WO2015192064A1

WO2015192064A1 - Graphene-based nanosensor for identifying target analytes

Info

Publication number: WO2015192064A1
Application number: PCT/US2015/035640
Authority: WO
Inventors: Qiao Lin; Yibo Zhu
Original assignee: The Trustees Of Columbia University In The City Of New York
Priority date: 2014-06-12
Filing date: 2015-06-12
Publication date: 2015-12-17

Abstract

A receptor capable of binding to the target analyte can be used in monitoring a target analyte in a bodily fluid or a sample. A microdevice in accordance with the disclosed subject matter can include a substrate and a conductance elements with receptors grafted on the surface of the conductance element. A microdevice in accordance with the disclosed subject matter can also include a substrate, a first and second conductance elements, and synthetic polymers grafted on the surface of the first and second conductance elements. The first conductance element can be grafted with a sensing polymer that binds the target analyte, and the second conductance element can be grafted with a reference polymer that is insensitive to the target analyte. Differential measurement of the graphene conductance can allow determination of target analyte concentration in a bodily fluid.

Description

GRAPHENE-BASED NANOSENSOR FOR IDENTIFYING TARGET

ANALYTES

CROSS REFERENCE TO RELATED APPLICATIONS

This application claims priority from United States Provisional

Application No. 62/01 1 ,481 , filed June 12, 2014, Provisional Application No.

62/ 100,366, filed January 6, 2015, and Provisional Application No. 62/100,379, filed January 6, 2015 each of which is incorporated by reference herein in its entirety. STATEMENT REGARDING FEDERALLY FUNDED RESEARCH

This invention was made with government support under 1DP3

DK101085-01 awarded by the National Institutes of Health. The government has certain rights in this invention. BACKGROUND

Several hundred million people in the world have diabetes, making it a leading cause of death. In addition, complications induced by diabetes, such as heart disease, stroke, hypertension, blindness, kidney failure, and amputation, impact many others. Tight control of glycemia can reduce diabetes-related complications by 50% or more among Type I diabetics, with similar results for Type II diabetes patients. Thus, it is important to closely monitor abnormal blood sugar levels in diabetes patients so timely treatments (e.g., insulin injection, exercise, and diabetic diet, intake of carbohydrates) can be administered.

Certain conventional glucometers with sparsely discrete measurements, however, do not allow tight blood sugar control. In contrast, continuous glucose monitoring (CGM) can effectively detect hypo- and hyperglycemic (i.e., low and high blood sugar, respectively) events. CGM can achieve this by taking repetitive measurements of physiological glucose concentrations to enable close monitoring and timely correction of problematic blood sugar patterns of patients with diabetes mellitus. CGM can reduce the risk of diabetes-related complications, but certain

CGM devices are not adequate because of limited stability, insufficient accuracy, and slow time.

CGM can be achieved via minimally invasive or noninvasive methods, such as those which use subcutaneously implanted devices that determine glucose concentration in interstitial fluid (ISF) via measurement of electrochemical enzymatic reactions or equilibrium-based affinity binding. Electrochemistry, however, can involve irreversible consumption of glucose and depends on the rate at which glucose reaches the electrodes, which often makes electrochemical CGM sensors susceptible to influences of reactant supply rates, electroactive interferences, and biofouling

(deposition of biological material such as proteins and cell debris on sensor surfaces). As a result, certain sensors can exhibit drifts, have delays, require frequent calibration, and have limited accuracy (especially at low glucose concentrations). Moreover, such devices do not have adequate accuracy in the hypoglycemic realm (blood glucose concentration below 70 mg/dL), and also typically exhibit significant delays.

Nonreactive methods that use equilibrium affinity binding of glucose to a specific receptor do not involve irreversible consumption of glucose and can offer improved stability. Affinity glucose sensing can use concanavalin A (Con A), a glucose-binding protein, which unfortunately lacks stability and whose toxicity generates safety concerns.

Glucose sensors using stable, nontoxic glucose-binding can be limited by issues in accuracy, time response or miniaturization. Micro/nanoscale glucose affinity sensors, using microelectromechanical systems (MEMS), as well as using nanoscale materials such as carbon nanotubes, offer improvement in accuracy and stability, but can suffer from slow time response, rigid construction, and poor sensitivity. Moreover, they can be invasive by requiring subcutaneous implantation.

Noninvasive devices can attempt detection of interstitial fluid (ISF) glucose across the skin use optical spectroscopy or transdermal ISF sampling, which can be susceptible to variations in skin conditions. Others can use a measurement of glucose in urine, saliva, and tears. Unfortunately, glucose concentration in urine does not necessarily accurately reflect that in plasma, in particular in the hypoglycemic regime. Electrochemical methods to detect saliva glucose levels can suffer from the presences of residual food in saliva which can cause interferences.

Certain correlations have been found between tear and blood glucose concentrations, with deviations attributable to artifacts such as inconsistent tear collection methods. For example, the absorption and emission spectra of boronic acid containing fluorophores for glucose sensing can cumbersome and prevent CGM. Contact lens-based electrochemical sensors using glucose oxidase can be error-prone because of the irreversible nature of the reaction which consumes glucose during the measurement process.

There is a need to develop noninvasive glucose monitoring systems that offer improved long term accuracy and stability, biocompatibility, resistance to biofouling, resistance to environmental parameter fluctuations, easier calibration, as well as the capability of providing real-time report of a subject's glucose level via wireless telemetry.

SUMMARY

The disclosed subject matter provides a microdevice and techniques for monitoring a target analyte. In one aspect, the disclosed subject matter provides a microdevice and techniques for monitoring a target analyte in a sample using a receptor capable of binding to the target analyte. In one aspect, the microdevice does not require a receptor that binds the target analyte.

In one aspect, a microdevice includes a nanosensor, on a substrate platform, including a pair of conductance sensors functionalized with a receptor. One receptor (sensing receptor) binds specifically to the target analyte, and the other (reference receptor) is insensitive to the target analyte. Target analyte binding of the sensing receptor changes the charge density on the sensor surface, inducing changes in the carrier concentration of the sensor. Meanwhile, the reference receptor does not bind to the target analyte and its associated sensor conductance would change due to fluctuations in environmental parameters. Thus, differential measurement of the target analyte conductance allows determination of the target analyte concentration in, for example, a sample or bodily fluid. In certain embodiments, the receptors bind reversibly with essentially all analytes. In certain embodiments, the sensing receptor binds reversibly with the target analyte. In certain embodiments, the receptor is a synthetic polymer. In certain embodiments, the sensor includes graphene.

In one aspect, a microdevice includes a nanosensor, on a substrate platform, including a single conductance sensor functionalized with synthetic receptors (sensing receptors), which bind specifically to the target analyte, on a substrate platform. Target analyte binding of the sensing receptor changes the charge density on the sensor surface, inducing changes in the earner concentration of the sensor allowing for the determination of the target analyte concentration in, for example, a sample or bodily fluid. In certain embodiments, the sensing receptor binds reversibly with the target analyte. In certain embodiments, the receptor is a synthetic polymer. In certain embodiments, the sensor includes graphene.

The disclosed subject matter also provides a microdevice for monitoring a target analyte in a bodily fluid using a polymer capable of binding to the target analyte. In certain embodiments, the microdevice includes a substrate platform and a nanosensor including a first conductance element functionalized with a sensing polymer for detecting the target analyte and a second conductance element functionalized with a reference polymer that is insensitive to the target analyte. Detecting a difference, if any, in the conductance of the first and second conductance elements can be used to determine the presence and/or concentration of the target analyte in the sample.

In certain embodiments, the microdevice includes a graphene nanosensor on a contact lens platform. A noninvasive contact lens-based nanosensor, which is miniaturized and mechanically flexible, can include graphene functionalized with a glucose-binding polymer and enable noninvasive CGM. By differential measurement of glucose binding-induced changes in graphene conductance as compared to measurements from a sensor insensitive to glucose, the device can allow specific, sensitive, and rapid detection of glucose concentration in tear fluid. In certain embodiments, the nanosensor consists of a pair of graphene conductance sensors on a flexible, contact lens-based substrate respectively functionalized with synthetic polymers. In certain embodiments, synthetic glucose-specific polymers and reference polymers can be grafted on graphene surfaces for differential

measurements. In certain embodiments, the polymer includes a PAAPBA based polymer. In certain embodiments, the polymer includes a plurality of boronic acid moieties.

In certain embodiments, the nanosensor can be coated with a biocompatible glucose-permeable hydrogel thin layer. Glucose binding of the sensing polymer changes the charge density on the graphene surface, inducing changes in the carrier concentration within the bulk of the atomically thin graphene and hence in the graphene electric conductance.

In certain embodiments, the microdevice can be adapted to be disposed on or coupled to a contact lens-based substrate, a dermal patch, an eye patch, a tattoo, jewelry, a watch, bandages, clothing, or a wireless body sensor. In certain embodiments, the bodily fluid is tears, blood, saliva, mucus, interstitial fluid, spinal fluid, intestinal fluid, amniotic fluid, lymphatic fluid, pericardial fluid, peritoneal fluid, pleural fluid, semen, vaginal secretions, sweat, or synovial fluid of a subject.

BRIEF DESCRIPTIONS OF THE DRAWINGS

Figure 1 : A contact lens-based graphene nanosensor (100) for noninvasive CG in tears according to some embodiments of the disclosed subject matter.

Figure 2: Design of a graphene based nanosensor according to some embodiments of the disclosed subject matter: (a) top and (b) cross-sectional views. The device consists glucose sensing (103) and reference modules ( 104), each including an FET sensing element with graphene (105) as the conducting channel and h-BN as the dielectric (106). ITO serves as the gate (below graphene) (107) as well as source (108) and drain (109) (on either side of graphene). The graphene (105) is grafted with a specific glucose-binding (1 1 1) or reference polymer (1 12) monolayer. Glucose concentration in tears is determined by differential measurement of the graphene conductance. The device is based on a flexible substrate (101 ) and coated in a biocompatible hydrogel (not shown).

Figure 3: A micro fabricated differential affinity glucose sensor (a) was implanted in a mouse (b), with its capacitance output (reflecting ISF glucose concentration) tracking blood glucose concentration measured with a glucometer (c).

Figure 4: A method for CVD graphene (205) synthesis and transfer procedure according to some embodiments of the disclosed subject matter, (a) CVD graphene synthesis in quartz tubing furnace, (b) CVD graphene (205) transfer onto the substrate (217).

Figure 5: Characterization of a graphene sheet according to some embodiments of the disclosed subject matter, (a) AFM micrograph, (b) Height profile. (c) Raman spectra (532 nm laser excitation).

Figure 6: Fabrication of planar electrodes according to some embodiments of the disclosed subject matter.

Figure 7: Fabrication of a PDMS microchannel according to some embodiments of the disclosed subject matter. Figure 8: Schematic of a graphene-based FET nanosensor (502) according to some embodiments of the disclosed subject matter. Graphene (505) serves as the conducting channel, while a 20-nm-thick Hf0₂ layer (506) between the graphene (505) and the substrate- supported gate electrode (507) serves as the dielectric layer.

Figure 9: Fabrication process of a graphene-based FET nanosensor (502) according to some embodiments of the disclosed subject matter, (a) Deposition and patterning of 5/45 nm Cr/Au gate electrode (507), (b) deposition of 20 nm Hf02 dielectric layer using ALD (506), (c) fabrication of drain (509) and source (508) electrodes using lift-off, and (d) transfer of graphene (508).

Figure 10: Micrographs of graphene nanosensors according to some embodiments of the disclosed subject matter: (a) Multiple devices batch-fabricated on the same substrate (left) and close-up view of a single device (right), (b) Detailed view of the source, drain and gate electrodes. Dashed box approximately indicates the region covered by graphene. (c) Raman spectrum of the graphene. (d) AFM measurements of the graphene thickness. Inset: AFM photo of graphene whose thickness was measured along the dash line.

Figure 1 1 : A plot showing: (a) Transfer characteristic for graphene in air. The ambipolar curve was observed with V_Sa_, DP of 0.7 V. (b) The

transconductance estimated at different pH levels. The value is approximately constant (23 μ8).

Figure 12: A plot showing the dependence of the nanosensor characteristics on pH. (a) Transfer characteristic curves obtained at varying pH values. The VSG, DP shifts linearly to higher gate voltages with increasing pH (57.6 mV/pH). (b) Dependence of the Dirac point voltage on pH. The solid line represents a linear fit.

Figure 13 : A plot showing the results of Figure 12 on a separate device, (a) Transfer characteristic curves obtained at varying pH values, (b)

Dependence of the Dirac point voltage on pH. The V_Sa_, DP shifts linearly to higher gate voltages with increasing pH (58.2 mV/pH). The solid line represents a linear fit.

Figure 14: A plot showing dependence of V_SG_,DP on V_LG with Eo chosen to be 0. The slope was approximately 1. Figure 1 5: A plot showing real time measurements of pH: the source- drain current Ins varied consistently and reversibly with pH at a fixed gate voltage {V_BG = Q.15 V).

Figure 16: A plot showing a polymer according to some embodiments of the disclosed subject matter: (a) Chemical structure of pyrene-PAAPBA. (b)

Fluorescence intensity of polymer solution at maximum emission wavelength vs. time of incubation with pyrene substrate, indicating full attachment of pyrene-PAAPBA to graphene in 2 h. (c) Fluorescence spectrum of pyrene-PAAPBA attached to graphene.

Figure 17: Synthesis of pyrene-terminated sensing polymers for graphene attachment (top) and structures of monomers (bottom) according to some embodiments of the disclosed subject matter.

Figure 18: Schematic of the affinity glucose sensor (602) configured as a solution top-gated graphene field effect transistor according to some embodiments of the disclosed subject matter. The Ag/AgCl electrode (607) is inserted into the electrolyte solution (621 ) in contact with graphene (605) to serve as the gate electrode, while the EDL formed at the solution-graphene interface provides the gate dielectric.

Figure 19: Fabrication of a nanosensor according to some embodiments of the disclosed subject matter, (a) Patterning of drain (609) and source (608) electrodes, (b) Transfer of graphene (605) on to an oxide-coated silicon substrate, (c) Bonding of the PDMS microchannel to the sensor chip.

Figure 20: A nanosensor according to some embodiments of the disclosed subject matter, (a) Optical micrograph of the graphene covering the source and drain electrodes b) Measurement setup, (c) Raman spectrum of the graphene. The G and 2D bands are indicative of the graphene consisting of a single layer of carbon atoms.

Figure 21 : Raman spectrum of the graphene according to some embodiments of the disclosed subject matter. The G and 2D bands are indicative of the graphene consisting of a single layer of carbon atoms.

Figure 22: AFM images of graphene (a) before and (b) after functionalization with the PAPBA polymer. There was a significant increase (~10 nm) in the apparent height of the graphene sheet after the functionalization, suggesting the presence of the polymer molecules. Figure 23: Transfer characteristics measured before (dashed blue line) and after (red line) functionalization of graphene with the PAPBA polymer. The left shift of the Dirac point indicates that the graphene was n-doped due to the attachment of the polymer molecules.

Figure 24: Transfer characteristics in different glucose solutions at varying glucose concentrations. In response to increases in the glucose concentration, the Dirac point position Vgs,Dirac shifted to higher gate voltages and the

transconductance decreased from 100 to 20 μΞ.

Figure 25: Control experiments using pristine graphene without functionalization of the polymer. The change in the Dirac point position and transconductance is insignificant compared to Fig. 22, implying that the changes in earner mobility and density in that figure was caused by the glucose-polymer binding.

Figure 26: A nanosensor according to some embodiments of the disclosed subject matter, (a) micrograph, (b) pH-induced changes in the source-drain current Ids (at a fixed gate voltage of 0.75 V) of a solid-gated graphene FET sensor.

Figure 27: A nanosensor design as used in Example 1 .

DETAILED DESCRIPTION

The disclosed subject matter provides for devices and techniques to monitor target analytes. More specifically, the disclosed subject matter provides for field-effect transistor (FET)-based sensors and systems that can be used for continuous analyte monitoring, including but not limited to continuous glucose monitoring (CGM).

As used herein, the term "analyte" is a broad term and is used in its ordinary sense and includes, without limitation, any chemical species the presence or concentration of which is sought in material sample by the sensors and systems disclosed herein. For example, the analyte(s) include, but not are limited to, glucose, ethanol, insulin, water, carbon dioxide, blood oxygen, cholesterol, bilirubin, ketones, fatty acids, lipoproteins, albumin, urea, creatinine, white blood cells, red blood cells, hemoglobin, oxygenated hemoglobin, carboxyhemoglobin, organic molecules, inorganic molecules, phannaceuticals, cytochrome, various proteins and

chromophores, microcalcifications, ions, electrolytes, sodium, potassium, chloride, bicarbonate, and hormones. In one embodiment, the analyte is glucose. In various embodiments, the analytes can be other metabolites, such as lactate, fatty acids, cysteines and homocysteines.

As used herein, the term "functionalized^" means to have a capability of being reactive to an analyte. For example, functionalized refers to a substrate that has a substance attached to it, wherein the substance has a functional group that is capable of reacting with an analyte. For example, the substance can be covalently attached or grafted to the surface of the functionalized substrate.

In certain embodiments, the disclosed subject matter provides a microdevice as described herein coupled with a wireless interface. The output signal is typically a raw data stream that is used to provide a useful value of the measured target analyte concentration. The wireless interface can include a capacitance digital converter coupled with the microdevice and adapted to produce a digital signal representing a measurement of the target analyte in the bodily fluid of the subject; a microcontroller coupled with the capacitance digital converter; and a transponder coupled with the microcontroller to transmit the digital signal received from the capacitance digital converter to an external reader.

The Nanosensor

The disclosed subject matter provides a nanosensor for monitoring a target analyte. In certain embodiments, the nanosensor utilizes a pair of conductance sensors on a substrate platform, wherein one of the pair of sensors is functionalized with receptors for binding the target analyte and the other sensor is functionalized with receptors that are insensitive to the target analyte. In certain embodiments, the nanosensor utilizes a single conductance sensor on a substrate platform, wherein the sensor is functionalized with receptors for binding the target analyte. In certain embodiments, the receptors are natural polymers, synthetic polymers, peptides, antibodies, aptamers, or small molecules. In certain embodiments, the nanosensor does not require the sensor to be functionalized with a receptor, wherein the target analyte itself changes the conductance of the sensor.

The nanosensor can be fabricated of biocompatible materials to prevent adverse responses of the surrounding tissue. These include the substrate and passivation materials (e.g., PET and parylene), the sensing and reference receptors (e.g., natural polymers, synthetic polymers, peptides, antibodies, aptamers, or small molecules), and the glucose-permeable hydrogel coating. In addition, the functional, dielectric and metallization materials (e.g., graphene, hexagonal boron nitride (h-BN) and gold) are also biocompatible when their uptake by individual cells is avoided. Thus, inflammatory, allergic, immunogenic, cytotoxic, and genotoxic responses of tissue to the microdevice is minimized.

In certain embodiments, the sensor is made of graphene. Graphene is a flat monolayer of carbon atoms tightly packed into a two-dimensional honeycomb lattice. In certain embodiments, the FET sensing element with graphene as the conducting channel has an electric resistance of about 0.1 kQ - about 3 kQ. In certain embodiments, graphene as the conductance channel has an electric resistance of about 0.1 kQ - about 3 kQ, about 0.25 kQ - about 2.75 kQ, about 0.5 kQ - about 2.5 kQ, about 0.75 kQ - about 2.25 kQ, about 1 kQ - about 2 kQ, about 1.25 kQ - about 1.75 kQ, about 1.5 kQ - about 2 kQ, or about 2 kQ - about 3 kQ. In certain embodiments, graphene as the conductance channel has an electric resistance of at least about 0.1 kQ, at least about 0.2 kQ, at least about 0.3 kQ, at least about 0.4 kQ, at least about 0.5 kQ, at least about 0.6 kQ, at least about 0.7 kQ, at least about 0.8 kQ, at least about 0.9 kQ, at least about 1 kQ, at least about 1 .2 kQ, at least about 1.4 kQ, at least about 1.6 kQ, at least about 1.8 kQ, at least about 2 kQ, at least about 2.2 kQ, at least about 2.4 kQ, at least about 2.6 kQ, at least about 2.8 kQ, at least about 3 kQ, at least about 4 kQ, at least about 5 kQ, at least about 6 kQ, at least about 7 kQ, at least about 8 kQ, at least about 9 kQ,or at least about 10 kQ. In certain

embodiments, graphene as the conductance channel has an electric resistance of no more than about 0.1 kQ, no more than about 0.2 kQ, no more than about 0.3 kQ, no more than about 0.4 kQ, no more than about 0.5 kQ, no more than about 0.6 kQ, no more than about 0.7 kQ, no more than about 0.8 kQ, no more than about 0.9 kQ, no more than about 1 kQ, no more than about 1.2 kQ, no more than about 1.4 kQ, no more than about 1.6 kQ, no more than about 1.8 kQ, no more than about 2 kQ, no more than about 2.2 kQ, no more than about 2.4 kQ, no more than about 2.6 kQ, no more than about 2.8 kQ, no more than about 3 kQ, no more than about 4 kQ, no more than about 5 kQ, no more than about 6 kQ, no more than about 7 kQ, no more than about 8 kQ, no more than about 9 kQ, or no more than about 10 kQ.

The graphene monoatomic sheet has the ultimate thinness (0.34 nm) of any known material and possesses unparalleled mechanical strength (Young's modulus: 1 TPa), adheres strongly to underlying substrates, and is highly flexible, optically transparent, and chemically stable, thereby holding the potential to enable new, transformative methods for detection of biological analytes. The monoatomic structure as well as exceptional electric conductivity (-1738 S/m) and charge carrier mobility (2 10^s cm2/Vs) of graphene can be exploited to enable highly sensitive analyte detection. Holding the potential to enable new, distinctly transformative methods for detection of biological analytes, graphene has been used for

electrochemical or affinity detection of analytes such as DNA and proteins, and for electrochemical detection of glucose.

In certain embodiments, the graphene sensor includes a single layer sheet. In certain embodiments, the graphene sensor includes a multilayered sheet. In certain embodiments, the graphene sensor includes at least one layer of graphene. In certain embodiments, the graphene sensor includes at least two layers of graphene, at least three layers of graphene, or at least four layers of graphene. In certain embodiments, the graphene sheet is formed by mechanical exfoliation, chemical exfoliation, chemical vapor deposition, or silicon carbide. In certain embodiments, the graphene sheet is formed by chemical vapor deposition (CVD). In certain embodiments, the graphene sheet is formed by, mechanical exfoliation, which can include the removal of a layer of graphene from a block of graphite using tape or other sticky substance. Exemplary techniques for fabrication of the graphene sheet is illustrated in Figures 4 and discussed in further detail in Example 2.

The nanosensors as disclosed herein are highly sensitive, as changes in surface charge due to the presence of the target analyte near the sensor or the target analyte binding effectively penetrates the bulk of the atomically thin graphene, leading to a detectable signal even at low target analyte concentrations.

In using graphene as a functional material on a flexible substrate, the nanosensors disclosed herein can be miniaturized and mechanically flexible for placement on a substrate platform. In certain embodiments, the substrate platform is rigid. In certain embodiments, the substrate platform is flexible. The monoatomic thickness allows the graphene nanosensor to be highly miniaturized. Miniaturization drastically reduces distances over which diffusive analyte (e.g. glucose) transport occurs, leading to a rapid response to target analyte changes in a bodily fluid. In certain embodiments, the change in target analyte levels are detected within about 7 to about 60 seconds. In certain embodiments, the change in target analyte levels are detected within about 10 to about 50 seconds, about 12 to about 45 seconds, about 15 to about 40 seconds, about 17 to about 35 seconds, or about 20 to about 30 seconds. In certain embodiments, the change in target analyte levels are detected within about 5 seconds, about 6 seconds, about 7 seconds, about 8 seconds, about 9 seconds, about 10 seconds, about 1 1 seconds, about 12 seconds, about 13 seconds, about 14 seconds, about 15 seconds, about 16 seconds, about 17 seconds, about 18 seconds, about 19 seconds, about 20 seconds, about 21 seconds, about 22 seconds, about 23 seconds, about 24 seconds, about 25 seconds, about 26 seconds, about 27 seconds, about 28 seconds, about 29 seconds, about 30 seconds, about 31 seconds, about 32 seconds, about 33 seconds, about 34 seconds, about 35 seconds, about 36 seconds, about 37 seconds, about 38 seconds, about 39 seconds, about 40 seconds, about 41 seconds, about 42 seconds, about 43 seconds, about 44 seconds, about 45 seconds, about 46 seconds, about 47 seconds, about 48 seconds, about 49 seconds, about 50 seconds, about 51 seconds, about 52 seconds, about 53 seconds, about 54 seconds, about 55 seconds, about 56 seconds, about 57 seconds, about 58 seconds, about 59 seconds, or about 60 seconds, or about 1 minute, 2 minutes, 3 minutes, 4 minutes, or 5 minutes. In certain embodiments, the device response time is within about 1 to about 10 minutes, about 2 to about 9 minutes, about 3 to about 8 minutes, about 4 to about 7 minutes, or about 5 to about 6 minutes. In certain embodiments, the response times of the device are at least 1 minute, at least about 2 minutes, at least about 3 minutes, at least about 4 minutes, at least about 5 minutes, at least about 6 minutes, at least about 7 minutes, at least about 8 minutes, at least about 9 minutes, or at least about 10 minutes. In certain embodiments, the response time is between about 1 .5 to about 2.5 minutes, which is at least 2-3 times as rapid as existing CGM devices.

The microdevices as disclosed herein can also be mechanically flexible because of the use of flexible materials, including graphene on a flexible substrate. Thus, the device can readily conform to the local tissue geometry, and minimize irritation and injury to a tissue or organ (e.g., an eye) during the sensor placement, operation and replacement.

The miniature size of the device also prevents the obstruction of vision when integrated on a contact lens. The intimate contact also facilitates exchange of glucose between the device and tears, leading to improved sensitivity and reliability. Therefore, because of the miniaturization and flexibility, the device is highly compatible with contact lens-based noninvasive CGM.

To reduce protein adsorption and biofouling, in certain embodiments the entire nanosensor can be coated in a biocompatible hydrogel. In certain embodiments, the nanosensor is coated with a thin hydrogel layer. In certain embodiments, the hydro gel is permeable to the target analyte. In certain embodiments, the hydrogel is permeable to glucose. In certain embodiments, the biocompatible hydrogen is synthesized in situ. Exemplary techniques for fabrication of the hydrogel is discussed in further detail in Example 10. Hydrogels can be made by any other commonly understood method.

In certain embodiments, the hydrogel can include, but is not limited to, at least one of polyQiydroxyethyl -methacrylate (PHEMA), hydroxyethyl methacrylate (HEMA), tetraethyleneglycol diacrylate (TEGDA), polyethyleneglycol methacylate (PEGMA), or N-[tris(hydroxymethyl)methyl]-acrylamide (HMMA). In certain embodiments, the hydrogel includes hydroxyethyl methacrylate (HEMA),

tetraethyleneglycol diacrylate (TEGDA), polyethyleneglycol methacylate (PEGMA) and N-[tris(hydroxymethyl)methyl]-acrylamide (HMMA). In certain embodiments, the hydrogel includes a combination of hydroxyethyl methacrylate (HEMA), tetraethyleneglycol diacrylate (TEGDA), polyethyleneglycol methacylate (PEGMA) and N-[tris(hydroxymethyl)methyl]-acrylamide (HMMA). In certain embodiments, the hydrogel includes hydroxyethyl methacrylate (HEMA), tetraethyleneglycol diacrylate (TEGDA), polyethyleneglycol methacylate (PEGMA) and N- [tris(hydroxymethyl)methyl]-acrylamide (HMMA) at a ratio of 86:3:5.5 mol%.

In certain embodiments, the thickness of the hydrogel layer ranges from about 20-50 μηι. In certain embodiments the hydrogel layer is no thicker than about 5 μιη, about 6 μηι, about 7 μηι, about 8 μηι, about 9 μιη, about 10 μηι, about 1 1 μηι, about 12 μηι, about 13 μηι, about 14 μιη, about 15 μιη, about 16 μηι, about 17 μιη, about 18 μιη, about 19 μιη, about 20 μηι, about 2^m, about 22 μηι, about 23 μηι, about 24 μηι, about 25 μιη, about 26 μηι, about 27 μηι, about 28 μηι, about 29 μιη, about 30 μηι, about 31 μιη, about 32 μιη, about 33 μιη, about 34 μηι, about 35 μιτι, about 36 μιη, about 37 μηι, about 38 μιη, about 39 μιη, about 40 μιη, about 41 μιη, about 42 μιη, about 43 μηι, about 44 μιη, about 45 μιη, about 46 μιη, about 47 μηι, about 48 μιη, about 49 μιτι, or about 50 μιη, In certain embodiments, the thickness of the hydrolayer is between about 20 μιη to about 50 μηι.

In certain embodiments, the nanosensor is constructed on a substrate platform. In certain embodiments, the substrate is fabricated on the final substrate (e.g., a contact lens). In certain embodiments, the substrate is fabricated on a thin flexible film and the nanosensor device (including the thin flexible film) is attached to the final substrate platform (e.g., a contact lens). In certain embodiments, the substrate platform and the thin layer film are made from the same material. In certain embodiments, the substrate platform and thin later film are made from different material. In certain embodiments, the nanosensor is fonned on a substrate platform that is then formed/molded (e.g. , by heating) into a contact lens shape.

In certain embodiments, the nanosensor is on the inside the contact lens (i.e., towards the eye). In certain embodiments, the nanosensor is on the outside of the contact lens (i.e., towards the eyelid).

In accordance with certain embodiment, the nanofilms can be transferred in a vacuum to prevent air bubbles from being trapped between the graphene, dielectric layer, and the substrate. In certain embodiments, the nanofilm surfaces can also be treated by heat or low-power oxygen plasma to further improve adhesion.

The design parameters can be varied without departing from the scope of the disclosed subject matter. Such design parameters include, for example, the number and dimensions of atomic layers and substrate platform layers, shape and dimensions of the graphene, molecular weight of the sensing receptor, and the thickness of the hydrogel coating.

Single Channel

A set of sensing receptors functionalized on the conductance sensor binds to the target analyte. In certain embodiments, the sensing receptor binds specifically to the target analyte. In certain embodiments, the sensing receptor binds reversibly or irreversibly to the target analyte. In certain embodiments, the sensing receptor binds reversibly to the target analyte. In certain embodiments, the reference receptor binds reversibly to essentially all analytes. In certain embodiments, the sensing receptor binds to more than one target analyte.

In certain embodiments, target analyte binding of the sensing receptor changes the charge density on the sensor surface, inducing changes in the carrier concentration of the sensor.

In certain embodiments, the sensing polymer binds specifically to the target analyte. Target analyte binding of the sensing polymer changes the charge density on the sensor surface, inducing changes in the carrier concentration of the sensor. In certain embodiments, the polymers bind reversibly with essentially all analytes. In certain embodiments, the sensing polymer binds reversibly with the target analyte. In certain embodiments, the nanosensor is configured as a solution- gated graph ene-based FET (GFET) in that the graphene (e.g., 605) is the conducting channel, which is formed between two electrodes (source (e.g. , 608) and drain (e.g. , 609)) on an insulating substrate surface (e.g. , 601 ) (Fig. 18) The substrate or platform surface can be one or multi-layered. For example, the substrate or platform surface can consist of two layers (see Fig. 1 8, 601 ) such as, but not limited to a silicon wafer based device. An example of a silicon wafer based device, the lower layer can be silicon as the substrate while the upper later is silicone oxide which can serve as an insulating layer. In certain embodiments, the substrate platform is a single layer. In certain embodiments the single later substrate platform is a polymer substrate.

A monolayer of the synthetic glucose responsive polymer (e.g. , 61 1 ) can be attached to the graphene surface via π-π stacking interactions (Fig. 17). A sample (e.g., 621) can be in contact with the polymer-functionalized graphene (e.g. , 61 1 , 605) in a microchannel, with an electrode wire (e.g., 607) inserted into the solution to serve as a gate electrode. An electrical double layer (EDL) can form at the interface of the graphene and solution, and serves as the gate dielectric layer. During operation, under the control of a voltage applied between the gate and source electrodes (gate voltage Vgs), a bias voltage applied between the drain (e.g., 609) and source (e.g., 608) electrodes (drain-source voltage Vds) generates a current through the graphene (e.g., 605) (drain-source current /^). In certain embodiments, this yields the transfer characteristics of the GFET (i.e., the functional dependence of Ids on Vgs) and allows the determination of the glucose concentration, because the binding of boronic acid moieties of the polymer PAPBA changes the electrical properties of the graphene as follows. In certain embodiments, the wire probe is replaced with a gate electrode nanolayer (Figs. 8 and 9).

In certain embodiments, the nanosensor is not functionalized with a receptor. The charge of the target analyte changes the charge density on the sensor surface, inducing changes in the carrier concentration of the sensor. In certain embodiments, such a sensor can be used to test a sample or bodily fluid. In certain embodiments, such a sensor can be used to test the pH or electrolyte concentration of a sample or bodily fluid. In certain embodiments, such a sensor can be used to test for the presence or absence of an analyze carrying a charge or capable of eliciting a charge by any means. For example, but not limited to, the analyte can interact with a substance present near the sensor that would allow a charge or change in a charge to occur.

In certain embodiments, the nanosensor can be configured as a solid- gated FET device, in which a graphene sheet, serving as the conducting channel, connected the source and drain electrodes on a dielectric layer_, which in turn lies above the gate electrode on the substrate (Figs. 8 and 9).

In certain embodiments, when the nanosensor is not functionalized with receptors, the presence of ions near the sensor are detected. The magnitude of the electric potential depends on the ion concentration (e.g., H⁺). In certain embodiments, the pH level can be determined by measuring the graphene' s electric properties such as its transfer characteristics and conductance, which is directly related to the carrier concentration. To allow a high level of integration while avoiding the need for high gate voltages, the nanosensor can use about 20 nm thick HfOi, a material with a high dielectric constant (κ ~ 20, compared to κ ~ 3.9 for Si0₂), as the dielectric layer to provide a high gate capacitance (~1 μΡ/cm²). This in general allows the Dirac point, at which the drain-source current IDS achieves its minimum, to be observed at a lower gate voltage.

In certain embodiments, the electrode wire can be, for example, but not limited to Ag/AgCl, Ag, Pt, or combinations thereof.

In certain embodiments, the dielectric nanolayer can be made from material such as, but not limited to hexagonal boron nitride (h-BN), Hf0_2, parylene, Si0₂, Si₃N_4i or combinations thereof.

In certain embodiments, the gate electrode can be made from material such as, but not limited to ITO, Ti/Pd/Pt, gold, copper, chrome, or mixtures thereof.

In certain embodiments, the source and drain electrodes, are separately made from material such as, but not limited to ITO, Ti/Pd/Pt, chromium, gold, chrome, or combinations thereof.

In certain embodiments, the polymer substrate and or thin layer film can be made from material such as, but not limited to polyethylene terephthalate (PET), polycarbonate polystyrene, polym ethyl methacrylate (PMMA), polymacon, silicones, fluoropolymers, silicone acrylate, fluoro-silicone/acrylate, poly

hydroxyethyl methacrylate, or combinations thereof.

In certain embodiments, the polymer coating can be made from material such as, but not limited to, parylene, polyimide, organic polymer, hydrophobic polymer, or combinations thereof. In certain embodiments, the nanosensor is covered with a polymer coating except for the functionalized part of the graphene sheet.

In certain embodiments, the sensing polymer can be, for example, but not limited poly(N-hydroxyethylacrylamide-ran-3-acrylamidophenylboronic acid) (PHEA-ran-PAAPBA) and pyrene-terminated poly(3-acrylamidophenylboronic acid) (py-PAPBA). In certain embodiments, the copolymer can be liner or branched.

Additional receptor polymers are discussed below.

In certain embodiments, the dimensions of the graphene conducting channel in the single channel graphene sensor can be a length of about 10 - about 20 μηι by a width of about 10 - about 20 μιη by a thickness of about 1 - about 10 nm. In certain embodiments, the graphene channel in the sensor can be about 20 x 10 μιη . In certain embodiments, the graphene conducting channel can be functionalized with a polymer.

In certain embodiments, the length of the graphene channel in the sensor can be about 10.2 - about 19.8 μηι, about 10.4 - about 19.6 μιτι, about 10.6 - about 19.4 μηι, about 10.8 - about 19.2 μιη, about 1 1 - about 19 μηι, about 1 1.2 - about 18.8 μιη, about 1 1.4 - about 18.6 μηι, about 1 1.6 - about 18.4 μηι, about 1 1.8 - about 18.2 μηι, about 12 - about 18 μηι, about 12.2 - about 17.8 μιη, about 12.4 - about 17.6 μιη, about 12.6 - about 17.4 μιη, about 12.8 - about 17.2 μηι, about 13 - about 17 um, about 13.2 - about 16.8 urn, about 13.4 - about 16.6 μιη, about 13.6 - about 16.4 μιη, about 13.8 - about 16.2 μηι, about 14 - about 16 μηι, about 14.2 - about 15.8 μιη, about 14.4 - about 15.6 μιη, about 14.6 - about 15.4 μιη, or about 14.8

- about 15.2 μηι. In certain embodiments, the width of the graphene channel in the sensor can be about 10.2 - about 19.8 μιη, about 10.4 - about 19.6 urn, about 10.6 - about 19.4 μηι, about 10.8 - about 19.2 μηι, about 1 1 - about 19 μηι, about 1 1.2 - about 18.8 μιη, about 1 1.4 - about 18.6 μηι, about 1 1.6 - about 18.4 μηι, about 1 1.8 - about 18.2 μιη, about 12 - about 18 μηι, about 12.2 - about 17.8 μηι, about 12.4 - about 17.6 μη , about 12.6 - about 17.4 μηι, about 12.8 - about 17.2 μιη, about 13 - about 17 μιη, about 13.2 - about 16.8 μηι, about 13.4 - about 16.6 μηι, about 13.6 - about 16.4 μηι, about 13.8 - about 16.2 μηι, about 14 - about 16 μιη, about 14.2 - about 15.8 μηι, about 14.4 - about 15.6 μιη, about 14.6 - about 15.4 μιη, or about 14.8

- about 15.2 μηι. In certain embodiments, the thickness of the graphene channel in the sensor can be about 1 .2 - about 9.8 nm, about 1.4 - about 9.6 nm, about 1.6 - about 9.4 nm, about 1 .8 - about 9.2 nm, about 2 - about 9 mn, about 2.2 - about 8.8 nm, about 2.4 - about 8.6 nm, about 2.6 - about 8.4 nm, about 2.8 - about 8.2 nm, about 3 - about 8 nm, about 3.2 - about 7.8 nm, about 3.4 - about 7.6 nm, about 3.6 - about 7.4 nm, about 3.8 - about 7.2 nm, about 4 - about 7 nm, about 4.2 - about 6.8 nm, about 4.4 - about 6.6 nm, about 4.6 - about 6.4 mn, about 4.8 - about 6.2 nm, about 5 - about 6 nm, about 4.2 - about 5.8 nm, about 4.4 - about 5.6 nm, about 4.6 - about 5.4 nm, or about 4.8 - about 5.2 nm.

In certain embodiments, the dimensions of the single channel nanosensor chip can be a length of about 5- about 10 mm by a width of about 5- about 10 mm by a thickness of about 50 - about 500 μηι. In certain embodiments the graphene sensor is about 25 x 10 μιη² In certain embodiments, the total nanosensor can be about 10 x 10 mm².

In certain embodiments, the length of the single channel nanosensor chip can be about 5.2 - about 9.8 mm, about 5.4 - about 9.6 mm, about 5.6 - about 9.4 mm, about 5.8 - about 9.2 mm, about 6 - about 9 mm, about 6.2 - about 8.8 mm, about 6.4 - about 8.6 mm, about 6.6 - about 8.4 mm, about 6.8 - about 8.2 mm, about 7 - about 8 mm, about 7.2 - about 7.8 mm, or about 7.4 - about 7.6 mm. In certain embodiments, the width of the single channel nanosensor chip can be about 5.2 - about 9.8 mm, about 5.4 - about 9.6 mm, about 5.6 - about 9.4 mm, about 5.8 - about 9.2 mm, about 6 - about 9 mm, about 6.2 - about 8.8 mm, about 6.4 - about 8.6 mm, about 6.6 - about 8.4 mm, about 6.8 - about 8.2 mm, about 7 - about 8 mm, about 7.2 - about 7.8 mm, or about 7.4 - about 7.6 mm. In certain embodiments, the thickness of the single channel nanosensor chip is about 75 - about 475 μηι, about 100 - about 450 μηι, about 125 - about 425 μιη, about 150 - about 400 μιη, about 175 - about 375 μηι, about 200 - about 350 μηι, about 225 - about 325 μηι, or about 250 - about 300 μηα.

Exemplary techniques for fabrication of the devices illustrated in, for example, Figures 4-10, 17-20, and 22 will be discussed in further detail in Examples 2-8.

Multi-Channel

In certain embodiments, one set of receptors (sensing receptors) lunctionalized on one of the pair of conductance sensors binds to the target analyte and the other set of receptors (reference receptors) functionalized on the a separate conductance sensor is insensitive to the target analyte. In certain embodiments, the sensing receptor binds specifically to the target analyte. In certain embodiments, the sensing receptor binds reversibly or irreversibly to the target analyte. In certain embodiments, the sensing receptor only binds reversibly to the target analyte. In certain embodiments, the reference receptor binds reversibly to essentially all analytes. In certain embodiments, the first and second conductive elements are essentially identical except for the first conductive element being functionalized with the sensing receptor and the second conductance element being functionalized with a reference receptor. In certain embodiments, binding of the target analyte to the sensing receptor induces changes in the electrical conductance of the graphene of the first conductive element and conductance of the second conductive element only changes due to fluctuations in environmental parameters. In certain embodiments, the sensing receptor binds to more than one target analyte.

In certain embodiments, at least two channels are present in the nanosensor. In certain embodiments, at least three channels, at least four channels, or at least five channels are present in the nanosensor. In certain embodiments, four channels are present. In certain embodiments, at least one of the multichannel sensors are functionalized with a reference receptor (e.g., polymer). In certain embodiments, at least two of the sensors are functionalized with sensing receptors (e.g., polymers). In certain embodiments, at least two of the sensors are functionalized with two different sensing receptors (e.g., polymers).

In certain embodiments, target analyte binding of the sensing receptor changes the charge density on the sensor surface, inducing changes in the carrier concentration of the sensor. Meanwhile, the reference receptor does not bind to the target analyte and its associated sensor conductance would change only due to fluctuations in environmental parameters. Thus, differential measurement of the target analyte conductance allows determination of the target analyte concentration in a bodily fluid.

In certain embodiments, one polymer (sensing polymer) binds specifically to the target analyte, and the other (reference polymer) is insensitive to the target analyte. Target analyte binding of the sensing polymer changes the charge density on the sensor surface, inducing changes in the carrier concentration of the sensor. Meanwhile, the reference polymer does not bind to the target analyte and its associated sensor conductance would change only due to fluctuations in

environmental parameters. Thus, differential measurement of the target analyte conductance allows determination of the target analyte concentration in a bodily fluid. In certain embodiments, the polymers bind reversibly with most analytes. In certain embodiments, the polymers bind reversibly with essentially all analytes. In certain embodiments, the sensing polymer binds reversibly with the target analyte.

Figures 1 and 2 illustrate the structure of an example two channel microdevice according to some embodiments of the disclosed subject matter. As shown in these figures, the microdevice e.g., 100 includes a substrate platform 101 (e.g., Si02, a contact lens or a thin film or layer that can be attached to the inside of the contact lens) and a two channel (103, 104) nanosensor 102 coupled to the substrate platform. The two channel nanosensor 102 design consists of two polymer- functionalized field-effect transistor (FET) modules of identical construction (except for their respective sensing/reference polymers) 103, 104 on the substrate platform 101. In each module a slender graphene strip 105 (the conducting channel) lies on a dielectric layer 106 (e.g., h-BN or parylene) (which minimizes substrate-induced charge carrier scattering in the conducting channel) passivating a transparent gate electrode 107 (e.g., ITO), and makes contact with source 108 (e.g., ITO) and drain electrodes 109 (e.g.. ITO). These nanolayers lie on a substrate platform 101 , and are covered with a thin polymer layer 1 10 (e.g., parylene) except the graphene, which is grafted with a monolayer of a glucose-binding polymer (the sensing module) 1 1 1 or a glucose-insensitive polymer (the reference module) 1 12. The entire nanosensor can be coated with a glucose-permeable hydrogel (not shown). In certain embodiments, the nanolayers and polymer layers are flexible and biocompatible In certain

embodiments, the polymer layer 1 10 can be a passivating layer. In certain

embodiments, an optional passivation layer 122 can be present. In certain

embodiments, a passivation layer 122 can be present if the substrate platform 101 is conductive (e.g.. S1O2). In certain embodiments, the passivation layer is not present if the substrate platform 101 is electrically insulated (e.g., PET). This passivation layer can be optionally found in either the single or multi-layer nanosensors.

In certain embodiments, the dimensions of the graphene conducting channel in the dual channel graphene sensor can be a length of about 1 0 - about 20 μι^η by a width of about 1 0 - about 20 μηι by a thickness of about 1 - about 10 nm. In certain embodiments, the graphene channel in the sensor can be about 20 x 10 μιτι²

In certain embodiments, the length of the graphene chamiel in the sensor can be about 10.2 - about 19.8 μιτι, about 1 0.4 - about 19.6 μηι, about 10.6 - about 1 9.4 μηι, about 10.8 - about 19.2 μη , about 1 1 - about 19 μιη, about 1 1 .2 - about 1 8.8 μτη, about 1 1.4 - about 18.6 μηι, about 1 1.6 - about 18.4 μηι, about 1 1.8 - about 18.2 μηι, about 12 - about 18 μηι, about 12.2 - about 17.8 μηι, about 12.4 - about 17.6 μιτι, about 12.6 - about 17.4 μιη, about 12.8 - about 17.2 μηι, about 13 - about 17 μηι, about 13.2 - about 16.8 μιτι, about 13.4 - about 16.6 μηι, about 13.6 - about 16.4 μηι, about 13.8 - about 16.2 μηι, about 14 - about 16 μηι, about 14.2 - about 15.8 μηι, about 14.4 - about 15.6 μιη, about 14.6 - about 15.4 μιτι, or about 14.8

- about 15.2 μιη. In certain embodiments, the width of the graphene channel in the sensor can be about 10.2 - about 19.8 μιη, about 10.4 - about 19.6 μηι, about 10.6 - about 19.4 μηι, about 10.8 - about 19.2 μηι, about 1 1 - about 19 μηι, about 1 1.2 - about 18.8 μηι, about 1 1.4 - about 18.6 μιτι, about 1 1.6 - about 18.4 μηι, about 1 1.8 - about 18.2 μπι, about 12 - about 18 μηι, about 12.2 - about 17.8 μηι. about 12.4 - about 17.6 μηι, about 12.6 - about 17.4 μηι, about 12.8 - about 17.2 μτη, about 13 - about 17 μηι, about 13.2 - about 16.8 μηι, about 13.4 - about 16.6 μηι, about 13.6 - about 16.4 μιη, about 13.8 - about 16.2 μηι, about 14 - about 16 μηι, about 14.2 - about 15.8 μιτι, about 14.4 - about 15.6 μηι, about 14.6 - about 15.4 μηι, or about 14.8

- about 15.2 μη . In certain embodiments, the thickness of the graphene channel in the sensor can be about 1.2 - about 9.8 nm, about 1.4 - about 9.6 nm, about 1.6 - about 9.4 nm, about 1.8 - about 9.2 nm, about 2 - about 9 nm, about 2.2 - about 8.8 nm, about 2.4 - about 8.6 nm, about 2.6 - about 8.4 nm, about 2.8 - about 8.2 nm, about 3 - about 8 nm, about 3.2 - about 7.8 nm, about 3.4 - about 7.6 nm, about 3.6 - about 7.4 nm, about 3.8 - about 7.2 nm, about 4 - about 7 nm, about 4.2 - about 6.8 nm, about 4.4 - about 6.6 nm, about 4.6 - about 6.4 nm, about 4.8 - about 6.2 nm, about 5 - about 6 nm, about 4.2 - about 5.8 nm, about 4.4 - about 5.6 nm, about 4.6 - about 5.4 nm, or about 4.8 - about 5.2 nm.

In certain embodiments, the dimensions of the dual channel nanosensor chip can be a length of about 8 - about 15 mm by a width of about 8 - about 15 mm by a thickness of about 50 - about 500 μηι. In certain embodiments, the total nanosensor chip can be about 15 x 15 mm².

In certain embodiments, the length of the dual channel nanosensor chip can about 8.2 - about 14.8 mm, about 8.4 - about 14.6 mm, about 8.6 - about 14.4, about 8.8 - about 14.2 mm, about 9 - about 14 mm, about 9.2 - about 13.8 mm, about 9.4 - about 13.6 mm, about 9.6 - about 13.4, about 9.8 - about 13.2 mm, about 10 - about 13 mm, about 10.2 - about 12.8 mm, about 10.4 - about 12.6 mm, about 10.6 - about 12.4, about 10.8 - about 12.2 mm, about 1 1 - about 12 mm, about 1 1.2 - about 1 1.8 mm, or about 1 1 .4 - about 1 1 .6 mm. In certain embodiments, the width of the dual channel nanosensor chip can about 8.2 - about 14.8 mm, about 8.4 - about 14.6 mm, about 8.6 - about 14.4, about 8.8 - about 14.2 mm, about 9 - about 14 mm, about 9.2 - about 13.8 mm, about 9.4 - about 13.6 mm, about 9.6 - about 13.4, about 9.8 - about 13.2 mm, about 10 - about 13 mm, about 10.2 - about 12.8 mm, about 10.4 - about 12.6 mm, about 10.6 - about 12.4, about 10.8 - about 12.2 mm, about 1 1 - about 12 mm, about 1 1 .2 - about 1 1 .8 mm, or about 1 1.4 - about 1 1.6 mm. In certain embodiments, the thickness of the dual channel nanosensor chip is about 75 - about 475 μιη, about 1 00 - about 450 μηι, about 125 - about 425 μηι, about 150 - about 400 μηι, about 175 - about 375 μιη, about 200 - about 350 μηι, about 225 - about 325 μι^η, or about 250 - about 300 μιη.

The slender graphene strips (e.g., 105) lie on a dielectric nanolayer (e.g., 106). In certain embodiments, the graphene strips are two separate strips. The dielectric nanolayer can passivate a gate electrode (e.g., 107) and make contact with a source (e.g., 108) and drain electrode (e.g., 109). In certain embodiments, these nanolayers lie on a polymer substrate/substrate platform (e.g., 101 ) and are covered with a thin polymer layer (e.g., 1 10) with the exception of the conducting-channel graphene (e.g., 103, 104), which is functionalized with a target specific receptor (e.g., glucose-binding polymer (the sensing module)) (e.g., 1 1 1 ) or a target insensitive receptor (e.g., glucose-insensitive polymer (the reference module)) (e.g., 1 12). In certain embodiments, nanolayers and polymer layers are all flexible and

biocompatible.

In certain embodiments, the dielectric nanolayer can be made from material such as, but not limited to hexagonal boron nitride (h-BN), parylene, Si0₂ and Si₃N₄, or combinations thereof.

In certain embodiments, the substrate platform (e.g., polymer substrate) and/or thin layer film can be made from material such as, but not limited to

polyethylene terephthalate (PET), polycarbonate polystyrene, polymethyl methacrylate (PMMA), polymacon, silicones, fluoropolymers, silicone acrylate, fluoro-silicone/acrylate, poly hydroxyethyl methacrylate, or combinations thereof.

Additional receptor polymers are discussed below.

During operation, the conducting-channel graphene can be subjected to a gate voltage (e.g., between ±300 mV) with respect to the gate electrode underneath, and the current through the graphene under a bias voltage (10 - 50 mV) between the source and drain, which can be kept sufficiently small to limit the potential leakage current into the hydrogel coating well below the threshold (-10 uA/mm2) required to prevent undesirable electrochemical effects in tissue. In certain embodiments, the bias voltage is between about 10 to about 50 mV. In certain embodiments, the bias voltage is between about 10 to about 50 mV, about 12 to about 48 mV, about 14 to about 46 mV, about 16 to about 44 mV, about 18 to about 42 mV, about 20 to about 40 mV, about 22 to about 38 mV, about 24 to about 36 mV, about 26 to about 34 mV, or about 28 to about 32 mV.

Exemplary techniques for fabrication of the devices illustrated in, for example, Figures 1 and 2 will be discussed in further detail in Examples 8, 9, and 1 1.

Receptors

In various embodiments of the disclosed subject matter, the sensor can be used to determine the level of a target analyte in the body, for example oxygen, lactase, insulin, hormones, cholesterol, medicaments, viruses, or the like. In certain embodiments, the receptors that identify the target analyte are natural polymers, synthetic polymers, peptides, antibodies, aptamers, or small molecules.

In certain embodiments, the receptor is a polymer. The sensing polymer includes one in which at least one monomer or moiety interacts with the target analyte. Interaction between the target analyte and the sensing polymer results in a change in conductance of the sensor. In certain embodiments, the reference polymer includes a polymer whose moieties or monomers do not interact with the target analyte. In certain embodiments, the polymer consists of at least two

monomers, at least three monomers, at least four monomers, at least five monomers, at least six monomers, at least seven monomers, at least eight monomers, at least nine monomer, or at least ten monomers.

In certain embodiments, the target analyte is glucose. When the target analyte is glucose, through proper adjustment of the composition percentage of the boronic acid moieties on the polymer and polymer concentrations, the polymer can detect and differentiate glucose from other monosaccharides and disaccharides.

Applying this polymer to the nanosensor as disclosed herein can enable highly reliable, continuous monitoring of glucose in bodily fluids.

As noted, the binding between the polymer and the analyte of interest can be reversible. For example, the binding and dissociation between the target analyte and the sensing polymer can be an equilibrium phenomenon driven by the concentration of the analyte in the conducting channel. The amount of the analyte bound with the sensing polymer depends on the concentration of the analyte in the conducting channel.

In one embodiment, a suitable polymer having boronic acid moieties can be formed as a copolymer of at least two monomers, where one of the monomers includes at least one boronic acid functional group. A copolymer can be synthesized with these monomers via classic free radical copolymerization processes. In certain embodiments, a suitable polymer includes, but is not limited to, a polymer that contains boronic acid groups, or other receptor groups that recognize the given analytes.. In one embodiment, the polymer is phenyl acrylamide (PAM). In another embodiment, the polymer is PAM-ra»-PAAPBA, which is an amphiphilic copolymer containing two components, hydrophilic polymer segment phenyl acrylamide (PAM) and hydrophobic polymer segment poly(3-acrylamidophenylboronic acid) (PAAPBA). In yet another embodiment, the polymer is PAM-rao-PPAM, which is an amphiphilic copolymer containing two components, hydrophilic polymer segment phenyl acrylamide (PAM) and hydrophobic polymer segment polymeric allylamine (PPAM). In certain embodiments, basic monomers, such as, but not limited to, N,N- dimethylacrylamide can be used to increase the ionization of boronate upon binding with diols.

In certain embodiments, the copolymer can be liner or branched. In certain embodiments, the copolymer includes hydrophilic motifs (e.g., polymeric allylamine (PAM), hydroxyethyl methacrylate (ΉΕΜΑ)). In certain embodiments, the copolymers can be in situ gelation pre-polymers. In certain embodiments, in situ gelation of polymers is one in which a solution of polymers can form a hydrogel on the sensor chip by methods of heating and UV irradiation.

In certain embodiments the polymer is conjugated to a substance that can immobilize the polymers on the graphene sensor. In certain embodiments, the substance-terminated polymers can be irreversibly attached to graphene with strength comparable to covalent attachment. In certain embodiments, the substance is pyrene. A pyrene-terminated polymer can be irreversibly attached to graphene using a sticky point for μ-μ stacking interactions without disrupting the graphene^" s conjugation or altering its electronic properties, can be synthesized.

The polymer sensors can undergo a viscosity change as well as a permittivity change when interacting with glucose molecules, as discussed in US Patent Application Publication No. 20120043203, assigned to the common assignee, the disclosure of which is incorporated herein by reference in its entirety.

The sensing polymer contains boronic acid groups, which bind to glucose molecules at a 2: 1 ratio to reversibly form cyclic esters of boronic acid, while having almost no response to other potential interferents, such as fructose, galactose, and sucrose. Below is a representative schematic of the reaction:

Boronic Acid Glucose Boronate

In certain embodiments, the polymer contains as least one boronic acid group, at least two boronic acid group, at least three boronic acid group, at least four boronic acid group, at least five boronic acid group, at least six boronic acid group, at least seven boronic acid group, at least eight boronic acid group, at least nine boronic acid group, at least 10 boronic acid group. The process causes the overall ionization equilibrium to shift from neutral/insoluble boronic acid moieties to anionic/hydrophilic boronate, thereby changing the surface charge density. This results in a change in the carrier concentration within the bulk of the atomically thin graphene in the sensing module, and hence a change in the graphene conductance. In certain embodiments, the reference polymer is glucose-insensitive, the graphene conductance in the reference module would change only due to fluctuations in environmental parameters. Thus, differential measurement of the graphene conductance allows determination of tear glucose concentration.

To estimate the nanosensor performance, Langmuir's adsorption isotherm relates the glucose concentration in ISF (c) to the surface density of boronic acid-bound glucose molecules. It can then be shown that the sensitivity of the graphene conductance (G) is approximately:

(dG I G) I dc * y_nm,K_d I [2N(K, + c)² ], (1 )

where N is the graphene' s charge carrier density, y_max the surface density of immobilized boronic acid groups, and Kd the equilibrium dissociation constant of the binding system. Thus, for typical material and binding properties (TV ~10¹⁵ irf², ymax ~ 10⁵ μιιη^"2 and Ki ~ 10 mM), over a desired glucose concentration range of 0.5-30 mg/dL (which encompasses the hypoglycemic, normoglycemic and hyperglycemic regimes), the nanosensor sensitivity - 0.5%-0.4%/(mg/dL). For an estimated conductance measurement resolution of 0.1 %, glucose concentration can be measured at a resolution of -0.7-0.02%, which allows accurate affinity detection of glucose in tears.

In certain embodiments, in order to screen out effects not caused by the target analyte, for example, environmental factors such as temperature or other analytes, the reference chamber includes a graphene sensor that is functionalized with another polymer (the reference polymer). The reference polymer does not bind with the target analyte. Also, the reference polymer should not bind with or otherwise react with any other substance in the bodily fluid to impact the property of the reference solution in a similar way as the target analyte impacts the corresponding property in the sensing polymer. The reference polymer and sensing polymer should, however, respond similarly to non-target analytes and environmental conditions. The reference polymer can be selected to have similar hydrophilic blocks to those in the sensing polymer, but have no phenylboronic acid moieties. For example, glucose- unresponsive PAA or PHEAA can be used as a reference polymer for glucose detection. The charge and viscosity of PAA (or PHEAA) polymers is glucose- independent. The analyte-free charge of the sensing polymer can be similar to that of the reference polymer.

In certain embodiments, the polymers are not immobilized on the graphene sensors. In these embodiments, the proximity of the change in the polymer charge is sensed by the graphene sensor. In certain embodiments, the polymers are sequestered near the sensors by the presence of a hydrogel.

In accordance with another embodiment, alternative, covalent attachment methods, e.g., by using residue hydroxyl on graphene to tailor the polymer design can be used. In accordance with a further embodiment, a more basic monomer, such as N.TV-dimethylacrylamide, can be used to prepare the polymer, which is known to enable the ionization of boronate upon binding with diols under physiologic conditions.

Target Analyte

In various embodiments of the disclosed subject matter, the sensor can be used to determine the level of a target analyte in the body, for example oxygen, lactase, insulin, hormones, cholesterol, medicaments, viruses, ions, or the like. The sensor can use any known method to provide an output signal indicative of the concentration of the target analyte. The output signal is typically a raw data stream that is used to provide a useful value of the measured target analyte concentration. In certain embodiments, before the device is used to detect or monitor a target analyte, it can be first calibrated using samples containing known amount of the target analyte to obtain correlations between sensor response (e.g., capacitance readout) and the known concentration of the calibration sample. Thereafter, in the monitor of the target analyte, the pre-established correlations can be used to interpret the output signals of the sensor and determine the presence and/or concentration of the target analyte in a test sample.

In certain embodiments, the nanosensor as disclosed herein can enable highly reliable monitoring of a target analyte in a sample. In certain embodiments, the sample is a bodily fluid, a non-bodily fluid liquid, or a laboratory sample. In certain embodiments, the nanosensor can be used to detect the change in the pH of a sample. In certain embodiments the nanosensor can be used to continuously monitor the change in pH of a sample. In certain embodiments, the nanosensor can be used to measure the amount or change in the amount of a target analyte in a sample. For example, the nanosensor can be part of a stand alone device that monitors the target analyte in a sample added to the device (e.g., a piece of lab equipment or home monitor).

In certain embodiments, the nanosensor as disclosed herein can enable highly reliable, continuous monitoring of glucose in bodily fluids. In certain embodiments, the bodily fluid is tears, blood, saliva, mucus, interstitial fluid, spinal fluid, intestinal fluid, amniotic fluid, lymphatic fluid, pericardial fluid, peritoneal fluid, pleural fluid, semen, vaginal secretions, sweat, and synovial fluid of the subject.

In certain embodiments, tear fluid is used to measure target analyte concentrations. Tear fluid is easily accessible, and the eye is an ideal candidate for the placement of a microdevice (e.g., a contract lens with a nanosensor coupled to it) to monitor analytes. In certain embodiments, tear fluid can be used to test, for example, lactate levels or glucose levels.

In certain embodiments of the disclosed subject matter, the nanosensor is used to monitor glucose as the target analyte. In these embodiments, the

nanosensor can measure a concentration of glucose or a substance indicative of the concentration or presence of the glucose by using a specific receptor (e.g., polymer etc..) in the nanosensor. In certain embodiments, the present microdevice includes a combination of graphene, a novel biosensing nanomaterial, and a contact lens platform to enable noninvasive CGM in tear fluid.

Glucose concentration in tears has been found to range from 1 -6.2 mg/dL in healthy individuals and up to 26 mg/dL in diabetic persons. Significant correlations have been found between tear and blood glucose concentrations, with deviations attributable to artifacts such as inconsistent tear collection methods. Tear fluid includes basal tear, which keeps cornea wet and nourished, and reflex tear, which is induced by irritation of the eye due to foreign particles or imtant substances. No significant differences in glucose concentration have been found in basal and reflex tears.

A contact lens-based graphene affinity nanosensor (Fig. 1 ) offers several potential advantages. For example, binding of glucose with an affinity polymer monolayer immobilized on graphene can cause a change in surface charge density, which can penetrate the atomically thin graphene to significantly change the graphene's conductance, leading to a significant detectable signal even at low, hypoglycemic glucose concentrations. Accurate detection of hypoglycemia is of great importance to diabetes care because this condition can cause acute, severe

complications to patients, but is challenging in tear where hypoglycemic

concentrations are 10s of times lower than those in blood. In certain embodiments, accurate detection of small amounts of glucose is achieved by the high sensitivity of graphene in combination with differential detection via two small graphene sensing elements placed in close proximity to achieve effective common-mode cancellation of noise and environmental disturbances.

The nanosensor can include the glucose-binding polymer with boronic acid groups (e.g., polymer poly(acrylamide-rarc-3-acrylamidophenyl boronic acid) (PAA-rarc-3PAAPBA)) that can bind with glucose at 2: 1 ratio to form cyclic esters. The resulting changes in physical (e.g., viscometric and dielectric) properties of the polymer can be measured with microfabricated devices, enabling specific

measurement of glucose concentrations. The nanosensor can thus be highly sensitive and accurate at clinically important glucose concentrations in tears, in particular in the low, hypoglycemic regime (below 1 mg/dL). In certain embodiments, the glucose concentrations can be measured at a resolution of about 2 μg/dL - about 3 μg/dL. In certain embodiments, the glucose concentrations can be measured with the disclosed microdevice at a resolution of about 1 μ /dL - about 10 μg/dL, 2 μg/dL - about 9 μg/dL, 3 μg/dL - about 8 μg/dL, 4 μg/dL - about 7 μg/dL, or 5 μg dL - about 6 μ /dL. In certain embodiments, the glucose concentrations can be measured with the disclosed microdevice at a resolution of at least about 1 μg/dL, at least about 2 μg dL, at least about 3 μg/dL, at least about 4 μg/dL, at least about 5 μg dL, at least about 6 μg/dL, at least about 7 μg/dL, at least about 8 μg dL, at least about 9 μg/dL, at least about 10 μ /dL, at least about 12.5 ^ig/dL, at least about 15 μg/dL, at least about 17.5 μg dL, or at least about 20 μg/dL.

The nanosensor can also be used for other applications. In addition to diabetes, the proposed CGM microdevice can also be used for glucose monitoring for other diseases (e.g., glycogen storage disease and hyperinsulinaemic hypoglycaemia).

The method can be extended to other metabolites, such as lactate, fatty acids, cysteines and homocysteines. For example, in emergency medicine, lactate monitoring can be used to predict possible organ failure of trauma patients, organ transplant patients, and patients with other critical conditions.

Further, the methods disclosed herein can be used as a reliable method for long-term monitoring of metabolites. Such methods can have great military significance. For example, a miniature device for glucose detection with fully electronic readout would have significant applications in protecting armed forces in the field. It can also provide a platform to enable the delivery of drug treatments and nutritional supplements to protect and enhance performance in military personnel.

Moreover, the disclosed method can be applied to the diagnosis of disease. For example, the development of boronic acid based glucose sensing systems can be extended to other analytes, such as human viruses and bacteria, as many of those microorganisms carry glycoproteins on the exterior surface that can be targeted by the boronic acid based binding motifs.

Additionally, the disclosed method can be applied to a noninvasive method for monitoring cancer treatment. For example, receptors can be designed to that bind to target microparticles associated with cell apoptosis. The receptors can be specific for apoptosis of the cancer cells. In certain embodiments, the microdevice can be used to track overall cell apoptosis to monitor when levels of cell death before they become too toxic. In certain embodiments, the microdevice is able to detect the level of cancer cells in the bodily fluid.

Furthermore, the methods disclosed herein can be used as a reliable method for determining the drug distribution of a treatment regimen in order to maximize the therapeutic effects of the drug. In such methods, the receptors of the device can designed to bind to the drug found in the bodily fluid. For example, ocular drug delivery is a major challenge to pharmacologists due to its unique anatomy and physiology (e.g., different layers of cornea, sclera, and retina including blood aqueous and blood-retinal barriers, choroidal and conjunctival blood flow, lymphatic clearance, and tear dilution). The microdevice is a useful tool to determine what level of the drug reaches the ocular surface.

Metabolic monitoring is of great utility to environmental monitoring. Changes in the concentrations of metabolites are the precursors and products of enzymatic activity, and can be associated with biological function and regulation. Metabolic monitoring hence can be used for environmental monitoring, e.g., risk assessment of chemicals and diagnosis of diseases in wild animals. It can also be used as a tool to better understand the underlying mechanisms of action of toxic compounds in the environment. Additional aspects and embodiments of the disclosed subject matter are illustrated in the following examples, which are provided for better understanding of the disclosed subject matter and not limitation. Example 1. In vivo Testing of a Microdevice

Microdevices with affinity polymers such as poly(acrylamide-ran-3- acrylamidophenyl boronic acid) (PAA-ran-3PAAPBA) were tested in mice. The polymers were constructed with phenylboronic acid groups that bind with glucose at 2: 1 ratio to form cyclic esters. The resulting changes in physical (e.g., viscometric and dielectric) properties of the polymer were measured with a microelectromechanical systems (MEMS) device (Fig. 3a), enabling specific measurement of glucose concentrations. Miniaturization led to device response times of (1.5-2.5 min) at least 2-3 times as rapid as existing CGM devices. Using a differential design, the devices were highly stable, with drifts reduced from ~0.3%/hr (non-differential measurement) to less than 50 ppm/hr. Testing the device in mice via subcutaneous implantation (Fig. 3b) yielded output (reflecting interstitial fluid (ISF) glucose concentration) that consistently tracked blood glucose concentration (Fig. 3c) and was clinically accurate or acceptable by Clarke error grid analysis.

The nanosensor used in this experiment is depicted in Figure 27.

Example 2. Construction of CVD Graphene

CVD graphene (205) was synthesized by heating annealed Cu foil (213) in a quartz tubing furnace (Fig. 4a). The Cu foil (213) was first sharply heated to 1000 °C in Argon (Ar) enviromnent (200 mTorr), and annealed in hydrogen (H2) environment (10 mTorr). The mixture of methane (CH₄) and H₂ were then introduced and allowed to react for 1 8 min (CH₄: 170 mTorr, H₂: 10 mTorr), after which the sample was cooled down to room temperature in Ar flow at200 mTorr and then retrieved from the tube. After graphene growth, 500nm PMMA (214) was spin coated on top for protection and a PDMS stamp (21 6) and glass slide (215) was attached by pressing. Cu (213) was removed by wet etching.

Graphene (205) was transferred onto the substrate (217) at 170 °C (Fig. 4b) to realize the graphene conducting channel stretching over drain/source electrodes. After the protective PMMA layer (216) on graphene surface was dissolved by acetone, AFM (XE- 100, Park System) and Raman spectroscopy (Renishaw, 532 nm laser) were used to verify the single-layer graphene sheet (Fig. 5).

Example 3. Construction of Sensor Substrate Fabrication

Cr/Au (5 nm/45 nm) layers (307) were deposited on the top of Si/Si02 substrate (BOC/Auto 306 thermal evaporator, Edwards) (318) (with Photoresist (S I 81 1 , Shipley) was then spin-coated on top (319)) and patterned to create the source/drain electrodes using photolithography (MA6 Mask Aligner, arl-Siiss) and wet-chemical etches (Fig. 6). Cu layer was directly deposited on the back side of substrate as the backgate electrode.

Example 4. Construction of MicroChannel Fabrication

Microfluidic channel was fabricated using standard soft lithography (Fig. 7). SU-8 photoresist (419) was spin-coated on Si wafer (418) and patterned by photolithography3. Mixture of PDMS precursor and curing agent (420) was poured onto the SU-8 mold (419). The curing reaction was earned at 75°C for 1 h. Then, cured PDMS microchannel was peeled off from the mold.

Example 5. A solid dielectric gated graphene nanosensor in electrolyte solutions

Illustrated herein is a GFET sensor that can allow accurate detection of a target analyte (e.g.,H⁺)-

Graphene has been used to form a conducting channel in field effect transistors (FET), allowing highly sensitive electric detection of analytes. Such graphene FET (GFET) sensors, when operating in liquid media, are generally constructed in a solution-gated or solid-gated configuration. In a solution-gated GFET sensor, a reference electrode is inserted into the electrolyte solution that is in contact with graphene to serve as the gate electrode, while the electric double layer (EDL) formed at the solution-graphene interface plays the role of the gate dielectric. Theses solution-gated sensors typically require an external electrode inserted into the electrolyte solution, which hinders the integration and miniaturization of the device. In addition, the gate capacitance, or the capacitance across the EDL dielectric layer is susceptible to disturbances in liquid media, which can result in fluctuations in electrical measurements of properties of graphene including the position of the Dirac point. In contrast, in a typical solid-gated GFET, the gate capacitance is provided by a Si(¾ dielectric layer sandwiched between graphene and the underlying silicon substrate, which serves as the gate electrode. By eliminating the need for the external wire insertion into the electrolyte solution, solid-gated sensors can be highly miniaturized and integrated. However, due to the intrinsically low capacitance of the Si0₂ layer, usually the solid-gated GFET sensors require undesirably high gate voltages (40-50 V), consequently impeding their application to biosensing in liquid media.

This example presents a GFET nanosensor in liquid media using a thin layer of Hf0₂ with a high dielectric constant (κ) as a gate dielectric layer (506). The HfO^"2 layer is sandwiched between the conducting-channel graphene (505) and a gate electrode (507) (Fig. 8) and is embedded within the sensor. This enabled a high level of integration in the construction and passivation of electrically conducting elements in the sensor, as is highly desirable for analyte detection in liquid media. The use of the high-¾: dielectric material (HfO?) provides two orders of magnitude higher specific capacitance than conventional Si0₂ solid-gated sensors, thereby rendering high transconductance and allowing the device to operate at low gate voltages. In addition, the gate dielectric was isolated from the liquid media, thus eliminating errors caused by disturbances (e.g., bulk motion of sample solution). Furthermore, the sensor was amenable to time- and cost-effective microfabrication using photolithography without the need for manual assembly of discrete components (e.g., electrodes) with graphene, thereby simplifying the fabrication process. pH sensing was demonstrated using this high-K GFET nanosensor. Experimental results show that the device is capable of measuring pH in a range of 5.3 to 9.3 with a sensitivity of -57.6 mV/pH, and at a gate voltage of less than 1.5 V, which is approximately a factor of 30 lower than that used in Si0₂ solid-gated sensors.

Nanosensor Design

The nanosensor was configured as a solid-gated FET device, in which a graphene sheet (505), serving as the conducting channel, connected the source (508) and drain electrodes (509) on a Hf0₂ dielectric layer (506)_, which in turn lies above the gate electrode (507) on the substrate (Fig. 8). When a buffer solution was introduced onto the graphene surface, the carrier concentration in the bulk of the graphene undergoes a change due to variations in the electric potential in the buffer next to the graphene. As the magnitude of the electric potential depends on the ion concentration (e.g., H¹ ), the pH level can be determined by measuring the graphene's electric properties such as its transfer characteristics and conductance, which is directly related to the carrier concentration. To allow a high level of integration while avoiding the need for high gate voltages, the nanosensor used 20 nm thick Hf0₂, a material with a high dielectric constant (κ ~ 20, compared to κ ~ 3.9 for Si0₂), as the dielectric layer to provide a high gate capacitance (~1 μ /cm²). This in general allows the Dirac point, at which the drain-source current IDS achieves its minimum, to be observed at a lower gate voltage.

Nanosensor Fabrication

The nanosensor (Figs. 8- 1 l a) was fabricated on a Si0₂-coated silicon substrate (518) by first depositing and patterning the gate electrodes (Cr/Au 5/45 nm) (507)(Fig. 9a). Subsequently, a 20 nm Hf0₂ layer (506) was deposited over the wafer using atomic layer deposition (ALD) (Fig. 9b). A lift-off process was used to create the drain (509) and source (508) electrodes, onto which a single-layer graphene sheet (505) synthesized by chemical vapor deposition (CVD) was transferred (Figure 10b). A microchamber (2.5 μί), fabricated in a polydimethylsiloxane (PDMS) sheet via soft lithography, was placed on the resulting nanosensor chip to confine sample liquid on the device.

In particular, the nanosensor was fabricated on a Si0₂-coated silicon substrate (51 8, 519). After cleaning by piranha, 5/45 nm Cr/Au was deposited using thermal evaporation (BOC 306 Thermal Evaporator, Edward) (Fig. 9a). Photoresist (S I 81 1 , Shipley) was then spin-coated on top of Au at 5000 rpm for 1 min, and baked at 1 15°C for 1 min. Photolithography (MA6, Suss MicroTec) was then used to pattern the shape of the gate electrode on the wafer. The wafer was then developed in developer (AZ MIF 300, AZ Electronic Materials) and local wet etched in gold and chrome etchant subsequently (Fig. 9a). The wafer was cleaned with piranha solution followed by oxygen plasma. Next, a 20 nm HfO? layer (506) was deposited on top of the gate electrode (507) using atomic layer deposition (ALD, Savannah 200, Cambridge Nano Tech) at 3.6x 10^"1 torr and the temperature as high as 200°C (Fig. 9b) (Fig. 9b). Another layer of photoresist (S I 81 1 , Shipley) was spin-coated and patterned to define the shape of source and drain electrodes, followed by deposition of 5/45 nm Cr/Au. Lastly, the wafer was immersed in photoresist stripper (AZ MIF 400 Stripper) and acetone sequentially to dissolve the photoresist and shape the drain and source electrodes (Fig. 9c). After the completion of the fabrication process, a single- layer graphene sheet (505) synthesized by chemical vapor deposition (CVD) was subsequently transferred onto the sensor to cover the source (508), drain (509) and gate electrodes (507) (Fig. 9d). A Raman spectrum was taken to confirm the monolayer graphene sheet throughout the conducting channel. A polydimethylsiloxane (PDMS)-based microchamber was used to confine the liquid sample on top of the graphene.

Results

Raman spectroscopy and atomic force microscopy (AFM) was first used to confirm that the graphene used to fabricate the device consisted of a single atomic layer. The Raman spectrum revealed G mode and 2D mode, which are characteristic of single-layer graphene (Figure 10c). AFM data were used to determine the thickness of the graphene to be 0.3-0.4 nm (Figure l Od), which reflected the van der Waals diameter of carbon and was, hence, also indicative of single-layer graphene.

The transfer characteristics of bare graphene in air was determined. I_DS was measured while the solid-gate voltage V_Sc, which makes a major contribution to the overall gate voltage, was varied sinusoidally from -0.2 to 1.9 V. An ambipolar curve was observed; the Dirac point solid-gate voltage, or the gate voltage value at which the I_Ds achieves the minimum (Fig. 1 l a), was hence determined to be V_SG. _DP = 0.7 V. This confirmed that the conductivity of the graphene was being altered by the field effect.

The nanosensor was tested for pH sensing in liquid media. Samples at various pH values (5.3 to 9.3) were prepared by mixing NaOH or HC1 with phosphate buffered saline (PBS) buffer (Life Technologies, ionic strength -150 mM). A sample solution was incubated with our nanosensor, during which I_DS values were measured while the gate voltage V_B was swept from 0.6 V to 1.6 V. V_SG.DP < 1.5 V at all pH levels (Fig. 12). These significantly reduced gate voltage values, compared to 40-50 V for Si0₂ based solid-gated sensor, can be attributed to the high gate capacitance and hence the high transconductance provided by the high-A; Hf0₂ dielectric layer. V_SG._D was found to linearly increase with the pH value at a sensitivity of -57.6 mV/pH (Fig. 12a), which would otherwise not be attainable by a conventional dielectric (Si0₂)- based device operating at similarly low gate voltages.

It was also determined that these measurements were reproducible, with a different device of the same design yielding a closely agreeing sensitivity of 58.2 mV/pH (Fig. 13). When pH increased, the electrostatic potential above graphene increased due to the decrease of H⁺; therefore, the curve shifted to the right to compensate for the increase in the electrostatic potential. The leakage current between drain/source and gate electrodes was found to be much smaller than I_DS and therefore negligible.

There are two possible physical processes that have been used to explain how adsorption of ions on graphene causes variations in the conductivity. The first process involves the charging of the EDL capacitor by adsorbed ions, thereby causing variations in the potential in the solution in contact with the graphene, and therefore changing the Fermi level and carrier density of graphene, e.g., the electric field tuning. In the second process, which is known as the surface charge transfer doping, adsorbed ions serve as dopants, from which electrons are exchanged into the bulk of the graphene. The solid-gated sensor, which avoids the influence of the externally applied top gate voltage on the EDL, can be used to investigate the effect of either the EDL capacitor charging or the surface charge transfer doping. From the transfer characteristics obtained at different pH levels (Figure 12a), the transconductance was found to be within 0.3 μ8 of a constant value of 23.2 μ8 (Figure l ib), implying that the earner mobility was also approximately a constant regardless of the pH variations. Therefore, the surface transfer doping is not a dominant effect, which would otherwise have altered the carrier mobility significantly.

To investigate the effect of the charging of the EDL capacitor the nanosensor was modeled as a dual-gate field effect transistor consisting of the solid gate (with Hf0₂ as the dielectric) below the graphene and a solution gate formed by the EDL above the graphene at its interface with the solution. The voltage on the top solution gate, V_{! G}, which was equal to the potential drop across the EDL capacitor, depended on the ion concentration in the electrolyte solution. This solution gate voltage, which leads to the charging of the EDL capacitor, can be estimated by the Nerast equation¹⁸, V_LG = E₀ - 2.3 log(H ⁺)RT/nF, where E₀ is a constant reference potential, R the universal gas constant, T the temperature of 298.15 K, n the ionic charge (1 for H⁺), F the Faraday constant. With pH= -log(H⁺), the following was obtained:

V_LC = Eo + (59.2 mV)pH (2)

Thus, the highly linear dependence of the experimentally determined V_SG._DP on pH (Fig. 12b) raised the conclusing that the right shift of with pH was due to the increase in V_LG- In addition, it was seen that V_SG,DP depended on V_Lc in a roughly linear manner in the pH range tested (Fig. 14). As the slope of this dependence (estimated to be ~1 ) was equal to the ratio of the solution-gate capacitance (Qc) to the solid-gate capacitance (C_SG), C_LC,/CSG ~ 1. That is, the solid- gate capacitance was comparable to the liquid-gate capacitance (typically on the order of 1 μΡ/cm²) in the nanosensor, representing a significance improvement over Si0₂ solid-gated GFET devices. This allowed the nanosensor to operate at the low gate voltages as demonstrated. In addition, it should be noted that at a given pH level, a significant increase in the ionic strength of alkali cations (e.g. Na⁺, K⁺) can decrease the measurement sensitivity. This is because the electrostatic gating effects produced by the alkali cations compete with the gating effects from the tf . Therefore, to measure the concentration of FT, the concentration of the nonspecific ions should be maintained at a constant level to obtain a constant sensitivity. Indeed, the concentrations of the alkali cations were approximately constant in the experiments, as the NaOH was added to the buffer at a very dilute concentration (- 0.1 mM) and hence had negligibly effects on the on the ionic strength of the buffer (150 mM). Therefore, the sensitivity in pH measurements was approximately constant and not affected by the addition of NaOH for control of pH values.

To demonstrate the ability of the nanosensor to perform real-time pH measurements, IDS was measured at a fixed gate voltage ( V_BG = 0.75 V), while successively introducing samples with different pH values (Fig. 15). As pH decreased from 9.3 to 5.3 (i.e., the solution becomes more acidic), the V_TG also decreased (Eq. (2)), and the EDL capacitor accordingly underwent partial discharging. This caused a decrease in the carrier concentration of the graphene, and I S hence correspondingly decreased. These phenomena were reversed as pH continued to change, but now in a reversed direction by increasing from 5.3 to 9.3. This reflected that the EDL capacitor underwent charging, thereby causing the carrier concentration in the graphene, and hence I s, to increase. Throughout the entire set of measurements, the values of I_Ds were found to be consistent at a given pH value, regardless of whether this value was reached by pH increasing from a lower value or decreasing from a higher value, with small deviations attributable to the hysteresis in the electronic transport in the graphene. Thus, it was concluded that pH measurements by the nanosensor were reversible, which is important for practical applications.

This example describes a high-/c solid-gated GFET nanosensor in liquid media. The embedded solid gate eliminates the need for an external gate electrode and is hence amenable to the complete integration of the nanosensor as is highly desirable for analyte detection in liquid media. The use of a high-κ dielectric allows the device to operate at low gate voltages and avoids errors caused by gate capacitance variations. Experimental data from the nanosensor showed measurements of pH in a range of 5.3 to 9.3 with a sensitivity of -57.6 mV/pH. The pH-dependent electrical responses of the nanosensor responsible for the measurements were found to be caused by the charging of the electric double layer capacitor, rather than surface transfer doping. These results suggest that the GFET nanosensor can be potentially used to enable highly integrated sensing of chemical and biological analytes.

Example 6. Polymer Synthesis and Graphene Immobilization

A series of pyrene-terminated polymers were synthesized, e.g., poly(7V- 3-acrylamidophenylboronic acid) (pyrene-PAAPBA), using a radical addition- fragmentation chain transfer (RAFT) polymerization (Fig. 16a and 17), using an established protocol for synthesis of sugar-responsive copolymers of PAAPBA and Ν,Ν-dimethylacrylamide with pyrene-derived RAFT agent. N-3- acrylamidophenylboronic acid (AAPBA) monomer (e.g., 8-10%) was mixed with acrylamide (AM) and RAFT polymerization was initiated with AIBN. The reference polymer can use AM as monomer or replace AAPBA in PAM-ran-PAAPBA with phenyl acrylamide (PAM). The chemical structure confirmed by Ή NMR. Three sizes of pyrene-PAAPBA were prepared with a molar mass dispersity less than 1.2 as determined by gel permeation chromatography.

Pyrene-terminated polymers were irreversibly attached to graphene with strength comparable to covalent attachment, using a sticky point for π-π stacking interactions without disrupting the graphene's conjugation or altering its electronic propertiea (Fig. 16b,c) The polymers (PAM-ran-PAAPBA, PAM or PAM-ran-PPAM) were attached to graphene surfaces via incubation with graphene for 30 minutes followed with extensive washing.

Example 7. A Graphene-Based Affinity Glucose Nanosensor

This example presents a synthetic polymer-functionalized graphene nanosensor for affinity-based, label-free detection of low-molecular-weight, low- charged glucose molecules. In this sensor, graphene was functionalized with a synthetic polymer monolayer derivatized with a boronic acid group (Fig. 18) whose reversible complexation with glucose generates a detectable signal. The binding of the polymer monolayer with glucose on the graphene surface induced changes in the carrier density and mobility in the bulk of graphene, thereby potentially offering a high detection sensitivity. The small size of the graphene as the transduction element allows miniaturization of the sensor dimensions. Moreover, the polymer functionalization of the graphene eliminates needs for physical barriers such as semipermeable membranes commonly used in existing glucose sensors, thereby simplifying the device design and enabling rapidly responsive measurements for reliable glucose monitoring.

Nanosensor Design

The nanosensor was configured as a solution-gated graphene-based FET (GFET) in that the graphene (605) was the conducting channel, formed between two gold electrodes (source (608) and drain(609)) on an insulating substrate surface (601 ) (Fig. 18). A monolayer of the synthetic glucose responsive polymer, pyrene- terminated poly(3-acrylamidophenylboronic acid) (py-PAPBA) (61 1), was attached to the graphene surface via π-π stacking interactions (Fig. 17). A glucose solution (621 ) in phosphate buffered saline (PBS) was held directly above the polymer- functionalized graphene in a polydimethylsiloxane (PDMS) microchannel, with an Ag/AgCl wire (607) inserted into the solution to serve as a gate electrode. An electrical double layer (EDL) forms at the interface of the graphene (605) and solution (621 ), and serves as the gate dielectric layer. During operation, under the control of a voltage applied between the gate and source electrodes (gate voltage Vgs), a bias voltage applied between the drain and source electrodes (drain-source voltage Vds) generates a current through the graphene (drain-source current Ids) that was measured. This yields the transfer characteristics of the GFET (i.e., the functional dependence of Ids on Vgs) and allows the determination of the glucose concentration, because the binding of boronic acid moieties of the polymer PAPBA changes the electrical properties of the graphene as follows.

First, the polymer-glucose binding changes the position of the Dirac point (Vgs,Dirac), i.e., the value of the gate voltage at which the charge carriers neutralize and the drain-source current Ids achieves its minimum. Cyclic esters of boronic acid form as a result of the binding of boronic acid groups to glucose molecules, and this causes the overall ionization equilibrium to shift from neutral/insoluble boronic acid moieties to anionic/hydrophilic boronate. Thus, the carrier density varies because of the electron exchanges between the graphene and the solution when the charge density in the solution changes. This alters the Fermi level of the graphene, thereby shifting the Dirac point position.

The polymer-glucose binding also changes the transconductance, gm, i.e., the drain-source current change rate with respect to the gate voltage (dlds/oVgs) in the linear region of the GEFT transfer characteristics. The charged polymer molecules lying on the graphene surface can be considered charged impurities, and induce electron scattering that degrades the carrier mobility, μ, of the graphene. This accordingly decreases the transconductance according to:

= L( WC_gV_dsyⁱg_m (3)

where W and L are respectively the width and length of the graphene conducting channel, and Cg is the gate capacitance per unit area.

Nanosensor Fabrication

The device was fabricated using micro and nanofabrication techniques

(Fig. 1 ) on an oxidized silicon wafer (601 ). After cleaning by piranha, a layer of 5/45 nm Cr/Au was deposited (609, 608) using thermal evaporation. A layer of photoresist was then spin-coated on top of Au layer and baked at 1 15 °C for 1 min. Photolithography was used to pattern the gate electrode, and the wafer was then developed and etched in gold and chrome etchant sequentially. Graphene (605) synthesized via chemical vapor deposition (CVD) on a copper sheet was transferred onto the substrate following an established protocol to cover the source (608) and drain (609) electrodes (Fig. 20a).

To perform PAAPBA polymer functionalization, the graphene and the underlying substrate were immersed in a solution of the pyrene-terminated polymer (py-PAAPBA/methanol 3% w/v) for 4 hours at room temperature, and then washed thoroughly using methanol. During testing of the nanosensor, glucose solution was placed directly above the graphene and held in a PDMS open microchannel (-2.5 μL· in volume), which was fabricated using soft lithography and reversibly bonded to the device. An Ag/AgCl reference electrode was inserted into the solution above the graphene to serve as the gate electrode for application of a gate voltage (Fig. 20b).

Results

It was verified that single-layer graphene was used in the nanosensor via Raman spectroscopy (Fig. 21 ). The G band at - 1580 cm^{" 1} in the Raman spectrum, characteristic of the planar geometry of sp bonded carbon, indicated that the material was graphene. Moreover, the sharp and symmetric 2D band at -2685 cm^"1 indicated that the graphene consisted of a single layer of carbon atoms. Next, the polymer functionalization of the graphene was verified. Atomic force microscope (AFM) was performed, and the resulted images (Fig. 22) showed that there was a significant increase (-10 nm) in the apparent height of the graphene sheet, suggesting the successful grafting of the polymer molecules.

The polymer functionalization of the graphene was also verified by measurement of the GFET transfer characteristics. The shape of the Id_s- V_gs curve was similar before and after the functionalization protocol, while the Dirac point position gj.Dirac ^was found to have shifted from 0.22 V to 0.18 V (Fig. 23). The lack of change in shape of the Id_S- V_gs curve suggested that changes in the earner mobility in the graphene were insignificant. This is consistent with the polymer, and in particular the boronic acid moieties, being electrically neutral and would cause little electron scattering to change the graphene' s carrier mobility. On the other hand, the shift in ^gs, Dirac ^was believed to be caused by w-doping (i.e., electron doping) of the graphene, and was consistent with the surface-attached polymer inducing electron transfer from the solution to the graphene. Therefore, concluded graphene had been successfully functionalized with the PAAPBA polymer.

The nanosensor was tested to obtain the graphene' s transfer characteristics at varying glucose concentrations. It was observed that the transfer characteristics changed consistently as the binding of glucose to the boronic acid shifted the electrically neutral boronic acid groups to anionic boronate esters (Fig. 24). In response to increases in the glucose concentration, the Dirac point position F^- Dirac shifted to higher gate voltages with a sensitivity of -2.5 mV/(mg/dL), while the Ids- Vgs curve broadened in shape. The shift in V_{gs D}i_nc indicated that the graphene was p- doped, or hole-doped, which could be caused by changes in the amount of electric charge on the EDL gate capacitor due to the formation of anionic boronate esters. On the other hand, the broadening of the transfer characteristic curve reflected a decrease in the transconductance from 100 to 20 μβ. According to the equation above, this corresponded to a decrease in the carrier mobility and was a consequence of the polymer-glucose binding, as the negatively charged boronate esters, in the role of charged impurities, caused electron scattering. To investigate the influence of the potential contributors other than polymer-glucose binding, control measurements were performed on pristine graphene that was not functionalized with the PAPBA polymer. Neither the Dirac point position nor the transconductance changed significantly as the glucose concentration was varied from 60 to 200 mg/dL (Fig. 25), indicating that when not functionalized the PAAPBA polymer, there was negligible response of the graphene to glucose concentration changes. Thus, the response of the polymer-functionalized nanosensor to the changes in glucose concentration resulted from the glucose-polymer binding.

The device was tested for the detection of glucose in a concentration range of 0 to 200 mg/dL with a sensitivity of -2.5 mV/(mg/dL). These results demonstrate the potential of the device for blood glucose monitoring and control in diabetes care.

Example 8. A Graphene-Based Affinity Glucose Nanosensor

In order to construct a nanosensor that can be integrated and

miniaturized enough to be coupled to a contact lens, graphene, HfO₂ and gold were successively laid on an Si0₂-coated silicon substrate. The gold layer, lying below graphene, served as the gate, and the 20-nm Hf0₂ layer as the dielectric to form a solid-gated FET (Fig. 26a), similar to the configuration to be used in Example 7.

Using such a nanosensor demonstrated real-time monitoring of pH changes (Fig. 26b).

Example 9. A Graphene-Based Affinity Glucose Nanosensor

This nanosensor is built to use a mechanically flexible, CVD-prepared hexagonal boron nitride (h-BN) nanolayer to replace Hf0₂ (which may be brittle and not flexible) as the dielectric.

The graphene of this nanosensor can be functionalize with an optimized synthetic polymer and coupled to a contact lens-based flexible substrate.

Example 10. Preparation of a hydrogel

A poly(hydroxyethyl-methacrylate) (PHEMA) based hydrogel, which has excellent biocompatibility and glucose permeability, can be synthesized in situ. A mixture of hydroxyethyl methacrylate (HEMA), tetraethyleneglycol diacrylate

(TEGDA), polyethyleneglycol methacylate (PEGMA) and N- [tris(hydroxymethyl)methyl]-acrylamide (HMMA) in a ratio 86:3:5:5 mol% can be prepared, and applied to the device surfaces to form a hydrogel under UV irradiation in N₂, which is then conditioned via sequential immersion in mixtures of ethanol and deionized water. The hydrogel synthesis can be optimized to enable high glucose permeability while preventing permeation of larger molecules (e.g., proteins) from tissue.

Example 11. Preparation of a two channel sensor

Nanosensor Design

The graphene nanosensor design consists of two field-effect transistor (FET) modules of identical construction on a contact lens-based flexible substrate (Figs. 1 and 2). In each module a slender graphene strip (the conducting channel) lies on a sheet of h-BN (106) passivating a Ti/Pd/Pt gate electrode ( 107) (with h-BN minimizing substrate-induced charge carrier scattering in the conducting channel)), and makes contact with Ti/Pd/Pt source (108) and drain electrodes (109). These nanolayers lie on a polymer substrate {e.g., PET) (101), and are covered with a thin polymer {e.g., parylene) layer ( 1 10) with the exception of the conducting-channel (103, 1 04) graphene, which is functionalized with a glucose-binding polymer (the sensing module) (1 1 1 ) or a glucose-insensitive polymer (the reference module) (1 12). The entire nanosensor is coated with a glucose-permeable hydrogel. The nanolayers and polymer layers are all flexible and biocompatible. During operation, the conducting-channel graphene is subjected to a gate voltage (e.g., between ±300 mV) with respect to the gate electrode underneath, and the current through the graphene under a bias voltage (-100 mV) between the source and drain, which can be kept sufficiently small to limit the potential leakage current into the hydrogel coating well below the threshold (-10 μΑΛηιτι2) required to prevent undesirable electrochemical effects in tissue.

The nanosensor is design as summarized in the Fig. 2. To estimate the associated performance, Langmuir s adsorption isotherm relates the glucose concentration in tears (c) to the surface density of boronic acid-bound glucose molecules. It can then be shown that the sensitivity of the graphene conductance {G) is approximately:

(dG I G) I dc * _YumK_d I [2N{K_d + c)² ], (4) where N is the graphene' s charge carrier density, y_max the surface density of immobilized boronic acid groups, and Kj the equilibrium dissociation constant of the binding system. Thus, for typical material and binding properties (TV

] ^ ^ 2 86 87

~10 ^" m^"~, y_max ~ 10^' μηΓ and K_d ~ 10 mM) ' , it is estimated that over a desired glucose concentration range of 0.5-30 mg/dL (which encompasses the hypoglycemic, normoglycemic and hyperglycemic regimes), the nanosensor sensitivity ~ 0.5%- 0.4%/(mg/dL). For an estimated graphene conductance measurement resolution of 0.1 %, hence it is anticipated that glucose concentration can be measured at a resolution of ~0.7-0.02%, which allows accurate affinity detection of glucose in tears.

The device's time response is determined by the diffusive transport of glucose between the tissue and the implanted device, over a distance primarily given by the hydrogel coating (thickness ~ 20-50 μιη and glucose diffusivity ~ 6x 10^"" m/s²). This allows the device time constant to be estimated in the range of 7-42 seconds, affording a rapid response to tear glucose concentration changes.

Nanosensor Fabrication

To fabricate the device, a Ti/Pd/Pt layer is deposited onto a substrate and then covered with CVD-deposited h-BN, forming the gate electrode ( 107), followed by deposition and patterning of another Ti/Pd/Pt metallization as the drain (109) and source electrodes (108). A single-layer CVD graphene sheet (505) is transferred using a protective poly(methyl methacrylate) (PMMA) layer onto the substrate and patterned to form the conducting channel. With graphene still protected by PMMA and photoresist, a thin layer (1 -2 μηι) of Parylene is deposited and patterned using oxygen plasma. After dissolving the PMMA and photoresist in organic solvent, the conducting-channel graphene is exposed and then grafted with the sensing or reference polymer, and the entire device is coated with a hydrogel. A nanosensor prototype can be fabricated on a rigid silicon substrate, and then on a flexible PET substrate. For the flexible nanosensor the device can be fabricated on a flat sheet of PET, which can then be thermally molded into a contact lens shape. PET and Parylene are chosen for their biocompatibility as well as their excellent

mechanical and chemical stability.

The nanosensor is fabricated in two steps with increasing complexity, first on a rigid Si0₂-coated silicon substrate, and then on a flexible polymer substrate. In the case of a rigid substrate, ITO can be deposited onto a substrate and then covered with CVD-deposited h-BN, forming a transparent gate electrode. This is followed by deposition and patterning of another ITO layer as the drain and source electrodes. A single-layer CVD graphene sheet (105) is then transferred using a protective poly(methyl methacrylate) (PMMA) layer onto the substrate and patterned to form the conducting channel. With graphene still protected by PMMA and photoresist, a thin layer (< 1 pm) of Parylene is deposited and patterned using oxygen plasma. After dissolving the PMMA and photoresist in organic solvent, the graphene is exposed and then grafted with the sensing or reference polymer, and the entire device is coated with a hydrogel.

Building on experience from rigid substrates, the nanosensor can then be fabricated on a flexible substrate. To identify suitable materials to serve as the substrate for our nanosensor, materials that have been used in existing contact lens sensors can be measured, and also investigate materials commonly used in commercially available contact lenses. Contact lens-based sensors in the literature have used polymers such as polyethylene terephthalate (PET), which is unfortunately opaque and not appropriate for practical use, and silicone. Commercially available contact lenses include "soft lenses^"' and "rigid lenses". Soft lenses are made of hydrogels or silicone-hydrogel co-polymers that have high water content but a low mechanical stiffness. In comparison, rigid lenses, with low water content, have significant stiffness to retain their shape. PMMA can be used as the substrate for fabrication of our nanosensor. Other materials include soft lens materials such as silicone (e.g., NuSil Technology MED-6015), and rigid lens materials such as silicone acrylates (e.g., Dow Coming* FA 4001 CM Silicone Acrylate) and fluorosilicone acrylates (e.g., XCEL Contacts Fluoroperm* 92).

PMMA is the material of which the first contact lenses were made, and is still in use in commercially available contact lenses today. It has excellent optical qualities, durability and biocompatibility, and micro- and nanofabri cation techniques using PMMA are well established and have been used in our preliminary studies as a sacrificial substrate for transfer of graphene. The fabrication process described above for rigid substrates is generally compatible with PMMA used as substrate. Using the process, the nanosensor is fabricated on a flat sheet (-50 pm thick) of PMMA placed on a carrier silicon wafer. Upon the completion of the fabrication process, the flat PMMA sheet along with the fabricated device can then be themially molded into the shape of a contact lens. The device could also be fabricated on suitable handling substrate (e.g., PET), and then transferred onto PMMA that has been pre-molded into a contact lens shape.

Graphene can be grafted with sensing and reference receptors (e.g., polymers), and the nanosensor can be coated with a biocompatible hydrogel.

Example 12. Understanding the Physics of Nanosensor Operation

To understand the sensing mechanisms and inform the optimal design of the graphene nanosensor, the conductance change of graphene due to polymer- glucose binding-induced changes in surface charge and compare the results with experimental data is investigated using a simplified model in which a rectangular graphene sheet (with length / and width w) sandwiched between an insulating substrate and a solution of an electrolyte, and is biased longitudinally with a voltage without being subject to a gate voltage. The graphene surface is charged (density: σ) due to the presence of charged (polymer and bound glucose) molecules.

The distribution is determined of the electric potential φ in the electrolyte solution near the graphene surface and then consider the charge balance at the graphene surface. While two- and three-dimensional effects (as well as those of a nonzero gate voltage) is considered, here for simplicity it is assumed that the potential is only a function of the coordinate nonnal to the graphene surface (x), and that the electrolyte consists of anions and cations of equal charge number z. More general cases can be considered in a conceptually similar manner. The potential distribution is governed by the Poisson-Boltzmann equation of the form

where λ = (ε kT 1 2zVc₀ )' ¹² is the Debye length and β = kT I ze . Here c₀ is the concentration (number of molecules per unit volume) of the electrolyte in the bulk solution, ε the solution's dielectric constant. T the temperature, k the Boltzmann constant and e the electron charge. Since άφ/άχ→0 in the bulk of the solution (x-∞), the equation can be solved to obtain φ =

+ r_oe^{" v ;} ) / (1 - T₀e ^!A )] where Γ₀ = - tanh( _s Ι Αβ) and φ₈ = φ(0) is the (yet-to-be-determined) potential at the graphene surface. The charge per area induced in the graphene by the surface potential is approximately given by u_gr = C_q(§_s - φ₀) where φ₀ is the surface potential at the Dirac

92

point, and C_q is the quantum capacitance of graphene per unit area :

C_q = {N I n)^m e² I V_j h (6) with Vf the Fermi velocity and the Planck constant. Here N is the number of charge carriers per unit area, which depends on a_g,. (and hence C_q):

Ν = ( Ν₀ ² ₊ [^( , - ₀) / β]² )^υΐ (7)

where No is the sheet charge carrier concentration due to disorder and thermal excitation .

On the graphene surface and within the graphene, conservation of charge (Gauss's law) requires that -εφ' 0) + C φ = σ , i.e.,

C_? {Φ, - Φ₀ ) = σ - 2 (βε I λ) sinh (φ_χ 12β) (8)

The graphene' s conductance, or equivalently resistance, can then be computed from the resulting charge carrier concentration by

where μ is the carrier mobility in graphene. Thus, the graphene resistance R can be measured to obtain the charge carrier concentration N. Then Eqs. 6, 7, and 8 can be solved to obtain the surface potential φ₈, quantum capacitance Q, and in particular the surface charge density σ, which is directly related to surface density of the captured glucose molecules, γ. Meanwhile, the volumetric glucose concentration (c) is related to γ by the Langmuir adsorption isotherm:

γ = ξ[ο Ι 2{Κ_{ά +} α)} (10)

where (as defined above) ξ is the surface density of immobilized polymer molecules and K_d the equilibrium dissociation constant of the binding system. Thus, using Eq. 10, measurement of the graphene resistance eventually allows determination of the glucose concentration c.

Example 13. In Vitro and In Vivo Characterization of the Graphene Nanosensor

The polymeric materials (including PMMA, Parylene, and the synthetic functional polymer monolayers and hydrogel coating) of the nanosensor are biocompatible, as are the transducing (graphene), dielectric (h-BN) and metallization (ITO) materials when their uptake by individual cells is avoided. The in vitro characterization does thus not involve compatibility testing, and focuses on the device's response to glucose concentration changes in the absence and presence of biofouling to assess its potential for accurate glucose monitoring. Verification that the device is biocompatible in in vivo studies is conducted, where in particular issues relevant to the contact lens platform, such as oxygen and ion permeability, wettability, water content, and cytoxicity, as well as protein adsorption and biofilm formation are determined.

The nanosensor's characteristics in glucose detection is determined by exposing the nanosensor of varying design parameters (e.g., the number of atomic layers, shape and dimensions of the graphene, molecular weight of the sensing polymer, and the thickness of the hydrogel coating) to glucose or unspecific mono- and disaccharides dissolved in buffer. Measurement of the graphene conductance in these experiments allows assessment of the specificity of the device in glucose detection. The sensitivity and dynamic range of the nanosensor is investigated by measuring the graphene conductance at physiologically relevant concentrations (0.5- 30 mg/dL), with an emphasis on hypoglycemic concentrations (below 1 mg/dL). The sensitivity can then be used along with an assessment of the intrinsic noise levels in the graphene sensor and in readout instrumentation to determine the nanosensor resolution in the glucose concentration range of interest. In addition, time-resolved data from the measurements can be used to determine the device's time response.

The characterization is also performed under varying operating parameters including temperature, pH, ionic strength and oxygen concentrations, which can significantly influence affinity binding and graphene's electronic properties. This can also be used to evaluate the ability of the device's differential design to reject the effects of variations of these parameters during sensor operation for reliable CGM.

The nanosensor response's to glucose in the presence of simulated biofouling is investigated by repeating the experiments above but with the device exposed to artificial tear fluid consisting of sodium chloride, potassium chloride, sodium bicarbonate, urea, ammonia chloride, lactic acid, pyruvic acid, citric acid, Vitamin C, albumins, γ-globulins, and lysozyme. The goal is to evaluate the effects of the accumulation of biological materials, via protein adsoiption and biofilm formation, on the nanosensor performance. Adsorption of proteins to the lenses leads to the buildup of unwanted bacteria and other materials, causing contamination and degradation of the sensor sensitivity. Protein adsorption is monitored using fluorescence spectroscopy and labeled model proteins incubated on lens surfaces. Biofilms are collections of microorganisms encased in a matrix that is often comprised of both bacterial and host materials. The ability of microorganisms to attach to abiotic surfaces and to grow in highly stable communities greatly confounds devices that have direct contact with tissue. On the contact lens sensor, the microorganism community can form a barrier between tear and sensor, preventing detection of glucose. The attachment of biofilms to lens surfaces is monitored using scanning electron microscopy methods while measuring the sensor response under simulated biofouling by incubating the device in bacterial cell suspension.

The foregoing merely illustrates the principles of the disclosed subject matter. Various modifications and alterations to the described embodiments will be apparent to those skilled in the art in view of the inventors' teachings herein. Features of existing methods can be integrated into the methods of the exemplary embodiments of the disclosed subject matter or a similar method. It will thus be appreciated that those skilled in the art will be able to devise numerous methods which, although not explicitly shown or described herein, embody the principles of the disclosed subject matter and are thus within its spirit and scope.

Various publications, patents and patent application are cited herein, the contents of which are hereby incorporated by reference in their entireties.

Claims

A microdevice for monitoring a target analyte using a receptor adapted for binding to the target analyte, the microdevice comprising: a substrate platform and a nanosensor, wherein the nanosensor is coupled to the substrate platform and comprises: a first conductance element functionalized with a sensing receptor for detecting the target analyte; and a second conductance element functionalized with a reference receptor that is insensitive to the target analyte.

The microdevice of claim 1 , wherein the first conductive element comprises an element having a surface adapted for a change in charge density thereon upon the binding of the receptor with the target analyte.

The microdevice of claim 1 , wherein the second conductance element comprises an element having a surface such that a charge density thereon does not change upon the binding of the reference receptor with a reference analyte.

The microdevice of claim 1 , wherein a differential measurement of the conductance of the first conductance element and the second conductive element provides for determination of target analyte concentration.

The microdevice of claim 1 , wherein the first conductive element and the second conductive element each comprise graphene with a receptor grafted thereon.

The microdevice of claim 5, wherein the graphene comprises a single layer sheet.

The microdevice of claim 1 , wherein the nanosensor further comprises a dielectric nanolayer.

The microdevice of claim 1 1 , wherein the dielectric nanolayer comprises hexagonal boron nitride or Hf0₂.

9. The microdevice of claim 1 , wherein the nanosensor further comprises a gate electrode formed from a nanolayer including at least one of ITO, Ti/Pd/Pt, gold, chrome or copper.

10. The microdevice of claim 6, wherein the nanosensor further comprises a

source electrode and a drain electrode, and wherein the graphene sheet makes contact to both the source electrode and the drain electrode.

1 1. The microdevice of claim 10, wherein the source and drain electrodes

comprise a nanolayer including at least one of ITO, Ti/Pd/Pt, chromium, gold, or chrome.

12. The microdevice of claim 1 , wherein the platform substrate comprises at least one of polyethylene terephthalate (PET), polycarbonate polystyrene, polymethyl methacrylate (PMMA), polymacon, silicones, fluoropolymers, silicone acrylate, fluoro-silicone/acrylate, or poly hydroxyethyl methacrylate.

13. The microdevice of claim 1, wherein the nanosensor is covered with a

polymer coating except for the functionalized part of the graphene sheet and wherein the polymer coating comprises at least one of parylene, polyimide, organic polymer, or hydrophobic polymer.

14. The microdevice of claim 1 , wherein the target analyte comprises glucose.

15. The microdevice of claim 1 , wherein the receptor comprises a polymer.

16. The microdevice of claim 15, wherein the polymer comprises poly(N- hydroxyethylacrylamide-ran-3-acrylamidophenylboronic acid) (PHEA-ran- PAAPBA).

17. The microdevice of claim 1 , wherein the sensing receptor comprises a

plurality of boronic acid moieties.

18. The microdevice of claim 1 , wherein the nanosensor is coated with a glucose- permeable hydrogel.

19. The microdevice of claim 18, wherein the hydrogel comprises at least one of poly(hydroxyethy]-methacrylate (PHEMA), hydroxyethyl methacrylate (HEMA), tetraethyleneglycol diacrylate (TEGDA), polyethyleneglycol methacylate (PEGMA), or N-[tris(hydroxymethyl)methyl]-acrylamide (HMMA).

20. The microdevice of claim 1 , wherein the substrate platform comprises a contact lens or a flexible thin film.

21. A microdevice for monitoring a target analyte comprising: a contact lens and a graphene nanosensor, wherein the graphene nanosensor is coupled to the contact lens and comprises: a first conductance element functional ized with a sensing polymer for detecting the target analyte; and a second conductance element functionalized with a reference polymer that is insensitive to the target analyte.

22. The microdevice of claim 21 , wherein a differential measurement of the

conductance of the first conductance element and the second conductive element provides for determination of target analyte concentration.

23. The microdevice of claim 21 , wherein the target analyte comprises glucose.

24. The microdevice of claim 21 , wherein the sensing polymer comprises a

plurality of boronic acid moieties.

25. A method for monitoring a target analyte using a polymer capable of binding to the target analyte, comprising: placing a nanosensor in contact with a bodily fluid, wherein the nanosensor comprises a first conductance element functionalized with a sensing polymer for detecting the target analyte and a second conductance element functionalized with a reference polymer that is insensitive to the target analyte; detecting a difference, if any, in the conductance of the first and second conductance elements; and based on the detected difference, determining a presence and/or concentration of the target analyte in the sample.

26. The method of claim 25, wherein the binding of the sensing polymer with the target analyte causes a change in the charge density on the first conductive element surface.

27. The method of claim 25, wherein the conductance of the second conductive element only changes due to fluctuations in environmental parameters.

28. The method of claim 25, wherein the conductance elements comprises a

graphene sheet, wherein the graphene sheet is formed by chemical vapor deposition.

29. The method of claim 25, wherein the sensing polymer reversibly binds with the target analyte.

30. The method of claim 25, wherein the detection is continuous over time.

31. The method of claim 25, wherein the bodily fluid is selected from the group consisting of tears, blood, saliva, mucus, interstitial fluid, spinal fluid, intestinal fluid, amniotic fluid, lymphatic fluid, pericardial fluid, peritoneal fluid, pleural fluid, semen, vaginal secretions, sweat, and synovial fluid of the subject.

32. The method of claim 25, wherein the nanodevice detects low concentrations of the target analyte.

33. A microdevice for monitoring a target analyte using a receptor adapted for binding to the target analyte, the microdevice comprising: a substrate platform and a nanosensor, wherein the nanosensor is coupled to the substrate platform and comprises a conductance element functionalized with a sensing receptor for detecting the target analyte.

34. The microdevice of claim 33, wherein the conductive element comprises an element having a surface adapted for a change in charge density thereon upon the binding of the receptor with the target analyte.

35. The microdevice of claim 33, wherein the conductive element comprises

graphene with a receptor grafted thereon.

36. The microdevice of claim 33, wherein the nanosensor further comprises: a dielectric nanolayer comprising hexagonal boron nitride or Hf0₂; a gate electrode formed from a nanolayer including at least one of 1TO, Ti/Pd/Pt, gold, chrome, or copper; a source electrode and a drain electrode, wherein the source and drain electrodes comprise a nanolayer including at least one of 1TO, Ti/Pd/Pt, chromium, gold, or chrome; and a polymer coating except for the functionalized part of the graphene sheet and wherein the polymer coating comprises at least one of parylene, polyimide, organic polymer, or hydrophobic polymer.

37. The microdevice of claim 33, wherein the target analyte comprises glucose.

38. The microdevice of claim 33, wherein the receptor comprises a polymer

comprising a plurality of boronic acid moieties.

39. The microdevice of claim 33, wherein the substrate platform comprises lab equipment or home monitor.

40. A microdevice for monitoring a target analyte using a receptor adapted for binding to the target analyte, the microdevice comprising: a substrate platform and a nanosensor, wherein the nanosensor is coupled to the substrate platform and comprises a conductance element for detecting the target analyte.

41. The microdevice of claim 40, wherein the conductive element comprises an element comprising graphene having a surface adapted for a change in charge density thereon upon.

42. The microdevice of claim 40, wherein the nanosensor further comprises: a dielectric nanolayer comprising hexagonal boron nitride or Hf0₂; a gate electrode formed from a nanolayer including at least one of ITO, Ti/Pd/Pt, gold, chrome, or copper; a source electrode and a drain electrode, wherein the source and drain electrodes comprise a nanolayer including at least one of ITO, Ti/Pd/Pt, chromium, gold, or chrome; and a polymer coating, wherein the polymer coating comprises at least one of parylene, polyimide, organic polymer, or hydrophobic polymer.