The symptoms of infection caused by the influenza virus (H1N1) are similar to those of infection with the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Both types of viruses share common characteristics, such as predominantly affecting the respiratory system, causing symptoms including cough, runny nose, sore throat, fever, headache, and fatigue [1]. Accurate diagnosis and treatment are therefore crucial. To this end, various detection methods and tools have been developed using diverse technologies such as surface-enhanced Raman spectroscopy, Raman, fluorescence, and electrochemical techniques to enable more precise diagnosis of influenza virus discriminated from SARS-CoV-2 [2–4].

Among these technologies, electrochemical-based field-effect transistors (FETs) have shown excellent performance, particularly when incorporating a two dimensional nanomaterial (graphene), due to their unique properties including thermal conductivity, sp²-hybridized carbon atoms, zero-band gap, mechanical stiffness, elasticity, and strength [5–8]. 1-pyrenebutyric acid-N-hydroxy-succinimide ester (PANHS) has been utilized to functionalize the graphene surface and immobilize bioprobes, due to its facile introduction method and availability as a commercial product [9–11]. However, PANHS, which relies on π-π interactions, has limitations, such as the exposure of back bonds, a low number of terminal functional groups, and a wide band gap [12–14]. Therefore, to overcome these limitations, alternative approaches and strategies are required.

In this study, we demonstrate a high-performance FET system by introducing a covalent bond-based interfacial chemical (N-heterocyclic carbene, NHC) and H1N1 influenza virus antibodies (flu Abs) onto side-gate FETs (flu bioelectronics). The superiority of the NHC-based interfacial chemical was confirmed through density functional theory (DFT) simulations. Specifically, the semiconducting properties including the interaction energy and orbital energy gap (OEG) were evaluated. Covalent bonding onto graphene was verified by a surface analysis and transmission electron microscopy (TEM) and Raman analyses. The performance of the flu bioelectronics was evaluated through real-time electrical monitoring, showing excellent sensing capabilities with a limit of detection (LOD) of 10⁰ pfu/mL. Additionally, the detection time was within 30 s, and the linear detection range was extended from 10¹ to 10⁴ pfu/mL. Based on these results, the flu bioelectronics demonstrate potential for use in the detection and monitoring of highly infectious diseases.

2.1. Fabrication of the electrodes

Single layer graphene was synthesized by the chemical vapor deposition method using hydrogen, methane, and argon gases and the synthesized graphene was transferred onto a silicon wafer via a wettransfer process [15]. The graphene was then micropatterned using photolithography and etching processes via a microelectromechanical system. The side-gate electrodes (Cr/Au) were formed using photolithography, E-beam deposition, and lift-off processes.

2.2. Fabrication of the flu bioelectronics

The graphene channel in the electrodes was functionalized by immersion in NHC/tetrahydrofuran (50 μM) for 30 m under inert gases. Potassium bis(trimethylsilyl)amide of 1 eq. was then injected into the reactor for 2 h. Finally, the NHC-introduced graphene electrodes were washed using tetrahydrofuran, methanol, and distillated water three times [16]. The flu Abs were conjugated with the azide functional group of the NHC terminus by forming an amide bond using copper free agent [(1R,8S,9s)-Bicyclo[6.1.0]non-4-yn-9-ylmethyl N-succinimidyl carbonate].

To compare the interaction between graphene and interfacial chemical, a DFT simulation was performed to investigate the band gap and electron density [Fig. 1(a)]. The interaction energy and the band gap were calculated by Eq. (1) as follows:

Figure 1. Comparison of the interfacial chemicals using DFT simulation. (a) The structures of HOMO and LUMO and the OEG for the interacted materials. (b) EDDM of PANHS-introduced graphene and NHC-introduced graphene.

Here, E_b, E_GM, and E_chem denote the absolute energies toward the interaction energy, graphene, and each interfacial chemical (PANHS and NHC) respectively. The net interaction energies (E_int) were – 0.2439 and –29.1621 eV for PANHS and NHC, respectively [17,18]. The interaction energy between NHC and graphene was found to be 10 times higher than that for PANHS-graphene, which indicates outstanding stability under exposure to external energy, owing to the covalent bond based on the [2+1] nitrene cycloaddition reaction [19]. Additionally, the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) were confirmed to calculate the OEG : –4.923 and –2.101 for PANHS and –3.687 and –2.194 eV for NHC [20,21]. Through the orbital energy, the OEGbased band gaps were calculated as 2.822 and 1.493 eV, which indicates NHC-introduced graphene has a lower band gap compared to PANHS-graphene. Moreover, the electron density difference map (EDDM) was obtained to investigate the electron movement through a DFT simulation [Fig. 1(b)] [22]. The electrons in PANHS-graphene showed a static state, with no electron transfer between PANHS and graphene. On the other hand, the electrons in NHC-graphene displayed movement from NHC to graphene owing to the formed covalent bonds [23,24]. This indicates the possibility of obtaining higher sensitivity owing to the dramatical density change of the charge carrier.

To confirm the covalent bonding between NHC and graphene, the morphology of the graphene surface was observed through measurement by using Cs-corrected high-resolution scanning TEM (CS-STEM) [Fig. 2(a)]. The image of pristine graphene displayed a hexagonal pattern, while a changed lattice was observed after the introduction of NHC onto the graphene surface because of the formation of vertical bonds with the backbone structure of NHC [25]. Moreover, a selectedarea electron diffraction (SAED) analysis was carried out to confirm the crystal structure before and after the introduction of NHC. Pristine graphene clearly showed a hexagonal pattern centered at the (111) plane, whereas the NHC-introduced graphene exhibited an additional pattern surrounding the (111) plane. Based on these results, the formation of the bond was clearly confirmed by a morphology analysis. Raman spectroscopy measurements were conducted to investigate the NHC bonds introduced onto graphene [Fig. 2(b)]. Pristine graphene (black line) exhibited an ideal single layer graphene structure, presenting approximately a two-fold higher ratio of I_2D versus I_G (1,583 and 2,639 cm^-1). Additionally, other peaks including a D peak presented no recognizable signal in pristine graphene, verifying there are no defects on the clear single layer graphene. NHC-introduced graphene displayed several peaks including D, D*, D+D**, and D+D* peaks. These various peaks appeared as a result of the covalent bonds between NHC and graphene based on the unshared electron pair ([2+1] nitrene cycloaddition reaction) (orange line). In particular, a broad D band was confirmed in a range of 1,277 to 1,379 cm−1 by defects of the hexagonal carbon atom rings, while the peak for (D+D*) appearing at 2,918 cm−1 is the defect band peak from the emitted intervalley phonons [26,27]. Additionally, D* and D+D** peaks exhibited defect effects by covalent bonding and visible light [28,29]. Based on these results, NHC covalent bonding with graphene was formed.

Figure 2. Surface analysis of NHC-introduced graphene. (a) TEM and SAED images before and after graphene depending on the NHC introduction. (b) Raman spectra of pristine graphene and NHC-introduced graphene.

The flu bioelectronics were fabricated by immobilization of the flu Abs, and the electrical properties were measured after flu Abs immobilization on the flu bioelectronics. Current-voltage (I-V) curves were performed in a range of – 1.5 to 1.5 V_ds to observe the variance of the contact resistance [Fig. 3(a)]. The slopes of the NHC-introduced graphene-based electrode and the flu bioelectronics were 0.6346 and 0.4928, respectively. The slope decreased by increased resistance according to the immobilization of flu Abs onto the graphene electronics. Although the increased contact resistance indicates decreased current, the linear trend was maintained following Ohm’s law. In addition, transfer curves were measured to examine the change of the charge carrier density [Fig. 3(b)]. The NHC-introduced graphene electrode showed the Dirac point at 0.16 V_g, and this phenomenon presented the general electrical property (ambipolar) of the graphene. After the immobilization of flu Abs, the Dirac point was positively shifted (to 0.32 V_g) owing to the negative net charge, based on the zeta potential results [30]. Based on these results, the flu bioelectronics can be utilized to detect flu virus in the p-type area as nanobiosensors.

Figure 3. Electrical properties of the flu bioelectronics. (a) I-V curves and (b) transfer curves before and after the influenza Abs conjugation on the NHC-introduced graphene electronics.

The sensing performance was evaluated through electrical measurements in real-time for influenza virus detection [Fig. 4(a)]. Real-time responses upon exposure to phosphate-buffered saline buffer confirmed the effect by the buffer solution and the results showed no significant change while maintaining a stable current level. Various concentrations of the virus from 10⁰ to 10⁵ pfu/mL were then injected into the flu bioelectronics to observe the LOD and the saturation concentration. The current was changed after the injection of 10⁰ pfu/mL concentration, showing a LOD of 10⁰ pfu/mL. The saturation concentration was investigated at 10⁴ pfu/mL by no change of the current level. The concentration curve was obtained from the results of the real-time monitoring [Fig. 4(b)]. The determined K value was 89.0462 mL/pfu, which was calculated following Langmuir isothermal absorption [31,32]. Additionally, a linear trend was observed in the range of 10¹ to 10⁴ pfu/mL and the flu bioelectronics showed detection performance in a broad range.

Figure 4. (a) Real-time monitoring of various concentrations using the flu bioelectronics. (b) Concentration curve based on the real-time response.

In summary, we have demonstrated enhanced bioelectronics by covalent bonding using [2+1] a nitrene cycloaddition reaction-based interfacial chemical. The proposed flu bioelectronics provided high sensitivity and rapid detection using a real-time monitoring FET system. In addition, improved performance based on the covalent bonding between the interfacial chemical compound and graphene was verified by a DFT simulation, which confirmed higher interaction energy, a lower band gap, and electron transfer. The formation of covalent bonding between NHC and graphene was confirmed by TEM and Raman analyses and the results clearly showed the formation of regular bonds in a large area. Moreover, the sensing performance showed a low LOD of 10⁰ pfu/mL concentration and the detection was conducted within 30 s. These results indicate the possibility of utilizing the fabricated bioelectronics for virus detection and monitoring of highly infectious diseases universally.

This research was supported by the Defense Acquisition Program Administration (ADD-911255202).

The authors declare no conflicts of interest.