WO2018204780A1

WO2018204780A1 - 3d chemical sensing on 2d surface

Info

Publication number: WO2018204780A1
Application number: PCT/US2018/031078
Authority: WO
Inventors: Liwei Lin; YuMeng LIU; Takeshi Hayasaka
Original assignee: The Regents Of The University Of California
Priority date: 2017-05-05
Filing date: 2018-05-04
Publication date: 2018-11-08

Abstract

This disclosure presents a new class of gas sensing scheme: the phase sensitive signal on the GFET (Graphene Field Effect Transistor) gas sensors aiming to boost the response speed.

Description

3D CHEMICAL SENSING ON 2D SURFACE

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority under 35 U.S.C. §119 from Provisional Application Serial No. 62/502,549, filed May 5, 2017, the disclosures of which are incorporated herein by reference.

TECHNICAL FIELD

[0002] The disclosure provides sensors and methods of use thereof.

BACKGROUND

[0003] Conventional chemical sensors or chemical resistors detect the molecule concentration by monitoring the resistance change caused by the reaction near the sensing material surface. Such sensing scheme (defined as 2D sensing) is straightforward and easy to be implemented yet it bears many problems. One problem is the drift issue or the so-called "poison effect", which result from the molecules irreversibly binding with the surface. This effect causes the baseline drift of chemical sensor. To overcome the "poison effect" and baseline drift rigorous sensor calibration of baseline is needed after certain operation periods. Recovery speed is affected by the time cost of a molecule leaving the sensing surface. The weaker the interaction between molecule and the surface, the faster the recovery speed.

SUMMARY

[0004] The disclosure provides a sensor, wherein the sensing plane is moved away from the near field surface where the molecule-surface interaction is strong, but rather are directed to molecules in the far field where the interaction is weak for a faster recovery speed of the sensing response. Such sensors and processes are referred to herein as "3D chemical sensors" or "3D chemical sensing" or specifically "3D Gas Sensing" for gas sensor application as it enable the chemical sensor to reach out vertically above its surface, and it can benefit the chemical sensors suffering from drift issue and low response speed for a better performance.

[0005] This disclosure presents a new class of gas sensing scheme: the phase sensitive signal on the GFET (Graphene Field Effect

Transistor) gas sensors aiming to boost the response speed.

Compared with the state-of-art gas sensing reports on measurements of DC resistance changes which are limited by trap states, three distinctive advancements have been achieved on graphene FETs in responses to water vapor: (1) experimental validations of a faster saturation and recovery speed; (2) in-situ measurements of field effect mobility μ and DC carrier concentration N_GFET, and (3) multiple charge transfer processes through the plot of N_GFET versus μ^" ¹. As such, the sensing scheme and results open up a new frontier for graphene FET-based gas sensor with fast sensing speed in practical usages.

[ 0006] The sensing methods and compositions use a gas vs. an electrolyte. Thus, the system described herein deals with the charge transfer between sensing material and gas molecule directly, without electrolyte medium and/or reference electrodes. In addition, the phase in the present disclosure is the phase lag between gate voltage and the graphene resistance, or the phase of the trans- impedance, which contains rich information of the charge transfers. Thus, the primary AC voltage is applied at the gate voltage.

[ 0007 ] The disclosure provides a method of operating a chemical sensor, comprising applying a variable voltage to a gate of a transistor that includes a sensing material in contact with a medium in a gaseous state; and monitoring a phase of a charge transfer current between the sensing material and the gas while applying the variable voltage. In one embodiment, the transistor is a

field-effect transistor. In another or further embodiment, the field-effect transistor is a graphene field-effect transistor. In still another or further embodiment, the sensing material is included in a channel of the transistor. In yet another or further embodiment, the sensing material is graphene. The still another embodiment of any of the foregoing, the method further comprises using the phase of the charge transfer current to measure a

molecular concentration in the medium and/or identify the molecular type of analyte. In still another or further embodiment, the variable voltage is an alternating current (AC) voltage. In yet another or further embodiment of any of the foregoing, the

alternating current (AC) voltage has a frequency of 0. lHz to 1 MHz. In still another embodiment of any of the foregoing, the method can further comprise sweeping a frequency of the AC voltage while monitoring the phase of the charge transfer current. In yet another or further embodiment, the method further comprises using the phase of the charge transfer current to determine a molecular

concentration of the medium at one or more non-zero distances above a surface of the sensing material. In still another or further embodiment, the method further comprises using the phase of the charge transfer current to determine a molecular concentration of the medium as a function of a distance above a surface of the sensing material. In yet another embodiment, the monitoring includes measuring.

[ 0008 ] The disclosure also provides a method of operating a chemical sensor, comprising applying a variable voltage to a gate of a transistor that includes a sensing material in contact with a medium in a gaseous state; and monitoring a phase lag between the voltage of the gate and a resistance of the sensing material while applying the voltage. In one embodiment, the transistor is a field-effect transistor. In another or further embodiment, the field-effect transistor is a graphene field-effect transistor. In still another or further embodiment, the sensing material is included in a channel of the transistor. In yet another or further embodiment, the sensing material is graphene. In still a further embodiment of any of the foregoing, the method further comprises using the phase lag to measure a molecular concentration in the medium and/or identify an analyte in the gas phase. In another or further embodiment, the variable voltage is an alternating current (AC) voltage. In still another or further embodiment, the method further comprises using the phase lag to determine a molecular concentration of the medium at one or more non-zero distances above a surface of the sensing material.

DESCRIPTION OF DRAWINGS

[ 0009 ] FIG . 1A-D shows a schematic diagram of the AC sensing circuity of the graphene FET (GFET) gas sensor. (B) The circuit illustrated in the right block is used to model the charge transfer behavior between graphene and the gas adsorbates or trap states. (B- D ) Show (B) schematic diagram illustrating the gas sensing

performances between the phase lag detection scheme with minimal baseline drift and fast recovery speed as compared with the conventional DC resistance scheme on graphene-based gas sensors. (C) The setup of phase lag <pAB,_g between VAB and Vg on a CVD graphene FET sensor after the exposure to a specific chemical vapor. (D) The vapor adsorption and desorption on the graphene surface - the AC sensing scheme is targeting weakly adsorbed gases with a short distance from the graphene surface while the DC sensing scheme is targeting molecules very close to the graphene surface. This results in fast recovery for the AC sensing scheme.

[0010] FIG. 2A-F shows (A) optical microscopic picture of the as- fabricated GFET and (B) SEM photo of a fabricated graphene FET sensor. (C) Raman spectroscopy of the original graphene received from the vendor (Graphenea Inc.) and three sampling spots after microfabrication on the as-fabricated GFET. (D) Schematic of the gas sensor test setup and the photograph of the test setup. The inset shows the small chamber manufactured with nylon. (E) Schematic diagram of the experimental setup for the phase lag detection on graphene showing (a) the electrical circuity and the flowchart for the data acquisition and processing, and (F) the gas routes of the test system.

[0011] FIG 3A-C shows (A) Real time measurement results of the admitance 1/Z_SD between the source and drain electrode of GFET under DC gate voltage v_G modulation. (B) A zoom-in view of region I in a) . (C) A zoom-in view of region II in a) .

[0012] FIG. 4A-C shows experimental results of (A) the phase and amplitude of zSD, and (B) field effect mobility μ and DC carrier concentration NGFET of GFET under sweeping gate voltage VG. (C) Measurement results of the phase signal at various amplitude of vg at 50Hz (recorded in real time) .

[0013] FIG. 5A-H shows (A) Experimental results of one-cycle water vapor sensing using two types of sensing signals: the phase signal

Φ ( z_Sd) , and the DC resistance signal Z_SD- (B) The response of the phase signal at various relative humidities (RH) . (C-H) The schematic of carrier scattering (hole) in graphene with charged impurity of gas molecules (C-E) and trap states (F-H) .

[0014] FIG. 6A-B shows (A) Extracted transient responses of the field effect mobility μ and DC carrier concentration N_GFET in high definition of time (dt=0.1s) from FIG. 5A. (B) The N_GFET versus μ^"1 plot shows five distinctive stages of the charge transfer processes. The inset shows stages II IV and III V are two pairs of inverse process .

[0015] FIG. 7A-B shows (A) a physical model of a device,

composition and method of the disclosure; and (B) an equivalent circuit used in the methods of the disclosure.

[0016] FIG. 8 shows plots of CO₂ sensing using the methods and compositions of the disclosure (x-axis is time in seconds) .

[0017] FIG. 9 shows plots of the minimal drift VOC (methanol vapor) sensing results (x-axis is time in seconds) .

[0018] FIG. 10 shows molecule concentration mapping above the sensing plane.

[0019] FIG. 11A-C shows measurement results of three vapors (water, methanol, and ethanol) using the graphene FET sensor at room temperature. (A) Schematic of vapor concentrations. (B) Dynamic responses of the phase lag Acp_AB,_g upon exposures to three different gases at a representative frequency of 1000 Hz. (C) Dynamic

responses of the relative change of DC resistance AR/Ro upon exposures to different gases.

[0020] FIG. 12A-F shows dynamic responses of phase lag changes at exemplary frequencies of 50 Hz, 100 Hz, 500 Hz, and 1000 Hz for (A) and (D) water, (B) and (E) methanol and (C) and (F) ethanol vapors under various concentrations (from 10% to 90%) at room temperature. The shading indicates the gas input and purging events in (A-C) nitrogen and (D-F) dry air environment.

[0021] FIG. 13A-F shows response amplitudes for (A) water, (B) methanol and (C) ethanol in the nitrogen and air environment at representative frequencies of 50 Hz and 1000 Hz. The 90%recovery time for (D) water, (E) methanol and (F) ethanol at different sensing frequencies with 90% gas vapor concentrations in nitrogen and air environment.

[0022] FIG. 14A-F shows representative features of the frequency versus time phase lag spectra upon exposures to three chemical vapors (A-C) at their medium concentration (60%) and (D-F) at their high concentration (90%), respectively, as the vapor starts to enter the sensing system at 50 s and being purged with nitrogen at 90 s. [0023] FIG. 15 shows dynamic responses of phase lag changes at the exemplary frequencies of 50 Hz and 1000 Hz for ethanol vapors under 60% RH condition in air at room temperature (24°C) .

[0024] FIG. 16A-D shows response amplitude for (A) methanol and (B) ethanol under different RH conditions in air at the frequency of 1000 Hz. The 90% recovery time for (C) methanol and (D) ethanol at different frequencies from 50 Hz to 1000 Hz with saturated gases under different RH conditions in air at room temperature (24°C) .

[0025] FIG. 17A-D shows (A) Schematic diagram illustrating the vapor adsorption and desorption process on the graphene surface - the AC sensing scheme is targeting weakly adsorbed gases a short distance away from the graphene surface while the DC sensing scheme is targeting molecules close to the graphene surface. (B) The RC model of the charge transfers pathway and process between adsorbed gas and graphene with a distance d away from the graphene surface. The resistance component scales up exponentially with d, while the capacitance scales down with d in a logarithmic manner. (D) The representative response of the sticking probability p and (D) the AC signal φ for the weakly adsorbed molecules (d = 4 r_mo) , and the strongly adsorbed molecules (d = 2 r_mo) .

[0026] FIG. 18A-B provides score plots of the developed PCA model (A) PCI vs PC2 and (B) PCI vs PC2 vs PC3. PCI and PC2 provide strong contributions to discriminate four vapors at their medium and high concentrations. PC3 also provides contributions correlated with concentrations .

DETAILED DESCRIPTION

[0027] As used herein and in the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a sensor" includes a plurality of such sensors and reference to "the gas" includes reference to one or more gases and equivalents thereof known to those skilled in the art, and so forth.

[0028] Also, the use of "or" means "and/or" unless stated

otherwise. Similarly, "comprise," "comprises," "comprising"

"include," "includes," and "including" are interchangeable and not intended to be limiting. [ 0029] It is to be further understood that where descriptions of various embodiments use the term "comprising," those skilled in the art would understand that in some specific instances, an embodiment can be alternatively described using language "consisting

essentially of" or "consisting of."

[ 0030 ] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.

Although many methods and reagents similar to or equivalent to those described herein can be used in the practice of the disclosed methods and compositions, the exemplary methods and materials are now described.

[ 0031 ] Industrial development in the last century has dramatically changed the emission of many gases that are critical to the health of human beings and the climate of earth, and it has become urgent to develop miniature gas sensors that can be deployed ubiquitously for monitoring the indoor and outdoor air quality. Among the major gas sensors, carbon dioxide (CO₂) sensors took the largest market share in 2012 (-25% of the global gas sensor market) , and the leading technology of CO₂ detection is based on nondispersive infrared (NDIR) spectroscopy. NDIR based CO₂ detectors offer several key merits including long lifespan (>2 years), fast response

(<2min) , and a ppm level of detection limit (LOD) ; however, its high unit cost ($100-1000) and bulky size make it very challenging to fit into consumer devices. Other CO₂ gas sensors based on

electrochemistry have principal advantages in unit cost, size and energy consumption, but suffer from short operation lifetime.

Furthermore, there are also challenges in the packaging process of such miniature gas sensors, where the package has to pass through the target gas while maintaining enough physical and chemical shield for the sensor at the same time.

[ 0032 ] Materials such as graphene offer good characteristics as the base sensing structures, such as large surface to volume ratio, low electrical noise, low power consumption and process compatibility with integrated circuits. Research has reported the ability to realize molecule-level detections using a graphene FET for high sensitivity as the emerging platform for chemical sensors. The most common sensing mechanism is the DC resistance measurement which correlates with charge variations on the graphene surface due to external gas vapors. The intrinsic slow process of the charge transfer and the adverse effect of defects on the graphene surface are two great challenges. For example, FET gas sensors made of CVD graphene with a film transfer process suffer from intrinsic slow dynamics in its interface trap states and defect-compensated charge transfer process. As such, the recovery speed of CVD graphene sensors is slow, especially at room temperature. For example, the recovery of DC electrical conductance could take more than 1000 s as vapors and charges are released from interface trap states/defects. The resulting baseline drift has been a bottleneck for practical applications. To address this issue, UV lights or heaters were previously used to help boost the recovery process, which inevitably increase the power consumption and system complexity.

[ 0033 ] The disclosure provides a class of sensing signals

comprising AC phase sensing on, for example, GFET gas sensors.

Instead of measuring the DC resistance that is vulnerable to trap states, the disclosure uses the highly stable phase change of the charge transfers as the sensing signal. The phase of the charge transfers process on the graphene surface, driven by the applied AC gate voltage inputs, can selectively bypass the slow

adsorption/desorption process from the trap states, yet still maintains good sensitivity to evaluate the charge transfer process from the gases.

[ 0034 ] The overall carrier (charge) concentration n_GFET of a GFET gas sensor is collectively determined by three parts at a given time: n_G+ n_c+ n_T, where n_G is the electrostatic charge

associated with gate voltage, n_c is the electrochemical charge associated with reactive gas adsorption, and n_T is the trapped charge associated with trap state. In principle, all the charges have their DC component (N_GFET, N_G, N_C, N_t) and AC component (n_gf_et , n_g, n_c, n_t) . A goal is to filter out n_t through the modulation of gate voltage v_G. At a proper frequency, the gate modulation will allow n_g and n_c to follow but quench n_t, making the AC sensing signal immune to the trap states. After some initial studies and measurements, the phase component of n_gf_et, φ (n_gf_et) , is chosen as a representative AC sensing signal for its superior stability.

[ 0035] To further illustrate the sensitivity of φ (n_gfet) towards gas concentration, the RC model of Faradic process was adapted to analyze the charge transfer process on GFET as shown in Figure 1. With n_t quenched, the overall phase $ (n_gf_et) = Φ (n_g+n_c) will approach to φ (n_c) as the gas concentration increases as I n_g | =C_G I v_g | where C_G being the gate capacitance. In the experiment, the phase signal φ(ζ_3£ι) was measured as the equivalent signal of φ (n_gf_et) , where z_sci is the AC component of the impedance between the source and drain electrodes Z_SD. All the symbols used are denoted in Table I.

[ 0036] List of Symbols

[ 0037 ] To alleviate the problems stemming from the trap states and defects on graphene, this scheme takes the advantages of the reversible and stable phase change signals instead of DC resistances. As shown in Fig. 1C, the phase lag between the channel resistance (point A and B between the source and drain of the FET) and the gate voltage is detected when an AC gate voltage at a moderate frequency is applied. Experimental results show that the phase lags of different vapors under various concentrations have fast recovery speeds in the ranges of 10 s, which are at least 10 times faster than those of DC resistance results with similar setups. Furthermore, the dynamic response of the phase lag is reversible with large dynamic range while the DC resistance tests suffer from baseline drift problems. Fig. ID illustrates differences between the AC and DC domain measurements, where the AC phase lag results are sensitive to the weak adsorption of vapor molecules above a distance to the graphene surface for fast gas adsorption and desorption processes, while the DC resistance results are sensitive to the strong adsorption and desorption process close to the graphene surface. These observations and analyses are explained by an analytical model with good match to the measurements.

[0038] The methods and systems described herein can be applied to sensor and sensor arrays comprising the same or different

composition of sensor material. In certain embodiments, the sensors comprises semiconductive material, or combinations of semiconductive material and conductive inorganic and/or organic material. In some embodiments, the sensor material comprises a semiconductive material interspersed in or between conductive or non-conductive regions. In certain embodiments, the sensing material is a semiconductor, e.g., metal oxide, silicon, germanium, graphene, M0S2 and the like in a FET structure. In other embodiments, the sensing material is metallic, e.g., Pt, Ag, carbon and the like in a three (reference, working, counter) electrode setup.

[0039] The sensor can be used in the identification of analytes, gases and the like in a portable, relatively inexpensive

implementation. Thus, a wide variety of analytes and fluids may be analyzed by the disclosed sensors, arrays and noses so long as the subject analyte is capable generating a detectable response across a sensor or a plurality of sensors of an array. Applications of the sensing methods and compositions of the disclosure include broad ranges of chemical classes such as organics including, for example, alkanes, alkenes, alkynes, dienes, alicyclic hydrocarbons, arenes, alcohols, ethers, ketones, aldehydes, carbonyls, carbanions, biogenic amines, thiols, polynuclear aromatics and derivatives of such organics, e.g., halide derivatives, etc., biomolecules such as sugars, isoprenes and isoprenoids, fatty acids and derivatives, etc. Accordingly, commercial applications of the sensors, arrays and noses include environmental toxicology and remediation, biomedicine, materials quality control, food and agricultural products

monitoring, anaesthetic detection, automobile oil or radiator fluid monitoring, breath alcohol analyzers, hazardous spill

identification, explosives detection, fugitive emission

identification, medical diagnostics, fish freshness, detection and classification of bacteria and microorganisms both in vitro and in vivo for biomedical uses and medical diagnostic uses, and the like. A wide variety of commercial applications are available for the sensors arrays and electronic noses including, but not limited to, environmental toxicology and remediation, biomedicine, materials quality control, food and agricultural products monitoring, heavy industrial manufacturing, ambient air monitoring, worker protection, emissions control, product quality testing, leak detection and identification, oil/gas petrochemical applications, combustible, gas detection, ¾S monitoring, hazardous leak detection and

identification, emergency response and law enforcement applications, illegal substance detection and identification, arson investigation, enclosed space surveying, utility and power applications, emissions monitoring, transformer fault detection, food/beverage/agriculture applications, freshness detection, fruit ripening control,

fermentation process monitoring and control applications, flavor composition and identification, product quality and identification, refrigerant and fumigant detection, cosmetic/perfume/fragrance formulation, product quality testing, personal identification, chemical/plastics/ pharmaceutical applications, leak detection, solvent recovery effectiveness, perimeter monitoring, product quality testing, hazardous waste site applications, fugitive;

emission detection and identification, leak detection and

identification, perimeter monitoring, transportation, hazardous spill monitoring, refueling operations, shipping container inspection, diesel/gasoline/aviation fuel identification,

building/residential natural gas detection, formaldehyde detection, smoke detection, fire detection, automatic ventilation control applications (cooking, smoking, etc.)_? air intake monitoring, hospital/ medical anesthesia & sterilization gas detection, infectious disease detection and breath applications, body fluids analysis, pharmaceutical applications, drug discovery and

telesurgery. Another application for the sensor-based fluid detection device in engine fluids is an oil/antifreeze monitor, engine diagnostics for air/fuel optimization, diesel fuel quality, volatile organic carbon measurement (VOC) , fugitive gases in refineries, food quality, halitosis, soil and water contaminants, air quality monitoring, leak detection, fire safety, chemical weapons identification, use by hazardous material teams, explosive detection, breathalyzers, ethylene oxide detectors and anesthetics.

[0040] Fig. 7A illustrates a schematic of the 3D chemical sensing methods and composition of the disclosure. Each molecule-surface interaction pair can be modeled as a DC voltage source (molecule) and an AC voltage source (sensing surface) connected by a series of charge transfer capacitor and charge transfer resistor. Here the voltage is referred as chemical energy, or work function of the molecule (steady, DC) and the sensing surface (modulated, AC) . For simplicity, the charge transfer resistance R(d) can be approximated as the tunneling resistance between molecular orbital and the surface electronic band, which scales with distance d exponentially, while the charge transfer capacitance C (d) can be approximated as parallel plate capacitance which scales with reciprocal distance d^"1 linearly. Therefore, the molecule-surface charge transfer time constant τ (d) =C (d) R (d) varies monotonically to the molecule-surface distance d (Fig. 7B) .

[0041] This means the single molecule charge transfer current i(a>,d) has an amplitude a(a>,d) and phase φ{ω,ά) depending on the d, and the overall charge transfer current from all molecules, 1{ω,ά), has an amplitude A(a>,d) and phase Φ(ω,d) also depending on the molecule number distribution c(d) .

[0042] Considering an evenly distributed molecule number profile c(d), one can group all the molecules into three categories according to their charge transfer time constants: a) r(d) much smaller than ω (strong interaction), b) r (d) relatively close to ω (weak interaction) , and c) r (d) much bigger than ω (no interaction) . The overall charge transfer current amplitude Α(ω,ά) is most sensitive to the molecules in the first category where molecule- surface charge transfer interaction is strong dominating the current amplitude, however, the phased of overall charge transfer current Φ(ω,ά) is most sensitive to the molecules in the second category where molecules have both moderate phase (-45) and amplitude instead of small phase with large amplitude (category 1) , or large phase but small amplitude (category 3) . In other words, the overall phase response is only dominated by a slack of molecules within certain range of distance from the sensing surface. When molecule number profile c(d) is not even, and the phase of charge transfer current Φ(ω,ά) is proportional to the gas concentration (molecule number) at distance d. By sweeping the gate modulation frequency ω and

monitoring the overall charge transfer current, a 3D map of molecular number at different distance can be derived through the overall phase response Φ(ω,ά) .

[0043] The alternating work function of the sensing material can be achieved by a) applying alternating gate voltage in a semiconductor FET structure, or b) applying alternating voltage on the working electrode in a 3 electrodes structure.

[0044] Figs. 8 and 9 show the experimental results with both conventional 2D sensing (resistance response) and the 3D Chemical Sensing (phase response) achieved on a graphene FET in CO₂ and VOC

(methanol) respectively. In one embodiment, alternating voltage is applied on the gate electrode of graphene FET to create an

alternating work function of the graphene sensing layer. In Fig. 8, the CO₂ sensing response speed (recovery time) is 10 times faster with 3D Gas Sensing compared to 2D sensing, and similarly in Fig. 9, the baseline level of 3D Gas Sensing has minimal drift compared to 2D sensing in VOC vapor (methanol) .

[0045] Fig. 10, for example, shows the potential of 3D Gas Sensing in mapping the chemical concentration above the sensing plane. Here, methanol vapor concentration is mapped using a graphene FET. By sweeping the gate AC voltage frequency from lOHz to 40Hz, the collected phase lag reveals the concentration of gas molecules at different distance away from the sensing surface. The dotted line illustrates the molecular concentration contour in real time and reveals the dynamics of molecular diffusion along the z axis

(vertical to the sensing plane) over time.

[0046] The AC impedance measurement in a GFET gas sensor offers many advantages over the DC resistance signal. Besides the immunity to trap state, one of the other advantages is that it enables the in situ reading of field effect mobility μ and DC carrier concentration N_GFET in high definition of time. Previously, the linear slope of the N_GFET versus μ^"1 plot was used to label the charge transfer process from different parties. Here, the same method was used to

demonstrate the power of AC sensing by revealing multiple hidden stages of the charge transfer processes embedded in the moisture sensing result.

[0047] Figs. 5C-H show the different carrier scattering process when 1) charges are directly transferred from gas molecules to graphene (Fig. 5c) , and 2) charges are transferred from gas

molecules to trap states (Fig. 5F-H) . Fig. 6A shows the high resolution (dt=0.1s) transient response of the field effect mobility μ and the DC carrier concentration N_GFET measured from the same run shown in Fig. 5A. Multiple peaks and valleys emerged in the two curves in the period of t=10-320s, which were further broken and labeled into five linear slopes in Fig. 6B. For stage I during t=10-12s, it is believed that the quick jump in μ and N_GFET were due to the rapid changes of the electrostatic screening environment when introducing water vapor in contrast to the nitrogen background. For stage II during t=14-30s, direct charge transfer (p-type doping) between adsorbed water molecule and graphene took place, followed by the charge transfer (n-type doping) between adsorbed water molecule and trap state in stage III (t=30-225s) . Obviously, the charge transfer in stage III had an impact on μ and N_GFET, and it shaped the slope of NG_FET-U^'1 plot into a dramatically different direction compared to that of stage II. After nitrogen was introduced at t=225s, the water molecules started to desorb. In stage IV (t=225- 228s), water molecules that interacted with graphene were fully desorbed from the surface of graphene, while in stage V (t=228- 320s), the water molecule that interacted with the trap states were desorbed partially. Moreover, the stages II & IV and stages III & V were of the inverse process pair were verified by plotting the N_GFET versus Δμ^"1 curve in the inset of Fig. 6B, showing each pair sharing similar slope but opposite direction. Interestingly, the inverse process of stage I (screening effect of water molecules on graphene) was missing in the plot, and this was probably due to the fact that there was still water molecules adsorbed on graphene at t=320s.

[0048] Moreover, the relationship between phase lag and the RC time constant of the model can be derived. Specifically, the model uses an effective distance between the adsorbed gas molecule and graphene surface to determine the values of R and C and the phase lag associated with the distance between gas molecules and graphene surface reflects the adsorption strength.

[0049] Each molecule-surface interaction pair is modeled as a DC voltage source (molecule) and an AC voltage source (sensing surface) connected by a series of charge transfer capacitor and charge transfer resistor. The approaching process of the gas molecule onto the graphene surface is illustrated in FIG. 17A with the RC model shown in FIG. 17B. The carriers in the channel will flow back and forth under the modulation by the applied AC gate voltage. The capacitor C has a phase lag between the channel resistance R_AB and the gate voltage V_g and it is defined as: <p(cot) . Analytically, an adsorption process with a large time constant would induce a smaller amplitude of phase lag in the RC model. The strength of adsorption process is distinguished by the parameter d which represents the distance between the gas molecule and graphene surface. The resistance is modeled as a tunneling resistance R ~ θχρ(Αψ°-⁵ά), where A is a constant with the value of 1.025 A^_1eV^"0-⁵, and ψ is the difference in electron affinity between the gas and graphene. The capacitance can be expressed as C~ log(l + r_mo/d) using the infinite conducting plate model, where r_mo is the diameter of an individual gas molecule. Thus, the time constant of the charge transfer process between graphene and adsorbed gas molecules can be analyzed. For weakly adsorbed gases which have small time constants during the gas adsorption process, a large phase lag induced in R_AB is expected. This explains the fast recovery time of the phase responses for weakly adsorbed molecules. Specifically, the phase responses at high frequency are used to target/sense gas molecules in a short distance (as compared to those very close) to the graphene surface.

[0050] To simulate the transient response of phase response, one would calculate the induced phase change from molecules weighted by the sticking probability, then evaluate the integral responses of all adsorbed molecules by accumulating over the relevant range of d < 30 r_mo . The profile of sticking probability of gas molecules on graphene surface agrees with the molecular number density

distribution along the direction perpendicular to graphene given by molecular dynamics (MD) method analysis. For simplicity, error functions were used to express the probability in the adsorption and desorption periods as p_a(t) = 0.5 - 0.5 erfc (a_adt^~°- ⁵r_mo ^_1) and pb (t) = 0.5 erfc ( ddt^-0- ⁵r_mo ^_1) , respectively, where t is time and _a,d are the fitting constants. Two representative profiles of the sticking probability and their phase lag responses are plot-ted in FIG. 17C and FIG. 17D, respectively. Here, d/r_mo= 2 is used to represent the strongly adsorbed gases very close to the graphene surface which are expected to have strong adsorption and weak desorption properties, while d/r_mo= 4 (or larger) is used to rep-resent the weakly adsorbed gases in a short distance away from the graphene surface which are expected to have slow adsorption and fast desorption properties. The representative responses of the sticking probability p in Fig. 10c show strong and fast responses during the adsorption process for gases close to the graphene sur-face (d/r_mo= 2) while during the desorption process, the gases at a short distance away to the graphene surface (d/r_mo= 4) have faster responses during the desorption process. Furthermore, the small dip in the phase lag testing results (such as those in FIG. 12B for high concentration methanol) in the adsorption process can be predicted by the model. Analytically, the initial protuberance of the phase lag results during the adsorption process for a short distance away from the graphene surface can be attributed to the large accumulations of gas molecules to result in the slow desorption process. This behavior is simulated numerically as shown in FIG. 17D for the case of d/r_mo= 4 curve. Nevertheless, the recovery speed for the d/r_mo= 4 curve is much faster than that for the case of the d/r_mo= 2 curve. These characteristics agree well with the experimental results in FIG. 14.

EXAMPLES

[0051] Fabrication Process of the GFET Gas Sensor. The monolayer graphene was purchased from Graphenea Inc, which was grown via a CVD method and transferred on to a silicon wafer with 300nm thermally grown S1O2 layer on top. Then the following fabrication steps were carried out: 1) the lithography and lift-off process on graphene to define the source and drain electrodes (Pd/Au 5nm/30nm) ,

alternatively, electrodes and contact pads (30 nm Pd/25 nm Au) were pat-terned on graphene by the lithography and lift-off process; 2) lithography and oxygen plasma etching (50W, 7s) to define the channel of graphene; and 3) wire-bonding to the custom build PCB. Optical microscopic pictures and SEM photo of the fabricated device are shown in Figs. 2A-B. To characterize the defects on graphene, Raman Spectroscopy was used to examine the quality of graphene before and after the fabrication process, and found that the D peak (signature of defects) emerged after the fabrication process as compared to the graphene received from the vendor, indicating that the lithography and etching process do create additional defects on graphene. These defects can result in slow response speed of gas detection as discussed above. The schematic diagram of the test setup is shown in Fig. 2D-F. Fig. 2E illustrates the electrical configuration of the test setup, where the four-point probe method was used to avoid the influence of contact resistance in the measurement. An AC voltage was applied on the gate electrode with a frequency, f_gi ranging from 50 to 1000 Hz. A lock-in amplifier (SRS 860, Stanford Research Systems) was used to reduce the noise level at the source electrode with a reference of 20 kHz AC current (2 μΑ) through the graphene channel and a 1 ΜΩ resistor. The voltage between electrode A and B was measured by the lock-in amplifier. A data acquisition device (PicoScope 5242B) was used to collect V_AB and V_G simultaneously. With this setup, the phase lag between V_AB and V_G can be extracted by fast Fourier transform (FFT) . Fig. 2F

illustrates the gas connections as all gas sensing measurements were carried out in a sealed nylon chamber (2 cm ^χ 2 cm ^χ 2 cm) with a waste gas treatment system. The concentration of a specific vapor was applied via the split-stream system with dry nitrogen (or dry air) and gas-saturated nitrogen (or air) . The flow rates of both gas paths were regulated by mass flow controllers (Omega Engineering) and the total flow rate to the gas sensor was controlled at 200 seem. To ensure no leakage around the seals, the flow rate at the outlet of the chamber was monitored by a mass flow meter (Omega Engineering) . All experiments were carried out in a laboratory with general ventilation. The environmental temperature was 24°C with fluctuations within l.C. The chemicals (DI water, anhydrous

methanol, and anhydrous ethanol)were purchased from Sigma Aldrich, and inert gas (nitrogen) and dry air were purchased from Praxair Inc .

[0052] Characterization of GFET with AC Modulation. The admittance between the source and drain electrode, 1/Z_SD, was characterized with electron carrier and hole carrier under a nitrogen environment

(setting n_c to zero) . To switch the carrier polarity, the gate voltage v_G was biased with a DC voltage V_G (-50-50V) and an AC small voltage |v_g|=lV (peak-to-peak) in serial. The modulation frequency was selected to be 50Hz in order to fit in the time constant in charging the gate capacitance (~us) and the gas adsorbates (~ms) , and filter out the time constant of trap state (~s) . As shown in Fig. 3A, the overall profile of the admittance 1/Z_SD between the source and drain electrode has a "V" shape separating the curve into two parts at the point of minimal conductance (Dirac Point) , with the electrical current of the left part carried by holes and the right part carried by electrons. Two regions from each carrier regime were further highlighted: 1/Z_SD oscillates with v_g in an opposite manner (Fig. 3B) , while it follows in phase with v_g when the carrier in graphene switches from hole to electron (Fig. 3C) . It is important to note that there was a drift in the profile of 1/Z_SD

(downward in Fig. 3B and upward in Fig. 3C) , which was likely due to the charging process of the trap states, N_T, driven by the sweeping V_G.

[0053] The Stability of Phase Signal †(z_sd). With both AC voltage v_g and DC voltage V_G applied on the gate, multiple electrical parameters can be measured simultaneously, including the AC

impedance

the DC resistance Z_SD=V_SD/ISD, the DC carrier concentration

as well as the field effect mobility (e-NcFET-Vso) · As shown in Fig. 4A, φ(ζ_3£ι) stabilized at 0 consistently when V_G was between 0V to 25V and jumped to 180 after V_G passing through the Dirac Point (25V), while |z_scil shows a highly nonlinear dependency with V_G. Fig. 4B illustrates the dependence of the field effect mobility

and DC carrier concentration with respect to V_G , respectively, where both quantities inevitably diverged from their physical value as z_sci vanishes to zero near the Dirac Point. Among these signal

candidates for gas sensing, it is found that the phase signal φ(ζ₃₀;) demonstrated the highest level of stability over V_G tuning and immunity over disturbance from trap state, therefore we further adopted φ(ζ₃₀;) as the primary sensing signal in the subsequent sensing event. As expected, even under the influence of a trap state, the phase signal φ(ζ₃₀;) was only responsible to the charge modulation from the gate electrode and maintained steadily at either

0 or 180, indicating the effectiveness of rejecting the slow charge transfer process with trap state at 50Hz. Furthermore, the SNR of the phase signal in the setup is highly related to the amplitude

1 v_g I as shown in Fig. 4C, therefore |v_g|=lV was chose to achieve a measurement noise within 0.2 degree of phase without sacrificing the small signal nature of v_g.

[0054] Performance of Phase Signal †(z_sd) in Moisture Sensing. The gas sensing experiment was carried out by sealing the GFET in a custom built nylon chamber (lmL), and injecting the moisture vapor at various relative humidities (RH) at a constant flow rate

(200sccm) . A nitrogen flow (200sccm) was used to purge away the residual moisture vapor in the recovery stage of the GFET sensor. As shown in Fig. 5A, there were three consecutive periods in a moisture sensing event: GFET exposure to nitrogen (t=0-10s), to RH=60 vapor (t=10-225s), and to nitrogen (t=225s-320s) .

[0055] By comparing the transient response of the conventional DC resistance signal Z_SD and the AC phase signal φ(ζ₃₀;) in these three stages, two striking differences were found: 1) the phase signal φ(ζ_3£ι) quickly saturated in moisture vapor starting at t=50s, while the DC resistance signal Z_SD kept drifting at a relative constant speed over the second stage; 2) the phase signal φ(ζ₃₀;) recovered to its baseline within the first 100s after introducing nitrogen into the chamber, while the Z_SD failed to recover within the same period. It is believed both the slow drift and the stagnant recovery of Z_SD was caused by the charge N_T associated with trap state S. It is worth noting that the DC resistance signal Z_SD was able to recover to its original baseline after sitting in nitrogen flow overnight. The sensitivity of φ(ζ_3£ι) towards different RH levels was also

characterized and found that the amplitude of phase change was indeed sensitive to the concentration of water vapor in the range of RH=10-60 as shown in Fig. 5B.

[0056] Presented differently, the responses of the AC phase lag change Acp_AB,_G and the changes of DC resistance AR/Ro are plotted under multiple sensing cycles of water, methanol, and ethanol vapors, respectively. Fig. 11A shows the inputs of vapor concentrations increasing from 10% to 90% stepwise and each cycle includes 40 s chemical vapor feeding followed with 160 s nitrogen purging. Fig. 11B shows the dynamic response of the phase lag at 1000 Hz, with minimal drift and the average recovery time is around 10 s. Fig. 11C shows the DC resistance changes which apparently have significant baseline drift issues as compared with phase lag responses.

Furthermore, the average recovery time is larger than 100 s. In principle, the channel resistance decreases after the exposure to the p-type dopant molecules as the device is working in the hole branch of the graphene for all three p-type chemical vapors (water, methanol, and ethanol) . After long-term exposures in high

concentration vapors, the drop of the charge mobility in the graphene channel becomes significant to pull up the baseline resistance. On the other hand, the phase lag responses in Fig. 11B are reversible upon the nitrogen purge process at room temperature as the phase lag response is immune to the strong adsorption gas reactions. For the three tested vapors, the amplitudes of DC resistance responses become saturated when the vapor concentrations are larger than60% while the phase lag detections can still operate with up to 90%of water or methanol, and 80% of ethanol.

[0057] The dynamic responses of phase lag upon exposures to water, methanol and ethanol vapors in Fig. 12A-C, respectively, at

exemplary frequencies were studied at 50 Hz, 100 Hz, 500 Hz andlOOO Hz in nitrogen environment. It was observed that the amplitude of the phase lag change decreases for water vapor as the frequency increases, while the recovery responses at different frequencies do not have a large differences at 20 s. For methanol and ethanol, the amplitude of the phase lag change decreases and the recovery time reduces as the frequency increases in general. Specifically, the recovery time for 90% ethanol decreases from 80 to 30 s as the frequency increases from 50 Hz to 1000 Hz. These results imply that the recovery time can be reduced by using high frequency AC phase lag tests. The same experiment in nitrogen environment were carried out with the device stored in air atmosphere for three months to evaluate the stability of the gas sensors. The sensing results illustrate slight variations of 2%, 3% and 5% for water, methanol and ethanol vapors, respectively with good reproducibility.

[0058] Further experiments were performed under different

atmosphere conditions and the sensor responses at exemplary frequencies of 50 Hz, 100 Hz, 500 Hz and 1000 Hz upon exposures to water, methanol and ethanol vapors are shown in Fig. 12D-F in the air environment. The resulting sensing patterns are similar to those responses in the nitrogen environment, which indicates the sensor is partially inert to the presence of the oxygen contents at room temperature. The performances are summarized and compared in Fig. 13. In Fig. 13A-C, it was found that the response amplitudes in the air environment at 1000 Hz reduce ca. 5% when compared with those in the nitrogen environment and the average relative amplitude differences increase to ca. 10% as the frequency goes down to 50 Hz. The 90% gas sensing recovery time for water, methanol and ethanol with 90% concentration are summarized in Fig. 13D-F. In general, the 90% gas sensing recovery time remain at about 20 s for water vapor while for methanol and ethanol vapors, they are reduced as the sensing frequency increases. Clearly, both the average response amplitude and recovery time decrease at higher frequency and there is a tradeoff between fast recovery speed and large sensing responses. To keep the 90% sensing recovery time to be within 40 s, the operating frequencies for water, methanol and ethanol vapors should be higher than 50 Hz, 1000 Hz and 500 Hz, respectively. [0059] Representative sensing results of AC phase lag spectra upon exposures to three chemical vapors are recorded in the frequency versus time plot, including data for 60% (Fig. 14A-C) and 90% (Fig. 14D-F) gas concentrations of water, methanol, and ethanol,

respectively. In these tests, the chemical vapors are injected into the system and nitrogen purge processes are conducted 50 s, and 90 s, respectively, after the start of the recording process. These phase lag spectra clearly highlight three important features. First, the recovery speed of the AC sensing scheme at high frequency is faster than those at low frequency as the dark color regions (no phase change) reappear faster during the nitrogen purge process. The nitrogen purge process helps the recovery and cleaning of the graphene surface, while weakly adsorbed gases (a short distance away from the graphene surface) can be removed easier than strongly adsorbed gases (close to the graphene surface) . The conventional DC sensing signals are dominated by the desorption reaction for gases close to the graphene surface that is slow at room temperature.

Second, the signal strength of the AC sensing scheme is smaller at high input frequency as the responsive gases are a short distance away from graphene surface. This can be considered as a tradeoff for the fast recovery speed by using the AC sensing scheme as it is more sensitive to the weakly adsorbed gases. Third, negative phase changes may occur under high concentrations of gases as observed in the case of 90% methanol under low frequency sensing in Fig. 12B. The negative values are not presented in Fig. 14E and represented as zero phase change but the trend is also clearly observed. This is believed to be the accumulation of gas molecules at a short distance above graphene surface near saturated conditions that are difficult to be cleaned. This phenomenon can greatly damage the recovery speed as shown at the low frequency range while the recovery speed at high frequency (such as 1000 Hz) still maintains within 40 s in the testing results. Furthermore, it is potentially reasonable that different gases can induce different adsorption processes on the graphene surface such that the phase change results can be used as a feature to selectively sense the gas types. The slope of the phase change data could also correlate to the dynamic details of the adsorption process. The characteristics of phase change at different frequencies can be differentiated visually while specific machine learning or big data analyses tools could be applied for better and comprehensive identifications. The response amplitude decreases as the frequency increases from 1 kHz to 5 kHz which is in accordance with the trend observed in the previous lower frequency testing results. However, results from 5 kHz to 10 kHz are similar (or even show amplitude increases in the case of 80% methanol) implying possible deviations from the proposed simplified model at ultra-high frequency .

[0060] AC sensing measurements under different humidity conditions from 20% to 60% RH were also performed by continuously feeding the water vapor into the chamber for 200 s before the input of target chemical vapors. All the experiments were carried at room

temperature (24°C) . Dynamic responses of the phase lag changes are shown in FIG. 15 at exemplary frequencies of 50 Hz and 1000 Hz for ethanol under the 60% RH condition in air at room temperature. In general, the response amplitudes are enhanced as the RH level increases as shown in FIG. 16A for methanol vapors and FIG. 16B for ethanol vapors, as the result of the nonlinear response for each chemical vapor as illustrated in FIG. 13A-C. The non-linearity increases sharply when the concentration of humidity is greater than 40%. With accumulated water molecules as back-ground, the

incremental chemical vapor molecules will result in a larger response amplitude than those in the dry air condition.

Interestingly, the phase change responses upon exposure to the RH background will mostly saturate within the first 50 s as shown in FIG. 15, such that one can clearly indicate the humidity level by comparing the sensing results with the initial reference sensing results in the dry air condition. In other words, the concentration of the target chemical vapor can be identified as the enhanced responses as the background RH level is first determined and extracted by the initial phase change results. Meanwhile, the 90% gas sensing recovery time increases as the RH level increases as shown in FIG. 16C for methanol and FIG. 16D for ethanol vapor with saturated concentrations, respectively, as the gas desorption process is suppressed by the background humidity when compared with the dry air condition. [ 0061 ] To further validate the selective sensing philosophy, the multivariate analysis of the phase change at different frequencies was performed using PCA. In this work, the initial 15-second data of phase spectra during each chemical vapor feeding cycle were projected onto a subspace of lower dimensionality and presented as the weighted sums of the original responses. Finally, the weighted sums were visualized as principal component (PC) scores plots.

Results of phase spectra PCA analysis are shown in FIG . 18 with the darker spots referring to higher concentrations. The PCI and PC2 provide strong contribution (95.9%) and discriminate between four vapors at their medium and high concentrations. In FIG . 18B, the PC3 also provides contributions correlated with vapor concentrations. These results illustrate the gas sensing selectivity provided by the phase spectra sensing method. Such discrimination ability using only a single CVD graphene FET is demonstrated here for the first time.

[ 0062 ] A number of embodiments have been described herein.

Nevertheless, it will be understood that various modifications may be made without departing from the spirit and scope of this disclosure. Accordingly, other embodiments are within the scope of the following claims .

Claims

WHAT IS CLAIMED IS :

1. A method of operating a chemical sensor, comprising:

applying a variable voltage to a gate of a transistor that includes a sensing material in contact with a medium in a gaseous state; and

monitoring a phase of a charge transfer current between the sensing material and the gas while applying the variable voltage.

2. The method of claim 1, wherein the transistor is a

field-effect transistor.

3. The method of claim 2, wherein the field-effect transistor is a graphene field-effect transistor.

4. The method of claim 1, wherein the sensing material is included in a channel of the transistor.

5. The method of claim 4, wherein the sensing material is graphene .

6. The method of claim 1, further comprising:

using the phase of the charge transfer current to measure a molecular concentration in the medium.

7. The method of claim 1, wherein the variable voltage is an alternating current (AC) voltage.

8. The method of claim 7, wherein the alternating current (AC) voltage has a frequency of 0. lHz to 1 MHz.

9. The method of claim 7, further comprising:

sweeping a frequency of the AC voltage while monitoring the phase of the charge transfer current.

10. The method of claim 9, further comprising: using the phase of the charge transfer current to determine a molecular concentration of the medium at one or more non-zero distances above a surface of the sensing material.

11. The method of claim 9, further comprising:

using the phase of the charge transfer current to determine a molecular concentration of the medium as a function of a distance above a surface of the sensing material.

12. The method of claim 1, wherein monitoring includes measuring.

13. A method of operating a chemical sensor, comprising:

monitoring a phase lag between the voltage of the gate and a resistance of the sensing material while applying the voltage.

14. The method of claim 13, wherein the transistor is a

field-effect transistor.

15. The method of claim 14, wherein the field-effect transistor is a graphene field-effect transistor.

16. The method of claim 13, wherein the sensing material is included in a channel of the transistor.

17. The method of claim 13, wherein the sensing material is graphene .

18. The method of claim 13, further comprising:

using the phase lag to measure a molecular concentration in the medium.

19. The method of claim 13, wherein the variable voltage is an alternating current (AC) voltage.

20. The method of claim 13, further comprising:

using the phase lag to determine a molecular concentration of the medium at one or more non-zero distances above a surface of the sensing material.