Plasmon-enhanced stimulated Raman scattering microscopy with single-molecule detection sensitivity

Cheng Zong,Yimin Huang,Chi Zhang,Chen Yang,Lawrence D Ziegler,Ranjith Premasiri,Bin Ren,Haonan Lin,Ji-Xin Cheng

doi:10.1038/s41467-019-13230-1

Abstract

Stimulated Raman scattering (SRS) microscopy allows for high-speed label-free chemical imaging of biomedical systems. The imaging sensitivity of SRS microscopy is limited to ~10 mM for endogenous biomolecules. Electronic pre-resonant SRS allows detection of sub-micromolar chromophores. However, label-free SRS detection of single biomolecules having extremely small Raman cross-sections (~10−30 cm2 sr−1) remains unreachable. Here, we demonstrate plasmon-enhanced stimulated Raman scattering (PESRS) microscopy with single-molecule detection sensitivity. Incorporating pico-Joule laser excitation, background subtraction, and a denoising algorithm, we obtain robust single-pixel SRS spectra exhibiting single-molecule events, verified by using two isotopologues of adenine and further confirmed by digital blinking and bleaching in the temporal domain. To demonstrate the capability of PESRS for biological applications, we utilize PESRS to map adenine released from bacteria due to starvation stress. PESRS microscopy holds the promise for ultrasensitive detection and rapid mapping of molecular events in chemical and biomedical systems.

Highlights

This sharp feature is close to the prominent adenine ring-breathing mode frequency observed in the normal Stimulated Raman scattering (SRS) spectrum of adenine powder (Fig. 1c, blue) and identical to the corresponding 733 cm−1 peak observed in the SERS results on Au substrates (Supplementary Fig. 2)[48]
These results collectively indicate that the observed vibrational plasmon-enhanced stimulated Raman scattering (PESRS) signal component originates from the surface adsorbed adenine
The standard SRS setup could not generate any Raman signal from a pure adenine powder, while PESRS could detect a thin layer of adenine adsorbed on Au nanostructures

Summary

Introduction

The reproducible spectra recorded at the same location demonstrate that the laser power in our experiment minimally damaged the substrate or induced molecular photodegradation during SRS imaging (~1.0 min per hyperspectral stack). These results collectively confirm the SRS origin of the vibrationally resonant component of the observed spectrum and the plasmonic enhancement of this signal. An adaptive iteratively reweighted penalized least squares (airPLS) algorithm (https://github.com/ zmzhang/airPLS), developed by Zhang et al.[59], is employed to subtract the baseline from the raw PESRS spectrum. We use a BM4D V3.2 (http:// www.cs.tut.fi/~foi/GCF-BM3D/index.html#ref_software) denoising algorithm, developed by Maggioni and Foi[50,51], to process the raw PESRS hyperspectral data cube. The BM4D algorithm relies on the so-called grouping and collaborative

Methods

Results

Conclusion