*4.2. Microfluidic Devices*

Microfluidics technology is considered to be a multidisciplinary technique interlinking several aspects of science including biochemistry, fluid dynamics, material science, physics, engineering, nanotechnology, chemistry, microtechnology, and biotechnology. It has been introduced as a novel point-of-care testing device in biosensing, providing large surface-to-volume ratio and making it a portable technology [88]. Fluorescence-based detection on microfluidic chips comprises bioluminescence, laser-induced fluorescence, immunofluorescence technique, and chemiluminescence, and the unique combination of these biochips with fluorescence detectors efficiently promotes sensitive detection of food-borne pathogens [89].

The fabrication of microfluidic-based devices comprises manufacturing technologies using silicon, glass, polymer (polydimethylsiloxane:PDMS) and ceramic that employs a photolithography method integrating mass production by micro electro-mechanical systems (MEMS). Generally, there are three versions of microfluidics: (a) continuous-flow, (b) droplet-based, and (c) digital, and their fabrication employs wet-etching, molding, sanding, laser, and milling techniques. Microfluidic fluorescence sensors need to maintain excitation spectra slightly different to the emissions in order to obtain complex spatial arrangement in glass-based microfluidics, while polymer microsystems often use PDMS that incorporates molding, layer structuration, or 3D printing which is an inexpensive

method. Lastly, ceramic was primarily utilized in microelectronics due to its significant features of designing 3D structures in low-temperature cofired ceramic (LTCC) [90]. PDMS were also applied as the surface for capturing bacteria. A recent study presented a 3D PDMS sponge fabrication utilizing salt crystals as the scarifying mold and the inner surface of the PDMS sponge was functionalized by apolipoprotein-H (ApoH), as universal ligand to capture both Gram-positive (*L. monocytogenes*) and Gram-negative (*Salmonella* spp.) bacteria, in combination with a microfluidic bioreactor. The capture proficiency was found greater than 70% for both targeted pathogens with an LoD of 103 and 104 CFU/mL for *Salmonella* spp. and *L. monocytogenes*, respectively [91].

Microfluidic devices have facilitated lab-on-chip (LOC)-integrating micropores, mixers enhancing capture efficiency, micropillars, and microfilters as additional modules combining these analytical procedures onto the same chip. The miniaturization, portability, instant detection, automation, and high-throughput are key advantages offered by microfluidics that are widely applicative in sensitive detection of food pathogens and toxins [92]. Recently, many smartphone microfluidic platforms integrating immunomagnetic nanoparticles or urease enzyme or paper-based/impedance electrochemical measurements have been introduced, offering high-end food sensing with multiplexed and rapid detection of pathogens [93]. A study has reported QD fluorescent-probe-based readout integrated with manganese nanoflowers as QD nanocarriers for signal amplification to detect *Salmonella typhimurium*. The bacterial load was determined with a low detection limit of 43 CFU/mL in food samples such as chicken, depending on the fluorescent intensity of released QDs [94]. Another sensor introduced immunomagnetic separation with fluorescent-labeling and video-processing smartphone for detection of *Salmonella*. The immunomagnetic particles separated and concentrated *Salmonella* followed by labeling with immunofluorescent microspheres to form fluorescent bacteria. This fluorescent *Salmonella* was injected into a biochip integrated with a smartphone fluorescent microscopic system. A low detection limit of 58 CFU/mL *Salmonella* was obtained by online counting of fluorescent spots using a smartphone App. (as presented in Figure 3) [95]. Shin et al., recently proposed a lateral-flow assay for multiplexed detection of *E. coli*, *Salmonella typhimurium*, *Staphylococcus aureus*, and *Bacillus cereus* in contaminated lettuce samples (Figure 3c) [96].

Paper-based devices are facile and flexible analytical biosensors as they offer a wide range of advantages over microfluidic chips in being cost-effective, with easy fabrication, great biocompatibility and high capillary action [97]. Lateral-flow assays (LFAs) and microfluidic paper-based analytical devices (μPADs) are the most common type of paperbased devices. LFAs or dipsticks are known for their facile handling, and rapid and naked-eye-visible readout without any additional equipment. Their cost-effectiveness and versatility in assay formats and user-friendliness offer their wide applicability in point-ofcare testing of food pathogens. LFAs simply comprise a sample pad where sample is added, a conjugate pad where the sample travels via capillary action activating the immobilized molecules, an absorbent pad, and a nitrocellulose membrane; all arranged on a plastic padding. The molecular components in the sample are separated as they travel across the membrane and produce a test line as positive-result output and a control line [98]. LFAs that are used for food-borne-pathogen detection incorporate monodispersed latex labels, gold colloid, and fluorescent/carbon tags for conjugate labeling. The colored particle, generally colloidal gold, binds to biomolecules (antigen/antibody/aptamer) immobilized onto test line that correlates with the amount of sample added [99,100]. Commercial LFA strips available in the market for bacterial sensing include *Listeria-*, *Salmonella-*, and *Escherichia coli* O157-Reveal test kits (Neogen®) Lansing, USA; *Listeria*, *Salmonella-* and *Escherichia coli*-VIP GOLD™ (BioControl Systems®) Bellevue, USA, and for *Listeria,* DuPont™ Lateral Flow System (DuPontQualicon) [4].

Paper-based μPADs generally utilize paper instead of chip microfluidics and are economical and efficient, removing the need for cleanroom facilities. Compared to siliconbased conventional biochips, paper-based chips are simple and highly porous, allowing physical absorption-generating devices that are easy to operate, modify and dispose of. These μPADs perform liquid transport, reactions, and even reagent storage on the hydrophilic porous paper that promotes transfer of liquids in the device. In this way, the designed flow-channels obviate the requirement for an external pump for running the assay [98]. The major component of paper μPADs is cellulose. Being biocompatible and flexible, it somehow absorbs the reagents dried onto it and arranged this in a cartridge integrated to a fluid delivery system viz. a droplet dispenser. Only the template has to be added to the kit and the start button is pressed, triggering the fluid delivery into μPAD. The fabrication of paper pads is categorized as patterning of hydrophobic barriers onto paper such as wax/laser/inkjet printing and shaping techniques, for instance, paper cutting/laser etching [98]. A recent work has developed an aptasensor integrating microfluidics paperbased multiplexed detection of *E. coli* O157:H7 and *S. typhimurium* (as presented in Figure 4). This novel sensor comprises single-input detection of more than single whole-cell food pathogen providing a quantitative signal readout as image analysis with a low detection limit of 103 and 10<sup>4</sup> CFU/mL, respectively [101].

**Figure 3.** (**A**) PDMS microfluidic-platform-based study reports QD fluorescent-probe-based detection of *Salmonella typhimurium*. (**i**) Schematic presentation of microfluidic channel with inlet and outlet and presentation of the experimental process and (**ii**) the bacterial load was determined with LoD of 43 CFU/mL in food samples using the laser b. Copyright (2020), with permission from MDPI [94]. (**B**) Immunomagnetic separation with fluorescent-labeled sample and (**i**) video-processed using smartphone for detection of *Salmonella* with an LoD of 58 CFU/mL and (**ii**,**iii**) the efficiency of salmonella detection compared to other bacteria and bacterial-capturing mechanism with nanoparticle, respectively. Copyright (2019), with permission from Elsevier [95]. (**C**) Shin et al. presented a (**i**) CD-disk-type microfluidic system for lateral-flow assay, (**ii**) the assembly of lateral-flow assay, and (**iii**) multiplexed detection of *E. coli*, *Salmonella Typhimurium*, *Staphylococcus aureus*, and *Bacillus cereus* in contaminated lettuce samples. Copyright (2018), with permission from the American Chemical Society [96].

**Figure 4.** (**A**) Paper-based microfluidics assembly for multiplexed assay. (**B**) Nanoparticle surface modification and building with ssDNA and blocking BSA protein for detection of *E. coli* O157:H7 and *S. typhimurium.* (**C**) Sensor detecting whole-cell food pathogen with an LoD of 10<sup>3</sup> and 10<sup>4</sup> CFU/mL, respectively. Copyright (2022), with permission from Elsevier [101].

Finally, we can say that the traditional approaches, such as PCR-based techniques and fluorescence detection on the surface are time-consuming and require specialized instrumentation. The microfluidic-based biosensor has shown its potential in research into rapid and sensitive detection with a very high limit of detection. Above, we discussed some examples of microfluidic biosensors for the detection of food contaminants. As add-ons to microfluidic systems and in integration with these methods, nanomaterials have become attractive in attaining selectivity. Nanomaterial provides a large surface area for binding of recognition molecules and enhances the signal for fluorescence. The use of nanomaterials in these biosensors makes them easy to use and feasible for point-of-care detection. In particular, the pros and cons of microfluidic-based biosensors include (i) high sensitivity in the analysis of small- and large-volume sample and (ii) high specificity and multiplexity to detect different analytes. In microfluidic systems, the challenges for food samples are is that some liquid samples are highly dense and cause blockage in the microfluidic device. Still, it is predicted that the future for microfluidic-based sensing of food samples is very promising.
