Fabrication of Plasmonically Active Substrates Using Engineered

Subscriber access provided by University of Florida | Smathers Libraries

Article

Fabrication of Plasmonically Active Substrates using Engineered Silver Nanostructures for SERS Applications Menekse Sakir, Sami Pekdemir, Ahmet Karatay, Betül Küçüköz, Hasan Hüseyin Ipekci, Ayhan Elmal#, Gokhan Demirel, and M. Serdar Onses ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b12279 • Publication Date (Web): 19 Oct 2017 Downloaded from http://pubs.acs.org on October 19, 2017

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 28

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

Fabrication of Plasmonically Active Substrates using Engineered Silver Nanostructures for SERS Applications Menekse Sakir,a Sami Pekdemir,a Ahmet Karatay,b Betül Küçüköz,b,c Hasan H. Ipekci,a Ayhan Elmali,b Gokhan Demirel, d,* M. Serdar Onsesa,* a

Department of Materials Science and Engineering, Nanotechnology Research Center

(ERNAM) Erciyes University, Kayseri, 38039, Turkey b

Department of Engineering Physics, Ankara University, 06100 Besevler, Ankara, Turkey

c

Department of Chemistry and Chemical Engineering, Chalmers University of Technology,

41296 Gothenburg, Sweden d

Bio-inspired Materials Research Laboratory (BIMREL), Department of Chemistry, Gazi

University, 06500 Ankara, Turkey * Address correspondence to: [email protected] (MSO), [email protected] (GD)

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 28

ABSTRACT Demanding applications in plasmonics, electronics and biotechnology requires fabrication of plasmonically active substrates (PAS). Herein, we demonstrate a bottom-up, versatile and scalable approach that rely on direct growth of silver nanostructures from seed particles that were immobilized on polymer brush grafted substrates. Our approach is based on i) uniform and tunable assembly of citrate-stabilized gold nanoparticles on poly(ethylene glycol) brushes to serve as seeds, ii) the use of hydroquinone as an reducing agent which is extremely selective to the presence of seed particles confining the growth of silver nanostructures on the surface of the substrate. The diameter of the seed particles, concentration as well as ratio of reactants and duration of the growth process are investigated for large-area growth of silver nanostructures with high surface coverage and plasmonic activity. The resulting silver nanostructures exhibit high levels of SERS activity at two different laser lines and allow detection of molecules at concentrations as low as 10 pM. The plasmonic properties of the silver nanostructures are further studied using ultrafast pump-probe spectroscopy. Spatially defined silver nanostructures are fabricated through the seed particles that are patterned via a soft-lithography, showing the capabilities of the presented approach in device applications. KEYWORDS: silver nanostructures, seed-mediated synthesis, hydroquinone, plasmonics, SERS,


2

Page 3 of 28

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


1. Introduction Silver nanostructures exhibit unique optical,1 electrical2 and antibacterial3 properties which are of significant interest in applications including plasmonics, electronics and biotechnology. The optical properties of silver nanostructures, in particular, have received great attention over the last two decades due to their high efficiency of plasmon excitation and strong electromagnetic enhancement in the visible range.4 Most applications such as electronic devices and sensors require uniform placement of the silver nanostructures on the surface of solids for fabrication of plasmonically active substrates (PAS).5-7 Surface-enhanced Raman scattering (SERS) is one of the most promising example to these applications.8, 9 The strong localization of electromagnetic fields in silver nanostructures leads to enhancement of the signals received from molecules in Raman spectroscopy.10 This enhancement is greatly intensified when the individual silver nanostructures are placed in close-proximity inducing the hot-spots formation in the gap regions.11 This enhancement enables detection of trace amounts of molecules down to single entities showing great promise for areas including food safety,12 point of care diagnostic,13 and anti-counterfeiting.14 Therefore, PAS should not only include plasmonic nanostructures but also support frequent generation of hot-spots leading to strong enhancement of the electromagnetic fields. The enormous interest in silver nanostructures have resulted in development of a rich menu of techniques for their utilization in PAS. Lithography techniques combined with physical vapor deposition of metallic films have been widely investigated for fabrication of silver nanostructures.15-17 Although the uniformity across a substrate and spatial control are impressive, the need for costly lithographic tools together with issues in the roughness and polycrystallinity of the structures limit their usage. Colloidal approaches have enabled massively parallel synthesis of silver nanostructures with precise control over their size, geometry and therefore their


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 28

properties.18,19 A challenge for fabricating PAS relates to the uniform placement of the nanoparticles (NPs) in a way to generate hot-spots. In this context, substrates that are functionalized or patterned with self-assembled monolayers and polymer brushes may be used to localize silver nanostructures;20, 21 however, such processes work best with simple building block such as spherical NPs. This limitation mainly relies on the need for tailoring the ligand of the particle to interact with the underlying functionality of the substrate. Tailoring the ligands to place the colloidal NPs on substrates not only impose issues from synthesis perspective, but also may hinder the interaction of the nanoparticle surface with the analyte molecules in SERS applications. Another approach involves direct growth of metallic nanostructures from the surface of the substrate either by incorporating metal ions on the functional films followed by reduction or by using immobilized reducing agents placed in a solution containing metal ions.22-24 Here the density of grown NPs is either low or continuous films with bulk behavior are obtained. These issues prompt the development of complementary approaches for fabrication of PAS for SERS and other applications. A highly promising approach for fabrication of PAS is seed-mediated growth of silver nanostructures from immobile seeds placed on the surface of the substrates. Seed-mediated synthesis of NPs in colloidal solutions has been widely studied to generate nanostructures with varying composition and geometry such as core-shell, rod, cuboid and star shaped NPs.25-27 Despite its promise in synthesis of metallic nanostructures with unique plasmonic properties, this approach has not been fully exploited using immobile seeds placed on the surface of solid substrates. Only recently, Kim and coworkers28 have shown that gold nanoparticle seeds fabricated by block-copolymer lithography can be used to generate high density arrays of Au@Ag core-shell particles. This work employed hydroquinone (HQ), a type of reducing agent which can only


4

Page 5 of 28

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


support the growth of NPs in the presence of seeds that consist of clusters of metal atoms.29 The block-copolymer template confined the growth process and minimized interaction between the neighboring seeds favoring growth of discrete arrays of core-shell nanostructures. Despite localization of electric fields in the shell region, the seeds were separated by tens of nanometer resulting in inability to generate hot spots near the gaps between individual NPs limiting the sensitivity of detection down to 100 nM levels. Therefore, it is still highly desirable to directly grow silver nanostructures on substrates through seed-mediated synthesis having much higher SERS enhancements and without need for nanoscopic templates. In this work, we develop a simple, scalable and low-cost fabrication of PAS by seedmediated synthesis from gold NPs immobilized on polymer brush grafted substrates for ultrasensitive sensing via SERS. Robust and tunable assembly of citrate-stabilized gold NPs on grafted chains of poly(ethylene glycol) (PEG) serve as seeds for subsequent growth of silver nanostructures. A unique advantage of this scheme is that colloidal gold NPs with narrow size distribution are uniformly assembled on the surface of the substrate, serving as tunable reaction centers. Subsequent growth of silver nanostructures is achieved by seed-selective reduction of silver ions by HQ. Here the growth of the silver nanostructures on the seed particles is expected to significantly enhance the Raman scattering from the substrate through strong localization of electromagnetic fields in between the silver nanostructures and seed gold NPs as well as the in between the silver nanostructures as a result of the surface growth process that evolves from multiple seeds that are closely placed. The density and size of surface bound seeds together with the growth conditions are optimized for fabrication of PAS with frequent generation of hot-spots leading to extremely high levels of SERS intensities allowing detection of molecules at very low concentrations. The fabricated PAS are systematically characterized for their structural,


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 28

compositional, crystalline, plasmonic and SERS properties. Patterned silver nanostructures can be readily accomplished using soft-lithography approaches by spatially defined localization of the seed particles, showing the promise of the presented approach in fabrication of plasmonic devices.

2. Experimental 2.1. Materials Silicon wafers () were purchased from Wafer World Inc. Silver nitrate (AgNO3; ≥99.5 %), hydroquinone (HQ; ≥99 %), cyclopentanone, propylene glycol monomethyl ether acetate (PGMEA), chloroform, chlorobenzene and rhodamine 6G were obtained from SigmaAldrich. Poly(ethylene glycol) (PEG; Mn = 35.0 kg/mol, polydispersity index = 1.15) was purchased from Polymer Source Inc. Polydimethylsiloxane (PDMS) prepolymer and curing agent (Syllgard 184) were ensured from Dow Corning. SU-8 2050 was supplied from Microchem Inc. Citrate stabilized gold NPs were provided from Ted Pella Inc. All chemicals were of analytical grade. All solutions were prepared with distilled water. Pure nitrogen gas was used for drying. 2.2. Immobilization of seeds Silicon and glass substrates were cleaned in a UV-ozone chamber prior to the experiments. The substrates were then functionalized with end-grafted PEG chains by spin coating a film of the polymer followed by thermal annealing and washing as described previously.30 Citrate-stabilized gold NPs were then deposited on the PEG grafted substrates by dropping a solution (50 µL/cm2) of the particles in a humidified atmosphere for 1 h, followed by washing in water under sonication for 2 min and drying with nitrogen. Unless otherwise stated, the diameter of the gold NPs was 10 nm.


6

Page 7 of 28

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


2.3. Growth of silver nanostructures from immobilized seeds A growth solution containing 0.1 mM AgNO3 and 0.2 mM HQ in water was mixed with a magnetic stirrer for 15 min. The substrates with immobilized gold NPs were then immersed into the growth solution for 4 h, under constant agitation provided by a shaker. The substrates were then rinsed with water and dried with nitrogen. The mixing of the solution and the growth was performed in dark. 2.4. Patterned growth of silver nanostructures Elastomeric molds were fabricated by pouring PDMS prepolymer and curing agent in a ratio of 10:1, over the masters that consisted of patterns of SU-8 on silicon fabricated by photolithography as described elsewhere.31 PDMS molds were cured by heating at 65 oC for 2 h. PDMS molds were then placed over a freshly cleaned silicon substrate. A solution of 1% PEG in methanol was placed near the open-end of the channels. PEG solution immediately filled the channels through capillary action. PDMS mold was removed following drying of methanol. A thermal annealing at 120 oC for 5 min followed by washing in chloroform resulted in patterned PEG brushes. The immobilization of seed particles and growth of silver nanostructures were then performed using the procedures described above. 2.5. Structural characterization The morphology of the substrates following the immobilization and growth was imaged by scanning electron microscope (SEM; ZEISS EVO LS10) at 25 kV. The chemical composition of the nanostructures was determined by energy-dispersive X-ray spectroscopy (EDS; Bruker) attached to the SEM. X-ray thin film diffraction (XRD) pattern was recorded with a Rigaku SmartLab diffractometer operating at 40 kV and 30 mA by using Cu Kα radiation source. UV-Vis spectroscopy was obtained from Perkin Elmer Lambda 25. The thickness of the PEG brush was


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 28

measured via an ellipsometer (Gaertner LSE Stokes). The refractive index of PEG was assumed as 1.46. 2.6. SERS measurements Two different laser lines were used in the Raman spectroscopy. SERS measurements with a laser line of a wavelength of 532 nm were performed using an Alpha M+ Raman spectrometer (confocal micro-Raman microscope, Witec, Germany) with a power of 0.5 mW and fine-focusing 50× microscope objective (N.A. = 0.85). The reporter molecule consisted of rhodamine 6G and deposited from an ethanolic solution at a concentration of 100 µM and volume of 4 µL/cm2. Raman spectra were taken following evaporation of ethanol. The intensity mapping conditions of the instrument were 80 × 80 points at 40 × 40 µm2 areas with a 500 nm step (0.1 s/point). We derived the reported average intensity of the SERS measurement from the area mapping following the baseline correction. SERS measurements with a laser line of a wavelength of 785 nm were performed using a Delta Nu Examiner Raman Microscopy system with a 150 mW laser power, 20x objective, 3 µm spot size, and 30 s acquisition time. The Raman signal enhancement ability of the fabricated platform was analyzed using methylene blue (MB) as the Raman probe. In a typical measurement, a 5 µL of aqueous solution of MB having varying concentrations (10-3 M – 10-11 M) was placed on the PAS and allowed to dry. The Raman spectra were collected from at least ten different spots across the entire dried spot area. To calculate the enhancement factor, we have employed the reference Raman spectrum obtained from 0.1 M concentration of MB on the silicon wafer substrate. The same volume (5 µL) of MB was placed on the silicon substrate and allowed to dry leading to a coin-shaped area with a diameter of 4.1 ± 1.2 mm, corresponding to 2.39x1010 MB


8

Page 9 of 28

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


molecules/µm2. The enhancement factor (EF) was calculated by applying the following equation and using the Raman peak intensity at 1622 cm-1 for MB: EF = (NReference x IPAS)/(NPAS x IReference) where IReference and IPAS are the Raman intensities of 0.1 M MB on the reference silicon substrate and 10-5 M MB on the PAS, respectively. In the case of 10-5 M MB with a volume of 5 µL on the PAS, the drying procedure resulted in a coin-shaped area with a diameter of 3.8 ± 0.6 mm, corresponding to 2.67x106 MB molecules/µm2. NReference (1.69x1011) and NPAS (1.88x107) are the total number of MB molecules located in the laser spot area (7.065 µm2) on the reference silicon substrate and on the PAS, respectively. Note that all Raman measurements were done under the same experimental conditions (e.g., laser wavelength, laser power, microscope objective/lenses, and spectrometer). Under the plausible assumption that all of the analyte molecules within the laser spot are illuminated and contribute to the SERS spectra, the average EF from the fabricated PAS was calculated. 2.7. Ultrafast pump-probe spectroscopy measurements Ultrafast pump-probe spectroscopy measurements were performed using a Ti:Sapphire laser amplifier, an optical parametric amplifier system with 45 fs pulse duration and 1 kHz repetition rate (Spectra Physics, Spitfire Pro XP, TOPAS). In order to investigate the effect of the silver growth time on the plasmonic properties of the fabricated PAS, commercial pump-probe experimental setup (Spectra Physics, Helios) with a white light continuum probe was used. The pump wavelength was chosen as 400 nm and 550 nm to excite Ag and Au plasmon bands Pulse duration was measured by cross-correlation as 120 fs inside the pump probe setup.


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 28

3. Results and Discussion Figure 1a schematically describes our approach to fabricate PAS by growth of silver nanostructures from immobilized seed particles. To this end, silicon and glass substrates are first functionalized with PEG by end-grafting the polymer through its hydroxyl end-groups leading to a uniform, ultrathin (12±0.3 nm) and covalently bound layer for immobilization of the seed particles. End-grafted PEG chains (i.e. PEG brushes) provide a powerful interface for immobilization of citrate-stabilized gold NPs with the ability to modulate the density of the particles and inter-particle spacing as a function of the molecular weight of the PEG, the size of the particle, and number of cycles used to deposit the particles.30 In our experiments, unless otherwise stated, we employed gold NPs with a diameter of 10 nm as the seed particles, since these particles immobilized with high surface coverage and formed closely packed assemblies which is a typical behavior of particles that are smaller than 20 nm on PEG brushes with molecular weights higher than 20.0 kg/mol (Figure 1b).32 The substrate with the immobilized gold NPs is then placed in the growth solution which contains silver nitrate and HQ. The choice of HQ as a reducing agent is critically important for this study, since HQ reduces silver nitrate selectively on the immobilized gold NPs confining the growth process on the surface of the substrate. We did not observe the growth of silver nanostructures on the bare silicon and PEG brush grafted substrates in the absence of immobilized gold NPs proving the seed selective reducing mechanism of HQ (Figure 1c). In the case of substrates with immobile seed particles, silver nanostructures directly grew on top of the substrate (Figure 1d). The growth process was macroscopically uniform over the whole substrate (see supporting Figure S1 for more images), whereas the diameter and geometry of the nanostructures showed a certain degree of distribution. The EDX mapping of the silver and gold presented in Figure 1e showed the presence of both metals on the surface of the substrate


10

Page 11 of 28

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


suggesting the preservation of the seed particles in the process. X-ray diffraction (XRD) pattern exhibited the characteristic peaks corresponding to the (111), (200), (220) and (311) planes of the silver with a face-centered cubic crystal structure.33 The (111) plane was preferred in the growth process, probably due to the low energy of this plane.34

Figure 1. Fabrication of PAS by seed-mediated growth of silver nanostructures from immobile seeds. a) Schematic description of the process. b) SEM image of the seeds that consist of immobilized gold NPs (10 nm in diameter), c-d) SEM images following treatment with the growth solution for 4 h (0.1 mM AgNO3 and 0.2 mM HQ) on top of PEG grafted substrates in the c) absence, d) presence of immobilized gold NPs. e) EDX mapping of silver and gold for the substrate shown in part d. f) XRD pattern of the silver nanostructures. To evaluate the effect of the diameter of the immobile seed particles on the growth of silver nanostructures, gold NPs with a diameter of 10 nm, 30 nm and 50 nm were deposited on PEGbrush grafted silicon substrates leading to 2500, 113 and 23 particles per square micrometer,


11


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 12 of 28

respectively. The decrease of surface coverage of immobilized particles with increasing size of the particle arises from the size of particle dependent interaction between end-grafted PEG and citrate stabilized gold NPs. SEM images (Figure 2a) of the substrates following the seed-mediated growth showed that the size of the silver nanostructures increased with the diameter of the seed particles. The average sizes of the silver nanostructures were 122 nm, 153 nm, and 207 nm for 10 nm, 30 nm and 50 nm seeds, respectively (see supporting Figure S2 for the distribution of the sizes and details about determination of the particle sizes). Significantly higher number of immobile seeds present in the case of 10 nm particles in comparison to 30 nm and 50 nm ones likely resulted in the separation of the growth process into multiple sites leading to smaller silver nanostructures. The uniformity of the silver nanostructures over microscopic areas was investigated by mapping the SERS response of rhodamine 6G on the fabricated PAS. A droplet of rhodamine 6G was placed on the substrate and excited with a laser source of 532 nm following evaporation of the solvent. The characteristic peak of rhodamine 6G at 1360 cm-1 was then mapped over an area of 40x40 m2 (Figure 2b). SERS response was received over the entire substrate area confirming the growth of silver nanostructures with high surface coverage. The intensity of the signal showed a certain degree of variation (see supporting Figure S3 for the individual spectra taken across the substrate) Regions with low intensities manifested themselves as dark areas in the mapping. We found a slightly higher occurrence of such regions, in the case of the silver nanostructures grown from 30 nm and 50 nm seed particles. Similarly, the average intensity of the peak at 1360 cm-1 was the lowest for the seed particles of 50 nm. These results imply that the high surface coverage and closer placement of the seed particles with small sizes is advantageous for the growth of silver nanostructures with close-proximity. We therefore chose 10 nm gold NPs as the seed particles for the fabrication of PAS in the rest of the study.


12

Page 13 of 28

50 nm Au

30 nm Au

10 nm Au

(a)

200 nm

200 nm

200 nm

4 m

4 m

(b)

4 m

(c)

10 nm 30 nm 50 nm

Intensity ( a.u.)

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


600

700

800

900

1000

1100

1200 1300 1400 1500

Raman Shift (cm-1)

Figure 2. The effect of the diameter of the immobilized seed particles. a) SEM and b) SERS mapping images of the silver nanostructures grown on top of gold NPs of the varying diameter. SERS mapping images were obtained using the characteristic peak of rhodamine 6G at 1360 cm1

. c) Raman spectra of rhodamine 6G on top of the silver nanostructures. All Raman spectra and

mappings shown in this figure were acquired by placing 2 µL droplet of 100 µM solution of rhodamine 6G in ethanol followed by evaporation of the solvent using a laser source of 532 nm. The diameter of the gold NPs is given on top of the images. The growth was performed for 4 h in a solution containing 0.1 mM AgNO3 and 0.2 mM HQ.


13


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 14 of 28

The concentration of the reactants and duration of the growth process have a profound impact in the growth of silver nanostructures and their optical properties. Therefore, we investigated the effect of concentration of the reactants and molar ratio of silver nitrate to HQ. The silver nanostructures grew with high surface coverage in the case of 0.1 mM silver nitrate and 0.2 mM HQ which corresponds to the stoichiometric ratio.29 High concentrations (e.g. 1 mM) of silver nitrate resulted in growth of large silver nanostructures with poor surface coverage (see supporting Figure S4 for the images). Fixing the concentration of the silver precursor and reducing agent at these values, different growth durations (0.25-4 h) were investigated. SEM images presented in Figure 3 reveal that both the diameter and surface coverage of the silver nanostructures increased with the growth duration. The increase in the surface coverage is further supported by EDX measurements (see supporting Figure S5). In the case of 0.5 h growth time, the average size of silver nanostructures was ~69 nm with a partial coverage of the surface (see supporting Figure S6 for the size distribution). The size of the nanostructures approached 98 nm and 122 nm for the growth of silver nanostructures at 2 h and 4 h, respectively. The latter had an almost complete coverage of the substrate and the average gap distance between the silver nanostructures was below 100 nm with many instances of nearly touching nanostructures. The extinction spectra of the silver nanostructures grown on glass substrates presented in Figure 4a showed that all samples with the silver nanostructures had a broad peak centered around 400 nm which showed an increase in the intensity and slight red shift with the growth duration as consistent with the previous studies performed in colloidal solutions.35 The reference sample containing only the seed particles had a peak at 560 nm which is slightly larger than the colloidal form of the particles probably due to the coupling between the individual particles on the substrate. The intensity of the plasmon resonance peaks for the silver nanostructures was significantly higher than the seed particles, which is likely


14

Page 15 of 28

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


a result of the size and surface coverage of the silver nanostructures. SERS measurements further confirm the fabrication of PAS. The growth of silver nanostructures for 0.5 h resulted in 40 fold increase in the intensity of the characteristic peak of rhodamine 6G in comparison to the seed particles (Figure 4b). The enhancement was strongest in the case of PAS with the growth time of 4 h, as consistent with the complete surface coverage of the nanostructures observed in SEM and highest intensities observed in the UV-vis extinction spectra. In all growth durations, the entire substrate exhibited plasmonic activity (Figure 4c, see supporting Figure S7 for the individual spectra taken across the substrate). We found that the SERS activity remained highly stable for several weeks with slight decrease in the intensity of the signals probably associated with the oxidation of the silver (see supporting Figure S8). 0.25 h

0.5 h

(a)

0.75 h

(b)

(c)

400 nm

400 nm

1h

400 nm

2h

(d)

4h

(e)

400 nm

(f)

400 nm

400 nm

Figure 3. The effect of the growth time on the fabrication of PAS. SEM images of the silver nanostructures grown for a) 0.25 h, b) 0.5 h, c) 0.75 h, d) 1 h, e) 2 h, f) 4 h. The diameter of the


15


seed particles was 10 nm and the growth was performed in a solution containing 0.1 mM AgNO3 and 0.2 mM HQ.

(a) Absorbance ( a.u.)

398.1 nm 403.1 nm

4h 2h 0.5 h 10 nm Au

410.9 nm

559.1 nm

320

(b)

370

420

470

Wavelength (nm)

520

570

44hhours 22hhours 0,5hhours 0.5 10nm nmAu Au 10

Intensity ( a.u.)

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 16 of 28

600

700

800 900 1000 1100 1200 1300 1400 1500 1600

2h

Raman Shift (cm-1)

4h

(c)

Figure 4. The effect of the growth time on the plasmonic properties of PAS. a) UV-vis extinction spectra, b) SERS spectra and c) SERS mapping images of the fabricated PAS grown for 2 h and 4 h, respectively. The diameter of the seed particles was 10 nm and the growth was performed in a


16

Page 17 of 28

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


solution containing 0.1 mM AgNO3 and 0.2 mM HQ. SERS measurements were performed with 100 µM concentration of rhodamine 6G using a laser source of 532 nm. In order to probe the plasmonic properties and ultrafast dynamics of the fabricated PAS, we used ultrafast pump-probe spectroscopy. This spectroscopy technique relies on observation of the absorbance changes in both the decay kinetics and transient spectra and informs about the electron transfer and possible other transitions in plasmonic nanostructures. The transient absorption measurements were carried out with femtosecond temporal resolution that is important for investigation of ultrafast energy and electron transfer between nanostructures. There are several investigations36, 37 of exciton plasmon coupling properties of hybrid nanostructures by using this ultrafast spectroscopic technique. These experiments give us the opportunity to investigate the electron transfers between gold and silver nanostructures. To excite the plasmons associated with the silver and gold nanostructures, we used pump wavelengths of 400 nm and 550 nm, respectively. Figure 5a-c presents the transient absorption spectra of fabricated PAS with different silver growth times upon excitation at 550 nm. There are two bleach signals around 475 nm and 650 nm which correspond to plasmon bands of silver and gold nanostructures, respectively. The intensity of plasmon bleach signal (~475 nm) for silver is stronger than that of gold (~650 nm) even though the pumping wavelength is near the plasmon band of gold. This result indicates that there is an electron transfer between gold and silver nanostructures. The electron transfer occurs in ultrafast time scale and the transfer rate is under time resolution of our ultrafast pump-probe experimental setup. Therefore, two bleach signals were observed simultaneously when the PAS was excited with 550 nm. In order to understand the origin of these electron transfers, a laser beam at 400 nm was used as a second pump wavelength to excite the plasmon bands of silver. Similarly, two bleach signals occur, but this time the intensity of the bleach signals are almost the same (see supporting


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

Page 18 of 28

Figure S9). When the substrate with just the immobilized gold NPs (10 nm) in the absence of any silver growth were excited with 400 nm, there are also similar transient characteristics with 550 nm excitation experiments. This result shows that 400 nm pump wavelength excites both silver and gold nanostructures. The decay curves presented in Figure 5d further supports the electron transfer between gold and silver nanostructures. The decay profiles show the PAS with silver nanostructures has a faster decay rate (~1.5 ps) than the just immobilized gold NPs (~5 ps) upon excitation at 550 nm. All these findings prove that there is an electron transfer between gold and silver nanostructures and the electron transfer makes decay profiles of the nanostructure faster. The highly rapid electron transfer (

Fabrication of Plasmonically Active Substrates Using Engineered

Recommend Documents