Aluminum Nanocrystals: A Sustainable Substrate for Quantitative

Communication

Aluminum Nanocrystals: a sustainable substrate for quantitative SERS-based DNA detection Shu Tian, Oara Neumann, Michael J. McClain, Xiao Yang, Linan Zhou, Chao Zhang, Peter Nordlander, and Naomi J. Halas Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.7b02338 • Publication Date (Web): 30 Jun 2017 Downloaded from http://pubs.acs.org on July 1, 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.

Nano Letters 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 25

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

Nano Letters

Aluminum Nanocrystals: a sustainable substrate for quantitative SERS-based DNA detection Shu Tian* §, Oara Neumann♯ §, Michael J. McClain* §, Xiao Yang¶ ♯, Linan Zhou* §, Chao Zhang♯ §, Peter Nordlander¶ ♯ and Naomi J. Halas* ¶ ♯ § † *

Department of Chemistry, ¶Department of Physics and Astronomy, ♯ Department of Electrical and Computer Engineering, §Laboratory for Nanophotonics and the Smalley-Curl Institute, Rice University, 6100 Main Street, Houston, Texas 77005

†

Corresponding Author: Naomi J. Halas

E-mail: [email protected]; Phone: (713)348-5612

1

ACS Paragon Plus Environment

Nano Letters

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 25

ABSTRACT Since its discovery in the 1970’s, surface-enhanced Raman scattering (SERS) has been primarily associated with substrates composed of nanostructured noble metals. Here we investigate chemically synthesized nanocrystal aggregates of Aluminum, an inexpensive, highly abundant, and sustainable metal, as SERS substrates. Al nanocrystal aggregates are capable of substantial near-infrared SERS enhancements, similar to Au nanoparticles. The intrinsic nanoscale surface oxide of Al nanocrystals supports molecule-substrate interactions that differ dramatically from noble metal substrates. The preferential affinity of the ssDNA phosphate backbone for the Al oxide surface preserves both the spectral features and nucleic acid cross sections relative to conventional Raman spectroscopy, enabling quantitative ssDNA detection and analysis.

KEYWORDS: Aluminum, nanocrystals, SERS, DNA, aggregates

Aluminum, the most abundant metal on earth, is attracting increasing interest as a plasmonic material due to its vivid plasmon response in the visible region of the spectrum,1-3 along with its greatly reduced cost and inherent sustainability. Al nanostructures have photocatalysis,4,

5

shown promising properties

photodetection,6 flat-panel displays,7,

in applications 8

such as

and sensing9-15, including

surface-enhanced spectroscopies. These include localized surface plasmon resonance (LSPR) sensing,9 surface-enhanced infrared absorption spectroscopy (SEIRA), 10 and surface-enhanced Raman spectroscopy (SERS).11-15 Thus far, SERS substrates based 2


Page 3 of 25

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

Nano Letters

on plasmonic Al have primarily focused on nanostructures fabricated using e-beam lithography, such as nanoparticle11 or nanohole arrays12 and bowtie structures,13 along with roughened films14 and gratings.15 Recently we reported the chemical synthesis of Al nanocrystals (NCs) of uniform shape and controlled size.16 The plasmon resonance of Al NCs shows a wide tunability, from the UV to the near-infrared region of the spectrum, as their size is increased. Since chemically synthesized noble metal nanoparticles condensed into aggregate substrates support strong SERS enhancements,17, 18 it is interesting to investigate whether Al NC aggregates have similar potential as SERS substrates, and could ultimately prove to be a more sustainable alternative SERS substrate. A distinguishing feature of Al NCs is their thin (2-4 nm), stable, amorphous surface oxide layer that inhibits further oxidation of the metal. This thin oxide surface layer can be seen as quite advantageous for SERS. For example, SiO2 and Al2O3 nanoscale coatings were grown around Au nanoparticles (NPs), and the resulting shell-isolated nanoparticles, known as “SHINERS”, demonstrated broadly versatile SERS detection.19, 20 This is because nanoscale surface oxide layers can provide binding sites for a variety of functional groups such as carboxylic/phosphoric acids, silanes, and amides, providing very different molecule-substrate interactions relative to conventional Au or Ag SERS substrates. For Al, its intrinsic nanoscale surface oxide layer expands the types of molecules that can be bound to its surface relative to noble metals, where molecules are commonly functionalized with thiol or amine groups to facilitate molecule-substrate binding. The oxide layer can enable the SERS detection of 3


Nano Letters

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 wider range of non-resonant molecules, including biomolecules such as DNA, where label-free detection is highly desirable. In this study, we investigate Al NC aggregates as near-infrared SERS substrates. Theoretical simulations of the wavelength-dependent response of an Al dimer were performed to estimate the electromagnetic field enhancement generated by the hot spots in Al NC aggregates under laser illumination, and compared with experimental dark field microspectroscopy of isolated Al NC dimers. The properties of Al NC aggregates are compared to those of Au NP aggregates, and the differing molecule-substrate affinities of these two aggregate systems were examined using near-IR SERS (785 nm laser illumination). We investigate the Al NC surface oxide-molecule binding through carboxyl and silane functional groups, then extend these studies to single stranded DNAs (ssDNAs). Al NC substrates enable us to perform the first label-free SERS detection of ssDNA without any modification to either the molecule or the substrate. A comparative study of the SERS spectra of ssDNA on Al and Au aggregate substrates reveals very different spectra and relative intensities for all four bases. Due to the affinity of the ssDNA phosphate backbone for the Al NC substrate, the SERS spectra and relative SERS cross sections for all four bases are very similar to the bulk Raman spectra. We show that label-free, non-functionalized, quantitative ssDNA analysis is possible with as-synthesized Al NC-based substrates.

4


Page 4 of 25

Page 5 of 25

A

B

Experimental Theoretical

Extinction (a.u.)

Extinction (a.u.)


Al 400

Au 600

800

1000

400

Wavelength (nm)

800

1000

D

1.0

Scattering Intensity (a.u.)


600

Wavelength (nm)

C 0.8 0.6 0.4 0.2 0.0 400

600

800

1.0 0.8 0.6 0.4 0.2 0.0 400

1000

600

800

1000

Wavelength (nm)

Wavelength (nm)

F

E

H

300

Field Enhancement (E/E0)

G Scattering 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

Nano Letters


400

500

600

700

70 60

3nm oxide layer no oxide layer

50 40 30 20 10 0 300

Wavelength (nm)

400

500

600

700

Wavelength (nm)

Figure 1. Experimental and theoretical extinction spectra of: (A) monodisperse Al NCs in isopropanol and (B) Au NPs in aqueous solution. Experimental dark field spectra of: (C) Al NC and (D) Au NP aggregates. Inserted are the SEM images (inserted, scale bar = 500 nm). (E) SEM image of an Al dimer (scale bar = 200 nm). (F) Scheme of an Al dimer with 2 nm gap. (G) Experimental and theoretical dark field spectrum of the Al dimer (with 3 nm oxide layer). (H) Theoretical near-field enhancement of Al dimers with/without 3 nm oxide layer. 5


Nano Letters

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

We begin by comparing the plasmonic properties of Al NCs, Au NPs, and their aggregates (Fig. 1). Al NCs were synthesized by a method previously reported (see Methods).16 The experimental and theoretical extinction spectra of monodisperse Al NCs in isopropanol and Au NPs in aqueous solution are shown (Fig. 1A, B). The extinction spectrum of monodisperse Al NCs exhibits a primary peak at ~570 nm corresponding to the dipolar plasmon mode, and a broadband absorption feature at ~850 nm (~1.4 eV) corresponding to the interband transition in Al. In the well-studied Au NP system,18, 21, 22 the dipolar plasmon peak at ~570 nm overlaps with the Au interband transition at ~500 nm. The theoretical extinction spectra (dashed lines) for both nanoparticles were obtained using Mie theory, assuming spherically shaped nanoparticles. The slight mismatch between the experimental and theoretical spectra of Al can be attributed to a lack of quantitative information concerning the dielectric function of the Al NC surface oxide (shown in Figure S1), and to the polyhedral morphology. Close-packed Al and Au aggregates with nano-sized gaps (Figs. 1C, D) were generated by drop-dry deposition onto a quartz substrate. The corresponding scanning electron microscopy (SEM) images and optical responses, obtained using dark field scattering spectroscopy, are shown. The Al NCs are primarily truncated trigonal bipyramids, with a few truncated octahedra and irregular crystals (Fig. 1C, inset). The average Al NC diameter is 144.5 ± 14.3 nm. The Au NPs are predominantly spherical (Fig. 1D, inset), with an average diameter of 57.5 ± 18.0 nm. The scattering spectra of 6


Page 6 of 25

Page 7 of 25

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

Nano Letters

both Al and Au aggregates show a broadband feature across the visible region. In the Al aggregate scattering spectrum, a dip at nominally 850 nm corresponds to the interband transition; for Au aggregates, the interband transition corresponds to reduced scattering at wavelengths below ~500 nm. To approximate the near-field properties of the Al NC aggregate, we examined an isolated Al dimer (Fig. 1E) using dark field microspectroscopy, and compared its properties with a theoretical model of the structure (Fig. 1F). A finite-difference timedomain (FDTD) simulation was performed on an Al dimer with a 3 nm oxide layer around each NC, with a 2 nm air gap between the two NCs, in an edge-to-edge configuration corresponding to the experimental structure. The theoretical spectrum corresponds relatively well to the experimentally obtained dark field spectrum (Fig. 1G). The broader features of the theoretical spectrum relative to the experimental spectrum are most likely due to the higher purity of Al nanocrystals relative to the Al thin films upon which the literature dielectric permittivity used in the theoretical calculation was based.3 Based on this agreement, we calculated the near-field enhancement as a function of wavelength for this structure, with and without the presence of the oxide layer (Figure 1H). The near-field enhancement calculated in the estimated 2 nm Al dimer gap is nominally 20-30 times throughout the visible region of the spectrum for this geometry, which would give rise to an enhancement factor of ~105106 in the junction. In reality, the enhancement factor may be slightly larger than this prediction, due to the recently measured reduced absorption in ultrapure Al at these wavelengths.3 While this SERS enhancement is smaller than what can be reached for 7


Nano Letters

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

noble metal nanostructures specifically optimized for single-molecule SERS,23, 24 it does suggest that Al NC aggregates with nano-sized gaps could be very useful for SERS applications. It is also apparent in this simulation that the presence of the oxide layer reduces the field enhancement, and also redshifts the maximum field enhancement beyond 700 nm for Al NCs in this size range.

8


Page 8 of 25

Page 9 of 25

A

Al

NR

Raman 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

Nano Letters

Au

B

Al

NR

Au

600

800

1000 1200 1400 1600 1800

Raman Shift (cm-1) Figure 2. (A) SERS spectra of p-aminophenyltrimethoxy silane (APhS) and (B) paminobenzoic acid (PABA) on Al (red) and Au (blue) substrates, with normal Raman spectra (black) as reference.

To investigate the differing surface chemistries between Au NP- and Al NC-based substrates, carboxylic acid and silane molecules with para-amino groups were chosen

9


Nano Letters

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

as analytes. These molecules should bind to the Al oxide through their carboxylic acid or silane groups, respectively, and bind to the Au NP surfaces through their amine groups. The SERS spectra of p-aminophenyltrimethoxy silane (APhS) on both Au NP and Al NC aggregate substrates are shown in Figure 2A, along with the normal Raman spectrum of p-APhS as a reference. The peaks at 1132 cm-1 and 838 cm-1, assigned to the Si-O-C stretching and Si-C rocking modes, appear more prominent on the Al NC aggregate substrate. For p-APhS on the Au NP substrate, the SERS spectrum is more complex, due to the binding of the amine group to Au, along with possible hydrolysis and polymerization of the silane groups of the molecule. The different binding mechanisms of p-aminobenzoic acid (PABA) are also observed by comparing SERS spectra on Au NP and Al NC aggregate substrates (Fig. 2B). Most features in both spectra correspond to the normal Raman spectrum of PABA. However, on the Al NC substrate, the peak at 1439 cm-1, assigned to the -COO- symmetric stretching mode, indicates binding of the carboxylate group to the oxide surface. This peak does not appear for PABA on the Au NP substrate. Instead, a peak at 1362 cm-1 appears, characteristic of a C-N stretching mode, enhanced due to the interaction of the nitrogen atom of the amine group with the Au NP surface. These features show that Al NC aggregates can serve as a near IR SERS substrate that binds functional groups other than thiol or amine moieties, potentially enabling SERS detection of a wider range of molecular species.

10


Page 10 of 25

Page 11 of 25

B

d(Cytosine)9 Al NR

Au 400

d(Adenine)9

Normalized Intensity


A

600

800 1000 1200 1400 1600 1800

400

Raman Shift (cm-1)

C

600

800 1000 1200 1400 1600 1800

Raman Shift (cm-1)

D

d(Guanine)9


d(Thymine)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

Nano Letters

400

600

800 1000 1200 1400 1600 1800

Raman Shift (cm-1)

400

600

800 1000 1200 1400 1600 1800

Raman Shift (cm-1)

E

F

A

C

T

Al2O3 Al

G

Figure 3. SERS spectra of 9-mer ssDNA molecules with four different bases: (A) dC9, (B) dA9, (C) dT9, and (D) dG9 on Al (red) and Au (blue) substrates, with normal Raman spectra (black) as reference. (E, F) Illustration of different surface binding mechanism of ssDNA molecules onto Au and Al substrates.

Another functional group that should exhibit a strong affinity with the oxide layer of Al NC aggregates is phosphate,25 which should facilitate SERS detection of DNA. Currently, Au and Ag nanostructures have been studied extensively as SERS substrates for ssDNA detection.26-28 Previous studies have shown that SERS of unlabeled ssDNA 11


Nano Letters

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 on Au substrates exhibits poor reproducibility; large spectral variations have been attributed to the differing affinities of the various nucleic acids towards the Au surface.29-32 This binding is typically through covalent interactions of the noble metal substrate with the base ring nitrogen and exocyclic amino/keto groups, resulting in the formation of randomly coiled configurations of ssDNA on the Au surface with multiple binding sites. To avoid non-specific interactions between ssDNA and the metallic substrate surface, either the molecule or the substrate needs to be modified. One strategy is to form a well-organized self-assembled monolayer of ssDNA on an Au substrate surface by terminating the ssDNA chains with a thiol group.33, 34 However, thiolation adds substantial complexity to the detection process: with this approach, SERS-based DNA detection is not truly a label-free method. Several strategies aimed at modifying the metal surface have been proposed, such as functionalizing the surface with a probe molecule,17 optimizing intermolecular electrostatic interactions by modifying the surface charge,35-37 or introducing a protective layer of oxide. 38 On the contrary, with its native oxide layer already in place, Al NC aggregate substrates seem ideal for labelfree SERS-based DNA detection. The SERS spectra of 9-mer ssDNA oligomers with four different bases (adenine, cytosine, thymine and guanine) on Al NC (red) and Au NP (blue) aggregate substrates are compared in Figure 3 (A-D), with normal Raman spectra (black) of ssDNA molecules on quartz as reference. All spectra are normalized for comparison of peak positions. The differences between the Al NC-based and the Au NP-based SERS spectra are quite striking. It is immediately apparent that there are many more spectral 12


Page 12 of 25

Page 13 of 25

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

Nano Letters

similarities between the normal Raman spectrum and the SERS spectrum of the ssDNA oligomers on the Al NC substrate than on the Au NP substrate. On the Au NP substrate, strong binding between the bases and the Au substrate occurs, shifting the characteristic peaks assigned to ring stretching and breathing vibrations by 10-40 cm-1. For example, the peaks corresponding to ring breathing mode of adenine at 727 cm-1 shifts to 738 cm1

for poly dA9, and the cytosine mode at 785 cm-1 shifts to 798 cm-1 for poly dC9. The

ring stretching mode shifts from 1369 cm-1 to 1337 cm-1 for poly dT9 and from 1365 cm-1 to 1347 cm-1 for poly dG9. However, no such peak shifts are observed for the ssDNA oligomers on the Al NC substrate. Another noticeable feature in this spectral comparison is that two peaks, at 1096 cm1

and 786 cm-1, can be observed for all SERS spectra on Al NC substrates. These

spectral features can be assigned to symmetric stretching and skeleton stretching vibrations of the phosphodioxy (PO2-) group, respectively. This provides strong evidence that the backbone phosphate groups of the ssDNA interact strongly with the Al NC surface. These symmetric stretching and skeleton stretching vibration modes of PO2- were not observed on the Au NP substrate. The numerous spectral differences of ssDNA on Au NP and Al NC substrates indicate that ssDNA binds to the Al NC and the Au NP surfaces through very different mechanisms. The proposed binding mechanisms of ssDNA on Al/Al oxide and Au substrates are illustrated in Figures 3E and 3F. On the Au surface, ssDNA binds predominantly through the nitrogen-constituent groups, substantially shifting their characteristic Raman peaks. On the Al NC oxide surface, however, ssDNA molecules 13


Nano Letters

bind through the ssDNA phosphate backbone, preserving the Raman features of the bases and allowing for reliable identification. The strong similarity between the SERS spectra on Al NC substrates and the normal Raman spectra may also possibly indicate that ssDNA oligomers form ordered assemblies on the oxide surface layer of the Al NC aggregate substrate.

A

C

B

ss20-2 ss20-1

35

Reference

Experimental

ss20-1 30

Base Percentage (%)

Sequences (5’-3’): ss20-1: ACG CTA ATG CAT GCT GAC TG ss20-2: TGC GAG ACG CTG CAC GCA CT


25 20 15 10 5 0

D

C

A

G

400 600 800 1000 1200 1400 1600 1800 -1

Raman Shift (cm )

T

45

ss20-2

40

Base Percentage (%)

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 25

35 30 25 20 15 10 5 0

C

A

G

T

Figure 4. (A) Sequences of two 20-mer ssDNA oligomers used to investigate SERSbased DNA quantification. (B) SERS spectra of the two 20-mer ssDNAs given in (A). All peaks were normalized to the intensity of the PO2- symmetric stretching mode at 1096 cm-1. (C, D) Quantification results of nucleobase content (%) in each 20-mer ssDNA oligomer (striped) compared with known composition (solid).

Since these observations strongly indicate that all four ssDNA oligomers bind to the Al NC substrate through the same mechanism, their relative SERS intensities should directly reflect the relative Raman cross sections of the constituent bases. Previous 14


Page 15 of 25

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

Nano Letters

studies of ssDNA SERS on Au substrate have shown that different bases adsorb on the Au surface competitively. For example, the relative adsorption affinity of bases on the Au surface is A ≫ G ≥ C ≫ T,39 with the adsorption of adenine strongly dominating over the other bases due to a stronger interaction with the Au surface. To validate the binding affinity of ssDNA on the Al NC oxide surface, the SERS spectrum of each oligomer was obtained (Fig. S4). All spectra were normalized to the peak intensity of the PO2- symmetric stretching mode at 1096 cm-1 as an intrinsic internal standard,36, 37 and peak intensities were compared. After normalizing, we find that the relative Raman intensity of four bases is C ≈ A > T > G, consistent with a recent study where the relative Raman intensities of the ring breathing modes of the four bases were investigated.37 In clinical diagnostics, DNA base quantification remains a significant challenge, due to lack of specific binding, poor reproducibility, and sample complexity. The competitive affinity of bases on Au NPs makes it virtually impossible to obtain quantitative information using Au NP-based SERS. However, Al NC-based SERS substrates enable the quantitative analysis of pristine ssDNA directly from SERS spectra. Here we applied a mathematical approach to determine the base contents in two random 20-mer ssDNA oligomers (Fig. 4A). A multivariate regression model was used to determine the relative base content from the SERS spectra of each random sequence (Fig. 4B). More details are described in the Supporting Information. The quantification results are in good agreement with the reference values for both sequences (Fig. 4C, D). The compositions obtained from SERS analysis agree well with

15


Nano Letters

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 25

the reference compositions of ss20-1 and ss20-2 with average errors of 1.6% and 2.8%, respectively. Besides the different surface chemistries of the Al and Au substrates demonstrated in this work, another important feature of SERS is the sensitivity. As shown in Fig. 1H, the enhancement factor of an Al dimer is on the order of 105-106, comparing to 106-107 for Au dimer based on previous research.33, 40 This difference is due primarily to the presence of the passivating oxide on the Al NC surface, but may also be affected by the electronic structure of Al at the laser excitation wavelength. Experimentally, this difference in enhancement factor is evident by the lower excitation power and shorter integration time on the Au substrate relative to the Al substrate. For the ssDNAs, the detection limit is estimated to be ~2 μM (see Fig. S5). However, this detection limit is based on the specific measurement conditions used in this paper, which were chosen for a comparison of Al with Au nanoparticle aggregate substrates. The sensitivity of Al NC substrates could be further improved by many methods.

For example, higher

excitation powers, longer integration times, and averaging more scans, are all possible with this robust Al NC-based substrate.

Importantly, the size of the Al NCs could be

optimized to blueshift the aggregate plasmon peak away from the energy of the Al interband transition, which should increase the hot spot intensities and further enhance the SERS signal when excited at a more optimal wavelength. In summary, we have demonstrated the potential use of Al NC aggregates as a plasmonic substrate for SERS. The native oxide layer serves as a valuable linker between molecules and substrate, prohibiting non-specific adsorption on the Al NC 16


Page 17 of 25

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

Nano Letters

surface. Al NC aggregates, as synthesized, are SERS substrates that enable the first quantitative label-free detection of ssDNA with no modification to either the ssDNA or the substrate surface. Al NC aggregates show outstanding potential as highperformance substrates that are ultimately low-cost, disposable and recyclable. In addition, they also show a unique promise for as-is detection and quantification of chemical and biological molecules that could be useful for a wide variety of sensing, assaying, and diagnostic applications.

Materials Required. N,N-dimethylethylamine alane, tetrahydrofuran, titanium (IV) iso-propoxide, 1,4-dioxane, and oleic acid were purchased from Sigma-Aldrich and dried and distilled under inert atmosphere before use. Hydrogen tetrachloroaurate trihydrate (HAuCl4∙3H2O), potassium carbonate (K2CO3), formaldehyde, stearic acid and p-aminophenyltrimethoxy silane were purchased from Sigma-Aldrich and used without further purification. All ssDNA molecules were purchased from Integrated DNA Technologies (IA, USA) and used without further purification.

Al Nanocrystal Synthesis. The Al NCs were synthesized according to previously reported protocols with minor modifications.16 10 mL of tetrahydrofuran (THF) and 15 mL of 1,4-dioxane were added to a dry Schlenk flask under Ar atmosphere. The reaction flask was heated to 40 °C in an oil bath. As the solvents were stirring, 6.5 mL of 0.5 M N,N-dimethylethylamine alane in toluene was syringed into the flask, followed by the 17


Nano Letters

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

addition of 0.5 mL of a 3.3 mM solution of titanium (IV) iso-propoxide in toluene. The colorless solution immediately turned to brown. After two additional hours, the solution color changed from brown to gray indicating the formation of Al NCs. The reaction was allowed to proceed for an additional hour at room temperature under vigorous stirring. Then the Al NCs were exposed to air for the formation of oxide layer, which allows the particles to stay stable in iso-propanol for months. The NCs were cleaned by centrifugation and sonication for three cycles with a mixture of THF and toluene followed by two cycles with iso-propanol. The shapes of Al NCs were a mixture of truncated trigonal bipyramids, truncated octahedra, and irregular crystals. The reaction schematic is shown below: (𝐶𝐻3 )2 𝐶2 𝐻5𝑁𝐴𝑙𝐻3

𝑇𝑖(𝑂𝑃𝑟)4 𝐴𝑙 + 𝐻2 + (𝐶𝐻3 )2 𝐶2𝐻5 𝑁 𝑇𝐻𝐹/1,4 − 𝑑𝑖𝑜𝑥𝑎𝑛𝑒

Au Nanoparticle Synthesis. Au NPs were synthesized by reducing HAuCl4 with formaldehyde. 50 mg of K2CO3 and 3 mL of 1% HAuCl4 solution were added to 200 mL of deionized water and shaken vigorously to mix well. The solution was stored in a dark cabinet at room temperature for 24h before use. 300 µL of formaldehyde was added to the solution and stirred for 1h. The colorless solution turned pink indicating the formation of Au NPs. The Au NPs were washed three times with deionized water.

SERS measurements. All normal Raman and SERS spectra were acquired in ambient atmosphere under 785 nm excitation wavelength. The SERS on Au, Al substrate and the normal Raman spectra were measured under 0.0038, 0.038, 3.8 mW 18


Page 18 of 25

Page 19 of 25

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

Nano Letters

laser power and 10, 40, 10 s integration time, respectively. The SERS spectra on Al substrate were averaged over 50 scans. All spectra are baseline subtracted based on a built-in function of the instrument. All peak assignments are based on previous studies.41

PABA and APhS functionalization. Al NCs and Au NPs were functionalized by mixing well with 10 mM PABA/APhS in IPA. After 60 minutes, they were washed three times by centrifugation and re-dispersion in IPA to remove non-specific binding molecules. The functionalized Au and Al were then immobilized onto a quartz substrate by drop-dry deposition. Quartz substrates were cleaned before use by sonicating for 20 min in acetone and ethanol, followed by 30 s Ar plasma cleaning. Films of functionalized aggregates supported on quartz were used as substrates for SERS, SEM, and dark field measurements.

ssDNA functionalization. 50 μL of as-synthesized Al NCs and Au NPs were centrifuged down and dried in vacuum to remove all solvent, respectively. The number of nanoparticles used in each sample was ~1013 particles/mL determined by dividing the experimental extinction intensity by the simulated extinction coefficient of Au and Al respectively. Based on previous study on maximum loading density of ssDNA on Au NPs42 and the relative surface area, the ratio of nanostructures and ssDNA is set to 1:10000 for fully coverage with an excess of ssDNA molecules. The ssDNA molecules were attached to the nanostructures by overnight incubation in 50 μL of 0.2 mM ssDNA 19


Nano Letters

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

aqueous solution. The mixture was sonicated until completely re-suspended. The functionalized nanoparticles were then immobilized on a quartz substrate by drop-dry deposition and dried under vacuum. The normal Raman samples were prepared by drop-drying ssDNA aqueous solution onto quartz substrate and dried under vacuum.

Instrumentation. Extinction spectra were measured on a Cary 5000 UV/Vis/NIR spectrometer. SEM images were obtained on a FEI Quanta 650 FEG Scanning Electron Microscope. Dark field measurements were performed on a Zeiss Microscope coupled with a Princeton Instrument SP2150 Spectrometer and a CCD detector. XPS spectrum was taken on a PHI Quantera SXMTM Scanning X-ray Microprobe system. TEM measurements were performed on a JEOL 2100 Transmission Electron Microscope. SERS spectra were measured on a Renishaw inVia Raman Microscope.

ASSOCIATED CONTENT Supporting Information HRTEM and XPS spectrum of Al NC. SERS spectra of equimolar mixture of ssDNAs. SERS spectra of ssDNAs of different lengths. SERS spectra of 9-mer ssDNAs with four different bases. SERS spectra of ssDNAs with different concentrations. Definition and method on multivariate regression.

ACKNOWLEDGMENTS The authors would like to thank David Renard and Hossein Robatjazi for helpful discussions, Wenhua Guo for guidance on TEM, and Bo Chen for guidance on XPS. 20


Page 20 of 25

Page 21 of 25

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

Nano Letters

This work was supported by the Army Research Office (ARO) MURI grant W911NF12-1-0407, the Air Force Office of Scientific Research (AFOSR) MURI grant FA955015-1-0022, the Defense Threat Reduction Agency (DTRA) grant HDTRA1-16-1-0042, the National Science Foundation (NSF) grant ECCS-1610229, the Robert A. Welch Foundation grants C-1220 (N. J. H.) and C-1222 (P. N.), and the J. Evans Attwell-Welch Fellowship grant L-C-0004.

REFERENCES 1.

Langhammer, C.; Schwind, M.; Kasemo, B.; Zoric, I. Nano Lett. 2008, 8, 1461-71.

2.

Knight, M. W.; King, N. S.; Liu, L.; Everitt, H. O.; Nordlander, P.; Halas, N. J. ACS

nano 2014, 8, 834-40. 3.

Cheng, F.; Su, P. H.; Choi, J. H.; Gwo, S.; Li, X. Q.; Shih, C. K. ACS Nano 2016,

10, 9852-9860. 4.

Zhou, L.; Zhang, C.; McClain, M. J.; Manjavacas, A.; Krauter, C. M.; Tian, S.;

Berg, F.; Everitt, H. O.; Carter, E. A.; Nordlander, P.; Halas, N. J. Nano Lett. 2016, 16, 1478-84. 5.

Peng, S. Y.; Sheldon, M. T.; Liu, W. G.; Jaramillo-Botero, A.; Goddard, W. A.;

Atwater, H. A. Appl. Phys. Lett. 2015, 106, 023102. 6.

Zheng, B. Y.; Wang, Y.; Nordlander, P.; Halas, N. J. Adv. Mater. 2014, 26, 6318-23.

7.

Huang, Y. W.; Chen, W. T.; Tsai, W. Y.; Wu, P. C.; Wang, C. M.; Sun, G.; Tsai, D.

P. Nano Lett. 2015, 15, 3122-7. 8.

Olson, J.; Manjavacas, A.; Liu, L.; Chang, W. S.; Foerster, B.; King, N. S.; Knight, 21


Nano Letters

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

M. W.; Nordlander, P.; Halas, N. J.; Link, S. Proc. Natl. Acad. Sci. U. S. A. 2014, 111, 14348-53. 9.

Liu, F.; Zhang, X. Biosens. Bioelectron. 2015, 68, 719-25.

10. Cerjan, B.; Yang, X.; Nordlander, P.; Halas, N. J. Acs Photonics 2016, 3, 354-360. 11. Jha, S. K.; Ahmed, Z.; Agio, M.; Ekinci, Y.; Loffler, J. F. J. Am. Chem. Soc. 2012, 134, 1966-9. 12. Yang, Z. L.; Li, Q. H.; Ren, B.; Tian, Z. Q. Chem. Commun. 2011, 47, 3909-11. 13. Li, L.; Fang Lim, S.; Puretzky, A. A.; Riehn, R.; Hallen, H. D. Appl. Phys. Lett. 2012, 101, 113116. 14. Dörfer, T.; Schmitt, M.; Popp, J. J. Raman. Spectrosc. 2007, 38, 1379-1382. 15. Mattiucci, N.; D'Aguanno, G.; Everitt, H. O.; Foreman, J. V.; Callahan, J. M.; Buncick, M. C.; Bloemer, M. J. Opt. Express 2012, 20, 1868-77. 16. McClain, M. J.; Schlather, A. E.; Ringe, E.; King, N. S.; Liu, L.; Manjavacas, A.; Knight, M. W.; Kumar, I.; Whitmire, K. H.; Everitt, H. O.; Nordlander, P.; Halas, N. J. Nano Lett. 2015, 15, 2751-5. 17. Braun, G.; Lee, S. J.; Dante, M.; Nguyen, T. Q.; Moskovits, M.; Reich, N. J. Am. Chem. Soc. 2007, 129, 6378-9. 18. Ghosh, S. K.; Pal, T. Chem. Rev. 2007, 107, 4797-862. 19. Li, J. F.; Tian, X. D.; Li, S. B.; Anema, J. R.; Yang, Z. L.; Ding, Y.; Wu, Y. F.; Zeng, Y. M.; Chen, Q. Z.; Ren, B.; Wang, Z. L.; Tian, Z. Q. Nat. Protoc. 2013, 8, 52-65. 20. Tian, X. D.; Liu, B. J.; Li, J. F.; Yang, Z. L.; Ren, B.; Tian, Z. Q. J. Raman. Spectrosc. 2013, 44, 994-998. 22


Page 22 of 25

Page 23 of 25

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

Nano Letters

21. Eustis, S.; el-Sayed, M. A. Chem. Soc. Rev. 2006, 35, 209-17. 22. Link, S.; El-Sayed, M. A. J. Phys. Chem. B 1999, 103, 4212-4217. 23. Xu, H. X.; Bjerneld, E. J.; Kall, M.; Borjesson, L. Phys Rev Lett 1999, 83, 43574360. 24. Rao, S.; Raj, S.; Balint, S.; Fons, C. B.; Campoy, S.; Llagostera, M.; Petrov, D. Appl. Phys. Lett. 2010, 96, 213701. 25. Bauer, T.; Schmaltz, T.; Lenz, T.; Halik, M.; Meyer, B.; Clark, T. ACS Appl. Mater. Inter. 2013, 5, 6073-80. 26. Harpster, M. H.; Zhang, H.; Sankara-Warrier, A. K.; Ray, B. H.; Ward, T. R.; Kollmar, J. P.; Carron, K. T.; Mecham, J. O.; Corcoran, R. C.; Wilson, W. C.; Johnson, P. A. Biosens. Bioelectron. 2009, 25, 674-81. 27. Lo, H. C.; Hsiung, H. I.; Chattopadhyay, S.; Han, H. C.; Chen, C. F.; Leu, J. P.; Chen, K. H.; Chen, L. C. Biosens. Bioelectron. 2011, 26, 2413-8. 28. MacAskill, A.; Crawford, D.; Graham, D.; Faulds, K. Anal. Chem. 2009, 81, 813440. 29. Barhoumi, A.; Zhang, D.; Tam, F.; Halas, N. J. J. Am. Chem. Soc. 2008, 130, 55239. 30. Liu, J. Phys. Chem. Chem. Phys. 2012, 14, 10485-96. 31. Harper, M. M.; McKeating, K. S.; Faulds, K. Phys. Chem. Chem. Phys. 2013, 15, 5312-28. 32. Sandstrom, P.; Boncheva, M.; Akerman, B. Langmuir 2003, 19, 7537-7543. 33. Thacker, V. V.; Herrmann, L. O.; Sigle, D. O.; Zhang, T.; Liedl, T.; Baumberg, J. 23


Nano Letters

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

J.; Keyser, U. F. Nat. Commun. 2014, 5, 3448. 34. Lim, D. K.; Jeon, K. S.; Hwang, J. H.; Kim, H.; Kwon, S.; Suh, Y. D.; Nam, J. M. Nat. Nanotechnol. 2011, 6, 452-60. 35. Masetti, M.; Xie, H. N.; Krpetic, Z.; Recanatini, M.; Alvarez-Puebla, R. A.; Guerrini, L. J. Am. Chem. Soc. 2015, 137, 469-76. 36. Xu, L. J.; Lei, Z. C.; Li, J.; Zong, C.; Yang, C. J.; Ren, B. J. Am. Chem. Soc. 2015, 137, 5149-54. 37. Morla-Folch, J.; Alvarez-Puebla, R. A.; Guerrini, L. J. Phys. Chem. Lett. 2016, 7, 3037-41. 38. Sendroiu, I. E.; Warner, M. E.; Corn, R. M. Langmuir 2009, 25, 11282-4. 39. Kimura-Suda, H.; Petrovykh, D. Y.; Tarlov, M. J.; Whitman, L. J. J. Am. Chem. Soc. 2003, 125, 9014-5. 40. Talley, C. E.; Jackson, J. B.; Oubre, C.; Grady, N. K.; Hollars, C. W.; Lane, S. M.; Huser, T. R.; Nordlander, P.; Halas, N. J. Nano letters 2005, 5, 1569-74. 41. Prescott, B.; Steinmetz, W.; Thomas, G. J., Jr. Biopolymers 1984, 23, 235-56. 42. Hurst, S. J.; Lytton-Jean, A. K.; Mirkin, C. A. Anal. Chem. 2006, 78, 8313-8.

24


Page 24 of 25

1200 1400 1000 800 Table of contents art: 1600 1800

Raman Shift (cm-1)

400

600

C

T

0.6 0.4

Al2O3 0.2 Al

500 G nm

600

0.8

800 1000 1200 1400 1600 1800

F

A

1.0

Raman Shift (cm-1)

E

800

0.0 400

1000

G

H


nhancement (E/E0)

F


600

Wavele

Wavelength (nm)

E

ing Intensity (a.u.)


600

D Scattering Intensity (a.u.)

Normalized Int

C

1 2 1.0 3 4 5 6 0.8 400 7 8 9 100.6 11 12 13 140.4 15 16 17 180.2 19 20 21 22 0.0 23 24 400 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

Wavele

Normalized Int

Wavelength (nm) Nano Letters

Page 25 of 25

70 60 50 40 30 20

3nm oxide la no oxide laye

Aluminum Nanocrystals: A Sustainable Substrate for Quantitative

Recommend Documents