical expression for the RIS- λ o relationship for a free particle, 2�𝐴𝐴𝜆𝜆𝑜𝑜 2 + 𝐵𝐵𝜆𝜆𝑜𝑜 + 𝐶𝐶� 𝑅𝑅𝑅𝑅𝑅𝑅 = , 𝑛𝑛𝑠𝑠 (2𝐴𝐴𝜆𝜆𝑜𝑜 + 𝐵𝐵)

(3)

where n s is the index of refraction of the substrate, and A, B, and C are fitting parameters of the real part of a quadratic dielectric function of the sensing material, 𝑅𝑅𝑅𝑅[𝜀𝜀] = 𝐴𝐴𝜆𝜆2 + 𝐵𝐵𝐵𝐵 + 𝐶𝐶.

(4) 17

ACS Paragon Plus Environment


Page 18 of 31

Thus, the real part of the dielectric functions for both C Ag = 100 and 90 at.%. were fit to Eq. (4) and are shown in the Supporting Information S11(c) & (d). For our thin film samples at C Ag = 100 at.%, A = (-5.1 ± 0.1)× 10-5 nm-2, B = (-3.0 ± 0.1)× 10-3 nm-1, C = 5.43 ± 0.04 and for C Ag = 90 at.%, A = (-4.3 ± 0.1)× 10-5 nm-2, B = (34.9 ± 0.4)× 10-3 nm-1, C = -10.1 ± 0.1. The quadratic form was used instead of the linear form suggested by Miller and Lazarides because of our larger wavelength range.43 The black dashed curve in Fig. 6(d) shows the prediction from Eq. (3) for C Ag = 100 at.%, the RIS increases monotonically with λ o , but the predicted RIS is much higher than that obtained experimentally. By considering the effect of the substrate, Saison-Francioso et.al. modified Eq. (3) to give the RIS as,

𝑅𝑅𝑅𝑅𝑅𝑅𝑠𝑠𝑠𝑠𝑠𝑠

2�𝐴𝐴𝜆𝜆𝑜𝑜 2 + 𝐵𝐵𝜆𝜆𝑜𝑜 + 𝐶𝐶� = + 𝑛𝑛𝑠𝑠 (2𝐴𝐴𝜆𝜆𝑜𝑜 + 𝐵𝐵)

�

𝑛𝑛𝑠𝑠 𝑎𝑎𝑥𝑥 𝑎𝑎𝑦𝑦 𝑎𝑎𝑧𝑧 𝑛𝑛𝑠𝑠 𝑛𝑛𝑛𝑛 2 𝑒𝑒𝑒𝑒𝑒𝑒 � 6𝑑𝑑 3 (𝑛𝑛2 + 𝑛𝑛2 )2 𝑛𝑛 𝑠𝑠 𝐿𝐿𝑥𝑥,‖ 2𝐴𝐴𝜆𝜆𝑜𝑜 + 𝐵𝐵

(5)

,

𝑒𝑒𝑒𝑒𝑒𝑒

where 𝑛𝑛𝑠𝑠 and 𝑛𝑛𝑛𝑛 are the index of refraction of the substrate and metallic NP respectively, 𝐿𝐿𝑥𝑥,‖ is a geometric factor that depends on the NP’s shape, 𝑎𝑎𝑥𝑥 , 𝑎𝑎𝑦𝑦 , and 𝑎𝑎𝑧𝑧 are the semi major axes of the

NP and d is the distance between the center of the NP and the substrate. The red dashed curves in Fig. 6(d) are the fitting data based on this model at both C Ag = 100 and 90 at.%, and the shadowed areas between the two dashed curves would be the expected RIS values for composites between 100 at. % and 90 at. %. Clearly the results cover most RIS data points obtained experimentally and can be used to better describe the RIS of composite nanostructures.

Conclusions In summary, Ag-MgF 2 NT patterns and thin films with a variety of Ag compositions were fabricated by nanosphere lithography and electron beam co-deposition. For Ag-MgF 2 with 18 ACS Paragon Plus Environment

Page 19 of 31 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


high Ag composition their optical properties are function of the Ag composition as well as the size of the NTs, while for low C Ag (< 90 at.%), a nearly constant LSPR peak appears in all samples, regardless of the Ag composition, NT size, or thin film, which is believed to be due to Ag NPs formed in the deposition. Thus, the Ag-dielectric composite nanomaterial can have its plasmonic resonance property tuned only at high Ag content. The addition of the dielectric material in Ag would make the dispersion relationship of the real dielectric constant vary slowly with the wavelength, which can be used to improve the sensitivity of LSPR sensor. Our experimental results confirm that when C Ag = 90 at.%, the NT samples at D = 500 nm PSNS can have a RIS of 696 RIU/nm, as compared to 312 RIU/nm for the same NTs of C Ag = 100 at.%. This result is, in fact, the consequence of tuning the dispersion relationship of the dielectric constant as confirmed by ellipsometry measurements. Our study clearly demonstrates that besides the size, shape, and arrangement of the plasmonic nanostructure, the appropriate composition of the plasmonic-dielectric mixture structures can be used to tune the LSPR properties as well as the sensitivity. Such a tunability is very significant and can play an important role in future plasmonic material design and applications.

Associated Content Supporting Information: The Supporting Information is available free of charge on the ACS Publication website at DOI: XXXXXXXXXX. Additional SEM/AFM images, XRD, extinction spectra, polarization dependent spectra, extracted dielectric constants from ellipsometry measurements, nanoparticle effective medium theory modeling extinction modeling, refractive index dependent LSPR measurements, and sample degradation measurements. 19 ACS Paragon Plus Environment


Page 20 of 31

Author Information Corresponding Author E-mail: [email protected] Department of Physics and Astronomy University of Georgia Athens, Georgia 30602 Phone: 706/542-6230 Fax: 706/542-2492 Contributions: The manuscript was written with contributions from all authors. All authors have given their approval to the final version of the manuscript. Notes: These authors declare no competing financial interests.

Acknowledgments This work is supported by The National Science Foundation under Grant no. CMMI-1435309 and ECCS-1611330. The authors would also like to thank Dr. Zhengwei Pan for the use of his SEM/EDS.


Page 21 of 31 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


Figure Captions: Figure 1: Maxwell-Garnett effective medium theory approximation for composite Ag and MgF 2 from 100 to 85 at.% Ag. Δλ 100 and Δλ 90 are the wavelength shifts expected from one dielectric environment to another for the pure Ag and 90 at.% Ag sample, respectively.

Figure 2: Representative SEM images of NT samples fabricated with D = 500 nm PSNS monolayer for C Ag = 0, 33, 50, 66, 80, 95, and 100 at.%.

Figure 3: (a)Nanoparticle size d and concentration m in NT samples versus C Ag for D = 750 nm 𝐸𝐸𝐸𝐸𝐸𝐸 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 PSNS template. (b) The measured Ag content 𝐶𝐶𝐴𝐴𝐴𝐴 versus 𝐶𝐶𝐴𝐴𝐴𝐴 calculated from the deposition 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝐸𝐸𝐸𝐸𝐸𝐸 = 𝐶𝐶𝐴𝐴𝐴𝐴 . The plot of resistivity ρ of rates. The black solid line is a guide to the eye where 𝐶𝐶𝐴𝐴𝐴𝐴 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 the thin film samples versus 𝐶𝐶𝐴𝐴𝐴𝐴 is also shown.

Figure 4: The extinction spectra for high C Ag samples: (a) D = 750 nm, (b) D = 500 nm, (c) D = 350 nm PSNS template and (d) thin films.

Figure 5: (a) The plot of the main LSPR peak wavelength λ o versus C Ag for different PSNS template samples. The inset images show SEM of samples C Ag = 95 at.% and 80 at.%, respectively.

Figure 6: (a) LSPR extinction peaks of the C Ag = 90 at.%, D = 500 nm PSNS samples immersed in different index liquids. (b) The plot of experimental RIS versus C Ag for different PSNS templates. (c) The measured real dielectric constant Re(ε eff ) for high Ag content samples and the corresponding calculated results from MG EMT (dashed curves). (d) The plot of RIS versus λ o from all the samples for various C Ag and templates. The dashed curves are results from Eqs. (3) and (5).



-2

∆λ90 > ∆λ100

100 at.% Ag 90 at.% Ag 85 at.% Ag

-4

Re(εeff)

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 22 of 31

-6 85 at.% Ag 90 at. % Ag

-8 -10

∆λ100

∆λ90

100 at.% Ag

-12 400

450

500

550

Wavelength λ (nm)

600

Figure 1


Page 23 of 31 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


Figure 2



NP density m (NPs/NT)

60

(a) 50

40

NP count NP size

30

40 30

20

20 10

10

0

0

(b)

0.8

60 0.6

40

0.4

20

0.2

Resistivity ρ (Ω⋅µm)

1.0

CEDS Ag Resistivity

80

CEDS Ag

NP diameter d (nm)

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 24 of 31

0.0

0 20

40

60

80

100

Calc Ag

C

Figure 3


Page 25 of 31

0.5

(b) D = 500 nm

(a) D = 750 nm

Extinction

0.4

100 at.% 97 at.% 95 at.% 90 at.%

85 at.% 0.6 80 at.% 75 at.% 0.5

0.4 0.3

0.3

0.2

0.2

0.1

0.1

Extinction

0.5

(d) Thin films

(c) D = 350 nm

3.0

Extinction

0.4

2.5

0.3

2.0

0.2

1.5

Extinction

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


1.0

0.1

0.5 0.0

300

600

900

1200

1500

1800

300

Wavelength λ (nm)

600

900

1200

1500

Wavelength λ (nm)

1800

Figure 4 25 ACS Paragon Plus Environment


PSNS Template size D = 750 nm D = 500 nm D = 350 nm Nanoparticle dependent

1600 1400 1200

λo (nm)

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 26 of 31

1000

80 at.% Ag

Nanotriangle dependent

800

95 at.% Ag

600 400 100

95

90

85

80

75

70

65

CAg Figure 5


Page 27 of 31

1.05

n = 1 Air n = 1.36 Acetone n = 1.42 1-Hexanol

(a)

800

n = 1.44 Chloroform n = 1.46 CCl4 n = 1.50 Toluene

PSNS Template Size D = 750 nm D = 500 nm D = 350 nm

(b) 700

RIS (nm/RIU)

Extinction

1.00 0.95 0.90

600 500 400

0.85

300

0.80 1000

200

0

1100

1200

1300

1400

Wavelength λ (nm)

1500

90

1600

(c)

1000

RIS (nm/RIU)

-20 100 at.% 97 at.% 95 at.% 90 at.% 85 at.% 80 at.%

-30 -40 -50 400

500

Ellipsometry Re(εeff) from MG EMT 600

700

800

Wavelength λ (nm)

92

94

96

98

100

CAg

-10

Re(εeff)

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


(d)

800

100 at.% 97 at.% 95 at.% 90 at.%

600 400

g .% A t a 90

100

200

900

1000

600

(3) D = 750 nm Eq. e l tic D = 500 nm Par e e r D = 350 nm F 5) q. ( E e trat ubs S h Wit

800

at.%

1000

Ag

1200

λo (nm)

1400

1600

Figure 6 27 ACS Paragon Plus Environment


Page 28 of 31

References 1. 2. 3. 4. 5. 6.

7. 8. 9.

10.

11. 12. 13. 14.

15.

16.

17.

18.

Mayer, K. M.; Hafner, J. H. Localized Surface Plasmon Resonance Sensors. Chem. Rev. 2011, 111, 3828-3857. Li, M.; Cushing, S. K.; Wu, N. Plasmon-Enhanced Optical Sensors: A Review. Analyst 2015, 140, 386-406. Stewart, M. E.; Anderton, C. R.; Thompson, L. B.; Maria, J.; Gray, S. K.; Rogers, J. A.; Nuzzo, R. G. Nanostructured Plasmonic Sensors. Chem. Rev. 2008, 108, 494-521. Armani, A. M.; Kulkarni, R. P.; Fraser, S. E.; Flagan, R. C.; Vahala, K. J. Label-Free, Single-Molecule Detection With Optical Microcavities. Science 2007, 317, 783-787. Anker, J. N.; Hall, W. P.; Lyandres, O.; Shah, N. C.; Zhao, J.; Van Duyne, R. P. Biosensing with Plasmonic Nanosensors. Nat. Mater. 2008, 7, 442-453. Svedendahl, M.; Chen, S.; Dmitriev, A.; Kall, M. Refractometric Sensing Using Propagating versus Localized Surface Plasmons: A Direct Comparison. Nano Lett. 2009, 9, 4428-4433. Wang, H.; Brandl, D. W.; Le, F.; Nordlander, P.; Halas, N. J. Nanorice: A Hybrid Plasmonic Nanostructure. Nano Lett. 2006, 6, 827-832. Underwood, S.; Mulvaney, P. Effect of the Solution Refractive Index on the Color of Gold Colloids. Langmuir 1994, 10, 3427-3430. Lee, J.; Hasan, W.; Odom, T. W. Tuning the Thickness and Orientation of Single Au Pyramids for Improved Refractive Index Sensitivities. J. Phys. Chem. C 2009, 113, 2205-2207. Jeong, H.-H.; Mark, A. G.; Alarcón-Correa, M.; Kim, I.; Oswald, P.; Lee, T.-C.; Fischer, P. Dispersion and Shape Engineered Plasmonic Nanosensors. Nat. Commun. 2016, 7, 11331. Chen, H.; Kou, X.; Yang, Z.; Ni, W.; Wang, J. Shape- and Size-Dependent Refractive Index Sensitivity of Gold Nanoparticles. Langmuir 2008, 24, 5233-5237. Nehl, C. L.; Liao, H. W.; Hafner, J. H. Optical Properties of Star-Shaped Gold Nanoparticles. Nano Lett. 2006, 6, 683-688. Mock, J. J.; Smith, D. R.; Schultz, S. Local Refractive Index Dependence of Plasmon Resonance Spectra from Individual Nanoparticles. Nano Lett. 2003, 3, 485-491. Jensen, T. R.; Schatz, G. C.; Van Duyne, R. P. Nanosphere Lithography: Surface Plasmon Resonance Spectrum of a Periodic Array of Silver Nanoparticles by Ultraviolet−Visible Extinction Spectroscopy and Electrodynamic Modeling. J. Phys. Chem. B 1999, 103, 2394-2401. Ramaccia, D.; Arcieri, S.; Toscano, A.; Bilotti, F. Core-Shell Super-Spherical Nanoparticles for LSPR-Based Sensing Platforms. IEEE J. Sel. Top. Quantum Electron. 2017, 23, 1-8. Sherry, L. J.; Jin, R. C.; Mirkin, C. A.; Schatz, G. C.; Van Duyne, R. P. Localized Surface Plasmon Resonance Spectroscopy of Single Silver Triangular Nanoprisms. Nano Lett. 2006, 6, 2060-2065. Lodewijks, K.; Van Roy, W.; Borghs, G.; Lagae, L.; Van Dorpe, P. Boosting the Figure-Of-Merit of LSPR-Based Refractive Index Sensing by Phase-Sensitive Measurements. Nano Lett. 2012, 12, 1655-1659. Bahramipanah, M.; Dutta-Gupta, S.; Abasahl, B.; Martin, O. J. F. Cavity-Coupled Plasmonic Device with Enhanced Sensitivity and Figure-of-Merit. ACS Nano 2015, 9, 7621-7633. 28 ACS Paragon Plus Environment

Page 29 of 31 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


19.

20.

21.

22.

23. 24.

25.

26.

27. 28.

29.

30.

31.

32. 33.

Rodal-Cedeira, S.; Montes-García, V.; Polavarapu, L.; Solís, D. M.; Heidari, H.; La Porta, A.; Angiola, M.; Martucci, A.; Taboada, J. M.; Obelleiro, F.; et al. Plasmonic Au@Pd Nanorods with Boosted Refractive Index Susceptibility and SERS Efficiency: A Multifunctional Platform for Hydrogen Sensing and Monitoring of Catalytic Reactions. Chem. Mater. 2016, 28, 9169-9180. Charles, D. E.; Aherne, D.; Gara, M.; Ledwith, D. M.; Gun’ko, Y. K.; Kelly, J. M.; Blau, W. J.; Brennan-Fournet, M. E. Versatile Solution Phase Triangular Silver Nanoplates for Highly Sensitive Plasmon Resonance Sensing. ACS Nano 2010, 4, 5564. Lee, K. S.; El-Sayed, M. A. Gold and Silver Nanoparticles in Sensing and Imaging: Sensitivity of Plasmon Response to Size, Shape, and Metal Composition. J. Phys. Chem. B 2006, 110, 19220-19225. Whitney, I.; Steven, L.; Daniel, C.; Yiping, Z. Ag–Cu Mixed Phase Plasmonic Nanostructures Fabricated by Shadow Nanosphere Lithography and Glancing Angle Co-Deposition. Nanotechnology 2017, 28, 015301. Henkel, A.; Jakab, A.; Brunklaus, G.; Sönnichsen, C. Tuning Plasmonic Properties by Alloying Copper into Gold Nanorods. J Phys. Chem. C 2009, 113, 2200-2204. Motl, N. E.; Ewusi-Annan, E.; Sines, I. T.; Jensen, L.; Schaak, R. E. Au−Cu Alloy Nanoparticles with Tunable Compositions and Plasmonic Properties: Experimental Determination of Composition and Correlation with Theory. J. Phys. Chem. C 2010, 114, 19263-19269. Rioux, D.; Vallières, S.; Besner, S.; Muñoz, P.; Mazur, E.; Meunier, M. An Analytic Model for the Dielectric Function of Au, Ag, and Their Alloys. Adv. Opt. Mater. 2014, 2, 176-182. Whitney, I.; Yizhuo, H.; Keenan, S.; William, M. D.; Dexian, Y.; Yiping, Z. Tuning the Plasmonic Properties of Silver Nanopatterns Fabricated by Shadow Nanosphere Lithography. Nanotechnology 2016, 27, 385301. Anuj, K. S.; Gupta, B. D. Fibre-Optic Sensor Based on Surface Plasmon Resonance with Ag–Au Alloy Nanoparticle Films. Nanotechnology 2006, 17, 124. Nimrodh Ananth, A.; Umapathy, S.; Sophia, J.; Mathavan, T.; Mangalaraj, D. On the Optical and Thermal Properties of In Situ/Ex Situ Reduced Ag NP’s/PVA Composites and its Role as a Simple SPR-based Protein Sensor. Appl. Nanosci. 2011, 1, 87-96. Takele, H.; Greve, H.; Pochstein, C.; Zaporojtchenko, V.; Faupel, F. Plasmonic Properties of Ag Nanoclusters in Various Polymer Matrices. Nanotechnology 2006, 17, 3499. Scalisi, A. A.; Compagnini, G.; D’Urso, L.; Puglisi, O. Nonlinear Optical Activity in Ag–SiO2 Nanocomposite Thin Films with Different Silver Concentration. Appl. Surf. Sci. 2004, 226, 237-241. Ichiro, T.; Masaru, Y.; Yoshio, M.; Takao, T.; Satoshi, S.; Takashi, T.; Arao, N. Preparation and Nonlinear Optical Properties of Ag/ SiO2 Glass Composite Thin Films. Jpn. J. Appl. Phys. 1994, 33, L1410. Li, Y. Plasmonic Optics: Theory and Applications; SPIE Press: Bellingham: 2017. He, Y. P.; Zhang, Z. Y.; Hoffmann, C.; Zhao, Y. P. Embedding Ag Nanoparticles Into MgF2 Nanorod Arrays. Adv. Funct. Mater. 2008, 18, 1676-1684.



34.

35.

36.

37. 38.

39.

40.

41.

42.

43.

Page 30 of 31

Wei, H.; Eilers, H. Electrical Conductivity of Thin-Film Composites Containing Silver Nanoparticles Embedded in a Dielectric Fluoropolymer Matrix. Thin Solid Films 2008, 517, 575-581. Balberg, I.; Azulay, D.; Goldstein, Y.; Jedrzejewski, J.; Ravid, G.; Savir, E. The Percolation Staircase Model and its Manifestation in Composite Materials. Eur. Phys. J. B 2013, 86, 1-17. Sun, Y.-L.; Wang, J.; Li, S.-Y.; He, X.; Wang, C.-W. The Preparation and Characterization of Nanoparticle Ag–SiO2 Composite Films with Super Low Refractive Index. Superlattices Microstruct. 2014, 73, 136-144. Guske, J. T.; Brown, J.; Welsh, A.; Franzen, S. Infrared Surface Plasmon Resonance of AZO-Ag-AZO Sandwich Thin Films. Opt. Express 2012, 20, 23215-23226. Chan, G. H.; Zhao, J.; Hicks, E. M.; Schatz, G. C.; Van Duyne, R. P. Plasmonic Properties of Copper Nanoparticles Fabricated by Nanosphere Lithography. Nano Lett. 2007, 7, 1947-1952. Garcı́a, M. A.; Llopis, J.; Paje, S. E. A simple Model for Evaluating the Optical Absorption Spectrum from Small Au-Colloids in Sol–Gel Films. Chem. Phys. Lett. 1999, 315, 313-320. Ai, B.; Basnet, P.; Larson, S.; Ingram, W.; Zhao, Y. Plasmonic Sensor with High Figure of Merit Based on Differential Polarization Spectra of Elliptical Nanohole Array. Nanoscale 2017, 9, 14710-14721. Saison-Francioso, O.; Lévêque, G.; Boukherroub, R.; Szunerits, S.; Akjouj, A. Dependence Between the Refractive-Index Sensitivity of Metallic Nanoparticles and the Spectral Position of Their Localized Surface Plasmon Band: A Numerical and Analytical Study. J. Phys. Chem. C 2015, 119, 28551-28559. Molly, M. M.; Anne, A. L. Sensitivity of Metal Nanoparticle Plasmon Resonance Band Position to the Dielectric Environment as Observed in Scattering. J. Opt. A: Pure Appl. Opt. 2006, 8, S239. Miller, M. M.; Lazarides, A. A. Sensitivity of Metal Nanoparticle Surface Plasmon Resonance to the Dielectric Environment. J. Phys. Chem. B 2005, 109, 21556-21565.


Page 31 of 31 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


TOC Graphic


The Localized Surface Plasmonic Resonance and Sensing Properties

Recommend Documents