Spatiotemporal Raman Probing of Molecular Transport in sub-2-nm Plasmonic Quasi-2D Nanochannels

TL;DR

The paper addresses the challenge of observing molecular transport in ultrathin nanoconfinement by showing that ligand-capped nanoparticle-on-mirror (NPoM) gaps inherently form open quasi-2D nanochannels. It combines wavelength-multiplexed SERS (WM-SERS) with dual-wavelength addressing to map centripetal diffusion of molecules from the gap edges to the center, supported by a three-zone kinetic/EM-field model and extended to micron-scale microplate-on-foil (MPoF) devices for real-space visualization. Key findings include transport over lengths exceeding $5$ μm within ~2 nm gaps, an average transition width Δr ≈ 50 nm, and detection down to $10^{-11}$ M in real time, validating NPoM as an open transport-and-probe platform. This work establishes a scalable, on-chip nanophotofluidics framework that enables super-resolved, in-situ molecular sensing and provides a foundation for studying transport phenomena under extreme confinement.

Abstract

Capturing molecular dynamics in nanoconfined channels with high spatiotemporal resolution is a key challenge in nanoscience, crucial for advancing catalysis, energy conversion, and molecular sensing. Bottom-up ultrathin plasmonic nanogaps, such as nanoparticle-on-mirror (NPoM) structures, are ideal for ultrasensitive probing due to their extreme light confinement, but their perceived sealed geometry has cast doubt on the existence of accessible transport pathways. Here, counterintuitively, we demonstrate that ubiquitous ligand-capped NPoM-type nanogaps can form a natural quasi-two-dimensional nanochannel, supporting molecular transport over unprecedented length scales ($\gtrsim5$ $μ$m) with an extreme aspect ratio ($>10^3$). Using wavelength-multiplexed Raman spectroscopy, we resolve the underlying centripetal infiltration pathway with a spatial resolving power of $\sim$20 nm. This redefines the NPoM architecture as a sensitive, \textit{in-situ}, all-in-one "transport-and-probe" platform, enabling real-time, reusable monitoring of analyte with $\sim$10$^{-11}$ M. This work establishes a versatile new platform for advancing super-resolved \textit{in-situ} molecular sensing, nanoscale physicochemical studies, and on-chip nanophotofluidics.

Figures (5)

• Figure 1: Sequential molecule-exchange SERS sensing in NPoM plasmonic nanochannels. (a) Top panel: 3D schematics of metal nanoparticles (nanosphere, nanocube, etc.) placed over an metal mirror to form well-defined plasmonic nanogaps (top panel); Bottom panel: Simulated field distribution of a nanocube-on-mirror (NCoM) with 1 nm gap, exhibiting pronounced localized field enhancement ($|\mathbf{E}| / |\mathbf{E}_0| =437$). (b) Schematics of a molecule exchanges in a 2D molecular nanochannel. (c) Schematic illustration of super-resolved wavelength-multiplexed (WM) SERS spectroscopy for sensing of sequential molecular exchange dynamics in 1-2 nm thick 2D plasmonic nanochannels. (d) Normalized SERS spectra (785 nm excitation) of a single PVP-capped NCoM before and after the sequential immersion in the solution of MB $\rightarrow$ BSe $\rightarrow$ BPT (from top to bottom). The characteristic peaks of each molecule are highlighted in red, yellow, green, and blue area, respectively. (e) Molecular structures of PVP, MB, BSe and BPT (top panel), and their corresponding characteristic SERS peak intensities (colored open circles, extracted from d) as a function of immersion duration, where the dash lines are the fitting results by Langmuir kinetic equation (See details in Supplementary Section S8).
• Figure 2: Wavelength-multiplexed (WM) SERS sensing of molecular transport in a singe NCoM nanogap. (a) Simulated scattering spectra of a 70-nm-diameter Ag nanocube on a gold mirror with gap distance g varied from 3 to 1.8 nm, showing three plasmon modes labeled as $J_{\text{+}}$, $J_{\text{-}}$ and $S_{\mathrm{11}}$, respectively. (b) 3D colormap of electric field distributions of LAP $J_{\text{+}}$ and TCP $S_{\mathrm{11}}$ modes, respectively. The central panel shows the cross-sectional field intensity profiles of the LAP and TCP modes, illustrating their complementary spatial distributions and defining 20 nm spatial resolving power of the WM-SERS technique. (c) Measured dark-field scattering and dual-wavelength Raman spectra (660 and 785 nm excitation, orange and blue dashed lines) of a single PVP-capped Ag NCoM after immersion of BPT ethanol solution with duration from 5 to 7200 s (from top to bottom panels), with the SEM image shown in the inset of d. (d) BPT Raman peak intensity at 1280 cm$^{\mathrm{-1}}$ as a function of immersion time under 660 nm and 785 nm excitations with Raman emission at 720.8 nm and 872.6 nm, respectively (also labelled in c with orange and blue areas). (e) Statistical SERS spectra from individual BPT-spaced NCoMs using conventional and our molecule-exchange methods, respectively. The yellow (purple) line, area and inset box represents the SERS mean value, error bar and the schematics of the conventional (molecule-exchange) methods, respectively.
• Figure 3: Molecular transport mechanism in plasmonic 2D nanochannels resolved by WM-SERS spectroscopy. (a) Representative dark-field scattering spectra of NCoMs with BPT and/or 2-NAT spacer, along with the statistical distributions of resonance wavelengths for the LAP ($J_{\text{+}}$) and TCP ($S_{\mathrm{11}}$) modes from 84 NCoMs. The grey area covers the spectral windows for 660 and 785 nm SERS shown in b. (b) Representative WM-SERS spectra measured on individual NCoMs, representing four typical replacement stages from BPT to 2-NAT molecules, as labeled in d. The insets illustrate the molecular spatial distribution (top plane, green: BPT, yellow: 2-NAT) within the nanogap mapped onto the near-field distributions of LAP and TCP modes (bottom planes). Red and blue shaded areas highlight the characteristic Raman peaks of BPT (1280 $\text{cm}^{\mathrm{-1}}$) and 2-NAT (1380 $\text{cm}^{\mathrm{-1}}$). (c) 3D schematics of the model for molecular exchange within a molecular-spaced NCoM nanochannel. The 3D area on top of the molecular layer represents the fractional occupancy of new molecules $\omega_{\text{new}}$ (ranging from 0 to 1). The projections of $\omega_{\text{new}}$ onto the molecular layer are colored green, green-yellow striped, and yellow, respectively, representing (i) original, (ii) mixed and (iii) replaced zones. The corresponding WM-SERS from these three zones are mapped from the near-field distributions of the LAP and TCP modes (illustrated beneath the metal film). (d) Correlation plot of $R_{\text{NAT}}^{\mathrm{785}}$ versus $R_{\text{NAT}}^{\mathrm{660}}$ derived from 820 individual NCoMs, where the area of the grey circles is proportional to the nanocube numbers. Colored curves show the theoretically predicted relationship between $R_{\text{new}}^{\text{785}}$ and $R_{\text{new}}^{\text{660}}$ for varied $\Delta r$ of the mixed zone.
• Figure 4: Micron-scale molecular infiltration and exchange dynamics. (a) Cross-sectional schematic of the molecular transport experiment in a microplate-on-foil (MPoF) nanochannel with micro length. The sample consists of a PVP-capped microplate placed on a 10-nm-thick Au foil, forming a $\sim$2-nm-thick gap. The MPoF was then immersed in BPT solution for varying durations to probe molecular exchange process via SERS mapping. (b) Bright field image of a representative MPoF structure. (c) SERS spectra of the MPoF collected at the position before (P1, black dot in d (iii)) and after (P2, blue dot in d (iv)) BPT replacement, respectively, compared with that at the BPT-adsorbed Au foil region (P3, yellow dot in d (iii)). For clarity, the blue and yellow spectra are vertically offset. (d) Raman intensity maps at 1280 cm$^{-1}$ (785 nm excitation) for the MPoF in b after immersion in BPT solution for (i) 0.5 min, $25 ^\circ$C, (ii) 12 min, $25 ^\circ$C, (iii) 40 min, $50^\circ$C, and (iv) 120 min, $50 ^\circ$C, respectively. The dashed black circle and gray boxes indicate the original PVP-spacer region and boundary of the microplate, respectively. $L$ denotes the length of the microplate region where molecular infiltration has occurred.
• Figure 5: In-situ ultrasensitive and real-time molecular post-sensing via a single NPoM plasmonic nanogap. (a) Schematic of the selective nanogap engineering process for molecular post-sensing: (i) A BPT-spaced 150-nm-diameter Au NPoM is formed via the molecular exchange method; (ii) Partial etching of edge BPT molecules using NaBH$_{4}$ solution preserves central BPT to maintain nanochannel integrity; (iii) Subsequent immersion in analyte (4-MBN) solution enables analyte adsorption into the etched nanogap regions. (b) Near-field distribution (at 880 nm) of the 150 nm NPoM, providing field enhancement over 449 times. (c) Representative 785-nm SERS spectra showing the characteristic 4-MBN peak at 2227 cm$^{-1}$ across concentrations varied from 10$^{-6}$ to 10$^{-11}$ M. The spectra are vertically offset for clarity. (d) Raman intensity map (2227 cm$^{-1}$) of 225 individual NPoM structures at 10$^{-8}$ M 4-MBN, arranged in a 15$\times$15 grid. (e) Statistical SERS intensity at 2227 cm$^{-1}$ of the NPoM as 4-MBN concentrations varied from 10$^{-6}$ to 10$^{-11}$ M. Green line denotes a linear fit. (f) Digital SERS map derived from (d) by applying an intensity threshold (see Supplymentary Section 6). (g) Ratio of positive voxels (RPV) based on digital SERS analysis (f) versus 4-MBN concentration. The green line denotes a linear fit. (h) Microfluidic experimental setup for real-time SERS: A CTAC-capped Au nanoparticle (150 nm diameter) on Au foil (10 nm thick) forms a nanoparticle-on-foil (NPoF) nanogap. 10$^{-6}$ M BPT solution is flowed through the microchannel while acquiring 785-nm SERS spectra from the coverslip side. (i) Time-dependent SERS map during microfluidic perfusion over 600 s. (j) Time-dependent SERS intensity (BPT peak at 1580 cm$^{-1}$) during microfluidic perfusion, which is fitted by an exponential function based on the Langmuir kinetic model.