Engineering 3D Nanoplasmonic Assemblies for High Performance

Subscriber access provided by MONASH UNIVERSITY

Article

Engineering 3D Nanoplasmonic Assemblies for High Performance Spectroscopic Sensing Sanghamitra Dinda, Vignesh Suresh, Praveen Thoniyot, Armandas Balcytis, Saulius Juodkazis, and Sivashankar Krishnamoorthy ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.5b07745 • Publication Date (Web): 02 Nov 2015 Downloaded from http://pubs.acs.org on November 29, 2015

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 free 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 accessible to all readers and 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.

ACS Applied Materials & Interfaces 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 17

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

ACS Applied Materials & Interfaces

Engineering 3D Nanoplasmonic Assemblies for High Performance Spectroscopic Sensing S. Dinda,a,b V. Suresha, P. Thoniyot,a,c A. Balčytis,d,e S. Juodkazis,d S. Krishnamoorthy*a,f a

Institute of Materials Research and Engineering (IMRE), Agency for Science, Technology and Research (A*STAR), 3, Research Link, 117602 Singapore b

Department of Biotechnology, School of Pharmaceutical Sciences, Siksha O Anushandan University (SOA), Bhubaneswar, 751030, India c

Singapore Bio imaging Consortium (SBIC), Biomedical Sciences Institutes 11 Biopolis Way, #02-02 Helios, Singapore 138667

d

Centre for Micro-Photonics, Faculty of Science Engineering and Technology Swinburne University of Technology Hawthorn, VIC 3122, Australia

e

Institute of Physics, Centre for Physical Sciences and Technology, 231 Savanoriu Avenue, LT02300 Vilnius, Lithuania f

Nano-Enabled Medicine and Cosmetics Group, Materials Research and Technology (MRT),

Luxembourg Institute of Science and Technology (LIST), 41, Rue du Brill, L-4422, Belvaux, Luxembourg

ACS Paragon Plus Environment

1


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 17

ABSTRACT: We demonstrate the fabrication of plasmonic sensors comprising of gold nanopillar arrays exhibiting high surface areas, and narrow gaps, through self-assembly of amphiphilic diblock copolymer micelles on silicon substrates. Silicon nanopillars with high integrity over arbitrary large areas are obtained using copolymer micelles as lithographic templates. The gaps between metal features are controlled by varying the thickness of the evaporated gold. The resulting gold metal nanopillar arrays exhibits an engineered surface topography, together with uniform and controlled separations down to sub-10 nm suitable for highly sensitive detection of molecular analytes by Surface Enhanced Raman Spectroscopy (SERS). The significance of the approach is demonstrated through the control exercised at each step, including template preparation and pattern-transfer steps. The approach is promising means to address trade-offs between resolutions, throughput and performance in the fabrication of nanoplasmonic assemblies for sensing applications. KEYWORDS Plasmonic Nanoarrays, Nanopillar, Surface enhanced Raman spectroscopy (SERS), Self-Assembly, Sensing 1. INTRODUCTION Sensitive transduction of molecular binding events on chip carries profound implications to the outcome of a range of molecular sensors sought for applications in security, food and environment control, and health care.

1

Amongst nanoscale sensors, plasmonic biosensing

involving electromagnetic (EM) near-field enhancements for detection and quantification of molecules relies critically on control over nanoscale geometries in noble metal structures at ultrahigh (or molecular) resolutions. At these resolutions the control over geometric attributes translates into an advantage in shaping near-field profile and enhancing intensity of EM field. 2-7


2

Page 3 of 17

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


This can be favorably taken advantage towards highly sensitive transduction of molecular binding events through surface-enhanced spectroscopies8, spectroscopy (SERS),

10, 11

9

such as surface-enhanced Raman

metal enhanced Fluorescence (MEF)

12, 13

and localized surface

plasmon resonance imaging (LSPR). 4, 6Among these SERS is attractive and is known to exhibit sensitivity down to single molecule level,

14-16

inherent specificity to vibrational spectra of

analytes (or their labels), significant ease of multiplexing, ,17 and possibility for ready implementation within portable configurations.

18

The intensities of vibrational Raman peaks of

molecules are strongly enhanced by electromagnetic ‘hot-spots’ arising at curvatures or gaps between metal nanostructures with resolutions down to sub-10nm regime.

19-21

This constitutes

compelling analytical capabilities for trace analyte detection through the use of metal array geometries tailored for high EM field enhancements. The requirement however imposes inevitable trade-offs between the resolutions, throughput and quality control in the fabrication and engineering of nanoplasmonic assemblies down to molecular length scales. We present molecular self-assembly driven fabrication21-25 of 3D SERS substrates that yield a large number of curvature and gap hot-spots, together with fine-control over the critical geometric attributes that allow rational design in fabrication. The 3D SERS substrates are composed of an array of nanopillars with a narrow range of size and spatial distributions and with programmed control over density and geometry of hot-spots. Nanopillar SERS substrates are known in literature26 through different means, e.g. electron-beam lithography, ,27, colloidal lithography,

30, 31

anodized alumina templates,

32

formed templates33, 34or metallization of natural surfaces.

28

interference lithography,

and lithography using randomly self35

The advantages common to these

approaches are that they enable creating hot-spots either due to sharp corners, separated gaps33,

34

29

and they hold the potential for rational design.

28, 29

32

or closely

These approaches


3


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 4 of 17

however carry one or more of the limitations, viz. difficulty to engineer substrate geometries, large standard deviations in spectral intensities across the substrate, high propensity for variation across batches, and low density of hot-spots. Our report demonstrates gold nanopillar substrates offering combined advantages due to high density of gap hot-spots (>109-1010/cm2), exhibiting gaps down to sub-10nm regime, fine-tunability in geometries and high uniformity within chip and across batches (tolerance 10%). In addition, our approach combines performance advantages with economy and throughput, making exploitation of the chips closer to reality. The correlation between fabrication parameters ↔ metal pillar geometry ↔ optical/spectroscopic properties is systematically investigated to maximize SERS performance. 2. EXPERIMENTAL 2.1 Materials: Polystyrene-block-poly (2-vinylpyridine) (PS-b-PVP) (57000-b-57000 g/mole) was purchased from Polymer Source Inc. (Montreal, Canada). Silicon substrates were purchased from Silicon Valley Microelectronics (SVM, CA, USA). m-Xylene, 1-naphthalenethiol and nHexane were purchased from Sigma Aldrich. 2.2 Nanofabrication: The silicon substrates were cleaned by 2-propanol and exposed to UV/ozone (UV 1, SAMCO, Kyoto, Japan) for duration of 5 minutes. The silicon pillar arrays were fabricated by lithographic pattern-transfer process sequence as reported earlier.22,26 Thin films of PS-b-P2VP reverse micelles were coated from m-xylene solutions at polymer concentration of 0.5% w/w on to silicon substrates by spin-coating at 5000 rpm at relative humidity of 45%. To prepare gold nanopillar arrays, metallic gold was sputter deposited on to the silicon pillar arrays (Denton Explorer® Coating System) at rates of 2A/s. The characterizations of surface nanostructures were performed using atomic force microscope in


4

Page 5 of 17

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


tapping mode (Nano scope IV Multimode AFM, Veeco Instruments Inc., NY, USA), scanning electron microscope (FESEM 6700F, JEOL, Tokyo, Japan). 2.3 Optical/Spectroscopic measurements: The optical properties of the Au metal nanopillar arrays on silicon substrates were recorded using a CRAIC micro spectrophotometer (CRAIC Technologies, CA, and USA). SERS experiments were performed using LabRAM Raman spectrometer (HORIBA Jobin Yvon), equipped with microscope, a peltier cooled CCD detector and an excitation laser of 785 nm with objective lens of 50X and numerical of 0.55. For the SERS validation experiments (Figures 3, 4) Au nanopillar arrays were incubated into a hexane solution of 1-Napthalene thiol for 20 h. Following this, the substrates were rinsed with the respective solvent and blow dried with N2 purging. The SERS measurements were recorded from at least five different random locations spaced at least 2-3 mm apart, with an exposure time of 10s using laser power of 6.5 mW. 2.4 Finite-difference time-domain (FDTD) simulations: Numerical investigation of light interaction with Au nanopillar arrays was conducted using a commercially available FDTD solver from Lumerical Inc. (Vancouver, Canada) Representative reconstructions of Si pillars sputtered with varying thicknesses of gold were generated based on cross section SEM images. The calculations were performed on a single period segment of the nanostructured surface rendered on a uniform mesh with a 1 nm step size. The scattering behaviour of simulated surfaces was probed using a linearly polarized total-field scattered-field pulsed plane wave source, impinging at normal incidence. The simulation region was terminated by periodic boundary conditions along x- and y-axes, and by perfectly matched layer boundaries in the zdirection of plane wave propagation. The simulations were conducted for two perpendicular


5


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 6 of 17

linear light polarizations for the purpose of, upon averaging these two results, obtaining unpolarised light scattering data.

Figure 1. (i-vii) Dependence of nanopillar geometry as a function of thickness (texp) of sputter coated gold films, as measured by scanning electron microscopy in cross-sectional view. The schematic defines the different components of the nanopillar geometry, that evolve as a function of the texp (value as indicated in the images). Refer to Figure S1 (ESI) for top-view of pillars. 3. Results and Discussion 3.1. Au Nanoarrays with Engineered 3D topography. Si pillar arrays were obtained using lithographic pattern-transfer of reverse micelle templates formed out of polystyrene-blockpoly(2-vinylpyridine) using approaches we had reported earlier.

23, 36

The resulting nanopillar

arrays exhibited a periodicity of ~114nm, the same as that of the micelle array templates. The pillars were found to have a positively tapered profile, with a width of 55nm ± 5 nm at the top. The nanopillar substrate was subsequently sputter coated with gold with systematically varying thickness to obtain gold nanopillars with different heights and widths. The gold deposition by sputtering was found to result in a preferential increase in metal thickness on the top of the silicon pillars (Figure 1 t) as compared to the pillar side walls. With a progressive increase in the


6

Page 7 of 17

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


sputter coated metal quantity (texp between 10-120 nm), the width of the pillars continued to increase, resulting in a systematic closure of the inter-pillar space. Two different types of interpillar spaces were observed, one that arose between the side walls of the metal coated silicon pillars (wi) and another that was observed between the gold structural projections at the top of silicon pillars (wo) as represented in Figure 1. With increasing texp, the wi gap was found to close and the wo gap was found to reach less than 10 nm for texp above 100 nm (Figure 1, Figure S1, ESI). Thus, the deposition thickness as determined by the sputter coating duration (factoring the sputter rates), could be used as a convenient determinant of the inter-pillar separation of the resulting 3D gold features. The preferential growth of the metal layer on the pillar tops yields gold nanoarrays with finger like projections that add to the height of the silicon pillars, resulting in a significant increase in surface area of the pillar arrays. The height of the gold pillars (h) increased systematically between 120 nm to 280 nm in steps of few nanometers at a time for sputter coating thickness (texp) varying between 0-120 nm respectively (Figure 2) .

Figure 2 Systematic increase in pillar heights, as a function of thickness (texp) of sputter coated gold films


7


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 8 of 17

Figure 3. Raman spectral intensities of NT were found to systematically increase with an increase in thickness of gold (texp, values indicated) as observed in (top) the spectra of NT and (bottom) a plot of intensities for the peak at 1372 cm-1 3.2. SERS performance. The volume of Au increases almost linearly with thickness, giving rise to an increase in surface roughness leading to enhanced light scattering and the associated coupling into surface plasmon polaritons, as well as by the formation of hot spots between the bulging nano-features and grains that are tens of nm in size. This in turn contributes to stronger scattering (extinction) and is expected to contribute to better SERS efficiency. Such metal thickness dependent effects have been observed for other self-organizing Au coated SERS substrates.

37

In addition, the positive tapering of the topography is seen to decrease the

reflectivity of the pillar arrays in relation to silicon pillars (Figure S2, ESI) which is expected to


8

Page 9 of 17

result in focusing and concentration of light in the inter-pillar gaps. 38 Due to the small size of the pillars the radiative decay rates are much weaker as compared to non-radiative decay rates resulting in better field penetration within the metal and higher SERS intensities.5 The gold nanopillar arrays were subsequently examined using 1-naphthalene thiol (NT) as a reporter molecule to probe SERS performance. The Raman intensity of peak of NT positioned at 1372 cm-1 (ring stretching mode) was followed as a function of the thickness of gold deposited, and the concentration of NT. The results show a systematic increase in the SERS intensities of NT with an increase of gold deposition thickness that appeared to saturate at texp of 100 nm. The standard deviation in SERS intensities was found to be below 10% in all cases.

1500 Raman Intensity (counts/mW/s)

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


1000

500

0 -1 0 1 2 3 4 5 6 7 10 10 10 10 10 10 10 10 10

[NT] (nM)

Figure 4. Concentration dependence of the Raman spectral intensities for NT (plotted for peak at 1372 cm-1) show lowest detection limits of 0.46 nM (74 ppb) concentrations (Refer to Figure S3, ESI for spectra) The gold nanopillar substrates with texp=120 nm that exhibited the maximum signal enhancement (Figure 3) were chosen to investigate the concentration dependence of Raman signal intensities, to determine the lowest detection limits. We systematically varied the concentration of NT in the millimolar to nanomolar range maintaining incubation duration of 20


9


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 10 of 17

h for all concentrations. Well-defined spectra of NT were attainable down to 1nM concentration of NT (Figure S3, ESI). The peak intensity at 1372cm-1 was found to vary linearly as a function of the concentration of NT plotted in logarithmic scale (Figure 4). The lowest detection limits (LOD) was estimated as 3*SD/b, where, SD indicates standard deviation and ‘b’, the slope of the calibration plot. LOD of 0.46nM (74 ppb) was obtained, assuming a standard deviation of 10% of the mean. The observed sensitivity is on par or better than instances reported in literature with similar molecules.

39-41

It should be noted that the range of concentrations used for the

investigations are of the order of, or smaller than the concentration needed to form a complete monolayer of NT on the gold pillar arrays. Such attributes of high sensitivity combined with a large dynamic range are especially useful for biosensing applications. The dependence of the SERS intensities on the thickness of gold deposited (texp) is in conformity with expectations set by the geometric attributes of the resulting pillars. The uniformity of the micellar template of

Engineering 3D Nanoplasmonic Assemblies for High Performance

Recommend Documents