Fiber optics-based surface enhanced Raman Spectroscopy sensors for rapid multiplex detection of foodborne pathogens in raw poultry

Abstract

A new high-sensitivity, low-cost, Surface Enhanced Raman Spectroscopy (SERS) sensor allows for the rapid multiplex detection of foodborne pathogens in raw poultry. Self-assembled microspheres are used to pattern a hexagonal close-packed array of nanoantennas onto a side-polished multimode fiber core. Each microsphere focuses UV radiation to a photonic nanojet within a layer of photoresist on the fiber which allows the nanoantenna geometry to be controlled. Optimizing the geometry for the excitation layer generates electric field concentrations- referred to as a hotspot- within the analyte, thereby maximizing the Raman signal and improving the signal-to-noise ratio. The side polished configuration with a larger surface area has significantly better performance than the SERS sensor on the fiber tip. The use of additive manufacturing for the fiber polishing jigs as well as the sample testing compartment simplifies the sensor development and testing. Experimental results demonstrate a sensitivity range of 0.4-0.5 cells/ml achieved using raw chicken rinsates spiked with Salmonella typhimurium. Additionally, the sensor demonstrated its capability for multiplex and specific detection of Salmonella and E. coli O157:H7 with an optimal detection time of 10 min. The new sensor addresses a major global foodborne pathogen that poses significant public health concerns and can be readily adapted for the detection of other bacterial and viral pathogens such as E. coli O157:H7, Campylobacter, Listeria, and avian influenza and in other food products, e.g., dairy, beef, and produce, as well as clinical applications.

Fig. 1

SERS sensor (a) 3D view, (b) sideview, (c) setup. Raman instrument is used in backscattered configuration

Fig. 2

Schematics of the MPL process

Fig. 3

SEMs of (a) a 3D printed structure along with the nanodisk arrays. b, c magnified views of the nanodisks. d nanodisk arrays patterned across the sensing surface area. e, f Optical images of the fabricated and tested SERS sensor. The chamber, which was 3D printed, is affixed to the 3D printed microstructure with an adhesive

Fig. 4

Relative intensity versus concentration for SERS sensor with surface area and disk diameter of (a) 105 µm × 4 mm and 780 nm, (b) 105 µm × 7 mm and 780 nm, respectively, and versus sensing area for fixed diameter of 780 nm and concentration of (c) 10 cells/ml, (d) 1395 cells/ml

Fig. 5

Relative intensity as a function of disk diameters for a fixed concentration of 10 cells/ml for (a) 105 µm × 4 mm (b) 105 µm × 7 mm, and 1395 cells/ml for (c) 105 µm × 4 mm and (d) 105 µm × 7 mm

Fig. 6

Relative intensity for specific peaks (634 cm^-1) versus (a, b) Salmonella concentrations, (c, d) disk diameters for sensing areas of 105 μm × 7 mm and 105 μm × 4 mm, respectively

Fig. 7

Raman spectra were obtained for Salmonella in raw poultry rinsate samples, where the relative Intensity versus volume for (a) 1–3 cells, the fiber has a disk diameter and sensing area of 615 nm and 105 µm × 7 mm. b 10 cells. The fiber has a disk diameter and sensing area of 770 nm and 105 µm × 7 mm

Fig. 8

Raman spectra were obtained for mixture of E. coli O157:H7 and Salmonella in raw poultry rinsate samples with a concentration of (a) 15 cells/ml and 205 cells/ml, and (b) 1800 cells/ml and 1395 cells/ml, respectively. The figure also plots the spectra for E. coli O157:H7 and Salmonella, prior of mixing with a concentration of 10 cells/ml and 205 cells/ml, and 1320 cells/ml and 1395 cells/ml, respectively. The tested device has an area and disk diameter of 105 µm × 7 mm and 615 nm

Fig. 9

An optimum detection time (a) 3D spectrum of the relative Intensity as a function of detection time and wave number, (b) relative intensity of specific peaks versus detection time. in this experiment, the fiber core diameter is 770 nm, and the sensing area is 105 µm × 7mm