key: cord-0771926-1ftqm3ki
authors: Masurkar, Nirul; Thangavel, Naresh Kumar; Yurgelevic, Sally; Varma, Sundeep; Auner, Gregory W.; Reddy Arava, Leela Mohana
title: Reliable and highly sensitive biosensor from suspended MoS(2) atomic layer on nano-gap electrodes
date: 2020-10-17
journal: Biosens Bioelectron
DOI: 10.1016/j.bios.2020.112724
sha: 375438c19a4857d1929cc9e6a953bb7b70800f03
doc_id: 771926
cord_uid: 1ftqm3ki

The uneven morphology and the trapped charges at the surface of the traditionally used supporting substrate-based 2D biosensors produces a scattering phenomenon, which leads to irregular signal from individually fabricated device. Though suspended 2D channel material has the potential to overcome scattering effects from the substrates but achieving reliability and selectivity, have been limiting the biosensor technology. Here, we have demonstrated nanogap electrodes fabrication by using the self-assembly technique, which provides suspension to the 2D-MoS(2). These nano-spacing electrodes not only give suspension but also provide robust strength to the atomic layer, which remains freestanding after coating of the Hafnium oxide (HfO(2)) as well as linkers and antibodies. For evaluating the electrical characteristics of suspended MoS(2) FET, gating potential was applied through electrolyte on suspended MoS(2) transistor, which achieved lower subthreshold swing 70 mV/dec and ON/OFF ratio 10(7). Afterward, pH detection has demonstrated at room temperature, which shows an impressive sensitivity of ∼880 by changing 1 unit of pH. Besides, we have successfully shown Escherichia coli (E. coli) bacteria sensing from suspended MoS(2) transistor by functionalizing dielectric layer with E. coli antibodies. The reported biosensor has shown the ∼9% of conductance changes with a lower concentration of E. coli (10 CFU/mL; colony-forming unit per mL) as well as maintain the constant sensitivity in three fabricated devices. This type of the architecture has a potential to detect range of biomolecules such as COVID-19 viruses also by altering the oxide surface.

Due to their potential of providing continuous and real-time information, biosensors are prominent devices in the medical field and garnering substantial interest in other domains such as forensic industries, national security, and ecological monitoring (Mehrotra and research 2016) .

Specifically, electrochemical field-effect transistor (FET)-based biosensors have been recognized as one of the promising candidates because of their compatibility with electronic devices, low power consumption, label-free detection of specific biomolecules, and low-cost mass production (Kim et al. 2016; Kuriyama and Kimura 1991; Suvarnaphaet and Pechprasarn 2017) . In FET based biosensor, the charged biomolecules generate the electrostatic effect and bring the variation in the conductivity of the channel, which can easily measure by transistor characteristics such as source to drain current (Kaisti 2017; Sarkar et al. 2014a) . To obtain better electrostatic effects from charged biomolecules, many semiconducting materials have been employed (Holzinger et al. 2014; Li et al. 2017; Zhu et al. 2012 ). Among the various materials, two-dimensional (2D) semiconducting materials are attracting much attention because of their tunable bandgap, a high surface to volume ratio that provides higher sensitivity for detection.

In the 2D domain, transition metal dichalcogenides (TMDs) like MoS 2 has gained much attention in the field of FET-biosensor due to its direct bandgap, biocompatibility, and high mobility (Kalantar-zadeh and Ou 2015b). Despite these merits, the performance and consistency of such atomic layer crystals are easily affected by supporting substrate interaction (Chen et al. 2018 ). The interaction of supporting substrate with atomic layers of MoS 2 degrade the transport properties by scattering, which implies the interface control is vital for the performance and reliability of biosensor devices (Du et al. 2008) . For instance, the surface of silicon dioxide(SiO 2 ) substrate is highly disordered as well as chemically active due to the trapped atmospheric gases, J o u r n a l P r e -p r o o f chemical adsorbates, and unknown functional groups (Zhang et al. 2009 ). Therefore, transferring another layer of MoS 2 on the top of SiO 2 or any other insulating substrate cannot contribute to carrier charge transport distinctly, which leads to the unreliable output of every single device. In recent years, many efforts have been employed to enhance the quality of the substrate, such as active layer interface by using Poly(methyl methacrylate) (PMMA) and polymer electrolytes [9] . These layers elude chemical bonding or surface roughness and improve carrier transport properties, but they cannot be employed in biosensor applications due to the fabrication and reproducibility issues. The other approach is creating suspension of an atomic layer in between electrodes to enhance the carrier transport and remove scattering effect by wet etching of the SiO 2 layer underneath the monolayer MoS 2 . Freestanding MoS 2 has shown better performance than the supporting on the SiO 2 substrate in terms of back gating electronic conduction (Jin et al. 2013 ). However, the existing SiO 2 requires hazardous chemical etchants such as hydrofluoric acid (HF), which is difficult to handle and affects the 2D film structure and purity ). Secondly, freestanding MoS 2 sags between the two electrodes because of the large spacing (~ 2 µm), which makes it impossible to coat dielectric layers such as hafnium oxide (HfO 2 ) and antibodies. Therefore, this structure impedes making top gate FET biosensors, which allows only back gating. However, the back-gate FET requires more power (input gate voltage) to turn ON the device than the top gate, which hinders making a low power and highly sensitive biosensor. In FET based biosensor, the sensitivity is inversely proportional to the subthreshold swing (SS), which is defined as a change in the current with respective dielectric surface potential ( = / log ; dV gate and I drain are a change in gate potential and drain current respectively). Thus, in back gate requires higher input voltage to turn on the device than the top gate, which reduces the sensitivity of detection (Sarkar et al. 2014a; Sarkar et al. 2014b ).

Given the importance of reliability and sensitivity of FET based biosensor, in this work, chemical vapor deposition (CVD) grown MoS 2 is transferred by using novel dry stamping method on self-assembled photolithographically patterned nano-gaps to achieve suspension. nanogap. Nano-spacing between electrodes rely on the rate of oxidation of the Cr layer, where temperature controls the rate of the oxidation. Therefore, it is essential to maintain the temperature as well as environmental gases of the wafer constant throughout the oxidation process. To achieve the proper pattern of the second layer, the thickness of the Ti/Au (second layer, whose thickness is 10 nm/ 50 nm) should not exceed more than 50% of chromium thickness. Liftoff the chromium as well as the second Au electrode is processed in chromium etchant solution (Chromium etchant, Sigma Aldrich).

Nano-gaps cannot be seen via microscope; therefore, before transferring the 2D material, it was characterized by the current-voltage method. I-V curve between two electrodes was measured by the source meter (Keysight B2912A). 

Bacterial specimens used in the study were prepared from culture plates.

An isolated singer colony was added to 5 ml of media in a 14 ml culture tube. The culture tube was placed on a shaker in a 37°C incubator and incubated overnight (18 hours). The overnight culture was centrifuged at room temperature for 5 minutes at 3500 rpm. The supernatant was removed, and the bacteria pellet was re-suspended in 5 ml of filtered sterilized tap water. The bacteria were centrifuged, and the washing process repeated once. After the final wash, filtered tap water was added to the bacteria pellet until the optical density (measured at a wavelength of 600nm using a Spectro Max PLUS spectrophotometer) of the solution reached the desired value.

After the optical density at 600nm was measured for the stock solution, serial dilutions were employed.

J o u r n a l P r e -p r o o f 3.Results and Discussions:

Our approach for patterning the nanogap electrode uses a self-assembly photolithography technique, as shown in Fig. 1 (a) . The metal lift-off method has been used for engraving the first electrode gold (Ti/Au) and chromium (Cr), the thickness of the Au kept less than the Cr because it acts as a sacrificial layer as represented in Fig. 1(a) 

The electrostatic effect of MoS 2 FET investigated to compare the electrical performance of the supported and suspended devices. Fig. 2 represents the current-voltage (I-V) characteristics by applying different gate voltages (V BG ) of fabricated devices in a dry and wet environment. Fig. 2 (a) and (b) illustrates the I D -V BG curve of supported and free-standing MoS 2 at 100 mV bias voltage (V DS ), where supported channel length is 2 µm and suspended is ~90 nm.

The suspended device represents excellent ON/OFF ratio 10 7 and threshold voltage (V T ) 3.9 V as compared to the supported one, which is equal to 10 6 and -9 V at room temperature. The improvement of the V T, i.e., switching of the device at low voltage, is due to the elimination of the supporting substrate, which confines the electrons mean free path by scattering as well as from trapped charges on the surface (Masurkar et al. 2020 ). Fig. 2 Table S1 (Supplementary Fig   S4) . Low SS, threshold voltage, high mobility and elimination of substrate scattering make these types of the FET structure more sensitive as well as consistency with better gate controllability in a wet environment.

After achieving excellent performance in the ionic gating effect, the suspended device developed on the dielectric surface. The change in ∆V T found to be 59.1 mV/pH, which is satisfied the Nernst limit at room temperature, i.e., 59.3 mV/pH and agree with early studies of ion-sensitive field-effect transistor (ISFET) characterization on HfO 2 (Van Hal et al. 1995; Zafar et al. 2011) . Therefore, HfO 2 does not require any functionalization for pH sensing as compared to the SiO 2, whose change in threshold voltage found to be 30-40 mV/pH (Bergveld and Physical 1996; Gao et al. 2009 ). Fig. 3 (c) represents the sensitivity of the pH at three different regions (Subthreshold, saturation, and linear), which is defined by the equation (1), where S n is the sensitivity of pH, I pH1 and I pH2 are the transistor current measured from two different pH values (where pH1>pH2).

(1)

The sensitivity at the subthreshold region found to be much higher because the drain current is an exponential function of the gate voltage. In contrast, the saturation and linear parts are quadratic and linear with the change in gate potential. In a suspended device, the sensitivity at the subthreshold region was found to be ~876 from pH 5 to 6 and ~880 from pH 7 to 8 respectively, which is far better than the previous reports of MoS 2 FET biosensor on the supporting substrate (Sarkar et al. 2014b ). The suspended MoS 2 FET possesses higher sensitivity than supported one because of the lower SS, where SS defined as the = log ⁄

. This equation elucidates that the change in subthreshold current by one decade is a function of the applied gate voltage, whereas the consistency in the sensitivity of the suspended device at two different ∆pH ranges is due to the elimination of the external scattering from the supporting substrate.

J o u r n a l P r e -p r o o f of the I D -V LG curve illustrated ( Supplementary Fig. 6 ). The overall process flow of the linkers binding, antibodies immobilization on suspended MoS 2 FET through macro storage fluidic channel to detect E. coli as shown in the schematic of Fig. 4 (a) (Gunda et al. 2014; Wu et al. 2013 ). The I D -V LG transfer curve of the linkers, antibodies, PBS buffer solution, and the 100 CFU/mL of E. coli bacteria was performed, as demonstrated in Fig. 4 (b) . It is essential to introduce a buffer solution again after antibodies immobilization to confirm there is no change of conductance (Cyan curve in Fig. 4 (b) ). A shift found in the I D -V LG curve on the left side after incubation of 100 CFU/mL of E. coli (Magenta curve in Fig. 4 (b) ).This illustrated the current in the MoS 2 channel deteriorated due to the increment of hole concentration induced on the dielectric surface by highly negative charged bacteria wall (Huang et al. 2011) .

After achieving considerable performance from the suspended MoS 2 FET device, different E. coli concentrations ranging from 0 to 10 3 CFU/mL were prepared and used for where S n(CFU/mL) sensitivity, Buffer is a buffer solution current, I n(CFU/mL) is current after bacteria bind to the FET biosensor. It was found that the shift in I D -V LG is more in the subthreshold region as compare to saturation and linear regions. The sensitivity of the sensor in the subthreshold The sensitivity presented in this report is found to be impressive as compare to the previous reports of biomolecule detection via 2D FET based technology. Graphene-based FET biosensor has shown the conductance change of 3.25% at 10 CFU/mL of E. coli, which is almost three times smaller than the presented value (9% at 10 CFU/mL) (Chang et al. 2013; Huang et al. 2011 ). The detection limits of the graphene are due to the zero band gap in nature, which leads to an increase in the off-state leakage current and the SS value (SS is inversely proportional to the sensitivity) (Schwierz 2010 ). On the other hand, FET based MoS 2 biosensor has shown promising sensitivity in biomolecule detection due to ~1.8 eV bandgap. However, the uneven morphology from the supporting substrate leads to an unreliable output from every single device due to different interface resistances caused majorly by trapped charges, chemical adsorbates, and unknown functional groups (Jin et al. 2013; Kaushik et al. 2018 ). This is a significant barrier in proliferating this technology in biosensor industries. Therefore, elimination of the scattering effect in the proposed biosensor shows the overall change of conductance in the 2D film is entirely by the change in concentration of the biomolecule. 

In summary, we provide an approach for a comprehensive investigation of the highly sensitive and reliable biosensor based on suspended MoS 2 FET to detect the pH as well as E. coli 

Three dimensional monolayer graphene foam for ultra-sensitive pH sensing

The future of biosensors

Ultrasonicassisted self-assembly of monolayer graphene oxide for rapid detection of

Dramatic switching behavior in suspended MoS2 field-effect transistors

Comprehensive identification of sensitive and stable ISFET sensing layer high-k gate based on ISFET/electrolyte models

Approaching ballistic transport in suspended graphene

Nanogaps with very large aspect ratios for electrical measurements

Subthreshold regime has the optimal sensitivity for nanowire

Optimization and characterization of biomolecule immobilization on silicon substrates using (3-aminopropyl) triethoxysilane (APTES) and glutaraldehyde linker 305

Nanomaterials for biosensing applications: a review 2

Graphene-based biosensors for detection of bacteria and their metabolic activities

Suspended single-layer MoS2 devices

Detection principles of biological and chemical FET sensors

Biosensors based on two-dimensional MoS2

Low-frequency noise in supported and suspended MoS 2 transistors

Directional dependent piezoelectric effect in CVD grown monolayer MoS2 for flexible piezoelectric nanogenerators

FET-based biosensors

Self-aligned immobilization of proteins utilizing PEG patterns

Recent advances in synthesis and biomedical applications of two-dimensional transition metal dichalcogenide nanosheets

Low-frequency noise in layered ReS2 field effect transistors on HFO2 and its application for pH sensing

Supported and Suspended 2D Material-Based FET

Biosensors and their applications-A review

Proposal for tunnel-field-effect-transistor as ultra-sensitive and label-free biosensors

MoS2 field-effect transistor for next-generation label-free biosensors

MoS2 field-effect transistor for next-generation label-free biosensors

Graphene transistors

Monolayer MoS2 and WSe2 double gate field effect transistor as super nernst pH sensor and nanobiosensor 11

Graphene-based materials for biosensors: a review

A novel description of ISFET sensitivity with the buffer capacity and double-layer capacitance as key parameters

Detection of Escherichia coli in drugs with antibody conjugated quantum dots as immunofluorescence probes

Optimization of pH sensing using silicon nanowire field effect transistors with HfO2 as the sensing surface

Etchant-free transfer of 2D nanostructures

Origin of spatial charge inhomogeneity in graphene

This article was supported in part by the National Science Foundation under grant no 1751472 and ACS Petroleum Research Fund (ACS PRF: 57647-DNI10)

The authors declare no competing financial interests. J o u r n a l P r e -p r o o f Highlights:1. A suspended MoS 2 atomic layer field-effect transistors is reported. 2. Self-assembled nanogap electrodes provide remarkable robust strength to the MoS 2 atomic layer.3. Suspended MoS 2 field-effect transistor device exhibits impressive pH sensitivity and Escherichia coli bacteria detection.4. The sensing mechanism highly selective to Escherichia coli bacteria.

☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.☐The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:J o u r n a l P r e -p r o o f