Multi-ion concentration detection chip and method for determining the concentration of multiple ions in solution

Abstract

This invention provides a multi-ion concentration detection chip, comprising an FPC interface, discrete first and second thin-film transistors, a reference electrode, and a multi-channel sensing array composed of multiple SC-ISEs. This invention also provides a method for determining the concentrations of multiple ions in a solution. The advantages of this invention are: it provides a multi-ion concentration detection chip with high integration and portability; and it provides a method for determining the concentrations of multiple ions in a solution, employing a decoupling method to remove interference and achieve highly robust and accurate ion concentration prediction.

Description

Technical Field

[0001] This invention relates to methods for measuring ion concentration, and more particularly to a multi-ion concentration detection chip and a method for determining the concentrations of multiple ions in a solution. Background Technology

[0002] Water is the source of life. Human society cannot live without water and has personalized requirements for water quality. The aquaculture industry has strict control over parameters such as dissolved oxygen, pH and temperature in water, which are important indicators for the survival of aquatic animals [4]. In the field of biomedicine, biomarkers such as K+, Ca2+, Mg2+, NH4+ and other ions or molecules in human body fluids will be detected to contribute to the preliminary detection of body metabolic status and complex cardiac function studies [4-6].

[0003] Water quality testing technology has developed rapidly driven by the demand for widespread applications. Detection technologies can be classified according to the target substance. For example, polymerase chain reaction technology is commonly used for pathogenic microorganisms in water[7], while electrophoresis technology is often used for biological macromolecules such as nucleic acids and proteins[8]. However, most substances in water exist in ionic form, and the detection technology for various ions in water has a wide range of applications. In basic terms, complexometric titration is commonly used for the qualitative and quantitative analysis of ions. Its principle is simple, but the titration endpoint error is large. There are reports of using spectrophotometry to detect reduced nitrite, and there are also spectrophotometry methods combined with atomic absorption spectrophotometry to detect serum calcium. However, these methods are susceptible to interference and the analytical equipment is expensive[9-10]. In addition, potentiometric titration, oscillometric polarography, etc. are used for ion detection, but they also have disadvantages such as requiring a laboratory environment and not being able to monitor in real time.

[0004] Ion-selective electrodes are an important branch of chemical sensors. Their detection principle is usually based on the relationship between the zero-current open-circuit potential of the ion-selective sensitive membrane and the activity of the analyte ion, which conforms to the Nernst equation. Therefore, the activity of the analyte ion can be calculated by obtaining the zero-current open-circuit potential. Currently, polymer membrane ion-selective electrodes are the most actively researched ion-selective electrodes, mainly including two types: liquid contact electrodes and all-solid-state electrodes.

[0005] The liquid contact electrode uses a selectively sensitive membrane attached to the bottom of the outer wall of the electrode tube between the filling liquid and the test solution. A solution of a specific concentration and the working electrode are placed inside the electrode tube. Another electrode without a sensitive membrane serves as a reference electrode. Both electrodes are simultaneously inserted into the test solution to provide a potential difference signal, and their other ends are connected to a potentiometer for reading.

[0006] All-solid-state electrodes cover the sensitive film with a material that performs electron sensing, such as metal or conductive carbon. The working area is immersed in the solution to be measured, and the voltage signal is directly conducted through the metal and connected to the potentiometer.

[0007] Both of these electrodes are isolated sensing devices, requiring external circuitry to read the voltage signal.

[0008] Solid-state ion-selective electrodes (SC-ISEs) have the potential to avoid the above-mentioned drawbacks and can meet the needs of rapid ion detection in environmental monitoring and biofluid analysis

[11] . This is a pollution-free, passive detection method that can convert the activity of primary ions into potential. Due to their advantages such as small size, good integrability, and low cost, ISEs have broad application prospects in the fields of medicine, environment, and wearable devices [12,13].

[0009] Based on the specificity of ISEs, multiple ISEs with different selectivities are often combined to achieve multi-ion sensing. However, since the selectivity of ISEs for ions is not ideal, interfering ions with similar physicochemical properties can affect the potential of the ISEs, leading to distortion of the calculated ion activities. Otto M's team proposed a method for matrix operations on the extended Nernst equation (Nikolskii formula), as follows:

[0010]

[0011] Or

[0012]

[0013] The activity vector and response parameter matrix were constructed using several ISEs, and the ion concentration was calculated with an error of 10%. The experimental calibration process was complicated and had the problem of insignificance when the concentration of the main analyte was low. The Duarte LT team used a Bayesian nonlinear blind source separation algorithm to separate the independent information of Na+, NH4+, and K+

[14] . This work assumes that the signal sources are independent, but in reality, each ion will affect each other, causing the generated potential to change with different activity distributions. The Gallardo J team used multiple non-selective ISEs to capture mixed signals and trained an artificial neural network to obtain the mapping law. This scheme ignores the contribution of the selectivity of ISEs to the independence of the signal sources. The Mimendia A team used ANN to train specific ISEs in the laboratory to realize real-time wireless monitoring of river water environment

[16] . The Cho WJ team also used specific ISEs in the hydroponic environment detection, used an automatic data acquisition system to automatically train an ANN, and combined it with the TPN normalization method to realize multi-ion calculation

[17] .

[0014] These methods almost all focus on decomposing mixed signals from a signal processing perspective, lacking exploration from the perspective of reaction phenomena and experimental kinetics. Furthermore, currently reported multi-ion combined detection schemes are typically based on liquid-junction ISEs, leaving room for improvement in integration.

[0015] In summary, existing methods for measuring ion concentration have the following drawbacks:

[0016] 1. Existing schemes for implementing ion-selective electrode arrays and algorithm correction almost always combine multiple liquid-contact electrodes. This results in large sensor sizes, requiring fixtures, and necessitates a large liquid environment for the test solution. Consequently, the sensors can only be tested in fixed laboratory environments, leading to poor integrability. Schemes using solid-state selective electrodes typically only isolate the sensing electrode to achieve target ion sensing, failing to eliminate interference.

[0017] 2. Existing methods for calculating ion concentration mostly employ neural network algorithms, requiring large training datasets and thousands of training iterations to achieve a good fit. This results in low efficiency and high computational resource consumption, making it difficult to implement on microcontrollers without expensive computing chips, thus limiting integration. Furthermore, statistical signal processing methods tend to overlook the selectivity of the electrode itself for ions.

[0018] This results in insufficient accuracy and complex theoretical calculations.

[0019] The references are as follows:

[0020] [1] Gunnarsdottir MJ, Gardarsson SM, Figueras MJ, et al. Water safetyplan enhancements with improved drinking water quality detection techniques[J].Science of the total environment,2020,698:134185.

[0021] [2]Wen

[0022] [3]Clark R M,Hakim S.Public–Private Partnerships and TheirApplication to US Drinking Water Systems[J].Public Private Partnerships:Construction,Protection,and Rehabilitation of Critical Infrastructure,2019:281-289.

[0023] [4]Huan J,Li H,Wu F,et al.Design of water quality monitoring systemfor aquaculture ponds based on NB-IoT[J].Aquacultural Engineering,2020,90:102088.

[0024] [5]Bellando F,Garcia-Cordero E,Wildhaber F,et al.Lab on skinTM:3Dmonolithically integrated zero-energy micro / nanofludics and FD SOI ionsensitive FETs for wearable multi-sensing sweat applications[C] / / 2017 IEEEInternational Electron Devices Meeting(IEDM).IEEE,2017:18.1.1-18.1.4.

[0025] [6]Heikenfeld J.Non-invasive analyte access and sensing througheccrine sweat:challenges and outlook circa 2016[J].Electroanalysis,2016,28(6):1242-1249.

[0026] [7]Toze S.PCR and the detection of microbial pathogens in water andwastewater[J].Water Research,1999,33(17):3545-3556.

[0027] [8]Ou X,Chen P,Huang X,et al.Microfluidic chip electrophoresis forbiochemical analysis[J].Journal of separation science,2020,43(1):258-270.

[0028] [9]Miranda K M,Espey M G,Wink D A.A rapid,simple spectrophotometricmethod for simultaneous detection of nitrate and nitrite[J].Nitric oxide,2001,5(1):62-71.

[0029]

[10] Yuwadee B,Chakorn C,Orawon C,et al.Simple spectrophotometricsequential injection analysis system for determination of serum calcium[J].American Journal of Analytical Chemistry,2012,2012.

[11] Shao Y,Ying Y,PingJ.Recent advances in solid-contact ion-selective electrodes:Functionalmaterials,transduction mechanisms,and development trends[J].Chemical SocietyReviews,2020,49(13):4405-4465.

[0030]

[12] Bakker E,Bühlmann P,Pretsch E.Carrier-based ion-selectiveelectrodes and bulk optodes.1.General characteristics[J].Chemical reviews,1997,97(8):3083-3132.

[0031]

[13] Zuliani C,Diamond D.Opportunities and challenges of using ion-selective electrodes in environmental monitoring and wearable sensors[J].Electrochimica Acta,2012,84:29-34.

[0032]

[14] Duarte L T,Jutten C,Moussaoui S.A Bayesian nonlinear sourceseparation method for smart ion-selective electrode arrays[J].IEEE SensorsJournal,2009,9(12):1763-1771.

[0033]

[15] Gallardo J,Alegret S,Munoz R,et al.Use of an electronic tonguebased on all-solid-state potentiometric sensors for the quantitation ofalkaline ions[J].Electroanalysis:An International Journal Devoted toFundamental and Practical Aspects of Electroanalysis,2005,17(4):348-355.

[0034]

[16] Mimendia A, Gutiérrez JM, Leija L, et al. A review of the use of the potentiometric electronic tongue in the monitoring of environmental systems [J]. Environmental Modelling&Software, 2010, 25(9): 1023-1030.

[0035]

[17] Cho WJ,Kim HJ,Jung DH,et al.Hybrid signal-processing methodbased on neural network for prediction of NO3,K,Ca,and Mg ions in hydroponicsolutions using an array of ion-selective electrodes[J].Sensors,2019,19(24):5508.

[0036]

[18] Otto M, Thomas JD R. Model studies on multiple channel analysis of free magnesium, calcium, sodium, and potassium at physiological concentration levels with ion-selective electrodes[J]. Analytical Chemistry, 1985, 57(13): 2647-2651. Summary of the Invention

[0037] One of the objectives of this invention is to provide a multi-ion concentration detection chip for measuring the concentration of multiple ions in unknown solutions. Compared with traditional liquid contact electrodes and all-solid-state electrodes, it has the characteristics of high integration and portability.

[0038] The second objective of this invention is to provide a method for determining the concentration of multiple ions in a solution. This method uses a decoupling approach to remove interference, resulting in highly robust and accurate ion concentration predictions. Compared to traditional prediction algorithms that employ deep learning and artificial intelligence neural networks, this method has the advantages of low computational requirements, low computational power consumption, and data redundancy.

[0039] This invention provides a multi-ion concentration detection chip, including an FPC interface, discrete first and second thin-film transistors, a reference electrode, and a multi-channel sensing array composed of multiple SC-ISEs. The SC-ISEs are solid-contact ion-selective electrodes. The first and second thin-film transistors, the reference electrode, and the multi-channel sensing array are respectively connected to the FPC interface through wires.

[0040] As a further improvement of the present invention, the multi-channel sensing array includes at least two sensing units, each sensing unit including a container and a solid contact ion-selective electrode. The solid contact ion-selective electrode is disposed inside the container, and a sensitive membrane is provided inside the container. There are at least two solid contact ion-selective electrodes that detect the voltage at different sites on the sensitive membrane.

[0041] This invention also provides a method for determining the concentration of multiple ions in a solution, comprising the following steps:

[0042] S1, Electrode reset;

[0043] S2, Electrode concentration calibration;

[0044] S3. Establishment of the scaling factor model;

[0045] S4, Unknown solution test;

[0046] S5. Solve the model equations using the surface method;

[0047] S6, Multi-channel information combination;

[0048] S7, Multi-ion Concentration Display.

[0049] As a further improvement of the present invention, in step S1, the ion-selective electrode is immersed in pure water to reset it in order to clean the residual ions on the membrane.

[0050] As a further improvement of the present invention, step S2 includes the following sub-steps:

[0051] S21. Prepare the calibration group solution;

[0052] S22. Prepare the test group solution;

[0053] S23. Immerse the detection chip sequentially into the pure solution of the calibration group from low concentration to high concentration. When immersed in the solution, the surface potential of the ion-selective electrode will increase, forming a circuit with the reference electrode that is simultaneously immersed in the solution. Read the potential difference between the ion-selective electrode and the reference electrode.

[0054] As a further improvement of the present invention, in step S3, based on the potential difference between the ion-selective electrode and the reference electrode obtained in step S2, a scaling factor model is established as follows:

[0055]

[0056] Wherein, the logarithm of the concentration of ion 1 is x1, the response potential difference of electrode 1 is y1, the potential difference drift of electrode 1 is a, the responsivity of electrode 1 to ion 1 is a1, and the responsivity of electrode 1 to ion 2 is a2; the logarithm of the concentration of ion 2 is x2, the response potential difference of electrode 2 is y2, the potential difference drift of electrode 2 is b, the responsivity of electrode 2 to its own target ion is b1, and the responsivity of electrode 2 to the target ion of electrode 1 is b2.

[0057] Among them, y1 and y2 are obtained directly through chip sensing during measurement, x1 and x2 are the concentrations to be solved, and a1, a2, a, b1, b2, b are the values ​​to be calibrated.

[0058] As a further improvement of the present invention, in step S4, the chip with the established scaling factor model is immersed in the solution to be tested and left for a predetermined time. After the readings stabilize, the host computer reads the real-time voltage values ​​of multiple channels.

[0059] As a further improvement of the present invention, in step S5, the concentrations of ion 1 and ion 2 are obtained by the surface method, as follows:

[0060] S51. In step S4, the parameters [a1,a2,a,b1,b2,b] are obtained to form the bivariate quadratic parametric equation established in step S3, i.e., equation (1).

[0061] S52. Plot the first equation in equation (1) into a three-dimensional coordinate graph to obtain a monotonically changing surface graph. In step S3, when the chip detects an unknown solution, it obtains a potential difference value, which is a potential difference plane. The potential difference plane and the performance plane will intersect at a curve. Thanks to the monotonicity of the plane, when this curve is projected onto the plane, all the x1 and x2 combinations obtained are the logarithms of the possible concentrations of ion 1 and ion 2 corresponding to this potential difference value.

[0062] S53. Using the second equation in equation (1) as the limiting equation, obtain the unique logarithmic solution of the two-ion concentration combination.

[0063] As a further improvement of the present invention, in step S6, step S5 is repeated to obtain the total ion concentration.

[0064] As a further improvement of the present invention, step S8, temperature correction, is also included, wherein a discrete first thin-film transistor and a second thin-film transistor are integrated in the detection chip to obtain two measurement temperatures respectively, and the average method is used on the two measurement temperatures to provide a calibration temperature for calibration of the ambient temperature.

[0065] As a further improvement of the present invention, before performing step S8, temperature calibration is performed, and the electrode voltage response curve as a function of temperature is obtained by fitting.

[0066] As a further improvement of the present invention, in step S8, temperature correction is performed based on the Nernst equation;

[0067] The Nernst equation is:

[0068]

[0069] in,

[0070] In response to electric potential, The cumulative potential is independent, R is the molar gas constant, T is the absolute temperature, z is the target ion charge number, F is the Faraday constant, and α is the target ion activity in the solution; the absolute temperature T is the calibration temperature obtained by the first thin-film transistor and the second thin-film transistor.

[0071] The beneficial effects of this invention are: it provides a multi-ion concentration detection chip with high integration and portability; it provides a method for determining the concentration of multiple ions in a solution, employing a decoupling method to remove interference, achieving highly robust and accurate ion concentration prediction; compared to traditional prediction algorithms using deep learning and artificial intelligence neural networks, it has the advantages of low computational requirements, low computational power consumption, and data redundancy. Attached Figure Description

[0072] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other solutions can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a schematic diagram of a multi-ion concentration detection chip according to the present invention.

[0074] Figure 2 This is a flowchart of a method for determining the concentration of multiple ions in a solution according to the present invention.

[0075] Figure 3 This is a schematic diagram of the surface method for determining the concentration of multiple ions in a solution according to the present invention.

[0076] Figure 4 This is a test curve of the multi-concentration combination performance of a single calcium electrode for determining the concentration of multiple ions in a solution according to the present invention.

[0077] Figure 5This is a comparison chart of the calculation results and true values ​​of a method for determining the concentration of multiple ions in a solution according to the present invention. Detailed Implementation

[0078] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other.

[0079] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the scope of protection of this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0080] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0081] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0082] The main challenge in solid-state contact-type ion-selective electrode detection is that a specific sensitive membrane is not only capable of binding target ions but also of binding other ions with similar properties. Therefore, the potential signal obtained by the specific electrode is actually the result of a coupled response of multiple ion concentrations. Thus, the key issue is how to extract the required concentration signal from the mixed potential response. To address this, this invention proposes a model characterizing the relationship between potential and multiple ions, and offers a scheme for facilitating ion extraction.

[0083] Meanwhile, a secondary challenge with ion-selective electrodes is that the test results are affected by temperature, and using a separate temperature sensor would result in a large sensor size and poor portability. This invention proposes integrating a temperature sensor composed of thin-film transistors onto a detection chip.

[0084] Example 1

[0085] like Figure 1 As shown, a multi-ion concentration detection chip includes an FPC interface 1, a discrete first thin-film transistor T1 and a second thin-film transistor T2, a reference electrode 3, and a multi-channel sensing array 2 composed of multiple SC-ISEs.

[0086] SC-ISE is a solid contact type ion selective electrode. In this embodiment, a solid contact type ion selective electrode is used as the sensing electrode, and it is circular in shape.

[0087] The first thin-film transistor T1, the second thin-film transistor T2, the reference electrode 3, and the multi-channel sensor array 2 are respectively connected to the FPC interface 1 via wires.

[0088] Solid contact ion-selective electrodes (SC-ISEs) achieve different ion-selectivity properties depending on the type of ion-sensitive film attached to the electrode surface. This invention integrates multiple SC-ISEs into a multi-channel sensing array, which is then integrated with a temperature sensor onto a glass chip.

[0089] The chip measures 52mm in length and 40mm in width, with an FPC interface 1 at the top. Four 2x2 circular sensing electrodes are integrated in the lower center of the chip, serving as detection areas for calcium, magnesium, sodium, and potassium ions, respectively. Each circular sensing area has a diameter of 4mm and is spaced 1mm apart, with each of the four channels calculated independently. Compared to traditional liquid-contact ion-selective electrodes that are over ten centimeters long, the chip is small in size and can be integrated with PCB control boards of different sizes to build ion sensors in the form of detection pens or modules, thus offering portability.

[0090] The circular sensing areas, reference electrode 3, wires and FPC interface 1 on the chip are patterned on the glass substrate using screen printing.

[0091] Four different ion carriers will be used for the four sensitive membrane solutions: ETH5234 for calcium ions, benzo-15-crown ether-5 for magnesium ions, ETH4120 for sodium ions, and valinemycin for potassium ions. These solutions will be prepared by dissolving solid ion exchangers in tetrahydrofuran solution of PVC.

[0092] Each unique sensitive membrane solution is dotted into a square region belonging to a different ion, filling the square region completely. After evaporation, a sensitive membrane is obtained. Four sensing electrodes of a single ion will detect the voltage at four different sites on the same sensitive membrane, and the charge will be transferred along the wires to the FPC interface.

[0093] The chip and FPC are pressed together for connection, and the FPC is connected to the FPC connector on the signal acquisition PCB board. The PCB board integrates a 16-channel analog-to-digital converter chip, corresponding to the sixteen sensing areas on the detection chip. The acquisition board is connected to a host computer via a serial port, and visualization and algorithm programs are written in LabVIEW on the host computer.

[0094] Example 2

[0095] Based on the detection chip provided in Example 1, a corresponding detection method is proposed.

[0096] like Figure 2 As shown, a method for determining the concentration of multiple ions in a solution includes the following steps:

[0097] S1, Electrode reset;

[0098] S2, Electrode concentration calibration;

[0099] S3. Establishment of the scaling factor model;

[0100] S4, Unknown solution test;

[0101] S5. Solve the model equations using the surface method;

[0102] S6, Multi-channel information combination;

[0103] S7, Multi-ion Concentration Display.

[0104] The specific process of this method is as follows:

[0105] (1) Electrode reset

[0106] After each use, the ion-selective electrode needs to be immersed in pure water to reset it. The purpose of this is to clean the residual ions on the membrane. The ion-selective electrode is the sensing electrode in the detection chip.

[0107] The response of an ion-selective electrode to ions changes over time, and its state is unknown. Therefore, it is necessary to obtain the current state of the electrode through a calibration procedure to represent the performance of the electrode in the short term.

[0108] The preferred ion-selective electrode is a solid contact type ion-selective electrode.

[0109] (2) Electrode concentration calibration

[0110] Preparation of calibration solutions: Four sets of ion solution concentration steps 10 -5 mol / L, 10 -4 mol / L, 10 -3 mol / L, 10 - 2 A pure solution of mol / L.

[0111] Test group: Five groups of solutions with a constant total ion concentration, each with a total concentration of 10... -5 mol / L, 10 -4 mol / L, 10 - 3 mol / L, 10 -2 The solutions were in groups of mol / L with different concentrations, and the percentages of the target ions were 100%, 70%, 50%, 30%, and 0%.

[0112] Immerse the connected detection chips into the pure solution of the calibration group in sequence from low concentration to high concentration. When immersing in the same solution, they can be immersed in order. When immersing in different solutions, they need to be reset before immersion.

[0113] When immersed in the solution, the ions specifically bind to the ion-selective carriers on the four ion-selective films. These binding molecules move within the PVC film, forming an electrical double layer on the film surface. The charge carried by the binding molecules induces a charge on the sensing electrode, thus increasing the surface potential of the sensing electrode. This forms a circuit with a reference electrode simultaneously immersed in the solution. The potential difference between the sensing electrode and the reference electrode is read out using a differential circuit. After the host computer obtains the voltage responses of the sixteen sensing channels in the calibration group's solution, the performance of the multi-ion electrode system is calibrated using a scaling factor model.

[0114] After the chip is immersed in the solution, the host computer obtains 5 sets of voltage values ​​for 16 channels. Now, it is necessary to calibrate the electrodes based on these values, which is to obtain the unknown parameters of the solution model.

[0115] (3) Establishment of the scaling factor model

[0116] After the host computer obtains the voltage value of the detection chip in the solution of the test group, it establishes a scaling factor model. This model represents the sensing performance and electrode state of the detection chip at this time. Once the model is established, the detection chip has the ability to calculate the concentrations of various ions.

[0117] The model principle is as follows:

[0118] Theoretically, the relationship between the potential difference between the working electrode and the reference electrode and the target ion activity of its sensitive membrane will satisfy the Nernst equation.

[0119] In reality, no sensing membrane can achieve a response solely to a target ion; it can only achieve a maximum response to that target ion. That is, all ions present in the solution will affect the voltage value. To quantify this effect, this paper proposes a scaling factor model to describe it.

[0120] Let k be the percentage of a certain ion in the total concentration:

[0121]

[0122] x1 is the logarithm of the concentration of a certain ion, and xn is the logarithm of the concentration of all ions.

[0123] Formula (2) is the source of the model.

[0124] Based on the Nernst equation, this invention proposes a formula (3) for the voltage response in a multi-ion system:

[0125]

[0126] y represents the potential difference between a working electrode and a reference electrode, an represents the independent response of the working electrode to each ion in the solution (response sensitivity in pure solution), xn represents the logarithm of the concentration of each ion, and c represents the potential drift of the system. Here, xn is the value to be solved, and an and c are values ​​to be calibrated. Clearly, equation (3) is an n-variable quadratic equation, and the answer cannot be obtained by calculating the determinant due to its non-homogeneity, resulting in high equation complexity.

[0127] Formula (3) is the original model.

[0128] To make it solvable and reduce it to a binary quadratic form, two electrodes with different target ions need to work together to find the response sensitivity to their own target ion and the other's target ion in their respective formulas.

[0129] As shown in equation (4):

[0130]

[0131] Let x1 be the logarithm of the concentration of ion 1, y1 be the response potential difference of electrode 1, a be the potential difference drift of electrode 1, a1 be the responsivity of electrode 1 to ion 1, and a2 be the responsivity of electrode 1 to ion 2. Let x2 be the logarithm of the concentration of ion 2, y2 be the response potential difference of electrode 2, b be the potential difference drift of electrode 2, b1 be the responsivity of electrode 2 to its own target ion, and b2 be the responsivity of electrode 2 to the target ion of electrode 1.

[0132] Among them, y1 and y2 can be directly obtained through chip sensing during measurement, x1 and x2 are the concentrations to be solved, i.e. target values, and a1, a2, a, b1, b2, b are values ​​to be calibrated, which can be obtained through the next process.

[0133] Formula (4) is the objective equation.

[0134] In comparison, using neural network regression to calculate ion concentration requires four times the number of calibrations to obtain response parameters and train the model to achieve the same effect.

[0135] The voltage values ​​are converted into a model parameter matrix as follows:

[0136] The host computer uses the response voltage values ​​of the 16 channels and five groups obtained in the electrode calibration steps under each pure solution, combined with the equivalent equation (5) of the Nernst equation under the single-ion system.

[0137] y = ax + c (5)

[0138] Where y is the potential difference between the working electrode and the reference electrode, x is the logarithm of the target ion concentration, a is the sensitivity of the ion-selective electrode to the target ion, and c is the potential difference drift of the system.

[0139] The above formula shows that a single electrode has a linear response to different concentrations of a single ion. The calibration groups included calibration solutions with gradient concentrations for each single ion. Group 1: Concentration 10... -5 mol / L, 10 -4 mol / L, 10 -3 mol / L, 10 -2 mol / L. Therefore, based on the voltage values ​​at these four points, the responsivity 'a' of this electrode to this ion and its own potential shift 'c' can be fitted. Similarly, the matrices of the responses of the four types of electrodes to the four pure ions and their own voltage shifts can be obtained:

[0140]

[0141] Electrode parameters obtained through calibration testing.

[0142] a represents the potential difference drift of electrode 1 itself, a1 represents the response sensitivity of electrode 1 to ion 1, a2 represents the response sensitivity of electrode 1 to ion 2, a3 represents the response sensitivity of electrode 1 to ion 3, a4 represents the response sensitivity of electrode 1 to ion 4, and so on.

[0143] To determine the concentration of a specific ion among four ions, the two electrodes with the highest responsiveness to two ions in the response matrix are selected. For example, when determining the concentration of ion 1, the two largest values ​​among a1, b1, c1, and d1 should be chosen, and their corresponding two electrodes are used. Then, the ion with the second best common response among these two electrodes is selected. For example, when determining the concentration of ion 1, the maximum response is a1 and b1, so electrodes a and b are selected as the solution electrodes. In this case, electrodes a and b have a relatively large response to ion 2, so the parameters a2 and b2 are selected to construct the equation. There are six parameters: a, a1, a2, b, b1, and b2. Therefore, the bivariate surface equation in equation (4) is constructed.

[0144]

[0145] The parameters to be determined in the model were improved by selection.

[0146] The values ​​to be determined are the concentrations of two ions, x1 and x2, which need to be obtained by solving surface equations.

[0147] (4) Unknown solution test

[0148] The fully constructed detection chip is immersed in the test solution for 2 minutes. After the readings stabilize, the host computer reads the real-time voltage values ​​of the sixteen channels [y1, y2, ..., y16].

[0149] (5) Solving model equations using the surface method

[0150] like Figure 3 As shown, Figure 3 (a) is a schematic diagram of a single-performance surface; (b) is a diagram of the intersection of a single-performance surface and a potential plane; (c) is a diagram of the intersection of a dual-performance surface and its respective potential plane; and (d) is a diagram of the intersection of projections of dual concentration curves.

[0151] First, the concentrations of the two ions are solved. Based on the optimal selection principle mentioned in step (4), the obtained voltage values ​​are used to select appropriate matrix parameters [a1,a2,a,b1,b2,b] to form a two-variable quadratic parametric equation, as shown in equation (4). Now, the corresponding ion concentrations x1 and x2 need to be solved.

[0152] For the sake of convenience, the logarithmic coordinates of the concentrations of the two components are each increased by 6 to the positive domain, and the voltage response range is shifted to [0.3.3V]. The first equation in (4) is plotted as a three-dimensional coordinate graph as follows: Figure 3 As shown in (a), the result is a monotonically changing surface plot.

[0153] When an unknown solution is detected by ISEs, a potential difference value, i.e., a potential difference plane, is obtained, as shown in 2(b). The potential difference plane and the performance plane will intersect on a curve, thanks to the monotonicity of the plane. When this curve is projected onto the plane of y=0, all the possible combinations of x1 and x2 are the logarithms of the concentrations of possible ion 1 and ion 2 corresponding to this potential difference value.

[0154] To obtain a unique logarithmic solution for the combination of two ion concentrations, a constraint equation needs to be added. At this point, the second equation in equation (4) is also plotted to obtain the following surface: Figure 3 The projection lines in (b) represent the possible logarithmic combinations of calcium and magnesium ion concentrations, such as... Figure 3 As shown in (c) in the figure.

[0155] Since these two electrodes will indicate the ionic status in the same solution, both projection lines must be satisfied simultaneously, i.e., the intersection of the projection lines must be determined. Figure 3As shown in (d), the sparsity of actual sampling points may result in no intersection points. In this case, we find the two closest points and solve for the geometric midpoint between them as a representative to characterize the solution that is most likely to satisfy the two ISE performance surfaces. The coordinates of this point are the corresponding logarithmic values ​​of the concentrations of the two ions. After logarithmic transformation, we get the required values, and finally obtain the concentrations of the two ions using the surface method.

[0156] (6) Obtaining multi-ion concentration

[0157] In step (4), two ion concentrations were obtained. According to the optimal matrix parameter selection principle, the concentration of the initially selected main ion was chosen as the result from these two ion concentrations. The other three ions were processed in the same way as in step four, resulting in a total of four ion concentrations.

[0158] Each ion sensing region contains four sensing circles, each independently performing its sensing function. Solving for a single ion concentration requires combining two electrodes, resulting in 256 possible combinations. By selecting 4 or 8 sets of parallel equations, outliers are extracted, and the average is taken to eliminate the issue of variations in single sensing electrodes. Sufficient data redundancy is provided to eliminate problematic options; the data processing will involve removing non-significant terms and averaging the remaining terms.

[0159] At this point, the ion concentration calculation is complete, and a total of four ion concentration values ​​are obtained, thus completing the multi-ion concentration calculation.

[0160] (7) Temperature calibration

[0161] An on-chip temperature sensor is immersed in the solution along with the chip. The temperature sensor transmits current to the FPC interface, which is then read from the PCB and transmitted to the host computer for recording. This step can be performed before or after concentration calibration.

[0162] Temperature performance calibration requires measurements of pure ionic solutions at seven temperatures: 5°C, 10°C, 15°C, 20°C, 25°C, 30°C, and 35°C. The voltage values ​​of the sixteen-channel electrode at these seven temperatures are obtained, and the electrode voltage response curve as a function of temperature is fitted.

[0163] (8) Temperature correction

[0164] It is known that when recording the concentration of the solution during concentration calibration, the solution temperature needs to be recorded when actually measuring the ion concentration of the solution. By comparing the difference between the actual temperature and the concentration calibration temperature, the final output of each ion concentration is corrected according to the response temperature change relationship.

[0165] principle

[0166] The Nernst equation is known to be:

[0167]

[0168] in In response to electric potential, The cumulative potential is independent of the value, R is the molar gas constant, T is the absolute temperature, z is the target ion charge, F is the Faraday constant, and α is the target ion activity in the solution.

[0169] The second term includes factors such as constants R and F, charge number z, and temperature T. Only the temperature factor T is affected by the environment; in practice, it means that the response potential changes with temperature when entering a solution environment at different temperatures. This characteristic requires the ion-selective electrode to have a temperature compensation term. Other technical solutions typically use isolated thermometers for measurement, but these solutions have the disadvantage of being unable to simultaneously measure and control the temperature, and the thermometer and sensor may have different positions, potentially resulting in different water temperatures. The integrated temperature testing scheme proposed in this invention can acquire parameters within the same control system, achieving calibration of the ambient temperature. The T1 and T2 dual-temperature zones will use an averaging method to provide a temperature as a calibration term.

[0170] The principle of thin-film transistor temperature detection is that the higher the temperature, the greater the degree of opening of a single thin-film transistor TFT1. Therefore, under a fixed source-drain voltage Vdd, the higher the temperature, the greater the current through TFT1, and the smaller the node voltage between the load resistor R and TFT1. Based on this principle, the voltage-temperature relationship can be obtained experimentally.

[0171] Model validation and field testing

[0172] Actual testing verified the model's correctness.

[0173] Group 2: Prepare five groups with the same total ion concentration, each with a total concentration of 10. -5 mol / L, 10 -4 mol / L, 10 -3 mol / L, 10 -2 Solutions with different concentrations (mol / L) and calcium ion percentages relative to magnesium ion concentrations were 100%, 70%, 50%, 30%, and 0%, respectively.

[0174] To facilitate understanding, the electrode was first calibrated by immersing it in a pure solution representing the concentration gradient of Group 1, thus obtaining the parameter matrix, i.e., the complete model parameters. Next, the electrode was immersed in a solution containing five ions from Group 2 for testing, obtaining five performance planes. Based on the data, five concentration combinations were calculated and compared with the actual concentration combinations to observe the error.

[0175] The experimental results of the calcium electrode are as follows: Figure 4As shown, it can be observed that, from the perspective of the stepwise change in total concentration with the same ion ratio, the potential values ​​at the four total concentrations fit the linear performance with a high degree of fit, and the response lines at different ratios converge to 10. -6.5 Total concentration (mol / L); from the perspective of the same total ion concentration but different ion proportions, the higher the proportion of the target sample, the closer its response is to the response at pure concentration. This pattern basically satisfies the Nernst equation and the principle of the scaling factor model.

[0176] Based on the experimental results of the magnesium single electrode, ion calculations were performed using the concentration array. The comparison between the calculated results and the true values ​​is shown in the figure below. Figure 5 As shown, the stars mark the predicted values, and the blue marks mark the true values. The degree of overlap is satisfactory, with an average prediction error of 0.037. This verifies the scientific validity of the model.

[0177] For multi-component systems, the simulation process only requires adding dopants of other ions to the solution environment. The model remains the same, and the influence of these other ions will be approximately normalized to the intercept term. For different ions, their corresponding specific electrodes will have a significant response to their own ions, thus resulting in smaller errors. The concentrations of non-target ions obtained from other electrodes are typically not used.

[0178] This invention provides a multi-ion concentration detection chip that combines multiple ISEs with different selectivities to measure the concentration of multiple ions in an unknown solution, based on the specificity of ion-selective electrodes (ISEs). However, since the selectivity of ISEs for ions is not ideal, interfering ions with similar physicochemical properties can affect the potential of the ISEs, leading to distortion of the calculated ion activity. Therefore, based on the above detection chip, this invention also proposes a decoupling method to remove interference, achieving highly robust and accurate ion concentration prediction. Compared with traditional prediction algorithms using deep learning and artificial intelligence neural networks, this method has the advantages of low computational requirements, low computational power consumption, and data redundancy.

[0179] The present invention provides a multi-ion concentration detection chip and a method for determining the concentration of multiple ions in a solution, which have the following advantages:

[0180] 1. A detection chip based on a multi-ion, multi-channel ion-selective electrode made of a special conductive material was designed. It has the potential to design a sensor array, enabling the ion sensing electrode to achieve online and portable measurement. It has the characteristics of high integration and portability, and can solve the problem of ion interference.

[0181] 2. Based on experimental dynamics and experimental laws, a solution model with simple calibration steps, low implementation difficulty, and convenient calculation is proposed.

[0182] 3. A temperature sensor unit composed of thin-film transistors is integrated on the same chip to serve as a temperature compensation parameter, thereby realizing an integrated temperature calibration function and improving detection accuracy;

[0183] 4. Signal processing circuitry is also integrated on-chip, eliminating the need for external processing. This reduces latency and noise.

[0184] 5. Multi-ion detection is achieved using chip-level SC-ISEs. Taking calcium and magnesium ions with similar physicochemical properties as examples, a scaling factor model is proposed to reduce the number of sampling points required for calibration.

[0185] 6. A multi-ion decoupling model for ion-selective electrode chip arrays, namely the scaling factor model, was established. The solvability in a binary system was experimentally verified.

[0186] 7. This invention does not require the collection of large amounts of training data to build the model. Neural network algorithms typically require sampling at very small concentration intervals, resulting in complex pre-calibration procedures that struggle to handle electrode drift. Furthermore, the electrode performance may change after data collection, rendering the constructed network meaningless.

[0187] 8. This invention is applied to microelectrode array chips and has high integrability;

[0188] 9. The computational complexity of this invention is two-dimensional, requiring minimal machine computing power, and can be integrated into a simple microcontroller to achieve control and computation.

[0189] 10. This invention integrates a unified temperature calibration, which more accurately and conveniently compensates for the effects of temperature.

[0190] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for determining the concentration of multiple ions in a solution, characterized in that: A multi-ion concentration detection chip is provided, comprising an FPC interface, discrete first and second thin-film transistors, a reference electrode, and a multi-channel sensing array composed of multiple SC-ISEs. The SC-ISEs are solid-state contact ion-selective electrodes. The first and second thin-film transistors, the reference electrode, and the multi-channel sensing array are respectively connected to the FPC interface via wires. The multi-channel sensing array includes at least two sensing units. Each sensing unit includes a container and a solid-state contact ion-selective electrode. The solid-state contact ion-selective electrode is disposed within the container, and a sensitive membrane is disposed within the container. There are at least two solid-state contact ion-selective electrodes, which detect the voltage at different sites on the sensitive membrane. Based on the multi-ion concentration detection chip, the following steps are performed: S1, Electrode reset; S2, Electrode concentration calibration; S3. Establishment of the scaling factor model; S4, Unknown solution test; S5. Solve the model equations using the surface method; S6, Multi-channel information combination; S7, Multi-ion concentration display; In step S1, the ion-selective electrode is immersed in pure water to reset it in order to clean the residual ions on the membrane; Step S2 includes the following sub-steps: S21. Prepare the calibration group solution; S22. Prepare the test group solution; S23. Immerse the detection chip in the pure solution of the calibration group in sequence from low concentration to high concentration. When immersed in the solution, the surface potential of the ion selective electrode will increase, forming a circuit with the reference electrode that is immersed in the solution at the same time. Read the potential difference between the ion selective electrode and the reference electrode. In step S3, based on the potential difference between the ion-selective electrode and the reference electrode obtained in step S2, a scaling factor model is established as follows: （1） The logarithm of the concentration of ion 1 is x1, the response potential difference of electrode 1 is y1, the potential difference drift of electrode 1 is a, the responsivity of electrode 1 to ion 1 is a1, and the responsivity of electrode 1 to ion 2 is a2; the logarithm of the concentration of ion 2 is x2, the response potential difference of electrode 2 is y2, the potential difference drift of electrode 2 is b, the responsivity of electrode 2 to its own target ion is b1, and the responsivity of electrode 2 to the target ion of electrode 1 is b2. Among them, y1 and y2 are obtained directly through chip sensing during measurement, x1 and x2 are the concentrations to be solved, and a1, a2, a, b1, b2, b are the values ​​to be calibrated.

2. The method for determining the concentration of multiple ions in a solution according to claim 1, characterized in that: In step S4, the chip with the established scaling factor model is immersed in the solution to be tested and left for a predetermined time. After the readings stabilize, the host computer reads the real-time voltage values ​​of multiple channels.

3. The method for determining the concentration of multiple ions in a solution according to claim 2, characterized in that: In step S5, the concentrations of ion 1 and ion 2 are obtained using the surface method, as follows: S51. In step S4, the parameters [a1,a2,a,b1,b2,b] are obtained to form the bivariate quadratic parametric equation established in step S3, i.e., equation (1). S52. Plot the first equation in equation (1) into a three-dimensional coordinate graph to obtain a monotonically changing surface graph. In step S3, when the chip detects an unknown solution, it obtains a potential difference value, which is a potential difference plane. The potential difference plane and the performance plane will intersect at a curve. Thanks to the monotonicity of the plane, when this curve is projected onto the plane, all the x1 and x2 combinations obtained are the logarithms of the possible concentrations of ion 1 and ion 2 corresponding to this potential difference value. S53. Using the second equation in equation (1) as the limiting equation, obtain the unique logarithmic solution of the two-ion concentration combination.

4. The method for determining the concentration of multiple ions in a solution according to claim 3, characterized in that: In step S6, step S5 is repeated to obtain the total ion concentration.

5. The method for determining the concentration of multiple ions in a solution according to claim 1, characterized in that: It also includes step S8, temperature correction, in which a discrete first thin-film transistor and a second thin-film transistor are integrated in the detection chip to obtain two measurement temperatures. The average value method is used to provide a calibration temperature for calibration of the ambient temperature.

6. The method for determining the concentration of multiple ions in a solution according to claim 5, characterized in that: Before proceeding to step S8, temperature calibration is performed, and the electrode voltage response curve as a function of temperature is obtained by fitting. In step S8, temperature correction is performed based on the Nernst equation; The Nernst equation is: （2） in, In response to electric potential, Let be the independent cumulative potential, R be the molar gas constant, T be the absolute temperature, z be the target ion charge number, and F be the Faraday constant. The activity of the target ion in the solution; The absolute temperature T is the calibration temperature obtained through the first and second thin-film transistors.

Images