A sweat multi-parameter ion detection sensor, system and method

Abstract

The present application relates to the technical field of biosensor, in particular to a sweat multi-parameter ion detection sensor, system and method, wherein the sensor comprises a substrate layer, an electrode layer and a packaging layer arranged in sequence from bottom to top, the electrode layer comprises a gate, at least three electrode pairs arranged in array around the gate, and a plurality of pins, the electrode pair comprises a drain and a source which are parallel to each other, a PEDOT:PSS film, a solid-state electrolyte and an ion-selective membrane are sequentially arranged on the electrode pair from bottom to top, the plurality of pins are arranged side by side, the drain, the source and the gate are electrically connected with one pin respectively, each electrode pair and the gate form an OECT sensing unit, the ion concentration in the sweat to be detected is detected, the solid-state electrolyte is [MTEOA][MeOSO3] ion gel combined with gelatin, which not only has good electrochemical performance, but also has good performance in mechanical properties (flexibility and tensile strength) and biocompatibility, and the bending resistance is obviously improved compared with the prior art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of biosensors, in particular to a sweat multi-parameter ion detection sensor, system and method. BACKGROUND

[0002] The Na⁺ (sodium ion), K⁺ (potassium ion), Ca²⁺ (calcium ion) and other ion concentrations in sweat can reflect various physical conditions such as body fatigue, mental stress, neuromuscular function, etc. By analyzing the concentrations of various ions in sweat, potential diseases can be analyzed, and real-time monitoring of the user's physical condition can be achieved, which has important value for optimizing sports performance and preventing damage caused by dehydration and electrolyte imbalance. Therefore, the detection of various ion concentrations in sweat and its application in wearable devices have become a research hotspot.

[0003] Organic electrochemical transistors (OECT) have become an ideal choice for building such flexible sensing platforms due to their inherent mechanical flexibility, excellent biocompatibility, high transconductance, large on / off ratio and high sensitivity, especially for collecting and amplifying weak biological signals. However, traditional OECTs rely on liquid electrolytes (such as NaCl solution, phosphate buffer, artificial sweat, etc.), which have limited operating temperature range and risk of electrolyte leakage, severely restricting the long-term stable packaging and reliability of the device, making it difficult to meet the requirements of wearable applications.

[0004] In contrast, solid-state electrolytes (such as hydrogels or ion solutions + solid polymer matrix) exhibit mechanical flexibility, environmental robustness and potential as a bio-electronic interface. However, solid-state electrolytes still face many challenges when applied to wearable devices in practice. On the one hand, wearable devices are long-term exposed to high temperature (close to human body temperature) and high humidity environments, where the performance of electrodes, ion-selective membranes and other components will change significantly after long-term use. On the other hand, wearable devices inevitably encounter bending and folding conditions, which have high requirements for the mechanical properties (toughness, tensile strength, etc.) of the sensor. The OECT sensors provided by the prior art lack high-performance sensors that are highly flexible in spacing, stretchable, and fully compatible with sweat environments. At the same time, there is no effective solution to the problem of how to simultaneously measure multiple ions in sweat. In some OECT sensors using solid-state electrolytes, the problem of separation between solid-state electrolyte and ion-selective membrane occurs. SUMMARY

[0005] In order to solve the technical problem that the measurement result of the existing OECT sensor deviates due to the large change of the contact state between the solid-state electrolyte and the ion-selective membrane under long-term bending working conditions, the application provides a sweat multi-parameter ion detection sensor, system and method, wherein the sensor comprises a substrate layer, an electrode layer and a packaging layer arranged in sequence from bottom to top.

[0006] The substrate layer is used to support the electrode layer and the packaging layer.

[0007] The electrode layer comprises a gate, at least three electrode pairs arranged in an array around the gate, and a plurality of pins, the electrode pair comprises a drain and a source parallel to each other, a PEDOT:PSS film, a solid-state electrolyte and an ion-selective membrane are sequentially arranged on the electrode pair from bottom to top, a plurality of pins are arranged side by side, the drain, the source and the gate are respectively electrically connected with one pin, each electrode pair forms an OECT sensing unit with the gate, so as to realize detection of ion concentration in the measured sweat.

[0008] The solid-state electrolyte is [MTEOA][MeOSO3] ion gel combined with gelatin.

[0009] The packaging layer is used to package the electrode layer, and the packaging layer is provided with through holes corresponding to the gate and the electrode pair, so as to make the measured sweat contact with the gate and the electrode pair.

[0010] The application further discloses a preparation method for preparing the above-mentioned sensor, comprising the following steps:

[0011] The substrate layer is prepared by a spin coating process using a PI film.

[0012] Photoresist is spin-coated on the substrate layer, and an electrode layer is formed on the substrate layer by using photolithography combined with magnetron sputtering.

[0013] Ag is sputtered on the gate by using photolithography combined with magnetron sputtering, and the gate surface is formed with an Ag / AgCl layer by using HCl solution immersion.

[0014] The packaging layer mixture is spin-coated on the electrode layer and the substrate layer, and the gate and all the electrode pairs are exposed by using photolithography.

[0015] PEDOT:PSS solution is spin-coated on the electrode pair, and is heated and solidified to form a PEDOT:PSS film.

[0016] The prepared [MTEOA][MeOSO3] gelatin solution is drop-coated on the surface of the PEDOT:PSS film, and annealing operation is performed to solidify the [MTEOA][MeOSO3] gelatin solution to form a solid-state electrolyte.

[0017] Different ion selective membrane solutions are respectively dropped on different electrode pairs, dried at room temperature to form ion selective membranes, and the sensor preparation is completed.

[0018] The system disclosed in the application comprises the above sensor, a temperature sensor, a humidity sensor, an ADC module and a data processor.

[0019] The temperature sensor is used to measure the temperature of the sensor, and the humidity sensor is used to measure the humidity of the air near the sensor.

[0020] The ADC module is used to periodically acquire real-time drain current, real-time gate voltage, real-time temperature and real-time humidity of the three electrode pairs, and transmit the data to the data processor.

[0021] The data processor is used to calculate the theoretical concentration of the target ion according to the real-time drain current, real-time gate voltage, real-time temperature and real-time humidity, and correct the theoretical concentration according to the standard concentration-current voltage data to obtain the corrected concentration.

[0022] The method disclosed in the application comprises the following steps:

[0023] Periodically acquire real-time drain current, real-time gate voltage, real-time temperature, real-time humidity, and perform smoothing processing.

[0024] According to the real-time temperature and real-time humidity, the motion state is classified, and the Nernst model is corrected according to the motion state classification result.

[0025] The real-time gate voltage and real-time drain current are substituted into the corrected Nernst model to calculate the theoretical concentration.

[0026] According to the real-time drain current, real-time gate voltage and theoretical concentration, a real-time state vector is constructed, and the standard concentration-current voltage data is corrected by multi-dimensional data interpolation according to the real-time state vector to obtain the corrected concentration.

[0027] Technical effects and advantages of the present application: the sensor of the present application adopts [MTEOA][MeOSO3] mixed with gelatin to prepare [MTEOA][MeOSO3] ionic gel as a solid electrolyte, which not only has good electrochemical performance, but also has good performance in mechanical properties (flexibility and tensile strength) and biocompatibility, and the bending resistance is obviously improved compared with the prior art. At the same time, the system provided by the present application is based on this sensor, by adding temperature sensor and humidity sensor to monitor the motion state of the user, and then correcting the Nernst model according to the motion state, eliminating the influence of sweat secretion rate on the measurement result. At the same time, according to the experimental data of the standard concentration, the theoretical concentration is corrected by multidimensional interpolation to judge whether the theoretical concentration is accurate, so as to filter out abnormal values, eliminate errors caused by poor contact, and give confidence, so as to help the user or the program designer to judge the accuracy of the compensation result, and then the subsequent program can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 It is the exploded view of the overall structure of the sensor of the present application.

[0029] Figure 2 It is the assembly view of the overall structure of the sensor of the present application.

[0030] Figure 3 It is the cross-sectional view of the sensor of the present application.

[0031] Figure 4 It is the electrode layer diagram of the sensor of the present application.

[0032] Figure 5 It is the bending times-transconductance curve diagram of the sensor of the present application.

[0033] Figure 6 It is the overall structure diagram of the system of the present application.

[0034] Figure 7 It is the exploded view of the structure of the system of the present application after integrating the film thermal resistance wire with the sensor.

[0035] Figure 8 It is the overall flow chart of the detection method of the present application.

[0036] Figure 9 It is the flow chart of correcting the theoretical concentration in the detection method of the present application.

[0037] The reference signs are: 1, base layer; 2, electrode layer; 21, electrode pair; 211, drain; 212, source; 213, PEDOT:PSS film; 214, solid-state electrolyte; 215, ion-selective membrane; 22, gate; 23, pin; 3, packaging layer; 4, temperature sensor; 100, sweat to be measured. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative labor fall within the protection scope of the present application.

[0039] Embodiment one

[0040] Reference Figures 1 to 4 Embodiment one of the present application provides a sweat multi-parameter ion detection sensor, which comprises, from bottom to top, a base layer 1, an electrode layer 2 and a packaging layer 3. The base layer 1 is a support structure of the sensor, the electrode layer 2 is arranged on the base layer 1, and the packaging layer 3 covers the electrode layer 2.

[0041] The base layer 1 as the support structure of the sensor can adopt a PI (Polyimide) film, which has excellent insulation performance and chemical stability, can effectively prevent the interference and erosion of external environmental factors on the internal circuit of the sensor, and protect the normal work of the gate 22 and the electrode pair 21. At the same time, the PI film base layer 1 layer can also provide certain mechanical support and enhance the overall strength of the base layer 1, so that the sensor is not easy to be damaged when subjected to certain external force.

[0042] The electrode layer 2 comprises at least three electrode pairs 21, a gate 22 and a plurality of pins 23. The gate 22 is arranged at the middle position on the base layer 1. The gate 22 comprises a conductive layer and a reaction layer. The conductive layer adopts a laminated structure of 20nm chromium and 100nm gold from bottom to top. The 20nm chromium is used to strengthen the adhesion between the gold and the base layer 1. The reaction layer is a layer of Ag / AgCl (silver / silver chloride) attached to the gold, so as to serve as a reference electrode. The gate 22 can be arranged as a circle with a diameter of 3000μm as shown in FIG. 2, or can be arranged in other shapes. Figure 1

[0043] ​Three electrode pairs 21 are arranged around the array of gate electrodes 22, in particular, two of the electrode pairs 21 are parallel to each other, and the other electrode pair is perpendicular to the other two electrode pairs 21, and the three electrode pairs 21 are respectively used to detect the concentrations of Na⁺, K⁺ and Ca²⁺ in the sweat 100 to be detected, and each electrode pair 21 includes a drain electrode 211 and a source electrode 212 which are parallel to each other, and the source electrode 212 and the drain electrode 211 are also arranged in a laminated structure of 20nm chromium and 100nm gold, and the shapes of the source electrode 212 and the drain electrode 211 are set as interdigital as shown in Figure 4 , and the lengths of the source electrode 212 and the drain electrode 211 are both L=2000μm, the widths are 200μm, and the distance between the drain electrode 211 and the source electrode 212 is W=100-300μm, and a channel of a semiconductor is formed between the drain electrode 211 and the source electrode 212.

[0044] Referring to Figure 3 , the upper surface of each electrode pair 21 is sequentially covered with a PEDOT:PSS (poly 3,4-ethylenedioxythiophene:polystyrene sulfonate) film, a solid-state electrolyte 214 and an ion-selective membrane 215, and the ion-selective membranes 215 on different electrode pairs 21 are different, and specific reactions of Na⁺, K⁺ and Ca²⁺ are realized through the ion-selective membranes 215, so that each electrode pair 21 can independently detect a specific target ion.

[0045] The drain electrode 211 and the source electrode 212 of each electrode pair 21 and the gate electrode 22 are respectively electrically connected to a pin 23, and a plurality of pins 23 are arranged side by side and used to connect with an external circuit to realize the provision of current to the source electrode 212, the output of current from the drain electrode 211 and the output of voltage from the gate electrode 22, and each electrode pair 21 and the gate electrode 22 can form an independent OECT sensing unit, and a plurality of OECT units formed by a plurality of electrode pairs 21 can simultaneously measure the concentrations of a plurality of ions. The connecting wires between the drain electrode 211, the source electrode 212 and the pin 23 are provided with a circular arc transition section, so that the cross section of the connecting wire remains unchanged, thereby ensuring that the resistance of the connecting wire at different positions does not change significantly, preventing oscillation of the electrical signal due to sudden changes in resistance during transmission, and improving the signal-to-noise ratio of the signal. When the gate electrode 22 adopts Figure 4 the circular design as shown in the figure, the connecting wires between the gate electrode 22 and the pin 23 are also designed to have a smooth transition section to avoid sudden changes in resistance in the connecting wires.

[0046] Specifically, the solid-state electrolyte 214 adopts [MTEOA][MeOSO3] ion gel combined with [MTEOA][MeOSO3] and gelatin. [MTEOA][MeOSO3] is a biocompatible ionic solution, which has good conductivity when used as an internal electrolyte, and can effectively improve the transconductance of the PEDOT:PSS film and the sensitivity of the sensor after being combined with the PEDOT:PSS film, and shorten the response time.

[0047] The PEDOT:PSS film serves as a channel between the drain 211 and the source 212, and the channel current is regulated by the ions in the [MTEOA][MeOSO3] gelatin. Specifically, the anion MeOSO3- has a small size, a high charge density, and a migration rate significantly higher than that of the cation, and becomes the main carrier. After being in contact with the PEDOT:PSS film, a high-efficiency ion-doped interface is formed. When the gate 22 voltage is negative gate voltage (i.e., V_G<0), MeOSO3- migrates into the PEDOT:PSS film, the PEDOT+ in the PEDOT:PSS film is compensated, the conductivity of the PEDOT:PSS film decreases, and the drain 211 current I_D decreases. When the gate 22 voltage is positive gate voltage (i.e., V_G>0), MeOSO3- migrates out of the PEDOT:PSS film, PSS- re-binds PEDOT+, the conductivity of the PEDOT:PSS film increases, and the drain 211 current I_D increases.

[0048] The quaternary ammonium structure of the cation MTEOA+ of [MTEOA][MeOSO3] is resistant to reduction, and the anion MeOSO3- is resistant to oxidation, so that the prepared solid-state electrolyte 214 can work in a larger gate 22 voltage range, avoiding decomposition of the electrolyte in a sweat environment for a long time, and effectively improving the stability of the sensor.

[0049] The present application mixes [MTEOA][MeOSO3] with gelatin to prepare [MTEOA][MeOSO3] ion gel as the solid-state electrolyte 214 of the sensor, which not only has good electrochemical performance, but also has good performance in mechanical properties (flexibility and tensile strength) and biocompatibility. The three hydroxyl groups (-OH) in [MTEOA][MeOSO3] form strong hydrogen bonds with the carboxyl / amino groups of gelatin, destroy the original hydrogen bond network of gelatin, the quaternary ammonium structure of MTEOA+ has electrostatic attraction with the carboxyl groups of gelatin, and MeOSO3- forms an ion pair with the amino groups (-NH3+) of gelatin, introduces dynamic crosslinking points, enhances the movement ability of gelatin segments, and at the same time, through the reconstruction of the network by dynamic ionic bond / hydrogen bond, the gelatin has self-repairing function.

[0050] The [MTEOA][MeOSO3] ionic gel can significantly improve the adsorption effect of the ion-selective membrane 215, so that the ion-selective membrane 215 is not easy to fall off after long-term use, and the service life of the sensor is improved. The carboxyl / amino in the [MTEOA][MeOSO3] ionic gel can anchor the ion carrier (such as valinomycin) through covalent bond or electrostatic interaction, and the [MTEOA][MeOSO3] ionic gel has a microporous structure, which can make the interface of the two fuse with each other when the ion-selective membrane 215 is prepared on the surface of the [MTEOA][MeOSO3] ionic gel, thereby enhancing the adsorption force of the ion-selective membrane 215. Through the action of the two mechanisms, the combination of the ion-selective membrane 215 and the [MTEOA][MeOSO3] ionic gel is more closely than that of the other solid-state electrolyte 214, and after 1000 bending experiments, the transconductance of the sensor does not change obviously, which proves that the ion-selective membrane 215, the solid-state electrolyte 214 and the PEDOT:PSS film do not have problems such as falling off and cracking.

[0051] The ion-selective membrane 215 can selectively allow a certain ion to pass through and block other ions, and through cooperation with the solid-state electrolyte 214 and the PEDOT:PSS film channel layer, the electrode pair 21 has a measurement function for a certain specific ion. There are many mature preparation processes for the ion-selective membrane 215 for Na⁺, K⁺ and Ca²⁺ respectively, which have been disclosed in many existing technologies.

[0052] The packaging layer 3 is covered on the electrode layer 2, and through holes are formed in the corresponding positions of each electrode pair 21 and the gate 22 to enable the sweat to contact the electrode pair 21 and the gate 22. The packaging layer 3 is also provided with through holes in the corresponding positions of each pin 23 to facilitate the connection of the pin 23 with the external circuit. The through holes on the electrode pair 21, the gate 22 and the pin 23 are not connected with each other, so as to ensure that each electrode pair 21 and the gate 22 are in contact with the sweat separately, thereby preventing the conductivity of the sweat from affecting the measurement of each electrode.

[0053] Specifically, the through hole corresponding to the electrode pair 21 can be set as a rounded rectangle as shown in Figure 1 The through hole corresponding to the gate 22 can be set as a circle to correspond to the shape of the gate 22.

[0054] Through the above design, the sensor of the present application can realize multi-parameter sweat detection, and the sharing of the gate 22 by the multiple electrode pairs 21 can realize the miniaturization and integration of the sensor, thereby reducing the volume of the sensor. On the one hand, the sensor can reduce the space occupied in the use process, facilitate the integration with other devices, and provide greater convenience for the design of wearable sweat detection devices. On the other hand, the smaller volume also makes the sensor more portable, so that the user does not feel too much burden when wearing, thereby improving the comfort and convenience of use.

[0055] In practical applications, such a miniaturized, high-precision sweat multi-parameter ion detection sensor can be widely used in the field of sports health monitoring. For example, during the training process, athletes can monitor various ion parameters in sweat, such as sodium ions and potassium ions, in real time by wearing a device integrated with the sensor, helping coaches and athletes understand the body's water and electrolyte balance, and adjusting training plans and nutritional supplements in a timely manner. In addition, in the medical field, the sensor can also be used for early diagnosis and health management of diseases, providing doctors with more diagnostic evidence through detection of specific ions in sweat.

[0056] The preparation process of the above sensor is as follows:

[0057] (a) The glass substrate is placed in an ultrasonic cleaning tank containing isopropyl alcohol for 15 min, then the glass substrate is rinsed with deionized water, and then slowly blown with a nitrogen gun to achieve rapid drying, and finally transferred to a UV light cleaning machine for cleaning for 15 min;

[0058] (b) The copper substrate is sequentially placed in ultrasonic cleaning tanks containing acetone, anhydrous ethanol, and deionized water, and ultrasonically cleaned for 5 min each time; after cleaning, it is placed on a 90°C hot plate for 30 min to dry;

[0059] (c) Use a double-sided thermal release tape that can release at 100°C to paste the copper substrate on the glass substrate;

[0060] (d) Spin-coat the PI film on the copper substrate that has been baked on a 100°C hot plate for 4 minutes, then bake in a 100°C oven for 1 minute to obtain the base layer 1;

[0061] (e) Spin-coat photoresist (positive photoresist AZ6420) on the base layer 1 (PI film), ablate the electrode pattern by laser, including multiple electrode pairs 21, gate 22, pins 23, and connecting wires, clean the photoresist, expose the electrode layer 2 channel on the photoresist, then sputter 20 nm chromium and 100 nm gold on the photoresist in sequence to form a metal layer, and finally clean and peel off the residual photoresist with acetone, leaving the electrode layer 2 on the base layer 1;

[0062] (f) Spin-coat photoresist on the base layer 1 containing the electrode layer 2, laser ablate the gate 22 pattern on the photoresist, then clean the photoresist to expose the metal layer of the gate 22, then sputter Ag on the gate 22 and soak in HCl solution to form an Ag / AgCl layer on the surface of the gate 22, then clean and peel off the residual photoresist with acetone to complete the preparation of the gate 22;

[0063] (g) spin-coat the entire device with the prepared encapsulation layer 3 mixture, and then expose the gate 22 and the three electrode pairs 21 by laser ablation of the areas corresponding to the gate 22, the three electrode pairs 21, and the pin 23, to complete the preparation of the encapsulation layer 3 of the device;

[0064] (h) after the device is baked on a hot plate at 100°C for 5 minutes, 7 μL of the prepared PEDOT:PSS solution is dropped on each electrode pair 21 to ensure that the solution completely covers the electrode pair 21, and then spin-coated at a speed of 1200 rpm and an acceleration of 500 rpm / s for 35 s, and then the device is transferred to a hot plate at 130°C for heating for 60 minutes to completely dry the PEDOT:PSS film and form a uniform thickness, to complete the preparation of the PEDOT:PSS film;

[0065] The PEDOT:PSS solution can be prepared by the following steps:

[0066] First, high-purity 93.75 vol% PEDOT:PSS stock solution (e.g., purchased from Heraeus Technologies, model Clevios P VP AI 4083) is precisely mixed with 5 vol% ethylene glycol (as a co-solvent and conductivity enhancer), 0.25 vol% dodecylbenzenesulfonic acid (DBSA, as a surfactant and dopant), and 1 vol% (3-glycidoxypropyl)trimethoxysilane (GOPS, as a cross-linking agent and adhesion promoter) (all reagents are purchased from Sigma-Aldrich), and the mixture is placed on a magnetic stirrer and continuously stirred at a speed of 500 rpm at room temperature (25°C) for 8 to 10 hours until the solution becomes transparent and uniform without any visible precipitate or stratification, to complete the preparation of the PEDOT:PSS solution;

[0067] (i) the prepared [MTEOA][MeOSO3] gel solution is heated at 80°C to remain liquid, and then cast on the surface of the PEDOT:PSS film, and the device is first annealed at 95°C for 15 minutes and then at 130°C for 15 minutes to gradually solidify the liquid [MTEOA][MeOSO3] gel solution into [MTEOA][MeOSO3] ionic gel to form a solid-state electrolyte 214;

[0068] The [MTEOA][MeOSO3] gel solution is prepared by the following steps:

[0069] Gelatin (Type A, purchased from Sigma-Aldrich) was mixed with deionized water (ultra-pure water) at a mass ratio of 1:2, annealed at 80 °C, and stirred with a centrifugal stirrer from time to time until completely dissolved to form a gelatin aqueous solution. 80 wt% [MTEOA][MeOSO3] ionic liquid (e.g., purchased from IoLiTec Ionic Liquids Technologies GmbH, purity greater than 99%) was added to the gelatin aqueous solution and annealed at 80 °C, stirred with a stirrer from time to time until mixed evenly, and then stored at 24 °C, 67% relative humidity for at least 16 hours to allow the [MTEOA][MeOSO3] to fully react with the gelatin to form a gelatin solution;

[0070] (j) drop-coat ion selective membrane 215 solution for Na+, K+and Ca2+onto three electrode pairs 21 respectively, so that the ion selective membrane 215 solution fully covers the three electrode pairs 21, and then place the device in a room temperature environment to dry overnight, to complete the preparation of the ion selective membrane 215;

[0071] The ion selective membrane 215 solution for Na+, K+and Ca2+respectively belongs to the prior art, and the specific preparation process has been disclosed by multiple prior arts.

[0072] The present application also discloses a preparation process of an ion selective membrane 215 for preparing an ion selective membrane 215 solution, which specifically comprises the following steps:

[0073] Preparation of Na+ion selective membrane 215 solution:

[0074] Mix 1 wt% Na+carrier X, 0.55 wt% sodium tetra(4-chlorophenyl)borate, 33% m / m polyvinyl chloride, and 65.45 wt% dioctyl sebacate, and stir for 1 hour to allow the components to fully react to produce a Na+ion selective membrane 215 mixture. Then, dissolve 100 mg of the ion selective membrane 215 mixture in 650 μl of tetrahydrofuran, and stir for 7 to 8 hours to complete the preparation of the Na+ion selective membrane 215 solution.

[0075] The preparation process of the K+ion selective membrane 215 solution is similar to that of Na+:

[0076] Mix 2 wt% K+carrier III, 0.55 wt% sodium tetra(4-chlorophenyl)borate, 33% m / m polyvinyl chloride, and 64.45 wt% dioctyl sebacate, and stir for 1 hour to allow the components to fully react to produce a K+ion selective membrane 215 mixture. Then, dissolve 100 mg of the ion selective membrane 215 mixture in 650 μl of tetrahydrofuran, and stir for 7 to 8 hours to complete the preparation of the K+ion selective membrane 215 solution.

[0077] The preparation process of the Ca2+ ion-selective membrane 215 solution is also basically the same:

[0078] After 1 wt% Ca2+ carrier II, 0.55 wt% sodium tetra(4-chlorophenyl)borate, 33% m / m polyvinyl chloride, and 65.45 wt% dioctyl sebacate are mixed and stirred for 1 hour to allow the components to fully react, a K+ ion-selective membrane 215 mixture is generated, 100 mg of the ion-selective membrane 215 mixture is dissolved in 650 μl of tetrahydrofuran, and stirring is performed for 7 to 8 hours to complete the preparation of the K+ ion-selective membrane 215 solution.

[0079] (k) The device is placed on a hot plate at 130°C to remove the thermal release tape, and then the device is placed in a copper etching solution to etch the copper substrate and release the device. The peeled PI film is thoroughly rinsed with deionized water to remove residual etchant, and then dried under a nitrogen stream to complete the preparation of the sensor.

[0080] To verify the performance of the sensor, electrochemical performance and mechanical bending experiments are performed.

[0081] Figure 5 Bending experiment data of the sensor of the present application are shown. The bending experiment is performed after the three electrode pairs 21 and the gate electrode 22 are all dropped with sweat. Electrochemical performance testing is performed once every 250 bends. A comparative experiment is also performed on a liquid electrolyte sensor. The liquid electrolyte sensor is identical to the sensor of the present application in structure, electrode material, geometric size, channel layer, and composition of the ion-selective membrane 215, and only uses a liquid electrolyte instead of the solid electrolyte 214 ([MTEOA][MeOSO3] ion gel). The liquid electrolyte layer is prepared by dissolving 2 wt% agarose in a 0.1 M potassium chloride (KCl) aqueous solution. The specific preparation method is to add 2 grams of agarose to 100 milliliters of 0.1 M KCl solution, heat in a microwave oven until completely dissolved, cool to 40°C, and then cover the PEDOT:PSS film 213 channel layer by drop casting, followed by gelation at room temperature to form a gel-like liquid electrolyte layer.

[0082] The test results show that the sensor provided by the present application has a slow decrease in transconductance with an increase in the number of bends. The transconductance decreases significantly after bending 2750 times, while the transconductance of the conventional liquid electrolyte sensor decreases significantly after bending 1500 times.

[0083] The following table shows the electrochemical performance test results of the sensor provided by the present application after the bending experiment. It can be seen that the sensor provided by the present application is significantly superior to the conventional liquid electrolyte sensor in terms of mechanical flexibility and stability, and has a small increase in ion sensitivity.

[0084] Performance parameters Ionic gel electrolyte Liquid electrolyte Na+ sensitivity (mV / dacade) 58.1 56.8 Na+ response time (s) ~5 ~8 K+ sensitivity (mV / dacade) 55.2 53.9 K+ response time (s) ~7 ~11 Ca2+ sensitivity (mV / dacade) 28.5 27.1 Ca2+ response time (s) ~10 ~15 Long-term stability (24h signal drift, mV) <5 <15 (8h after >5) Risk of leakage None High Transconductance after 1000 bends (compared to the start of the experiment, mS) -2 -6

[0085] Embodiment Two

[0086] Based on the sensor provided in Embodiment One, the present application further provides a detection system, referring to Figure 6 , comprising the above sensor, a temperature sensor 4, an ADC module, a power supply, a data processor, and a host computer and / or a display.

[0087] The power supply is used to provide power to the sensor, the temperature sensor 4, the ADC module, the data processor, and the host computer.

[0088] The temperature sensor 4 is used to measure the temperature of the sensor for temperature compensation of the measurement results. The temperature sensor 4 can adopt a thin film thermal resistance wire, which can be specifically referred to Figure 7 , by depositing platinum on the side of the packaging layer 3 away from the electrode layer 2 through photolithography technology, the thin film thermal resistance wire is insulated from the electrode layer 2 to avoid mutual interference, and then the thin film thermal resistance wire is re-packaged. At the same time, two pins 23 are expanded on the original pin 23 for external circuit (power supply, ADC module and data processor) and thin film thermal resistance wire to make electrical connection.

[0089] The ADC module is used to discretely sample the analog signal at a high frequency, convert the analog signal into a digital signal, and deliver the digital signal to the data processor to obtain the real-time drain current I_DNa, I_DK, I_DCa of the three electrode pairs 21, the real-time gate voltage V_G, and the real-time temperature of the sensor.

[0090] The data processor is used to calculate the real-time concentration of the three ions according to the real-time drain current I_DNa, I_DK, I_DCa and the real-time gate voltage V_G, and correct the calculation results according to the real-time temperature of the sensor. The corrected real-time concentration is delivered to the host computer and / or the display.

[0091] Specifically, the calculation formula derivation process of the ion concentration (Na⁺, K⁺, Ca²⁺) in the sweat to be measured is as follows:

[0092] First, the ions (Cl - ) in the sweat will react with Ag / AgCl on the gate to produce a membrane potential, thereby changing the gate voltage:

[0093]

[0094] In the formula, is the effective gate voltage, i.e. the gate voltage after the reaction of chloride ions with the gate, is the voltage applied by the power supply to the gate, is the Cl- The generated membrane voltage, which obeys the Nernst equation:

[0095]

[0096] where, is the gas constant, is the thermodynamic absolute temperature of the device (in Kelvin), is the number of charges carried by the target ion, is the Faraday constant, is the reference ion activity, which is a fixed value, usually measured by standard experiments, is the target ion activity in sweat, reflecting the "effective concentration" of ions in the solution, which can be expressed as:

[0097]

[0098] where, is the activity coefficient, usually taken as .

[0099] Effective voltage of the gate Drive the ion-doped PEDOT:PSS thin film channel in solid-state electrolyte, at this time the drain current obeys the saturation region equation:

[0100]

[0101] where W and L are the width and length of the channel, respectively, is the carrier mobility, related to the materials of the channel and solid-state electrolyte, obtained by experimental fitting combined with physical models, is the unit volume capacitance, calculated by electrochemical impedance spectroscopy combined with film volume, is the threshold voltage of the gate, that is, the voltage at which the gate turns on the drain and source, determined by the transfer curve through experiments, is the drain voltage applied by the power supply to the drain, with a fixed value.

[0102] Through the above formula, the relationship between the drain current and the ion activity of the target ion can be obtained, so that the ion concentration can be determined according to the drain current:

[0103]

[0104] From the above formula, it can be seen that the relationship between the drain current and the ion activity is logarithmic linear:

[0105]

[0106] where, , represents the intercept of the fitting straight line, , represents the slope of the fitted straight line.

[0107] According to the above formula, the relationship between the drain current and the ion concentration can be obtained by using a standard concentration solution.

[0108] According to the above formula, it can also be seen that the temperature of the sensor has an important influence on the drain current, and therefore the present application integrates a thin film thermal resistance wire on the sensor to measure the temperature of the sensor.

[0109] In practical applications, since the multiple electrode pairs in the sensor are distributed in the horizontal plane, when in contact with the skin, different electrode pairs may have different contact states with the skin, thereby causing the amount of sweat contacted by the electrode pairs to be inconsistent, and at the same time, the secretion rate of the sweat will also have an influence on the detection result of the sensor. The response time of the sensor to ions is between 5-10s, and within this time range, the changes in the amount of sweat and the secretion rate will cause fluctuations in the detected ion concentration data, affecting the accuracy and stability of the detection result.

[0110] Therefore, based on the above analysis, the detection system of the present application also includes a humidity sensor, specifically, a humidity-sensitive resistor can be used, which has a small size and can be integrated with the sensor of the present application. Similar to the thin film thermal resistance wire, thereby improving the integration of the sensor and reducing the volume of the sensor.

[0111] When analyzing the ion concentration in sweat, the measurement results are corrected by temperature and humidity. Since the three electrode pairs can independently determine the concentrations of Na⁺, K⁺ and Ca²⁺ ions, the determination methods for the concentrations of the three ions are the same, and refer to Figure 8 , each of which includes the following steps:

[0112] S1, periodically acquire real-time drain current, real-time gate voltage, real-time temperature, real-time humidity, and perform smoothing processing;

[0113] The ion concentration in sweat does not change suddenly, so the changes in drain current and gate voltage should also be smooth. For the collected discrete data, smoothing processing can be performed by the following steps:

[0114] S11, band-pass filter the original discrete data of all parameters to obtain first filtered data of all terms;

[0115] S12, perform quadratic curve fitting on the first filtered data of each parameter respectively, calculate the first filtered data closest to the fitted quadratic curve, and mark this first filtered data as candidate data;

[0116] S13, timestamp alignment is performed on all candidate data of parameters, and the candidate data of parameters which are timestamp aligned are selected to participate in the calculation of the theoretical concentration, so that the sampling time of each candidate data of parameters participating in the calculation is consistent.

[0117] The original discrete data of the above parameters refer to one or all of the real-time drain current, real-time gate voltage, real-time temperature, and real-time humidity. The real-time drain current refers to one or all of I_DNa, I_DK, and I_DCa.

[0118] The original discrete data is smoothed by performing quadratic curve fitting, and the mutation data is filtered out, so as to reduce noise interference, make the data more stable and reliable, and improve the accuracy of subsequent theoretical concentration calculation. In the subsequent calculation process, the theoretical concentration can be calculated according to each candidate data which is screened after smoothing, or the theoretical concentration can be calculated according to multiple candidate data when the multiple candidate data change little, so as to reduce the influence of signal noise on the calculation result.

[0119] After obtaining reliable discrete data, the theoretical concentration is calculated according to the discrete data:

[0120] S2, the motion state is classified according to the real-time temperature and the real-time humidity, and the Nernst model is corrected according to the motion state classification result;

[0121] Specifically, since different motion states result in different sweating rates, and the sweating rate will affect the measurement result of the concentration, the Nernst model is corrected by the following steps:

[0122] S21, when the motion state is the resting state, the sensitivity (i.e. the slope in the above formula) of the Nernst equation is increased by 1.5% for every 1℃ increase in temperature;

[0123] S22, when the motion state is the light exercise state, the calculation result of the Nernst equation is multiplied by a compensation coefficient of 1.1, and the sensitivity is increased by 3% to correct the slight dilution of ion concentration caused by the accelerated sweat flow rate;

[0124] S23, when the motion state is the moderate exercise state, the sensitivity of the Nernst equation is increased by 0.5% per minute to compensate for the cumulative dilution effect of continuous sweating;

[0125] S24, when the motion state is the high-intensity exercise, the theoretical concentration is segmented and compensated: when the theoretical concentration is less than <20 mM, the calculation result of the Nernst equation is multiplied by 1.3 for severe dilution compensation, when the theoretical concentration is 20-40 mM, the calculation result is multiplied by 1.2, and when the theoretical concentration is >40 mM, an abnormal concentration prompt is issued, which may be caused by excessive sweat leading to short circuit between electrodes.

[0126] Specifically, the grading method of the motion state can refer to the following table:

[0127] Temperature Humidity <60% 60-75% 75-85% >85% 28-32 Resting Resting Abnormal Abnormal 32-36 Resting Light exercise Moderate exercise Moderate exercise 36.5-37 Light exercise Moderate exercise Moderate exercise Moderate exercise >37 Abnormal Moderate exercise High intensity exercise High intensity exercise

[0128] S3, substituting the real-time gate voltage and the real-time drain current into the corrected Nernst model to calculate the theoretical concentration;

[0129] The specific calculation process can refer to the above calculation formula.

[0130] S4, constructing a real-time state vector according to the real-time drain current, the real-time gate voltage and the theoretical concentration, and performing multi-dimensional data interpolation on the standard concentration-current voltage data set according to the real-time state vector, correcting the theoretical concentration according to the interpolation result to obtain the corrected concentration.

[0131] In the standard experimental environment, the ion concentration-current voltage relationship calibration experiment is carried out, the temperature of the sweat sensor is controlled in the human body temperature range, that is, between 36℃ and 37℃, the same simulated sweat is added to the three electrode pairs and the gate, the concentration of each ion (Na⁺, K⁺, Ca²⁺, Cl -

[0132] In the experiment, since the sweat can ensure sufficient contact with the gate and each electrode pair, it is not necessary to consider the influence of sweat secretion rate and the contact state of the electrode and the sweat, and it is not necessary to record the humidity of the environment around the sensor.

[0133] V_G (V) I_DNa (mA) I_DK (mA) I_DCa (mA) Cl - Concentration (mM) Na+ concentration (mM) K+ concentration (mM) Ca2+ concentration (mM) -0.40 1.46 1.40 1.27 375 300 50 24 -0.36 1.04 1.25 0.94 240 200 24 16 -0.32 0.83 1.04 0.62 128 100 16 12 -.025 0.68 1.05 0.97 100 80 12 8 -0.20 0.5 0.62 0.54 72 60 8 4 -0.14 0.41 0.77 0.41 46 40 4 2 -0.09 0.31 0.58 0.25 22.8 20 2 0.8 -0.05 0.20 0.41 0.1 11.4 10 1 0.4

[0134] According to the above standard concentration-current voltage data set, the theoretical concentration calculated by the Nernst model can be corrected.

[0135] Specifically, when constructing the real-time state vector, the theoretical concentration should also be included therein in order to perform interpolation operation and improve the accuracy of interpolation. A preferred form of the real-time state vector can be (taking Na⁺ as the target ion):

[0136] S_st=[V_G, I_DNa, C_Nath]

[0137] Wherein, C_Nath is the theoretical concentration of sodium ion, when the theoretical concentration of other ions is corrected, this component should be the theoretical concentration of other ions.

[0138] Based on this real-time state vector, the standard concentration-current voltage data set can be interpolated in multiple dimensions, and the interpolation result can be used to correct the theoretical concentration.​Figure 9 , specifically comprising the following steps:

[0139] S41, constructing a standard concentration-current voltage vector according to each sample in the standard concentration-current voltage dataset, each component of the standard concentration-current voltage vector should have the same meaning as the corresponding component of the real-time state vector;

[0140] S42, selecting one standard concentration-current voltage vector at a time, calculating the Euclidean distance between the standard concentration-current voltage vector and the real-time state vector, if the Euclidean distance is less than the matching threshold, marking the standard concentration-current voltage vector as matched;

[0141] S43, calculating the weighted concentration according to all matched standard concentration-current voltage vectors, correcting the theoretical concentration according to the weighted concentration, and outputting the confidence.

[0142] The weighted concentration can be calculated according to the Euclidean distance between the real-time state vector and the standard concentration-current voltage vector. Assuming that there are n matched standard concentration-current voltage vectors, the reference weight coefficient is 1 / n. The standard concentration-current voltage vectors are sorted according to the Euclidean distance, the weight of the smallest Euclidean distance is 1, and the weight of the largest Euclidean distance is 1 / n. The concentrations in all standard concentration-current voltage vectors are weighted and calculated to obtain the average value, which is the weighted concentration.

[0143] Correcting the theoretical concentration according to the weighted concentration can be directly outputting the weighted concentration as the corrected theoretical concentration, or calculating the absolute difference between the two, when the absolute difference is less than the confidence threshold, it means that the deviation between the two is not large, and the theoretical concentration can not be corrected, when the absolute difference is greater than the confidence threshold, it means that the deviation between the two is large, and the weighted concentration is output as the corrected concentration, and the confidence is output.

[0144] The confidence is determined according to the ratio of the number of matched standard concentration-current voltage vectors to the highest confidence threshold, or according to the number of matched standard concentration-current voltage vectors. The highest confidence threshold can be determined according to experience and the number of samples in the standard concentration-current voltage dataset. When the number of samples in the standard concentration-current voltage dataset is large, the highest confidence threshold is set relatively simply, and a higher threshold can be used to improve the accuracy expectation of the output result. When the number of samples in the standard concentration-current voltage dataset is small, the highest confidence threshold should be reduced to reduce the accuracy expectation of the calculation result.

[0145] In summary, by the above method, the detection system provided by the application judges the motion state of the user by monitoring the temperature and humidity of the sensor, and corrects the Nernst equation according to the motion state, thereby eliminating the influence of the sweat secretion rate on the measurement result. Meanwhile, the experimental data of the standard concentration are used as the basis for judging whether the calculation result is accurate, the theoretical calculation result is corrected through multi-dimensional interpolation, when the measurement result deviates greatly from the sample in the standard concentration-current voltage data set due to poor contact of a certain electrode pair or gate of the sensor, the method can well identify and compensate, and the confidence of the compensation result is given, thereby helping the user or the program designer to judge the accuracy of the compensation result, and the subsequent program can be improved.

[0146] Finally: the above only for the preferred embodiments of the application, and not for limiting the application, any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the application, should be included in the protection scope of the application.

Claims

1. A sweat multi-parameter ion detection system, characterized in that, The application relates to a sweat ion detection sensor, which comprises an ion detection sensor, a temperature sensor, a humidity sensor, an ADC module and a data processor. The ion detection sensor comprises, from bottom to top, a substrate layer, an electrode layer and a packaging layer. The substrate layer is used for supporting the electrode layer and the packaging layer. The electrode layer comprises a gate, at least three electrode pairs arranged in an array around the gate, and a plurality of pins, the electrode pair comprising a drain and a source which are parallel to each other, a PEDOT:PSS film, a solid-state electrolyte and an ion-selective membrane being sequentially arranged on the electrode pair from bottom to top, the plurality of pins being arranged side by side, the drain, the source and the gate being electrically connected to one pin respectively, each electrode pair and the gate forming an OECT sensing unit, and the ion concentration in the sweat to be detected being detected. The solid-state electrolyte is [MTEOA][MeOSO3] ion gel combined with [MTEOA][MeOSO3] and gelatin. The packaging layer is used for packaging the electrode layer, and the packaging layer is provided with through holes at positions corresponding to the gate and the electrode pairs, so that the sweat to be detected can contact the gate and the electrode pairs. The temperature sensor is used for measuring the temperature of the sensor, and the humidity sensor is used for measuring the humidity of the air near the sensor. The ADC module is used for periodically acquiring real-time drain current, real-time gate voltage, real-time temperature and real-time humidity of the three electrode pairs, and transmitting data to the data processor. The data processor is used for calculating the theoretical concentration of the target ion according to the real-time drain current, the real-time gate voltage, the real-time temperature and the real-time humidity, correcting the theoretical concentration according to the standard concentration-current voltage data, and obtaining the corrected concentration. The theoretical concentration is obtained by the following method: Periodically acquiring real-time drain current, real-time gate voltage, real-time temperature and real-time humidity, and performing smoothing processing. Classifying the motion state according to the real-time temperature and the real-time humidity, and correcting the Nernst model according to the motion state classification result. Substituting the real-time gate voltage and the real-time drain current into the corrected Nernst model to calculate the theoretical concentration. The smoothing processing is performed by the following steps:

2. The system of claim 1, wherein, Band-pass filtering the original discrete data of all parameters to obtain first filtering data of all parameters; Respectively performing quadratic curve fitting on the first filtering data of each parameter, calculating the first filtering data closest to the fitted quadratic curve, and marking the first filtering data as candidate data; Aligning the time stamps of the candidate data of all parameters, and selecting the time stamp aligned candidate data to participate in the calculation of the theoretical concentration. The Nernst model is corrected by the following steps:

3. The system of claim 1, wherein, When the motion state is the resting state, the sensitivity of the Nernst equation is improved by 1.5% for every 1 DEG C rise in temperature; When the motion state is the mild exercise state, the calculation result of the Nernst equation is multiplied by a compensation coefficient of 1.1, and the sensitivity is improved by 3%; When the motion state is the moderate exercise state, the sensitivity of the Nernst equation is improved by 0.5% per minute of sampling time; When the motion state is the high-intensity exercise state, the theoretical concentration is segmented and compensated according to the theoretical concentration. ​ When the concentration is less than <20 mM, the calculation result of Nernst equation is multiplied by 1.3; When the concentration is between 20-40 mM, the calculation result is multiplied by 1.2; When the concentration is greater than 40 mM, an abnormal concentration prompt is issued.

4. The system of claim 1, wherein, The theoretical concentration is corrected by the following steps: A standard concentration-current voltage vector is constructed according to each sample in the standard concentration-current voltage data set, and the meaning of each component of the standard concentration-current voltage vector is the same as that of the corresponding component of the real-time state vector; One standard concentration-current voltage vector is selected one by one, the Euclidean distance between the standard concentration-current voltage vector and the real-time state vector is calculated, and if the Euclidean distance is less than the matching threshold, the standard concentration-current voltage vector is marked as matched; The weighted concentration is calculated according to all matched standard concentration-current voltage vectors, the theoretical concentration is corrected according to the weighted concentration, and the confidence is output.

5. A method of sweat multi-parameter ion detection, using the sweat multi-parameter ion detection system according to any one of claims 1-4, characterized in that, The following steps are included: Periodically acquire real-time drain current, real-time gate voltage, real-time temperature, and real-time humidity, and perform smoothing processing; Classify the motion state according to the real-time temperature and the real-time humidity, and correct the Nernst model according to the motion state classification result; The real-time gate voltage and the real-time drain current are substituted into the corrected Nernst model to calculate the theoretical concentration; A real-time state vector is constructed according to the real-time drain current, the real-time gate voltage, and the theoretical concentration, and the standard concentration-current voltage data is corrected by multi-dimensional data interpolation according to the real-time state vector to obtain the corrected concentration.

Images