Plasma Functionalized Carbon Interfaces for Biosensor Application: Toward the Real-Time Detection of Escherichia coli O157:H7.

Nonthermal plasma, a nondestructive, fast, and highly reproducible surface functionalization technique, was used to introduce desired functional groups onto the surface of carbon powder. The primary benefit is that it is highly scalable, with a high throughput, making it easily adaptable to bulk production. The plasma functionalized carbon powder was later used to create highly specific and low-cost electrochemical biosensors. The functional groups on the carbon surface were confirmed using NH3-temperature-programmed desorption (TPD) and X-ray photoelectron spectroscopy (XPS) analysis. In addition, for biosensing applications, a novel, cost-effective, robust, and scalable electrochemical sensor platform comprising in-house-fabricated carbon paste electrodes and a miniaturized E-cell was developed. Biotin-Streptavidin was chosen as a model ligand-analyte combination to demonstrate its applicability toward biosensor application, and then, the specific identification of the target Escherchia coli O157:H7 was accomplished using an anti-E. coli O157:H7 antibody-modified electrode. The proposed biosensing platform detected E. coli O157:H7 in a broad linear range of (1 × 10-1-1 × 106) CFU/mL, with a limit of detection (LOD) of 0.1 CFU/mL. In addition, the developed plasma functionalized carbon paste electrodes demonstrated high specificity for the target E. coli O157:H7 spiked in pond water, making them ideal for real-time bacterial detection.

Introduction

Due to their cost-effectiveness, easy manipulation, and fast response, electrochemical biosensors have emerged with innovative designs to diagnose different diseases and detect various targeted biological agents.[1−4] In addition, they also offer the possibility of miniaturization for their use as point-of-care analysis, and the development of these sensors has been vastly improved using carbon materials.[4−6] The usage of different forms of carbon, such as graphite,[7] activated carbon,[8] graphene, carbon nanofibers/nanotubes,[9] fullerenes,[10] porous carbon,[11] and their composites,[12] is commonplace in electrochemical biosensors owing to their inherent advantages such as high chemical stability, excellent conductivity, superior mechanical strength, high surface-to-volume ratio, and biocompatibility. Furthermore, carbon materials can perform the dual role of a biorecognition element and a transduction element simultaneously.[13−15]

Bioreceptors are critical components that specifically bind with the target analyte of interest and govern the selectivity and sensitivity of the biosensor. The immobilization of such bioreceptors on top of the electrode surface is one of the critical steps in developing an ultrasensitive electrochemical sensor.[4−6,16] This is typically achieved by various surface functionalization techniques, which essentially bind the bioreceptor using noncovalent physical adsorption, electrostatic interactions, or covalent chemisorption. Furthermore, any surface functionalization method influences the electronical properties, signal transduction, and material interface. It is imperative to choose the right surface functionalization protocol to extract an optimum response from the sensor.[1,13−17] The throughput of this protocol governs the scalability of the sensor, an aspect that is key for commercialization.

Noncovalent interactions, in general, are simple and include the physical adsorption of the bioreceptors to the electrode surface but pose the limitation of nonspecific binding and long-term stability due to their weak interaction. In contrast, covalent interaction overcomes these limitations and provides stable attachment through covalent and specific cross-linking of the bioreceptors with the surface of the electrode. Among different covalent interactions, COOH terminated surfaces are frequently used in biosensor applications to bind bioreceptors through a covalent/ionic interaction. Self-assembled monolayers (SAMs) with COOH termination are broadly used for such purposes, but they are limited to specific substrates such as thiol for Au/Ag substrates and organosilane for silicon oxide substrates. Furthermore, such functionalization methods require hours to form SAMs and are likely to oxidize under certain conditions, making them unstable for prolonged use.[16,18]

The other modification methods involve the conventional oxidation of carbon materials through chemical treatment methods such as acid treatment. They involve a tedious process of multiple washing and filtration steps, followed by hot-air oven-drying. In contrast, the plasma-assisted formation of carboxylic groups is more stable and can also be used for different substrates. Plasma functionalization, which is a characteristic of generating active species under ambient conditions, possesses several advantages such as chemical interaction, stability, density, and coverage of the functionalized groups. In addition, this process is highly scalable, offering high throughput, and is readily amenable to bulk production for industrial applications.[13,16,17,19−21]

In general, plasma, the fourth state of matter, comprises partially or entirely ionized gas particles, such as electrons, ions, atoms, and molecules. The plasma technique involves thermal/nonthermal plasma and is regarded as a promising technique for changing the chemical surface properties of carbon materials because of the formation of active chemical species/functional groups on the surface. Due to its high-temperature operation, thermal plasma is a destructive technique and is not considered for such applications. In contrast, due to the low degree of ionization, nonthermal plasma (NTP) techniques have recently attracted significant attention toward the surface modification and engineering of different materials such as carbon, polymer, ceramic, and metallic surfaces without altering the bulk material properties. The surface functionalization of carbon materials through NTP can provide excellent reproducibility with remarkable specificity and sensitivity, unveiling different possibilities for biosensor applications. In this regard, one cost-effective NTP is dielectric barrier discharge (DBD), which uses AC voltage for operations due to a dielectric material present between electrodes, ensuring uniform discharge throughout the discharge volume.[15,20,22−24]

This work reports the usage of NTP as a nondestructive and scalable methodology for surface functionalization of carbon powders to generate carboxylic groups on the surface. Electrodes made of these functionalized carbon powders can be used to bind the bioreceptor and also transduce the interaction between the target analyte and the bioreceptor using well-known electroanalytical techniques. The focus is to maximize the carboxylic groups on the surface. To this extent, the carbon surface was functionalized using a DBD-plasma setup with different plasma treatments such as CO2, H2O, and CO2 + H2O treatments. Rigorous postcharacterizations such as NH3-temperature-programmed desorption (TPD) analysis and X-ray photoelectron spectroscopy (XPS) revealed that CO2 + H2O gives a higher number of carboxylic groups on the surface. As a proof of concept, the proposed surface functionalization protocol is utilized to develop an ultrasensitive biosensor for Escherichia coliO157:H7.

Identifying pathogenic bacteria is a serious concern worldwide; swift and specific identification, therefore, becomes crucial to ensure the safety of food, the environment, and people. Among various pathogenic bacteria, E. coliO157:H7 is considered one of the most dangerous foodborne pathogens causing life-threatening diseases such as hemorrhagic colitis, hemolytic-uremic syndrome, and severe gastrointestinal infections in some cases. Due to being simple, label-free, cost-effective, fast, sensitive, and specific, electrochemical sensors based on specific binding of the antigen and antibodies have emerged as excellent alternatives to traditional methods such as culture and colony-counting methods, enzyme-linked immunosorbent assays (ELISAs), surface-enhanced Raman scattering (SERS), colorimetric assays, and polymerase chain reaction (PCR) to identify pathogenic bacteria.[25−32] Recently, a few groups have already reported the identification of E. coliO157:H7 with a limit of detection (LOD) of as low as 2–10 CFU/mL, but a lower limit of detection is still needed for bacterial analysis in food and clinical samples.[25,28,33,34]

Subsequently, novel and cost-effective carbon paste electrodes comprising plasma functionalized carbon were fabricated and used as working electrodes. In addition, a robust and cost-effective miniaturized E-cell was also engineered to reduce sample wastage and improve the electrochemical reversibility of the electrode. The physical picture of our fabricated miniaturized E-cell comprising the electrodes is attached as Figure S2 in the Supporting Information. The electrochemical experimentation of the detection of the well-known and straightforward Streptavidin with Biotin-modified electrodes was done in view of its applicability toward biosensors. Later, the swift and specific identification of the target E. coliO157:H7 was accomplished using the bioelectrode having anti-E. coliO157:H7 antibodies on its surface. Further, the selectivity and interference of the bioelectrode were studied using the nontarget E. coli, Staphylococcus aureus, and the mixture of both with the target E. coliO157:H7.

Experimentation

Materials and Methods

In this work, graphite, poly(vinylidene difluoride) (PVDF), N-methyl-2-pyrrolidone (NMP), phosphate-buffered saline tablets (PBS (pH 7.4)), N-(3-dimethylaminopropyl)-N′-ethyl carbodiimide hydrochloride (EDC), N-hydroxy succinimide (NHS), and bovine serum albumin (BSA) were purchased from Sigma Aldrich. All of the chemicals obtained were of analytical grade and were used as received without any further purification. For all solutions prepared, deionized water (resistivity 18.2 MΩ/cm–1) from the Millipore water purification system was also used for the cleaning and rinsing process. Brain heart infusion broth (BHI Broth) and Luria Bertani broth (Miller) were purchased from Sisco Research Laboratories (SRL) Pvt. Ltd. Mouse anti-E. coliO157 antibody (E10) was obtained from The Native Antigen Company (U.K.), whereas E. coliO157:H7 (ATCC 43888) were obtained from Microbiologics. S. aureus (MTCC 3160) were purchased from MTCC, and E. coli (ATCC 23724) were obtained from CCMB.

Instruments Used for Characterization and Detection

The DBD-plasma setup was used for different plasma treatments, including CO2, H2O, and CO2 + H2O treatments. NH3-TPD analysis was performed to confirm the acidic sites present on the activated carbon using MicrotracBEL Corp., Japan, at a heating rate of 10 °C/min up to a maximum of 700 °C. Prior to the analysis, the samples were preheated at 150 °C for 1 h under a constant He flow to remove the physisorbed species. XPS was utilized to identify the elements present within the carbon or covering its surface along with their chemical states. All electrochemical measurements were performed using the CHI760E (CH Instruments, TX) electrochemical system as an electrochemical analyzer comprising a miniaturized E-cell, a carbon-paste-coated working wire-electrode, a platinum (Pt) wire counter electrode, a carbon paste quasi-reference electrode (CPE), and 0.1 M PBS electrolyte containing a 2.5 mM redox couple ([Fe (CN)6]3–/4–).a

Protocol for Plasma Functionalization of Carbon

Prior to different plasma treatments, carbon powder was pretreated at 800 °C temperature for 4 h under an Ar flow (75 kg/cm2) to confirm that there was no impurity on the surface of the carbon. The treated carbon was then stored in a controlled (airtight) environment to avoid moisture. Later, carbon was functionalized under different gaseous environments in a DBD-plasma reactor. The DBD-plasma setup is displayed in Figure and has been discussed in detail in our earlier publication.[35]

Figure 1

Schematic diagram of the experimental setup for plasma carbon treatment.

In brief, 5 g of activated carbon was placed in the discharge region of the DBD reactor, and N2 gas was flown (30 mL/min) through a water bubble for H2O treatment. Similarly, for CO2 treatment, 30 mL/min pure CO2 was introduced into the reactor, and 30 mL/min CO2 was sent through a water bubble into the reactor for the combined treatment of CO2 and H2O. Later, after 10 mins, for each gaseous treatment, the plasma was operated at 20 kV for 4 h for surface functionalization. NH3-TPD analysis was performed to confirm the acidic sites present on the activated carbon. Finally, the functionalized carbon was collected and tested for biosensing applications.

Protocol for Biosensing Studies

Preparation of Bacterial Samples

Our collaborator research group provided the bacterial cultures (E. coliO157:H7, E. coli, and S. aureus). All of the bacterial cultures consisting of E. coliO157:H7, E. coli, and S. aureus were cultured in BHI broth and incubated at 37 °C. Cells were then subinoculated in sterile BHI media and grown to an optical density (OD) of 0.1. The inoculum’s turbidity was adjusted to standard 0.5 McFarland and diluted in saline, corresponding to bacterial cells around 1.5 × 108 CFU/mL (E. coliandS. aureus) and 1 × 108 CFU/mL (E. coliO157:H7). Further, different concentrations of E. coliO157:H7 (1 × 101, 1 × 102, 1 × 103, 1 × 104, 1 × 105, 1 × 106, and 1 × 107 CFU/mL) and E. coliandS. aureus (1.5 × 101, 1.5 × 102, 1.5 × 103, 1.5 × 104, 1.5 × 105, 1.5 × 106, and 1.5 × 107 CFU/mL) to be tested were prepared in PBS by the process of serial dilution.

Fabrication of the Plasma Functionalized Carbon Paste Electrodes and Preparation of Bioelectrodes toward Streptavidin and E. coli O157:H7 Detection

In view of developing highly specific and cost-effective electrochemical sensors, plasma functionalized carbon interfaces were employed to fabricate the carbon paste electrodes. The protocol for the fabrication of electrodes has been discussed in our previous communication.[6] In brief, the in-house-developed carbon paste (made with the plasma functionalized carbon) was used to fabricate simple, novel, and cost-effective working wire-electrodes. The coating of carbon was done onto the surface of the stainless-steel wire with the help of a well-known and straightforward extrusion process followed by drying in a hot-air oven. The electrodes thus formed were stored and used as working and quasi-reference electrodes for electrochemical experimentation. The comparison of the various plasma functionalized CPEs was done through Biotin–Streptavidin, considering their affinity toward each other.

For the bioelectrode preparation, the well-known EDC-NHS chemistry (0.4 M, 4 h) was initially used to activate the carboxylic groups present on the plasma functionalized CPE (CO2, H2O, CO2 + H2O treated). The electrodes were then washed with DI water and were immersed overnight in a 100 ng/mL solution of Biotin to form a bioelectrode via amide bonding formed between the carboxylic groups and the amine groups present on the surface of the electrodes and in Biotin, respectively. Subsequently, the nonspecific sites of these surface-functionalized electrodes were blocked by BSA treatment and were then used as a bioelectrode to detect Streptavidin of different concentrations. The comparison of the various plasma functionalized CPEs through Biotin–Streptavidin detection studies is discussed in detail in the subsequent sections.

Further, as shown in Figure A, the preparation of a bioelectrode for E. coliO157:H7 detection was also done similarly, wherein the COOH functional groups present in the CO2 + H2O plasma-treated carbon paste electrodes were first activated with EDC-NHS chemistry, followed by the immersion of these activated electrodes at 4 °C in the anti-E. coliO157:H7 antibody of various concentrations (1, 5, and 10 μg/mL). The amine groups present in the antibody helped create an amide bond with the activated COOH terminated carbon paste electrodes. Finally, the nonspecific binding sites of these Ab immobilized electrodes were blocked by BSA treatment and were then used as a bioelectrode for E. coliO157:H7 detection.

Figure 2

(A) Schematic showing the steps required to form a bioelectrode to detect E. coli O157:H7. (B) Demonstration of bacterial detection in a miniaturized E-cell. (C) Experimental setup with the miniaturized E-Cell.

Protocol for Bacterial Detection, Interference, and Selectivity Studies

The electrochemical detection of E. coliO157:H7 was done with the help of a three-electrode system comprising the prepared bioelectrode as a working electrode, CPE as a quasi-reference, and Pt wire as a counter electrode (Figure B,C). A total of 15 μL of each concentration of E. coliO157:H7 was mixed in the miniaturized electrochemical cell containing 1.5 mL of the 2.5 mM redox couple ([Fe(CN)6]3–/4–) to obtain the final concentration ranges of 1 × 10–1, 1 × 100, 1 × 101, 1 × 102, 1 × 103, 1 × 104, 1 × 105, and 1 × 106 CFU/mL. EIS analysis was done to attain charge transfer resistance (Rct) after an incubation time of 30 min for each concentration of E. coliO157:H7 to bind with the antibody present on the surface of the bioelectrode. At least three bioelectrodes were used for the EIS analysis of each measurement, and the results are discussed in the subsequent sections.

Additionally, the analytical performance of the biosensor was estimated by analyzing the selectivity and interference of the bioelectrode toward the different combinations of E. coliO157:H7 with E. coli and S. aureus of various concentrations. The various combinations of all bacteria were selected in an equal ratio of 1:1 with concentrations of 1.5 × 100, 1.5 × 103, and 1.5 × 106 CFU/mL for E. coli and S. aureus and 1 × 100, 1 × 103, and 1 × 106 CFU/mL for E. coliO157:H7.

Results and Discussions

Characterization Studies

For DBD-plasma treatment, the discharged power was calculated with the help of the Lissajous figure (Figure A), and it was found that a maximum of 1.1 W discharged power was utilized for the treatment. The aim of different plasma treatments on carbon powder was to create carboxylic groups onto the surface of the carbon to finally bind with an amine to generate an amide bond. All of these treatments helped create the carboxylic groups onto the surface of the carbon, but the number of carboxylic groups with each treatment was different, and the NH3-TPD analysis revealed the same. Figure B describes the NH3-TPD of different functionalized carbons. As shown in Table , CO2 + H2O-treated carbon showed the highest number of acidic sites, which is expected due to the simultaneous dissociation of acidic surface functional groups. The maximum weak acidic sites were observed in the case of only CO2-treated carbon, which may be due to the moisture impurity present in the CO2 source. The lowest acidic group was found with H2O-treated carbon because of the least probability of COOH formation with OH radicals and solid carbon coupling.

Figure 3

(A) Lissajous figure. (B) NH3-TPD for different functionalized carbons. (C) XPS analysis.

Table 1

Number of Acidic Sites Present on Each Carbon Sample

sample	weak (μmol/g)	medium (μmol/g)	strong (μmol/g)	total (μmol/g)
raw carbon	0	0	0	0
H2O-treated carbon	7	1	0	8
CO2-treated carbon	6	2	1	9
CO2 + H2O-treated carbon	8	3	3	14

Plasma functionalization under various plasma gases is known to produce varying amounts of acidic and basic groups on the surface. XPS analysis as shown in Figure C was hence utilized to analyze the surface chemistry of the plasma-treated carbon with different methods. Deconvolution of C 1s confirms the presence of graphitic carbon (284.5 eV); phenolic, alcohol, or ether groups (286.1 eV); carbonyl or quinone groups (287.4 eV); and carboxyl acidic groups (289.3 eV). Similarly, the O 1s peak position for oxygen was around 528 eV for raw carbon and 529 eV for plasma-treated carbon. In addition, it also shows that the plasma-treated carbon had a higher oxygen content (highest for CO2 + H2O-treated carbon) when compared to the untreated/raw carbon, which indicates that the different plasma treatments successfully grafted oxygen species onto the surface of the carbon. XPS analysis thus confirmed the successful incorporation of acidic groups through different plasma treatments (highest for CO2 + H2O-treated carbon), which are later utilized for biosensing applications. NH3-TPD analysis, as discussed earlier, also confirmed the highest amount of acidic groups onto the surface of the CO2 + H2O-treated carbon.[18,36,37]

Streptavidin Detection Studies

For the application of plasma functionalized bioelectrodes toward electrochemical biosensors, we need to have a significant number of carboxylic groups onto the surface of the carbon powder to have more numbers of receptors binding to the surface with the help of amide bonding. The electrochemical experimentation of a well-known and straightforward Biotin–Streptavidin detection was done in view of its applicability toward biosensors, as shown in Figure A. Biotin–Streptavidin, having a strong affinity toward each other, is used extensively as an ideal pair for validating the biosensing mechanism and so was an appropriate choice for biosensor applications. Moreover, these studies were performed to investigate the influence of carboxylic groups created onto the carbon surface with different functionalization techniques. To explain in detail, the Biotin-modified bioelectrodes formed using the plasma functionalized carbon (different treatments) were used to detect Streptavidin of different concentrations. Figure S1 in the Supporting Information shows the DPV response of Biotin-modified plasma-treated electrodes with different target Streptavidin concentrations. The results shown in Figure S1 display the reduction in peak current with the increase in the target Streptavidin concentrations. The reduction in peak current was due to the successful binding of Streptavidin to Biotin, indicating the deterioration in reaction kinetics at the working electrode/electrolyte interface. Figure A and Table correspond to the normalized current response obtained from DPV analysis (Figure S1) of each plasma-treated electrode, and the results displayed different current responses for the different plasma-treated carbon at different concentrations of Streptavidin. For a 10 ng/mL target Streptavidin concentration, the normalized current responses obtained for CO2, H2O, and CO2 + H2O plasma-treated carbon bioelectrodes were 0.12 ± 0.01, 0.097 ± 0.0088, and 0.15 ± 0.0094, respectively. The CO2 + H2O-treated carbon bioelectrode undoubtedly showed a significantly higher response compared to their counterparts. The different responses obtained for different plasma-treated carbon were due to the different numbers of carboxylic groups present on the surface of the electrodes. The higher number of carboxylic groups present on the CO2 + H2O-treated carbon paste electrode resulted in a higher response, whereas the response obtained with the CO2- and H2O-treated carbon paste electrode was comparatively low, suggesting a lower number of bindings with Streptavidin. The results obtained validate the initial findings and results obtained from XPS and NH3-TPD analyses. The bioelectrodes accounted for a very low detection limit of 0.24 fg/mL, calculated using 3.3 σ/slope, where σ is the standard deviation of the blank response and slope = 0.011 was obtained from the linear fit of the calibration graph of Figure A (for the CO2 + H2O-treated electrode), showing the applicability of the plasma-treated carbon for biosensor application. Later, these electrodes were also subjected to control studies with PBS, and the results obtained from Figure B showed no significant response with different concentrations of PBS. The response received with PBS was almost negligible compared to the response obtained after Streptavidin binding to Biotin, offering an excellent selectivity of the bioelectrodes toward Streptavidin.

Figure 4

Biotin–Streptavidin detection. (A) Normalized current response obtained from DPV analysis for different concentrations of Streptavidin. (B) Normalized current response obtained from DPV analysis for different concentrations of PBS.

Table 2

Biotin–Streptavidin Detection: Normalized Current Responses Obtained for Different Plasma-Treated Electrodes at Different Concentrations of Streptavidin

Streptavidin concentration	CO2 treatment	H2O treatment	CO2 + H2O treatment
1 fg/mL	0.049 ± 0.005	0.04 ± 0.0009	0.053 ± 0.0008
100 fg/mL	0.061 ± 0.0005	0.057 ± 0.0028	0.075 ± 0.0024
10 pg/mL	0.08 ± 0.005	0.075 ± 0.008	0.098 ± 0.0069
1 ng/mL	0.098 ± 0.007	0.08 ± 0.008	0.12 ± 0.0082
10 ng/mL	0.12 ± 0.01	0.097 ± 0.0088	0.15 ± 0.0094

Bacterial Detection Studies

The results obtained from Biotin–Streptavidin showed a superior response toward CO2 + H2O-treated CPE, and therefore, the CO2-H2O-treated CPE was selected for E. coliO:157 detection. For this, the CPE was first fabricated from the CO2 + H2O-treated carbon, and later after DI water wash, these electrodes were stored and used to prepare a bioelectrode for E. coliO:157 detection. The schematic in Figure A shows the different steps required for a bioelectrode preparation to detect the different concentrations of E. coliO:157. The bioelectrodes formed were used for detecting various concentrations of E. coliO:157. DPV was used to optimize the incubation time and concentration of the anti-E. coliO:157 antibody, while EIS was used to detect various concentrations of E. coliO:157, followed by the interference and selectivity studies. Figure A,B shows the normalized current response obtained for the optimization studies of incubation time and concentration of the anti-E. coliO:157 antibody, respectively.

Figure 5

E. coli O157:H7 (EcO) detection, selectivity, and interference studies. (A) Normalized current response obtained from DPV analysis for various concentrations of E. coli O157:H7 at different incubation times (5 μg/mL anti-E. coli O157:H7 antibody concentration). (B) Normalized current response obtained from DPV analysis for various concentrations of E. coli O157:H7 at various concentrations of anti-E. coli O157:H7 antibody immobilized on a bioelectrode (incubation time of 30 min). (C) EIS analysis for various concentrations of E. coli O157:H7 detection (inset showing the modified Randle’s circuit). (D) Calibration graph showing the normalized charge transfer resistance obtained from EIS analysis for various concentrations of E. coli O157:H7 (5 μg/mL anti-E. coli O157:H7 antibody concentration with an incubation time of 30 min). (E) Selectivity and interference studies. (F) Bar graph showing the normalized charge transfer resistance obtained from EIS analysis for various concentrations of E. coli O157:H7 spiked in PBS and pond water.

As shown in Figure A, for an antibody concentration of 5 μg/mL, the normalized current response obtained for various concentrations of E. coliO:157 was more prominent when the electrode was incubated for 45 min. It goes without saying that a higher incubation time led to more numbers of bindings when compared to its counterparts, but a larger incubation time also led to a larger detection time, and hence, the optimized incubation time of 30 min was selected, which was sufficient enough to bind and detect the E. coliO:157 with a good resolution. Later, the optimization studies of the anti-E. coliO:157 antibody concentration were done at an incubation time of 30 min, as shown in Figure B. The anti-E. coliO:157 antibody concentration of 10 μg/mL accounted for the higher response and sensitivity, significantly higher than the response obtained for 5 μg/mL at an incubation time of 45 min. It, therefore, was selected as an optimized concentration with an incubation of 30 min.

Figure C corresponds to the Nyquist plot obtained from electrochemical impedance spectroscopy of the EcOAbmCHCPE bioelectrode, before and after binding the target E. coliO157:H7 at different concentrations of 1 × 10–1–1 × 106 CFU/mL. The results displayed in Figure C demonstrate an increment in charge transfer resistance (Rctn) with every increasing target bacterial concentration, suggesting that the reaction kinetics at the bioelectrode/electrolyte interface is degrading with an increase in the target bacterial concentrations of E. coliO157:H7. This can be attributed to the fact that once the target bacterial concentrations of E. coliO157:H7 are added to the electrolyte solution, they start binding with the immobilized anti-E. coliO157:H7 antibody onto the surface of the bioelectrode.

The electrical behavior of the sensing mechanism is determined with the help of a modified Randle’s circuit shown in the inset of Figure C, wherein the experimental values of the circuit elements (Table S2 in the Supporting Information) are obtained by fitting the obtained impedance data. In this circuit, the parameters Rs, Rct, W, and CPE represent the solution resistance, the charge transfer resistance, the Warburg impedance, and the constant phase element indicating the behavior of a double-layer capacitance, respectively (refer to Section S1 in the Supporting Information for additional information). Figure D corresponds to the calibration curve where the normalized change in charge transfer resistance (ΔRct/Rct) is plotted at different target concentrations of E. coliO157:H7. Here, ΔRct/Rct was calculated as (Rctn – Rct)/Rct, where Rct and Rctn (n = −1 to 6) represent the charge transfer resistance values of the bioelectrode without the target and the values obtained at different concentrations of target E. coliO157:H7, respectively. The error bars associated with each data point shown in Figure D represent the standard deviation (N = 3). A linear regression model was used to fit the curve shown in Figure D, which produced a linear equation (eq ) with an adj. R-square value of 0.976.Here, x represents the target concentrations of E. coliO157:H7 and y is the dependent variable, i.e., ΔRct/Rct. The values 0.096 and 0.469 represent the slope and the intercept of the linear curve, respectively. The proposed sensing platform accounted for a sensitivity of 0.98 ((ΔRct/Rct)/(CFU/mL))/cm2 with a wide linear detection range of 1 × 10–1–1 × 106 CFU/mL having an LOD of 0.1 CFU/mL. The limit of detection (LOD) and the sensitivity of the proposed biosensor were calculated using 3.3 σ/slope and slope/area, respectively, where slope = 0.096 was obtained from the linear fit of the calibration graph and σ is the standard deviation of the blank response. The LOD achieved with the proposed sensing platform is substantially low compared to that in the previous literature (Table S1 in the Supporting Information).

For point-of-care applications, the repeatability, interference, and selectivity studies of the proposed biosensing platform were also performed, and the same is shown as a bar graph in Figure E. The repeatability of the proposed sensing platform was investigated by successfully detecting different target E. coliO157:H7 concentrations by taking three different electrodes. The results obtained were consistent throughout the experimentation, showing excellent repeatability and reliability of the proposed biosensing platform. An extensive study was done to understand the response of the bioelectrode at various concentrations of nontargeted S. aureus, E. coli, and their mixture with the target concentration (E. coliO157:H7). For this, the concentrations of S. aureus and E. coli were chosen in the range of 1.5 × 100, 1.5 × 103, and 1.5 × 106 CFU/mL as well as 1 × 100, 1 × 103, and 1 × 106 CFU/mL for the target E. coli. The mixture consists of all three bacteria in equal ratios with the above-mentioned concentrations. The normalized change in charge transfer resistance is presented as a bar graph in Figure E. As illustrated in Figure E, for each concentration range, the normalized change in charge transfer resistance was the highest for the target E. coliO157:H7 when compared to S. aureus and nontarget E. coli, alone, suggesting the excellent selectivity of the proposed platform. The dependency of the target E. coliO157:H7 concentration can be seen in the mixture, implying that the proposed platform has a dominant response toward the target E. coli. The results indicated the superior response for E. coliO157:H7, suggesting its excellent selectivity and interference toward the interfering species, thus demonstrating its applicability toward swiftly identifying the target bacteria. In addition, we also demonstrated the applicability of the proposed biosensing platform toward real samples, and the same results are posted in Figure F. This was carried out by performing the experiments with bacterial samples prepared in pond water, and the results of the same were compared with the samples prepared in PBS. For this study, the concentrations of target E. coliO157:H7 in PBS and pond water were chosen as 1 × 10–1 and 1 × 100, respectively, and the normalized change in charge transfer resistance obtained is presented as a bar graph in Figure F. As illustrated in Figure F, the normalized change in charge transfer resistance was the highest for the target E. coliO157:H7 prepared in PBS when compared to its concentration prepared in pond water. The difference in the data values was probably due to the matrix effect present in pond water as it may contain several ions, bacteria, and other species. Although there was a difference between the absolute values of normalized change in charge transfer resistance, the difference was minimal, and the proposed biosensor could still identify the target E. coliO157:H7 concentration in pond water with a dominant response. This suggested the excellent selectivity of the proposed platform, thus demonstrating its applicability toward identifying the target bacteria in real samples.

Conclusions

Nonthermal plasma functionalization is a straightforward, quick, and widely used surface modification technique for a variety of materials. Furthermore, it is a nondestructive and scalable surface functionalization methodology that is widely used to overcome the limitations of wet chemistry. Plasma functionalized carbon interfaces were used to create highly specific, low-cost electrochemical sensors. A DBD-plasma setup was used to functionalize the carbon surface, which included CO2, H2O, and CO2 + H2O plasma treatments. The presence of functional groups on the carbon surface was determined using NH3-TPD and XPS analyses. Later, plasma-treated carbon was used to create novel and low-cost carbon paste electrodes. In addition, a robust and cost-effective miniaturized E-cell was built and engineered to reduce sample waste. Biotin–Streptavidin detection was performed due to its applicability to biosensors due to their strong affinity to each other. As a result of the superior response to the CO2 + H2O-treated bioelectrode, CO2 + H2O-treated CPEs modified with anti-E. coliO157:H7 antibodies were later used to identify the target E. coliO157:H7. The proposed biosensing platform detected E. coliO157:H7 with a wide linear detection range of 1 × 10–1–1 × 106 CFU/mL, accounting for an LOD and sensitivity of 0.1 CFU/mL and 0.98 ((ΔRct/Rct)/(CFU/mL))/cm2, respectively. Furthermore, the bioelectrode’s selectivity and interference were investigated using nontargeted E. coli, S. aureus, and a mixture of both with the target E. coliO157:H7. The superior response for E. coliO157:H7 in the mixture of interfering species suggested its excellent selectivity and interference, along with excellent repeatability and reliability. In addition, the applicability of the proposed biosensing platform toward real samples (E. coliO157:H7 spiked in pond water) was also demonstrated. The results suggested high specificity for the target E. coliO157:H7, making this methodology suitable for real-time bacterial detection.