Multi-sensor array, multi-sensor system, and related methods

Abstract

A multi-sensor array is described comprising a plurality of sensors for detecting one or more analytes in a gas phase or a liquid phase, where the plurality of sensors comprises: a first sensor configured to detect a change in an electrical parameter based on a dipole-dipole interaction with the analytes; a second sensor configured to detect a change in an electrical parameter based on hydrogen bond interaction with the analyte; and a third sensor configured to detect a change in an electrical parameter based on a dispersion interaction with the analyte. Further, a multi-sensor system is disclosed comprising the aforementioned multi-sensor array and a processing unit comprising a pattern recognition and machine learning analyzer, where the pattern recognition and machine learning analyzer is configured to receive an output signal of the multi-sensor array and to compare it to stored data, to classify and / or identify one or more analytes; wherein the stored data comprises Hansen solubility parameters obtained from a class library database, thereby enabling classification of unknown analytes. Methods of analyzing a sample present in a gas or liquid phase by using the above multi-sensor array are also described.

Description

Technical Field

[0001] The present invention generally relates to the field of multi-array sensors for classifying, classifying and / or identifying one or more analytes in a gas or liquid phase.

[0002] In certain embodiments, the present invention relates to a multisensor array for detecting one or more analytes in a gas phase or a liquid phase and a multisensor system comprising the multisensor array. In other embodiments, the present invention relates to a method for analyzing a sample present in a gas phase or a liquid phase by using the aforementioned multisensor array. Background Art

[0003] A chemical sensor is a device that converts chemical information (ranging from the concentration of a specific sample component to an overall compositional analysis) into an analytically useful analog or digital signal. To this end, such a device comprises a sensing material (or receptor) and a transducer as basic components. For example, WO 01 / 44796 describes a nanotube device for use as a chemical or biological sensor, the nanotube device comprising at least one nanotube (e.g., a carbon nanotube) whose ends are electrically connected to first and second conductive elements, such as electrodes. In order to tune the sensitivity of the device to various molecular species, the nanotubes can be modified by coating or decorating with one or more sensing agents to impart sensitivity to specific substances in their environment.

[0004] For example, multisensor arrays have found widespread use in the detection and identification of chemical compounds in the chemical industry, medicine, environmental monitoring, security, military applications, and public use.

[0005] Conventional multisensor arrays typically employ receptors chosen to specifically bind to the analyte of interest. While such a "lock and key" design may be advantageous in terms of its selectivity for the analyte, its application does not allow for the identification of similar chemical substances or classes of compounds, and the necessary custom synthesis of the receptor materials is often complex and associated with high costs.

[0006] Other examples of multi-sensor arrays include so-called electronic noses (e-noses) and electronic tongues for sensing analytes in the gas and liquid phases, respectively, which typically combine chemical sensing with pattern recognition systems (see, e.g., US 2017 / 199159 A1).

[0007] In addition, US2008 / 050839 A1 and US2019 / 317079 A1 disclose a method for identifying an analyte, which comprises: receiving image data from a sensor array that has been exposed to an analyte and is configured to respond to the presence of the analyte, processing the image data to derive one or more input image features, and using a trained machine learning (ML) classification technique to detect and classify the analyte based on the one or more input image features.

[0008] However, in the above-mentioned methods, machine learning techniques such as neural networks and statistical pattern recognition are used to process and analyze data received from the sensor array by comparing it with a database established by previously measuring defined analytes as a reference. Therefore, in order to identify and / or classify unknown analytes, the ML algorithm must be extensively trained on the same feature vectors to pre-establish the required reference database, and analytes not recorded in the database tend to remain unidentified, even if they are similar in chemical properties or composition.

[0009] In view of the above, it remains desirable to provide cost-effective multi-sensor arrays and methods that enable rapid detection and / or classification of one or more analytes in the gas or liquid phase without relying exclusively on extensive data generation and training.

[0010] At the same time, it would be desirable to provide a multi-sensor device that can be manufactured in a compact and lightweight form suitable for field deployment (e.g. as a handheld device) and that can be used by untrained individuals without specific methodological knowledge or analytical skills. Summary of the Invention

[0011] The present invention solves this object with the subject matter of the claims as defined herein.The advantages of the present invention will be explained in further detail in the following sections and further advantages will become apparent to those skilled in the art upon consideration of the present disclosure.

[0012] Generally speaking, in a first aspect, the present invention provides a multi-sensor array comprising a plurality of sensors for detecting one or more analytes in a gas phase or a liquid phase, wherein the plurality of sensors comprise: a first sensor configured to detect a change in an electrical parameter based on a dipole-dipole interaction with the analyte; a second sensor configured to detect a change in an electrical parameter based on a hydrogen bond interaction with the analyte; and a third sensor configured to detect a change in an electrical parameter based on a dispersed interaction with the analyte.

[0013] In a second aspect, the present invention provides a multi-sensor system for classifying and / or identifying one or more analytes in a gas phase or a liquid phase, comprising: a multi-sensor array according to the first embodiment; and a processing unit comprising a pattern recognition and machine learning analyzer, wherein the pattern recognition and machine learning analyzer is configured to receive an output signal of the multi-sensor array and compare the output signal with stored data to classify and / or identify the one or more analytes; wherein the stored data comprises Hansen solubility parameters obtained from a library database.

[0014] In a third aspect, the present invention relates to a method for analyzing a sample present in a gas phase or a liquid phase, comprising the following steps: a) introducing the sample into a multi-sensor array according to the first embodiment described above; and b) applying a pattern recognition and / or machine learning algorithm to data received from at least one of the first sensor, the second sensor, and the third sensor to classify and / or identify one or more analytes in the sample; wherein, in step b), the received data is correlated with Hansen solubility parameters obtained from a class library database. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is the Teas diagram that projects the three-dimensional Hansen solubility parameters onto the triangular space. Figure 2 Exemplary detector channels as part of a multi-sensor array are schematically depicted.

[0016] Figure 3A A top view of a channel system having a Hilbert curve shape is schematically illustrated.

[0017] Figure 3B Schematically illustrates the Figure 3A A sensor configuration with functionalized nanotubes randomly deposited thereon.

[0018] Figure 4 An exemplary configuration of a multi-sensor system according to the present invention is schematically illustrated. DETAILED DESCRIPTION

[0019] For a more complete understanding of the present invention, reference is now made to the following description of illustrative embodiments thereof:

[0020] Multi-sensor array

[0021] In a first embodiment, the present invention generally relates to a multi-sensor array comprising a plurality of detector channels for detecting one or more analytes in a gas phase or a liquid phase, wherein the plurality of sensors comprises: a first sensor configured to detect a change in an electrical parameter based on a dipole-dipole interaction with the analyte; a second sensor configured to detect a change in an electrical parameter based on a hydrogen bonding interaction with the analyte; and a third sensor configured to detect a change in an electrical parameter based on a dispersed interaction with the analyte.

[0022] Advantageously, by functionalizing the sensor to respond to non-bonded intermolecular forces, including dipole-dipole interactions (first sensor), hydrogen bonding interactions (second sensor), and London dispersion forces (also known as transient dipole-induced dipole forces) (third sensor), the sensor data can be correlated to Hansen solubility parameters.

[0023] The Hansen solubility concept (see C. Hansen, "The Three Dimensional Solubility Parameter and Solvent Diffusion Coefficient and Their Importance in Surface Coating Formulation". PhD Thesis 1967, Copenhagen: Danish Technical Press) was originally developed as a method to predict whether a material dissolves in another material based on the idea that if the parameters are similar, good solubility is established. For this purpose, each molecule is given three Hansen parameters, each parameter is usually expressed in MPa at 298.15 K. 0.5 (equivalent to joules / cubic centimeter; 2.0455×(cal / cc) 1 / 2 )Measured by:

[0024]

[0025] Here, δ D , δ P and δ H Represent dispersion interactions, polar interactions, and hydrogen bonding interactions, respectively. These three parameters can be viewed as the coordinates of a point in three dimensions (also known as Hansen space). The closer the Hansen parameters of two substances are in this three-dimensional space, the more likely they are to dissolve in each other. To calculate the distance (R) between Hansen parameters in Hansen space, a ), using the following formula:

[0026] (R a ) 2=4(δ D1 -δ D2 ) 2 +(δ P1 -δ P2 ) 2 +(δ H1 -δ H2 ) 2 (Formula 2).

[0027] In this paper, δ D1 , δ P1 and δ H1 represent the dispersion interaction, polar interaction and hydrogen bond interaction of the first substance, respectively, and δ D2 , δ P2 and δ H2 represent the dispersion interaction, polar interaction and hydrogen bonding interaction of the second substance, respectively.

[0028] The molecules in the fluid phase generally interact via such intermolecular forces, and each compound class has functional groups that contribute to one or the other of the forces and provide an indication of the molecule or compound class (e.g., water, hydrocarbons, alcohols, ethers, esters, chlorinated liquids, nitrogen compounds, aromatic hydrocarbons, or fluorinated liquids), as indicated by Figure 1 For example, Figure 1 Depicted is a Teas fractionation solubility diagram showing solubility parameters for common solvents and solvent families with similar properties, where F h represents hydrogen bond force, F p represents polar force, and F d The dispersion force is expressed in [%].

[0029] An extensive database of Hansen solubility parameters has been compiled for a large number of compounds (see Hansen Solubility Parameters: A User's Handbook, 2007 2 nd Ed., Boca Raton, CRC Press), and the solubility parameters of an unknown solvent or solute can be calculated by using the software "Hansen Solubility Parameters in Practice (HSPiP)" (available at https: / / www.hansen-solubility.com / buy-HSPiP-software.php).

[0030] Exemplary Hansen parameters for a number of selected analytes are shown in Table 1 below (data retrieved from Hansen Solubility Parameters: A User's Handbook, 2007 2 ndEd.,Boca Raton,CRC Press)

[0031] Table 1: Hansen solubility parameters of selected analytes.

[0032]

[0033]

[0034]

[0035]

[0036] The present invention advantageously exploits the fact that upon exposure to a specific analyte substance, the analyzable electrical parameters of each sensor are altered and produce patterns that can be analyzed, and the produced patterns can be correlated with an already available database of tabulated Hansen solubility parameters, allowing identification of classes of substances, potentially even specific substances, without the need for exhaustive prior training of pattern recognition algorithms using similar feature vectors.

[0037] The expression "electrical parameter" as used herein refers to the electrical signal measured when each sensor is exposed to a test sample and is generally selected from the group consisting of resistance, conductance, alternating current (AC), capacitance, impedance, inductance, potential and voltage threshold.

[0038] According to the present invention, the first sensor is configured to detect changes in electrical parameters based on dipole-dipole interactions with the analyte, the second sensor is configured to detect changes in electrical parameters based on hydrogen bond interactions with the analyte; and the third sensor is configured to detect changes in electrical parameters based on dispersion interactions with the analyte.

[0039] Each function is achieved by a different substituent or compound chemically bound or physically fixed to the solid support, and the substituent or compound can be independently selected for each sensor of the array.

[0040] The solid support of the sensor array can be based on silicon (e.g., polycrystalline and single crystal silicon, silicon dioxide, silicon nitride, silica, glass, controlled pore glass, silica gel), metals, metal alloys or metal oxides (e.g., gold, platinum, silver, titanium, vanadium, chromium, iron cobalt, nickel, copper, palladium, aluminum, gallium, germanium, indium, zinc and their oxides, alloys (such as cadmium telluride and copper-indium-selenide)), synthetic or natural polymers, carbon (e.g., glassy carbon, carbon paste, graphite, carbon nanostructures).

[0041] In particularly preferred embodiments, one or more than one of the first, second and third sensors, and more preferably all three sensors, comprise functionalized carbon nanostructures.

[0042] As used herein, the term "carbon nanostructure" includes single-walled carbon nanotubes, multi-walled carbon nanotubes, amorphous carbon, 2D-3D graphene structures, fullerites, fullerene molecules, buckyballs, carbon nanowires, carbon nanowhiskers, carbon nanohorns, carbon nanocones, carbon nanofibers, carbon nanobelts, carbon nanospheres, carbon nanorods, carbon nanobowls, carbon peapod structures, carbon nanoonions, carbon nanotori, carbon nanobuds, or any combination thereof. From the viewpoint of ease of synthesis and functionalization, carbon nanotubes (CNTs) or graphene are preferably used as the carbon nanostructure.

[0043] For the purposes of the present invention, unless otherwise indicated, the term "carbon nanotube (CNT)" refers to any type of carbon nanotube. CNTs typically exist as a single layer or multiple layers of cylindrical layers of graphene sheets. Individual sheets can differ in terms of layering and functionality. For example, CNTs can exist as single-walled CNTs (SWCNTs) and multi-walled CNTs (MWCNTs). The CNTs of the array can also be chiral, achiral, beginning, capped, bud-shaped, coated, uncoated, functionalized, pure, anchored, unanchored, basal, edge-plane, stepped, or have any other known configuration. For the purposes of the present invention, single-walled CNTs (SWCNTs) are particularly preferred. However, without limitation thereto, CNTs typically have a diameter ranging from about 0.6 nm to about 100 nm and a length ranging from about 50 nm to about 10 mm. In a preferred embodiment, CNTs having a diameter ranging from about 0.7 nm to about 20 nm and a length ranging from about 0.5 μm to about 250 μm are used.

[0044] As used herein, the term "functionalized" refers to the presence of one or more substituents bound to, complexed to, or otherwise associated with a solid support, such as a carbon nanostructure.

[0045] For example, CNTs can be chemically functionalized by methods known to those skilled in the art through internal functionalization, non-covalent functionalization, sidewall functionalization, and defect functionalization. For example, exemplary methods are disclosed and reviewed in JL Bahr et al., Chem. Commun. 2001, 193; S. Niyogi et al., Acc. Chem. Res. 2002, 35, 1105; A. Hirsch, Angew. Chem. Int. Ed. Engl. 2002, 41, 1853; S. Sinnott, Journ. Nanosci. Nanotechn. 2002, 2, 113; and A. Hirsch et al., Top. Curr. Chem. 2005, 245, 193. Among these methods, non-covalent functionalization and sidewall functionalization may be preferred. Non-covalent functionalization is based on the ability of the extended π system of the carbon nanotube sidewall to bind guest molecules via π-π stacking interactions. Functionalization of fullerenes can be achieved by covalent means due to the electrophilic properties of C60 (e.g., nucleophilic addition) or non-covalent means including hydrogen bonding, π-π interactions, and metal ion coordination. Graphene can be functionalized by covalent functionalization based on the attachment of organic functional groups such as free radicals (e.g., benzoyl peroxide, nitrobenzyl) and dienophiles (e.g., methylenediylide, phenyl azide, alkyl azide) to pristine graphene or by chemical action of oxygen groups of graphene oxide.

[0046] Note that desired functional groups (examples of which are discussed further below) can also be incorporated into polymer chains, which can be grafted onto the carbon nanostructures by covalent bonding, or can be physically bonded to the carbon nanostructures by adsorbing or wrapping the polymer onto the surface of the carbon nanostructure, thereby providing a polymer-carbon nanostructure composite.

[0047] Suitable functional substituents are selected by one skilled in the art by their ability to induce a change in a measurable electrical parameter (ie, resistance, voltage, current, capacitance, or impedance) in a given sensor upon interaction with an analyte.

[0048] According to the present invention, the first sensor is sensitive to dipole-dipole interactions, which is preferably achieved by functionalizing the carbon nanostructure with solvatochromic or dipolar groups or reagents.

[0049] As used herein, the term "solvation color" refers to the phenomenon that the spectrum of a substance shifts when it interacts with an analyte. Different analytes induce different effects on the electronic ground state and excited state of the solvation color group, so that the size of the energy gap between them varies depending on the analyte. This is reflected in the absorption or emission spectrum of the solute as differences in the position, intensity and shape of the spectral bands. When the spectral bands appear in the visible part of the spectrum, solvation color is observed as a change in color. Negative solvation color corresponds to a blue shift (or blue shift) when the polarity of the analyte increases, and positive solvation color corresponds to a red shift (or red shift) when the polarity of the analyte increases.

[0050] Examples of solvatochromic groups or reagents include, but are not limited to, 4-dicyanomethylene-2-methyl-6-(p-dimethylaminophenyl)-4H-pyran (DCM), 6-propionyl-2-(dimethylamino)-naphthalene (PRODAN), 9-(diethylamino)-5H-benzo[a]phenoxazin-5-one (Nile Red), 4-(dicyanovinyl)julolidine (DCVJ), phenol blue, stilbazolium dyes, coumarin dyes, ketocyanine dyes, Reichardt's dyes (e.g., Reichardt's betaine dye (2,6-diphenyl-4- (2,4,6-triphenylpyridinyl)phenolate), merocyanine dyes (e.g., merocyanine 540), N,N-dimethyl-4-nitroaniline (NDMNA), N-methyl-2-nitroaniline (NM2NA), 4-(4'-hydroxyphenylvinyl)-N-methylpyridinium iodide, 4,4'-bis(dimethylamino)fuchsinone, Nile blue, 1-anilinonaphthalene-8-sulfonic acid (1,8-ANS), dapoxylbutylsulfonamide (DBS), rhodamine B, auramine O, and derivatives thereof. In a preferred embodiment, the solvatochromic dye is selected from Reichardt's dyes. Note that the solvatochromic group can be covalently attached to the carbon nanostructure or non-covalently bound or immobilized in the first sensor.

[0051] As an alternative to or in addition to the solvatochromic group, a dipole group may be incorporated which increases the adsorption of dipolar analytes by directional dipole-dipole interactions. Preferred examples of dipole groups include substituents containing a cyano group, such as cyanoalkyl or cyanoalkoxy groups.

[0052] According to the present invention, the second sensor is sensitive to hydrogen bonding interactions with the analyte, which is preferably achieved by functionalizing (e.g., the carbon nanostructure) with functional groups including one or more than one and preferably a plurality of highly electronegative atoms, such as halogen atoms (e.g., F, Cl, I, and Br), oxygen atoms, sulfur atoms, and / or nitrogen atoms, etc.

[0053] Preferred examples of halogen-containing functional groups are fluorinated groups such as perfluoroalkyl, perfluoroaryl, or perfluorophenyl azide (PFPA) groups. PFPA and its derivatives are particularly preferred for covalent bonding to carbon nanostructures because they have highly reactive azide groups that promote bond formation with graphene (LH Liu et al., Nano. Lett. 2009, 9, 3375-3378; LH Liu et al., J. Mater. Chem. 2010, 20, 5041-5046), CNTs (SJPastine et al., J. Am. Chem. Soc. 2008, 4238-4239), and fullerenes (M. Yan et al., J. Org. Chem. 1994, 59, 5951-5954).

[0054] Preferred nitrogen-containing groups are selected from primary, secondary and tertiary amines, amides, amine-N-oxides and phosphazenes and their derivatives. Examples of sulfur-containing groups include, but are not limited to, sulfoxides. Non-limiting examples of oxygen-containing groups include hydroxyl groups (including, for example, fluoroalcohols, phenols or substituted phenols) and ester groups.

[0055] The third sensor is configured to detect the change of the electrical parameter based on the dispersion interaction with the analyte. Suitable substituents for this purpose include but are not limited to aliphatic hydrocarbons, which can be open chain (branched or straight chain) or cyclic, saturated or unsaturated. As specific non-limiting examples, propyl group, isopropyl, n-butyl, tert-butyl, isobutyl, hexyl, heptyl, cyclohexyl, 2-ethylhexyl, octyl, isooctyl, nonyl, different nonyl, decyl, isodecyl, undecyl, dodecyl, tridecyl, tetradecyl, butadienyl, isoprenyl, cyclohexyl, ethynyl, propynyl, butynyl group and combination thereof can be mentioned.

[0056] The functional groups mentioned above in conjunction with the first to third sensors can also be incorporated into a polymer that is immobilized on a solid support by chemical bonding, physical methods, or a combination thereof. Examples of polymer-CNT and polymer-graphene composites and their use for chemical sensing are reviewed in HJ Salavagione et al., J. Mater. Chem. A 2014, 2, 14289 and can be appropriately used for the purposes of the present invention. Suitable physical deposition techniques include, for example, vapor deposition or layer deposition using aqueous solutions.

[0057] It will be appreciated that a single sensor may be sensitive to more than one type of interaction with the analyte (e.g., basicity and dipolarity). However, in the present invention, it is preferred to provide such a hybrid sensor in addition to the first to third sensors to minimize interference with a single Hansen solubility parameter while providing additional information about the analyte.

[0058] In a preferred embodiment of the present invention, the multi-sensor array further comprises a fourth sensor comprising pH-sensitive functionality, which provides additional information about the analyte. Suitable pH sensors are not particularly limited and can be appropriately selected by those skilled in the art. As examples, CNT-based pH sensors (e.g., see US2014 / 0330100 A1 and CN 114624302 A), graphene-based pH sensors (e.g., see US2019 / 0079043 A1), hydrogel-based pH sensors (e.g., see A. Richter et al., Sensors 2008, 8, 561-581; and NF Sheppard, Sens. Actuat. B 1995, 28, 95-102) and pH sensors based on ion-sensitive field-effect transistors (ISFETs), etc. can be mentioned. Preferred examples include CNTs modified with functional groups, which are used as Acid or base for pH estimation.

[0059] In a preferred embodiment of the present invention, the multi-sensor array further includes a fifth sensor, which is configured to detect changes in electrical parameters based on a redox reaction with an analyte. This redox sensor is not particularly limited and can be appropriately selected by those skilled in the art. As an example, CNT-based redox sensors (e.g., see US2007 / 0278111 A1), carbon nanostructured redox sensors (e.g., see WO2019 / 074618 A2), etc. can be mentioned. Preferred examples include CNTs modified with redox-active functional groups.

[0060] By simultaneously employing the first to fifth sensors, a 5-dimensional feature space is opened up by the measurement data, thereby allowing for a more accurate classification of substances based on existing literature tables as data input to a pattern recognition algorithm.

[0061] To minimize errors and thereby improve the selectivity and / or sensitivity of the multi-sensor array, several of each of the first to third sensors (and fourth and fifth sensors, if present) are preferably used to determine the mean and standard deviation based on multiple readings.

[0062] A preferred embodiment of the present invention relates to a multi-sensor array comprising an independent detector channel for each sensor,

[0063] wherein the first detector channel comprises a substrate including an insulating layer, a pair of electrodes formed on the insulating layer, and a semiconductor material coupled to respective ones of the electrodes and comprising the first sensor;

[0064] wherein the second detector channel comprises a substrate including an insulating layer, a pair of electrodes formed on the insulating layer, and a semiconductor material coupled to respective ones of the electrodes and comprising the second sensor;

[0065] wherein the third detector channel comprises a substrate including an insulating layer, a pair of electrodes formed on the insulating layer, and a semiconductor material coupled to respective ones of the electrodes and comprising the third sensor;

[0066] Wherein, the semiconductor material is selected from functionalized carbon nanostructures, preferably functionalized carbon nanotubes;

[0067] wherein each electrode pair is configured as a conductive layer that is patterned to form a channel separating one electrode from another electrode; and

[0068] The channels between the electrode pairs are partially or completely filled with corresponding semiconductor materials to enable current conduction.

[0069] Optionally, the aforementioned multi-sensor array may additionally include one or two additional independent detector channels selected from the following:

[0070] a fourth detector channel comprising a substrate including an insulating layer, a pair of electrodes formed on the insulating layer, and a semiconductor material coupled to each of the electrodes and comprising the fourth sensor; and

[0071] A fifth detector channel includes a substrate including an insulating layer, a pair of electrodes formed on the insulating layer, and a semiconductor material coupled to each of the electrodes and including the fifth sensor.

[0072] Improved selectivity and / or sensitivity of a multi-sensor array can be achieved by incorporating a multiplicity (e.g., three or more) of sensors into the first through third detector channels (and the fourth and fifth detector channels, if present) to identify outliers, sensor defects, and to improve measurement accuracy by averaging multiple readings.

[0073] Note that the substrate and its components can be selected independently for each detector channel. For example, the insulating layer can be selected from a dielectric inorganic material (e.g., SiO2) or an organic material (e.g., PET, PEN, paper, and / or epoxy resin). The electrodes can be independently selected from any conductive material, including metals (e.g., copper, gold, palladium, silver, aluminum, chromium, etc.) and organic conductors (e.g., conductive polymers).

[0074] However, without limitation thereto, channels separating electrodes from each other may be formed by, for example, laser patterning, lift-off, lithographic methods, direct laser patterning, inkjet printing, or screen printing. These methods are preferred because they enable low-cost and high-throughput manufacturing.

[0075] Figure 2 An example of a single detector channel (1) is illustrated in Figure 1, comprising a substrate comprising an insulating layer (2), wherein a channel (3) is formed in a conductive layer (4) to provide a pair of electrodes separated from each other. Semiconductor material, exemplarily in the form of functionalized carbon nanotubes (5), is used as the sensing element.

[0076] like Figure 2 As depicted, the channel (3) separating one electrode from the other can have a serpentine shape. However, in a preferred embodiment, the channel has the geometry of a FASS curve, where the acronym FASS stands for space-filling, self-avoiding, simple and self-similar curve. The feature of space-filling refers to the fact that the curve passes within a small distance from all points of a given two-dimensional figure (such as a square) surrounding the curve. This distance can be arbitrarily reduced by continuing the recursive construction until the appropriate level is reached. Self-avoiding means that the segments of the curve do not touch or intersect. Simplicity indicates that the curve can be drawn by a single stroke of the pen without lifting the pen or drawing any segment more than once. The feature of self-similarity indicates that the curve can be constructed by recursive application of a set of rules for connecting components. In most cases, it is additionally assumed that the curve consists of straight line segments extending vertically or horizontally along the edges of the square grid. Examples of FASS curves and methods for their construction are described in the document “Synthesis of Space-filling Curves on the Square Grid” by P. Prusinkiewiczl, A. Lindenmayer and F.D. Fracchia, published in the textbook “Fractals in the Fundamental and Applied Sciences”, edited by H.-O. Peitgen, J.M. Henriques & L.F. Penedo for Elsevier Science Publishers BV, 1991.

[0077] Advantageously, channels shaped as FASS curves maximize the probability that functionalized CNTs will contact the channel and bridge the electrode pair, even if the CNTs are randomly deposited on the substrate (e.g., Figure 2 The increased number of contacts between the nanotubes and the channel also increases the current flowing between the electrodes and optimizes the signal-to-noise ratio.

[0078] According to another preferred embodiment of the present invention, the FASS curve comprises a Hilbert curve, which can be advantageously designed so that the ends of the Hilbert curve are on opposite sides of the electrode contact area in a manner that allows the formation of two different electrodes. In addition, a predefined extension can be obtained by simply increasing the number of iterations when constructing the curve.

[0079] Figure 3A A top view of a channel system having a Hilbert curve shape (Lindenmayer number equal to 2) is schematically illustrated, wherein a channel (10) is formed on a substrate layer (11) and separates an electrode pair, namely a first electrode (12) and a second electrode (13). The channel (10) can be partially or completely filled with a semiconductor material. For example, Figure 3B As illustrated, multiple functionalized CNTs (14) can be randomly deposited on electrodes (12) and (13). Since the Hilbert curve is a space-filling curve, the probability that one or more CNTs (14) overlap with the channel (10) at least at one point is maximized. Depending on the length of the CNT (14), it is also possible for one or more individual CNTs (14) to overlap with the channel (10) at multiple points, thereby enhancing the performance of the device. In addition, the extension or width of the curve can be increased by plotting the Hilbert curve without affecting the area (A) available for the electrodes in the detector channel.

[0080] In another preferred embodiment, the FASS curve comprises a Peano curve, which may also be constructed in a manner that allows for rounded edges, which may simplify the manufacturing process and / or make the manufacturing process more efficient.

[0081] According to another preferred embodiment of the present invention, one or more of the electrode pairs may form side walls facing the channels separating them, wherein the side walls are configured (in side view) to have an inclined or curved profile. This configuration has the advantage of optimizing the contact between the side walls of the electrodes and the semiconductor material filling the channels.

[0082] The multi-sensor arrays described in the above embodiments have the advantage that they can be manufactured in an inexpensive, compact and lightweight manner.Therefore, in a particularly preferred embodiment, the multi-sensor array is provided within a low power handheld device that can be deployed for field measurements.

[0083] Multi-sensor system

[0084] To enhance measurement accuracy, to achieve improved discrimination between multiple target compounds or classes in a single analyte sample (e.g., to reduce crosstalk) and / or to account for necessary calibration shifts, multi-sensor arrays according to the present invention can be combined with machine learning tools to analyze the data retrieved from the sensing elements.

[0085] In a second embodiment, the present invention therefore relates to a multi-sensor system for classifying and / or identifying one or more analytes in a gas phase or a liquid phase, comprising: a multi-sensor array according to the first embodiment; a processing unit comprising a pattern recognition and machine learning analyzer, wherein the pattern recognition and machine learning analyzer is configured to receive the output signal of the multi-sensor array and compare it with stored data to classify and / or identify the one or more analytes; wherein the stored data comprises Hansen solubility parameters obtained from a library database.

[0086] The term "library database" as used herein includes the library database from the textbook Hansen Solubility Parameters: A User's Handbook, 2007 2 nd A set of solubility parameters retrieved from Hansen Solubility Parameters in Practice (HSPiP) or calculated using the software "Hansen Solubility Parameters in Practice (HSPiP)". Ed., Boca Raton, CRC Press.

[0087] It should be understood that the output signal from the first sensor of the multi-sensor array is related to the Hansen parameter δ for polarity P The output signal of the second sensor from the multi-sensor array is related to the Hansen parameter δ for hydrogen bonding. H The output signal of the third sensor from the multi-sensor array is related to the Hansen parameter δ for dispersion D Related.

[0088] The multi-sensor system may also include additional components, including, for example: a communication circuit configured to receive data collected based on changes in electrical parameters from at least one of the first, second, and third detector channels; a pre-processing unit configured to generate a response pattern; a memory configured to store the data and pattern recognition and machine learning algorithms; and a processing circuit configured to apply the pattern recognition and machine learning algorithms to the data and probabilistically classify and / or identify one or more analytes. Figure 4 An exemplary scheme illustrating an example of a multi-sensor system is shown in FIG.

[0089] The multi-sensor system can communicate data to an application executing on a user device (such as a smartphone, tablet, or a device that enables a virtual assistant) via wireless and wired communications. For example, the multi-sensor system can include a wireless network interface (e.g., or etc.) to, for example, enable the system to communicate with other sensor devices (e.g., in an Internet of Things (IoT) setting), applications executing on a user's computing device, or a cloud-based server.

[0090] The multi-sensor system may be provided as a standalone device or as a kit of parts, wherein the constituent elements or combinations of constituent elements are provided individually. For example, the processing unit or its components may be provided independently of the multi-sensor array, for example by means of a computing device (PC or smartphone) connected to the multi-sensor array by wire or wirelessly.

[0091] The standalone device may include a display configured to view analytical data and measurement results (including identified analytes or analyte classes). For example, information regarding operating modes (e.g., detection mode or low power mode), network connectivity, and other information associated with the multi-sensor array may also be displayed. Additionally, the standalone device may be battery powered.

[0092] Signal processing to determine analytes interacting with the sensor array may include a probabilistic determination of the class of the analyte rather than an exact determination of the analyte, thereby enabling rapid assessment of the presence of classes of matter of interest, such as classes of potentially hazardous or harmful compounds.

[0093] Correlating the data received from the multi-sensor array with Hansen solubility parameters further enables probabilistic determination of compound classes in the analyte that are unknown to the pattern recognition and machine learning analyzer (i.e., the pattern recognition and machine learning analyzer has not been previously trained on the compound class), thereby providing added value over conventional sensor systems that utilize a combination of chemical sensors and machine learning tools.

[0094] The pattern recognition and machine learning analyzer employed in the multi-sensor system according to the present invention is configured to apply at least one algorithm to probabilistically classify and / or identify one or more analytes.

[0095] In a preferred embodiment, the at least one algorithm is selected from the group consisting of an artificial neural network (ANN) algorithm, principal component analysis (PCA), support vector machine (SVM), decision tree (DT), random forest (RF), voting classifier (VC), multi-layer perception (MLP), generalized regression neural network (GRNN), fuzzy inference system (FIS), self-organizing map (SOM), radial bias function (RBF), neuro-fuzzy system (NFS), adaptive resonance theory (ART), partial least squares (PLS), linear regression (LR), multiple linear regression (MLR), principal component regression (PCR), discriminant function analysis (DFA), linear discriminant analysis (LDA), cluster analysis and nearest neighbor algorithm.

[0096] In a further preferred embodiment, the at least one algorithm is selected from one or more of a support vector machine (SVM), a nearest neighbor algorithm (e.g., K-nearest neighbor (KNN)), a decision tree (DT), a random forest (RF), and a voting classifier (VC) algorithm.

[0097] In a machine learning-enabled sensor system, the data generated by the sensors is collected and compiled into a raw dataset, which is then preprocessed to extract relevant features from the dataset to create a so-called intelligent model. Once sufficient features have been extracted, it can be recompiled into sub-datasets for training the ML-enabled sensor system and sub-datasets for validating and testing the ML-enabled sensor system. Therefore, the algorithm selected to define the intelligent model for a specific sensor application should ideally be validated and tested with real-life samples.

[0098] Methods for analyzing chemical samples

[0099] In a third embodiment, the present invention is directed to a method for analyzing a sample present in a gas phase or a liquid phase, comprising the steps of: a) introducing the sample into a multi-sensor array according to the first embodiment described above; and b) applying a pattern recognition and / or machine learning algorithm to data received from at least one of the first, second and third sensors to classify and / or identify one or more analytes in the sample; wherein in step b), the received data is correlated with Hansen solubility parameters obtained from a class library database.

[0100] The method of sample introduction is not particularly limited. For example, in the gas phase, the multi-sensor array can be exposed to saturated vapor of the analyte, or a flow cell and pump can be used to direct the vapor into individual detector channels. In the liquid phase, the multi-sensor array can be immersed in the sample liquid to introduce the analyte into the detector channels.

[0101] In a preferred embodiment, the method utilizes pattern recognition and machine learning algorithms as described above in conjunction with the second embodiment.

[0102] By comparing the output signal from the first sensor with the Hansen parameter δ for polarity P The output signal of the second sensor from the multi-sensor array is correlated with the Hansen parameter δ for hydrogen bonding. H The output signal of the third sensor of the multi-sensor array is correlated with the Hansen parameter δ for dispersion. D Relatedly, the methods of the present invention advantageously enable rapid detection and / or classification of one or more analytes in the gas phase or liquid phase without relying exclusively on data generation and training via machine learning.

[0103] It will be appreciated that the present invention may employ any of the preferred features specified above in relation to the description of the first to third embodiments, and that the preferred features may be combined in any combination, except combinations where at least some of the features are mutually exclusive.

[0104] Typically, there is no particular limitation on the sample that can be analyzed. The multisensor array according to the present invention is sensitive to a large number of analytes and mixtures thereof. Exemplary analytes and mixtures thereof include aliphatic or aromatic hydrocarbons, compounds with groups comprising halogens, oxygen, nitrogen, sulfur, phosphorus, boron and / or silicon atoms. Typical analytes include ammonia, amines, amides, amidines, formamides, amidines, imides, azides, nitrates, nitrites, isonitriles, pyridine derivatives, carbamates, thiols, sulfides, sulfoxides, disulfides, sulfones, sulfinic acids, sulfonates, thiocyanates, thioketones, phosphine, phosphonic acid, phosphates, borate esters, alcohols, aldehydes, acyl halides, carboxylic acids, organic anhydrides, lactones, peroxides, hydroperoxides, ethers, amino acids, alkanes, aromatics, aliphatic groups, halogen compounds, inorganic acids and gases (e.g., CO ).

[0105] The multi-sensor array and system according to the present invention are particularly effective for the classification and / or identification of liquid analytes and analyte mixtures. As an example thereof, volatile organic compounds (VOCs) may be mentioned, which are understood to be organic compounds having a vapor pressure of 0.01 kPa or higher at 293.15 K.

[0106] For example, the present invention can be used to rapidly detect and identify VOCs, aromatic compounds, explosives, toxic compounds, toxins, drugs, drug precursors, anesthetics, waste gases, and environmental pollutants. Thus, the fields of use are broad and include, but are not limited to, industry, security, public services, military, construction work, and environmental monitoring. Furthermore, because they do not require any specific methodological knowledge or analytical skills, the multi-sensor arrays and multi-sensor systems according to the present invention are well-suited for home use.

[0107] Numerous other features, modifications and improvements will become apparent to those skilled in the art once given the above disclosure.

[0108] Reference numerals

[0109] 1 detector channel

[0110] 2 Insulation layer

[0111] 3 Channels between electrodes

[0112] 4 Conductive layer

[0113] 5 Functionalized carbon nanotubes (CNTs)

[0114] 10 Channels between electrodes

[0115] 11 substrate layer

[0116] 12. First electrode

[0117] 13. Second electrode

[0118] 14 Functionalized carbon nanotubes (CNTs)

[0119] Area A

Claims

1. A multi-sensor array comprising a plurality of sensors for detecting one or more analytes in a gas phase or a liquid phase, in, The plurality of sensors include: a first sensor configured to detect a change in an electrical parameter based on a dipole-dipole interaction with the analyte, a second sensor configured to detect a change in an electrical parameter based on hydrogen bonding interactions with the analyte, and A third sensor is configured to detect a change in an electrical parameter based on the dispersed interaction with the analyte.

2. The multi-sensor array according to claim 1, in, The first sensor is functionalized with a solvatochromic group, preferably one or more of the following: 4-dicyanomethylene-2-methyl-6-(p-dimethylaminophenyl)-4H-pyran, i.e., DCM; 6-propionyl-2-(dimethylamino)-naphthalene, i.e., PRODAN; 9-(diethylamino)-5H-benzo[a]phenoxazin-5-one, i.e., Nile Red; 4-(dicyanovinyl)julolidine, i.e., DCVJ; phenol blue; stilbazolium dyes; coumarin; cyanine dyes, ketocyanine dyes, Reichardt's dyes, merocyanine dyes, N,N-dimethyl-4-nitroaniline (NDMNA), N-methyl-2-nitroaniline (NM2NA), 4-(4'-hydroxystyryl)-N-methylpyridinium iodide, 4,4'-bis(dimethylamino)fuchsinone, Nile blue, 1-anilinonaphthalene-8-sulfonic acid (1,8-ANS), dapoxybutylsulfonamide (DBS), rhodamine B, auramine O and derivatives thereof, more preferably selected from Reichardt's dyes; and / or Wherein, the first sensor is functionalized with a dipole group, and the dipole group is preferably selected from one or more than one of cyano, cyanoalkyl and cyanoalkoxy groups.

3. The multi-sensor array according to claim 1 or 2, in, The second sensor is functionalized with a group comprising one or more than one, and preferably a plurality of, electronegative atoms selected from halogens, oxygen atoms, sulfur atoms and / or nitrogen atoms; The halogen-containing functional group is preferably selected from fluorinated groups, more preferably selected from perfluoroalkyl, perfluoroaryl or perfluorophenyl azide groups, i.e., PFPA groups; The oxygen atom-containing group is preferably selected from hydroxyl groups and ester groups; The nitrogen-containing groups are preferably selected from primary, secondary and tertiary amines, amides, amine-N-oxides and phosphazenes and derivatives thereof, and The sulfur-containing groups are preferably selected from sulfoxides.

4. The multi-sensor array according to any one of claims 1 to 3, in, The third sensor is functionalized with aliphatic hydrocarbon groups. 5 . The multi-sensor array according to claim 1 , further comprising a fourth sensor comprising a pH-sensitive functional group.

6. The multi-sensor array according to claim 5, in, The fourth sensor is selected from any one of a carbon nanostructure-based pH sensor, a hydrogel-based pH sensor, and an ion-sensitive field-effect transistor (ISFET)-based pH sensor. 7 . The multi-sensor array according to claim 1 , further comprising a fifth sensor configured to detect a change in an electrical parameter based on a redox reaction with the analyte.

8. The multi-sensor array according to claim 7, wherein: The fifth sensor includes a carbon nanostructure modified with redox-active functional groups.

9. The multi-sensor array according to any one of claims 1 to 8, wherein: At least one, and preferably at least two, and more preferably all three of the first, second and third sensors comprise a functionalized carbon nanostructure as a sensing element.

10. The multi-sensor array according to claim 9, wherein: The functionalized carbon nanostructure comprises single-walled carbon nanotubes, multi-walled carbon nanotubes, amorphous carbon, 2D graphene, 3D graphene, fullerite, fullerene, carbon nanowires, carbon nanofibers, carbon nanobelts, carbon nanospheres, carbon nanorods or any combination thereof, and the functionalized carbon nanostructure preferably comprises single-walled carbon nanotubes.

11. A multi-sensor array according to any one of claims 1 to 10, comprising an independent detector channel for each sensor, in, a first detector channel comprising a substrate including an insulating layer, a pair of electrodes formed on the insulating layer, and a semiconductor material coupled to respective ones of the electrodes and comprising the first sensor; wherein the second detector channel comprises a substrate including an insulating layer, a pair of electrodes formed on the insulating layer, and a semiconductor material coupled to respective ones of the electrodes and comprising the second sensor; wherein the third detector channel comprises a substrate including an insulating layer, a pair of electrodes formed on the insulating layer, and a semiconductor material coupled to respective ones of the electrodes and comprising the third sensor; Wherein, the semiconductor material is selected from functionalized carbon nanostructures, preferably functionalized carbon nanotubes; wherein each electrode pair is configured as a conductive layer that is patterned to form a channel separating one electrode from another electrode; and The channels between the electrode pairs are partially or completely filled with corresponding semiconductor materials to enable current conduction.

12. The multi-sensor array according to claim 11, wherein: At least one of the channels separating one electrode from another electrode has a shape of a FASS curve, and the FASS curve preferably comprises a Peano curve or a Hilbert curve.

13. A multi-sensor system for classifying and / or identifying one or more analytes in a gas or liquid phase, comprising: The multi-sensor array according to any one of claims 1 to 12; as well as a processing unit comprising a pattern recognition and machine learning analyzer, wherein the pattern recognition and machine learning analyzer is configured to receive output signals of the multi-sensor array and compare the output signals with stored data to classify and / or identify the one or more analytes; The stored data includes Hansen solubility parameters obtained from the library database.

14. The multi-sensor system according to claim 13, wherein: The pattern recognition and machine learning analyzer is configured to apply at least one algorithm to probabilistically classify and / or identify the one or more analytes, wherein the at least one algorithm is selected from the group consisting of artificial neural network algorithm, i.e., ANN algorithm, principal component analysis, i.e., PCA, support vector machine, i.e., SVM, decision tree, i.e., DT, random forest, i.e., RF, voting classifier, i.e., VC, multilayer perception, i.e., MLP, generalized regression neural network, i.e., GRNN, fuzzy inference system, i.e., FIS, self-organizing map, i.e., SOM, radial bias function, i.e., RBF, neuro-fuzzy system, i.e., NFS, adaptive resonance theory, i.e., ART, partial least squares, i.e., PLS, linear regression, i.e., LR, multivariate linear regression, i.e., MLR, principal component regression, i.e., PCR, discriminant function analysis, i.e., DFA, linear discriminant analysis, i.e., LDA, cluster analysis, and nearest neighbor algorithm, and preferably is at least one algorithm selected from support vector machine, i.e., SVM, nearest neighbor, decision tree, i.e., DT, random forest, i.e., RF, and voting classifier algorithm, i.e., VC algorithm.

15. A method for analyzing a sample in a gas phase or a liquid phase, comprising the steps of: a) introducing the sample into a multi-sensor array according to any one of claims 1 to 12; as well as b) applying pattern recognition and / or machine learning algorithms to data received from at least one of the first sensor, the second sensor, and the third sensor to classify and / or identify one or more analytes in the sample; Wherein, in step b), the received data is correlated with the Hansen solubility parameters obtained from the class library database.

16. The method according to claim 15, wherein The pattern recognition and / or machine learning algorithm is applied to the data received from the first sensor, the second sensor, and the third sensor to classify and / or identify the one or more analytes in the sample.

17. The method according to claim 16, wherein The data received from the first sensor is compared with the Hansen polarity parameter δ P The data received from the second sensor are related to the Hansen hydrogen bond parameter δ H and the data received from the third sensor of the multi-sensor array is related to the Hansen dispersion parameter δ D Related.

Images