Requirements for fast multianalyte detection and characterisation via electrochemical-assisted SERS in a reusable and easily manufactured flow cell

Abstract

Surface-enhanced Raman spectroscopy (SERS) is a highly sensitive analytical technique that captures vibrational spectra of analytes adsorbed to rough coin metal surfaces with remarkable signal intensities. However, its wider application is limited by challenges in substrate range, quantification, and the disposable nature of SERS substrates partly due to irreversible analyte adsorption-commonly referred to as the 'memory effect'. Overcoming these limitations and achieving real-time analysis in flow-through systems remains a key challenge for the advancement of SERS. This study presents a SERS flow cell incorporating an Ag-based SERS substrate and a Pt counter-electrode, enabling the investigation of how electrochemical methods can address existing challenges. Our approach demonstrates that signal intensities can be both enhanced and spectroelectrochemically modified. Additionally, the combination of constant solvent flow and electrochemical potentials enhances the longevity of the SERS substrate, facilitating multianalyte measurements while mitigating the memory effect. Key parameters have been systematically studied, including SERS substrate materials (silver and copper), solvents, buffers, supporting electrolytes, and electrochemical protocols. We achieved consistent and reproducible electrochemical tuning of SERS signals by using halogen-free electrolytes in polar solvents commonly used in techniques like HPLC. The versatility of the system was validated through the analysis of several model compounds and the sequential detection of multiple analytes. We also successfully applied the system to detect and characterise contaminants and pharmaceuticals, highlighting its potential for a wide range of analytical applications.

Keywords: Electrochemical SERS; Flow cell; HPLC; SERS substrates; Surface-enhanced Raman spectroscopy.

Fig. 1

Experimental setup

Fig. 2

Schematic sketch of the used chip design

Fig. 3

Intensity of the Raman signal of 10 µM CV in 66 mM PBS at pH 7 Period duration 70 s, Potential 0 V to − 3.5 V. Analyte was flushed in constant flow: 200 µl/min. (473 nm; 2.5 mW; 600 lines/mm; 40 × objective; 1 s integration time) A Intensity at 1620 cm⁻¹ in dependency of the applied potential. B 3D plot of the Raman spectrum of CV over time

Fig. 4

Power source: frequency generator with triangular potential between 0.0 and − 3.5 V and a period duration of 70 s. Measurement parameters: 473 nm; 2.5 mW; 600 lines/mm; objective: 40-fold; integration time: 1 s; flow: 200 µL/min; indicational band: 1620 cm⁻¹. A Influence of different supporting electrolytes on the progression of the SERS-Intensity of 10 µM CV in aqueous solution. B Influence of the concentration of Bu₄NOAc on the SERS-Intensity of 10 µM CV in aqueous solution. With * marked measurement was recorded with 0.25 mW laser power and signal intensity corrected by multiplying tenfold

Fig. 5

Intensity curve of 10 µM CV dissolved in different solvent mixtures containing 50 mM Bu₄NOAc. Power source: frequency generator with triangular potential between 0.0 and − 3.5 V and a period duration of 70 s. Measurement parameters: 532 nm; 4.1 mW; 600 lines/mm; objective: 40-fold; integration time: 1 s; flow: 200 µL/min; indicational band: as indicated

Fig. 6

Progression of the SERS signal of 10 µM CV on different substrate materials, while a triangular potential (0.0 to − 3.5 V; period duration: 70 s) is applied. Experimental parameters: Excitation wavelength: 473 nm; laser power: 2.5 mW; integration time: 1000 ms; flow: 200 µL/min; indicational band: 1620 cm⁻¹

Fig. 7

Intensity curves of the SERS-Spectra of 10 µM CV dissolved in 50/50 MeOH/H₂O influenced by different functions of applied potential. Current and potential are indicated. A potential between 0.0 and −3.5 V was used with a period duration of 70 s. Measurement parameters: (532 nm; 4.1 mW; 600lines/mm; objective: 40-fold; integration time: 100 ms; flow: 200 µL/min; indicational band:1607 cm⁻¹). Top: Triangular Potential. Bottom: Sawtooth-potential

Fig. 8

Intensity curve of CV in a microfluidic SERS sensor influenced by alternating frequency of a sawtooth voltage potential. A frequency generator was used as power source with a sawtooth potential varying between 0.0 and − 3.5 V. The period duration was gradually lowered. 10 µM CV dissolved in MeOH/H₂O 50/50 containing 50 mM Bu₄NOAc were flushed in a constant flow through the chip. (532 nm; 8.1 mW; 600 lines/mm; objective: 40-fold; integration time: 100 ms; flow: 200 µL/min; indicational band:1607 cm⁻¹)

Fig. 9

Detection with the following alternating analytes: 100 µM adenine, 10 µM malachite green, 100 µM methylene blue, and 500 nM CV dissolved in dissolved in MeOH/H₂O 50/50 containing 50 mM Bu₄NOAc. A potentiostat in 2-electrode configuration as powersource giving a sawtooth potential varying between 0.0 and − 3.5 V and a period duration of 10 s. Measurement parameters: 532 nm; 8.1 mW; 600 lines/mm; objective: 40-fold; integration time: 2 s; flow: 200 µL/min; indicational band: as indicated. In each measurment, 500 µL of sample were sampled via Hexaport ventile. A 3D diagram containing the whole Raman intensity curve of the measurement over time. B Experimental setup for sampling the analytes containing the Hexaport. WE, working electrode used as SERS substrate. CE, counter electrode. C Intensity curve at wavenumber as indicated together with the applied potential. Analytes are indicated for every group of peaks

Fig. 10

EC-SERS signal progressions of cytosine, L-Dopa, melamine, and guaifenesin. 1 mM concentration and dissolved in 50 mM aqueous Bu₄NOAc were used. Spectra were recorded under constant flow of 200 µL/min using the 473 nm setup. Power source: frequency generator between 0.0 and − 3.5 V, period duration: 70 s. Parameters: (analyte, laser power, integration time) cytosine, 7.9 mW, 1 s; L-Dopa, 7.9 mW, 1; melamine: 7.9 mW, 3 s; guaifenesin, 7.9 mW, 3 s. 600 lines/mm; objective: 40-fold