Tunable platforms by coupling gold nanorectangles and pNIPAM for surface-enhanced Raman scattering

$N{{\rm \text -}}$isopropylacrylamide monomer (NIPAM) 99% and $N,N,{{N}^{\prime}},{{N}^{\prime\prime}},{{N}^{\prime\prime}}{{\rm \text -}}$pentamethyldiethyltriamine (PMDETA) were purchased from Acros Organics. Triethylamine (TEA) 99% was purchased from Merck. Tert-butyl nitrite 90%, 2-(4-aminophenyl)ethanol 98%, CuBr 98%, and nile blue A dye were purchased from Sigma Aldrich. 2-bromoisobutyryl bromide 97% was bought from Alfa Aesar. The 4-(2-hydroxyethyl)-benzene diazonium tetrafluoroborate salt, termed HEBDT $\left({\rm +}{{{\rm N}}_{{\rm 2}}}{{\rm \text {--}}}{{{\rm C}}_{{\rm 6}}}{{{\rm H}}_{{\rm 4}}}{{\rm \text {--}}{\rm C}}{{{\rm H}}_{{\rm 2}}}{{\rm \text {--}}{\rm C}}{{{\rm H}}_{{\rm 2}}}{{\rm \text {--}}{\rm OH}} \right)$ was chemically synthesized following the approach described in our previous report [16].

The lithographic gold nanorectangles (AuNRs) array produced by electron-beam lithography (EBL) was grafted pNIPAM by following the multi-steps strategy relying on three major steps (figure 1): (i) the grafting of the aryl group derived from 4-hydroxyethylbenzene diazonium tetra-fluorborate salt (HEBDT). In this step, the clean AuNRs array was incubated in the HEBDT salt solution 3 mM at room temperature for 5 h; (ii) Then, the obtained AuNRs array with terminal hydroxyl groups was immersed in the toluene solution containing 0.1 M 2-bromoisobutyryl bromide and 0.12 M TEA for 5 min. The terminal groups of bromo ester were thus obtained onto the gold surface (Au-Br); (iii) pNIPAM brushes were grafted from Au-Br surface by atomic transfer radical polymerization (ATRP). A catalyst solution was produced by injection a 3 ml solution of PMDETA (0.3 M) in methanol to 20 mg of CuBr. Then, the ATRP was carried out by injecting the catalyst solution to 9 ml of deionized water containing 2.0 g NIPAM monomer under an argon stream. Then, the obtained solution of ATRP was transferred to a Schlenk flask containing the substrate of functionalized gold nanorectangles (Au-Br). After stirring for 15 min under the argon stream at room temperature, the substrate was taken out of the flask and carefully washed with water and ethanol. Finally, the substrate was dried under the argon flush.

Zoom In Zoom Out Reset image size

The extinction spectra were recorded by the LOT ORIEL spectrometer model MS 260i, combined with an optical microscope (OLYMPUS BX 51). Tapping-mode atomic force microscopy (AFM) of the samples before and after the modification was performed using SPM Nanoscope III (Veeco, Bruker) set-up in water. The SERS spectra were measured on a Jobin-Yvon LABRAM HR 800 Raman micro spectrometer with the excitation laser at 632.8 nm.

Recent efforts have been focused on identifying appropriate nanoplasmonic materials which have sensitive spectra plasmon upon a variation of the local environment. Wherein, AuNRs are interesting materials due to their optical properties with the plasmon wavelength that can be finely tuned by modification of AuNRs aspect ratio. Therefore, we focused our attention on AuNRs which have unique optical properties with two localized surface plasmon (LSP) bands. The longitudinal and transversal LSP band corresponds to the oscillation of the electrons parallel to the major and the short axis, respectively. AuNRs remain popular for plasmonic applications due to the adjustable ability of the LSP resonance wavelength by varying the aspect ratio, defined as the long axis (length) divided by the short axis (width) of a rod-shaped particle, their strong scattering efficiency and a small LSP damping.

In this work we consider an AuNRs array with a small aspect ratio $r=1.2$ (200 nm long, 167 nm wide and 30 nm high). The AFM images of these AuNRs are shown in figure 2(a). The extinction band around 596 nm corresponds to the transversal LSP mode (black dashed line in figure 2(c)). The other extinction band occurs at higher wavelengths around 616 nm corresponding to the longitudinal LSP mode (black dashed line in figure 2(d)). For such anisotropic particles, a strong red-shift is expected when the refractive index changes. The surrounding medium of AuNRs is changed by grafting the polymer film. The AFM images of AuNRs array before and after polymer grafting show that the pNIPAM brushes are grafted only on the AuNRs and not on the substrate (figures 2(a) and (b)). The dry polymer thickness in the air was ${{L}_{{\rm dry}}}\sim 5\,{\rm nm}$. The presence of polymer layer on the AuNRs leads to the location of peak absorption red-shift in wavelength of LSP due to an increase of refractive index: $\Delta {{\lambda}_{{\rm LSP}}}=31\,{\rm nm}$ for the transversal LSP mode, and $\Delta {{\lambda}_{{\rm LSP}}}=41\,{\rm nm}$ for the longitudinal LSP mode (figures 2(c) and (d)).

Zoom In Zoom Out Reset image size

The dynamic swelling ratio $(\alpha)$ of pNIPAM was determined as $\alpha ={{{L}_{{\rm swollen}}}}/{{{L}_{{\rm dry}}}}\;$ (where ${{L}_{{\rm swollen}}}$ and ${{L}_{{\rm dry}}}$ are the thickness of pNIPAM brush at swelling and dry states, respectively). The swelling ratio was deduced from our report to $\alpha \sim 2$ [17]. Thus, the pNIPAM thickness in the swollen state can be deduced to ${{L}_{{\rm swollen}}}\sim 10\pm 2\,{\rm nm}$. When the AuNRs were coated by pNIPAM, the phase transition of the polymer can change the refractive index of the surrounding medium of the AuNRs and lead to the optical properties of plasmonic nanostructures modified. The pNIPAM thickness can decrease from 10 to 5 nm and the refractive index can increase from 1.37 to 1.43 by raising the temperature [18]. In order to characterize the thermosensitive properties of this system, we considered its optical response as a function of temperature. The longitudinal mode of AuNRs is highly sensitive in wavelength to the local environment. Thus, in optical studies, the longitudinal LSP mode of AuNRs has received main attention. Extinction spectra on the longitudinal LSP mode of the AuNR@pNIPAM sample recorded in water at 20 and $40\,{}^\circ {\rm C}$ are shown in figure 3(a). From these spectra, it is clear that red-shift of the LSPR band is observed $\sim 4\,{\rm nm}$ upon an increase of the temperature, due to a change in the medium index in vicinity of AuNRs by the conformational change of the pNIPAM. The LSP red-shift as a function of temperature is plotted in figure 3(b). The sigmoidal curve fitting (red solid line) based on the experimental data (black dot) showed a sharp transition between two plateau areas below and above the LCST of pNIPAM.

Zoom In Zoom Out Reset image size

The SERS activity based on the synthesized hybrid substrate was determined by Nile blue A (NBA) dye as a probed molecule. Firstly, the substrate was immersed in NBA 0.73 ppb solution for 5 min. Next, it was washed with water and ethanol, and then dried. The temperature dependence of the SERS activity of AuNR@pNIPAM substrate was investigated by measurement their Raman spectra below and above the LCST of pNIPAM. SERS spectra of NBA recorded in water at 20 and $40\,{}^\circ {\rm C}$ were shown in figure 4. The enhanced Raman signals of NBA probe were characterized by bands from C–C–C and C–N–C deformations at $498\,{\rm c}{{{\rm m}}^{-1}}$ and $591\,{\rm c}{{{\rm m}}^{-1}}$; C–C–C and N–C–C in-plane bending at $665\,{\rm c}{{{\rm m}}^{-1}}$; C–H bending at 1256 and $1186{\rm c}{{{\rm m}}^{-1}}$; and ring stretching vibrations at 1643, 1542, 1493, 1435, 1419 and $1357\,{\rm c}{{{\rm m}}^{-1}}$. The bands are agreed upon by the literature [19].

Zoom In Zoom Out Reset image size

Recently, the sensitivity studies of several types of SERS substrates based on silver/gold nanostructres on 2D substrates demonstrated that the detection limit can be down to 0.01 ppm for parquat [8, 9] and 1 ppb for methamidophos [10]. In our report, hybrid substrate based on lithographic gold nanorectangles array and pNIPAM presents a high sensitivity which allows to detecting the NBA molecules at ultra-low concentration (0.73 ppb). Attentively, the results demonstrate an increase of the SERS response of NBA molecules at the temperature above the LCST of pNIPAM (compared to the SERS spectra registered below the LCST). The band intensities increase up to three-fold. This can be explained by an increase of the local field intensity when the distance between the AuNRs surface and the analyte is decreasing. Hence, the distance between the AuNRs surface and the molecular probes have influence on the SERS signal. It is due to the pNIPAM conformation change upon the temperature variation. Interestingly, figure 4 shows that SERS signal decreases back to the first spectra recorded at $20\,{}^\circ {\rm C}$ when cooling down the substrate to below the LCST ($20\,{}^\circ {\rm C}$). Especially, the modifications of the Raman signal intensity of NBA obtained for AuNR@pNIPAM substrate are fully reversible upon the variation of external temperature.

The temperature dependence of the intensity variation of Raman band at $1181\,{\rm c}{{{\rm m}}^{-1}}$ was plotted in figure 5. A steep increase was observed when the temperature increases from 28 to $34\,{}^\circ {\rm C}$ corresponding to the phase transition zone of pNIPAM. It is remarkable that this thermo-driven variation was fully reversible during repeating cycles of the temperature decrease from 40 and $20\,{}^\circ {\rm C}$. Interestingly, the intersection between two curves with increasing and decreasing temperatures is located around $32\,{}^\circ {\rm C}$, the LCST of pNIPAM. The full reversibility of the system indicates that our thermoresponsive system allows us to dynamically tune the distance between the AuNRs surface and analyte.

Zoom In Zoom Out Reset image size