Sensing system for rapidly detecting halogenated hydrocarbon pollutants and preparation method thereof

Abstract

The invention belongs to the field of pollutant detection, and particularly relates to a sensing system for rapidly detecting halogenated hydrocarbon pollutants and a preparation method of the sensing system. The sensing system comprises a detection assembly, a signal acquisition assembly and a signal processing assembly, the detection assembly is used for being combined with to-be-detected pollutants to form detection signals, and the signal acquisition assembly is used for acquiring the detection signals and sending the detection signals to the signal processing assembly for processing so as to obtain concentration and type information of the to-be-detected pollutants; the detection assembly comprises a substrate, a detection area is arranged on the substrate, and the detection area is provided with quantum sheets and / or quantum dots; the signal acquisition assembly is used for shooting picture information of quantum sheets and / or quantum dots on the detection assembly and sending the picture information to the signal processing assembly; and the signal processing assembly is used for carrying out image processing on the picture information to obtain image data information, and acquiring the concentration of the halogenated hydrocarbon pollutants according to the image data information. The sensing system is simple in structure, short in analysis time and capable of meeting the requirement for high-throughput screening.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of pollutant detection, and particularly relates to a sensing system for rapid detection of halogenated hydrocarbon pollutants and a preparation method thereof, in particular to a portable quantum sheet array sensing system for rapid quantitative detection of halogenated hydrocarbon pollutants and a preparation method thereof. BACKGROUND

[0002] With the oil and chemical industry, oil or its products in the process of mining, transportation, processing, storage or use of accidental or intentional release into the environment, causing serious harm to the ecological environment, human health and economic activities. Among them, halogenated hydrocarbon compounds as the main oil pollutants, is one of the important reasons to destroy the ecological environment and threaten human health. This kind of chlorine / fluorine containing organic pollutants not only has environmental persistence and biological accumulation, the halogen atom in its molecular structure also easy to occur electrophilic reaction with biological macromolecules, so that the endocrine system of human body is interfered, the nervous system is damaged, endangering human health. However, the traditional chemical detection method of halogenated hydrocarbon such as gas chromatography, high performance liquid chromatography, ion chromatography, spectroscopy, etc. Although the accuracy is high, but there are complex operation, low efficiency, dependent on large instrument and other limitations, can not meet the needs of the related emergency disposal field of halogenated hydrocarbon pollutant concentration rapid detection.

[0003] Therefore, we urgently need an efficient, portable halogenated hydrocarbon pollutant concentration rapid detection technology to solve this problem.

[0004] At present, the electronic nose sensor system can realize the rapid identification and semi quantitative analysis of halogenated hydrocarbon by integrating metal oxide semiconductor (MOX), conductive polymer (CP), quartz crystal microbalance (QCM) and other multi-element sensing array, combined with pattern recognition algorithm (such as principal component analysis, artificial neural network). Specifically, when the halogenated hydrocarbon molecules and the MOX sensor surface active material (such as SnO2, WO3) occur redox reaction, the resistance change can be captured; CP sensor can realize detection by the physical adsorption / chemical action of polymer film and halogenated hydrocarbon; QCM sensor can realize detection by the frequency shift caused by mass loading effect.

[0005] In addition, Shahar H et al. provided an innovative solution for halocarbon detection based on a reflective biosensor of halogenated alkane dehalogenase (DhlA), which covalently immobilized DhlA enzyme on polyacrylate microspheres and physically adsorbed a chromium ion carrier (NBC) as a proton indicator. When halocarbon (such as 1,2-dichloroethane) was contacted with immobilized DhlA, the enzymatic hydrolysis reaction released protons, causing the NBC to change from purple to blue, and quantification was achieved through the change in reflected spectrum; compared with traditional GC-ECD (gas chromatography-electron capture detector) and whole-cell biosensor, the advantages were that there was no need for complex sample pretreatment, and the sensitivity was improved by 2 orders of magnitude compared with the microbial method, and verification showed that it had statistical equivalence with the detection results of GC-ECD.

[0006] Compared with existing optical detection methods, the electronic nose sensor system and the enzyme-based reflective biosensor still have several limitations: the main disadvantage of the electronic nose sensor system is that the sensing unit is easily disturbed by environmental temperature and humidity, causing baseline drift, and the metal oxide semiconductor sensor is prone to response attenuation due to aging of the sensitive material in long-term use. In addition, although the cross-sensitivity of the multi-element sensing array helps to improve the recognition ability, it also increases the complexity of algorithm training, and a large number of standard samples are needed for model calibration.

[0007] The enzyme-based reflective biosensor is limited by the inherent properties of biological molecules, and the activity of immobilized DhlA enzyme will decrease with the extension of storage time, and usually only 2-4 weeks of stable detection performance can be maintained, at the same time, the sensitivity of the reaction system to pH and ionic strength also limits its application in complex matrix (such as high-salt wastewater). Although its sensitivity is significantly improved compared with the microbial method, the detection throughput is still restricted by the enzyme reaction kinetics, and a reflectance spectrometer is needed for actual spectrum measurement and analysis, and the single analysis time is relatively long, which is difficult to meet the demand of high-throughput screening. These limitations make the two methods still inferior to optical technology in real-time monitoring and field detection scenarios. SUMMARY

[0008] In order to improve the shortcomings of the prior art, the present application provides a sensing system for rapid detection of halocarbon pollutants and a preparation method thereof. The sensing system of the present application has a simple structure, short analysis time, and can meet the demand of high-throughput screening.

[0009] In a first aspect, the present application provides a sensing system for rapid detection of halocarbon pollutants, comprising a detection component, a signal acquisition component and a signal processing component, wherein the detection component is used to combine with the pollutants to be detected to form a detection signal, the signal acquisition component is used to acquire the detection signal and send it to the signal processing component for processing to obtain the concentration and type information of the pollutants to be detected.

[0010] The detection component comprises a substrate, and a plurality of detection areas are arranged on the substrate, wherein the detection areas are provided with quantum sheets and / or quantum dots capable of being combined with the to-be-detected pollutants, and the fluorescence spectrum of the quantum sheets and / or quantum dots changes after being combined with the to-be-detected pollutants.

[0011] The signal acquisition component is used for shooting picture information of the quantum sheets and / or quantum dots and sending the picture information to the signal processing component.

[0012] The signal processing component is used for image processing the picture information to obtain image data information, and obtaining the halogenated hydrocarbon pollutant concentration according to the image data information.

[0013] According to an embodiment of the present application, the picture information is preferably color picture information.

[0014] According to an embodiment of the present application, the quantum sheets and / or quantum dots are formed by dropping a quantum sheet and / or quantum dot solution on the substrate.

[0015] According to an embodiment of the present application, after the quantum sheet and / or quantum dot solution is dried on the substrate, an array of quantum sheets and / or quantum dots, i.e., a detection area, is formed.

[0016] According to an embodiment of the present application, the quantum sheets are selected from at least one of CdSe quantum sheets, MoS2 quantum sheets or perovskite quantum sheets, for example, CdSe quantum sheets or CsPbBr3 quantum sheets.

[0017] According to an embodiment of the present application, the quantum dots are selected from at least one of CdSe or perovskite quantum sheets, for example, CdSe quantum dots.

[0018] According to an embodiment of the present application, the light-emitting wavelength of the quantum sheets and / or quantum dots is within the wavelength range that can be acquired by the signal acquisition component, for example, within the visible light or near-infrared light region.

[0019] According to an embodiment of the present application, the concentration of the quantum sheet solution is 5-20 mg / ml, for example, 6.25 mg / ml, 13.5 mg / ml or 15 mg / ml.

[0020] According to an embodiment of the present application, the concentration of the quantum dot solution is 4-20 mg / ml, for example, 13.5 mg / ml.

[0021] According to an embodiment of the present application, the halogenated hydrocarbon is selected from C1-C10 halogenated hydrocarbons, preferably C1-C6 halogenated hydrocarbons, wherein the degree of substitution is one or more, and the halogen is Cl, Br or I, for example, 1-1-1-2 tetrachloroethane or 1.2 dichloroethane.

[0022] According to an embodiment of the present application, the sensing system further comprises an ultraviolet light source for providing ultraviolet light excitation quantum sheet luminescence.

[0023] According to an embodiment of the present application, the signal acquisition component is selected from a multispectral imaging device capable of visible light, ultraviolet light or near-infrared light imaging, such as a mobile phone with a visible light multispectral imaging device.

[0024] According to an embodiment of the present application, the processing component comprises a constructed inversion model, which is a secondary nonlinear inversion model based on existing image data information, established by a machine learning algorithm, capable of obtaining the corresponding halogenated hydrocarbon pollutant concentration according to the image data information.

[0025] According to an embodiment of the present application, the processing component comprises a data preprocessing module and a calculation module, the data preprocessing module is used for preprocessing the image data information and sending it to the calculation module, and the calculation module is used for calculating the halogenated hydrocarbon pollutant concentration according to the preprocessed image data information.

[0026] According to an embodiment of the present application, the processing component further comprises an output module for outputting the halogenated hydrocarbon pollutant concentration calculated by the calculation module.

[0027] According to an embodiment of the present application, the data preprocessing module comprises a preprocessing unit, a binarization unit, a multiplication processing unit, an RGB average value calculation unit and a normalization unit for sequentially performing data transmission;

[0028] The preprocessing unit is used for filtering the image data information;

[0029] The binarization unit is an Otsu binarization processing unit, which is used for separating the preprocessed data information from the background to obtain binarized data;

[0030] The multiplication processing unit is used for multiplying the binarized data with a color image to obtain enhanced data;

[0031] The RGB average value calculation unit is used for extracting the RGB feature information of the enhanced data after multiplication processing, calculating the average values of the RGB three channels of the picture respectively, and according to the fluorescence spectrum range of the quantum sheet and / or quantum dot, if the quantum sheet and / or quantum dot fluorescence peak is at about 600 nm, the data of the R channel needs to be processed, if its excited fluorescence range is about 500 nm, for example, the quantum sheet and / or quantum dot fluorescence peak is at about 508 nm, only the G channel data is used, and the average value of the G channel data is taken at this time.

[0032] The normalization unit is configured to normalize the average value of the required processing channel data to an effective range, for example, to the range of [50, 250] by clipping, and the average value of the normalized channel data is x.

[0033] According to an embodiment of the present application, the construction of the inversion model comprises the following steps:

[0034] Based on the fluorescence quenching effect, a quadratic nonlinear regression model is established with the normalized average value of the channel data x as the independent variable and the dropwise addition amount y of the halogenated hydrocarbon as the dependent variable, and the fitted quadratic nonlinear inversion model can be expressed as:

[0035] y = ax 2 + bx + c

[0036] wherein y is the dropwise addition amount of the halogenated hydrocarbon, a, b, and c are the coefficients of the quadratic term, the linear term, and the constant term, respectively.

[0037] According to an embodiment of the present application, the quadratic nonlinear inversion model further comprises an accuracy prediction mechanism, which is realized by dividing all the normalized average values of the channel data into a test set and a training set in a ratio of 3:7, using the training set to fit the quadratic nonlinear inversion model by machine learning, using the test set to test the prediction accuracy of the inversion model, and calculating the root mean square of the predicted dropwise addition amount and the actual dropwise addition amount. 2

[0038] According to an embodiment of the present application, the determination coefficient r 2 of the inversion model is greater than or equal to 0.85, preferably the determination coefficient r 2 of the inversion model is greater than or equal to 0.90, for example, the determination coefficient R 2 of the inversion model is 0.87, 0.91, or 0.987.

[0039] In a second aspect, the present application provides a method for detecting the concentration of halogenated hydrocarbon pollutants using the above-mentioned sensing system, comprising the following steps:

[0040] The halogenated hydrocarbon pollutants to be detected are added dropwise to the surface of the quantum sheet and / or quantum dots, the signal acquisition assembly acquires image information of the quantum sheet and sends it to the signal processing assembly for processing to obtain the concentration of the halogenated hydrocarbon pollutants to be detected.

[0041] According to an embodiment of the present application, the halogenated hydrocarbon pollutants to be detected are liquid pollutants or solid pollutants, and when the halogenated hydrocarbon pollutants to be detected are solid pollutants, the halogenated hydrocarbon pollutants to be detected are dissolved in a solvent to form a solution before being added dropwise to the surface of the quantum sheet and / or quantum dots.

[0042] Advantages ​

[0043] 1) Build a halogenated hydrocarbon pollutant concentration detection sensor system based on quantum sheet and / or quantum dot fluorescence quenching effect and visible light multispectral imaging technology, after quantum sheet and / or quantum dot is mixed with halogenated hydrocarbon, the spectrum of quantum sheet and / or quantum dot will change, and the change of spectrum intensity and peak value is related to the type and concentration of halogenated hydrocarbon, based on this, the constructed sensor system can accurately determine the concentration of halogenated hydrocarbon pollutant.

[0044] 2) The sensor system of the present application can directly use the equipment of light multispectral imaging, such as mobile phone with visible light multispectral imaging device, to shoot the image of quantum sheet and / or quantum dot, which will be input into the data preprocessing module and processed in the preprocessing unit, binary unit, multiplication processing unit, RGB average value calculation unit and normalization unit in turn to obtain the normalized channel data average value. The normalized average value can not only reflect the type of halogenated hydrocarbon pollutant, but also has a quadratic nonlinear relationship with the concentration of halogenated hydrocarbon pollutant, so that the concentration of halogenated hydrocarbon pollutant with high accuracy can be obtained. And through the processing of the picture, the detection result is not affected by the surrounding light, and the accuracy is high.

[0045] 3) The sensor system of the present application has high stability of quantum sheet and / or quantum dot, and can still maintain good detection effect after being stored for more than 6 months. The structure of the sensor system is simple, and it is convenient to carry to the detection site. It has the advantages of easy to obtain and strong portability, and the whole detection process is more environmentally friendly and rapid. Compared with the traditional detection method of large instrument such as "gas chromatography" and "liquid chromatography", the detection system has wider application range, more simple operation and stronger portability, can accurately adapt to various emergency detection scenes, and can meet the urgent needs of emergency detection.

[0046] 4) The inversion model in the present application innovatively combines the fluorescence characteristics of quantum sheet and / or quantum dot with visible light imaging technology, breaks through the dependence of traditional chemical detection methods on large instruments, realizes the convenient and rapid detection of halogenated hydrocarbon pollutant concentration in the environment, and has significant innovation in project design idea and technical route. The present application innovatively applies quantum sheet and / or quantum dot fluorescence characteristics, visible light imaging technology and machine learning algorithm to halogenated hydrocarbon pollutant detection, realizes the deep combination and innovative development of optical, electronic, chemical and computer technologies, and provides a new idea and effective solution for solving the current rapid detection problem of halogenated hydrocarbon pollutants. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 It is the principle diagram of the halogenated hydrocarbon pollutant concentration rapid detection system in the present application;

[0048] Figure 2 It is the technical route diagram in the present application;

[0049] Figure 3 Figure 1 is a plot of the fluorescence intensity of CdSe quantum dots as a function of the amount of 1,2-dichloroethane added;

[0050] Figure 4 (a) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, Figure 4 (b) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, Figure 4 (c) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane.

[0051] Figure 5 (a) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, Figure 5 (b) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, Figure 5 (c) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane.

[0052] Figure 6 (a) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, Figure 6 (b) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, Figure 6 (c) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane.

[0053] Figure 7 (a) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, Figure 7 (b) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, Figure 7 (c) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane.

[0054] Figure 8 Figure 1 is a plot of the fluorescence intensity of CdSe quantum dots as a function of the amount of 1,2-dichloroethane added;

[0055] Figure 9 Figure 1 is a plot of the fluorescence intensity of CdSe quantum dots as a function of the amount of 1,2-dichloroethane added;

[0056] Figure 10 (a) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, Figure 10 (b) is a plot of the model predicted values versus actual values for the CdSe quantum dots formed from a 6.25 mg / ml solution of CdSe quantum dots and 1,2-dichloroethane, DETAILED DESCRIPTION

[0057] The system and method of the present application will be further described in conjunction with specific embodiments. It should be understood that the following embodiments are merely exemplary and explanatory of the present application and should not be interpreted as limiting the scope of the present application. Any technology implemented based on the above description of the present application is encompassed within the scope of the present application.

[0058] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0059] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0060] The CdSe quantum plate used in the following examples was prepared according to the following reference: Yeltik A, Delikanli S, Olutas M, et al. Experimental Determination of the Absorption Cross-Section and Molar Extinction Coefficient of Colloidal CdSe Nanoplatelets [J]. The Journal of Physical Chemistry C, 2015, 119(47). DOI: 10.1021 / acs.jpcc.5b09275.

[0061] Example 1

[0062] The present application provides a sensing system for rapid detection of halogenated hydrocarbon pollutants, comprising a detection assembly, a signal acquisition assembly and a signal processing assembly, wherein the detection assembly is used to combine with the pollutants to be detected to form a detection signal, the signal acquisition assembly is used to acquire the detection signal and send it to the signal processing assembly for processing, and the concentration and type information of the pollutants to be detected are obtained.

[0063] The detection assembly comprises a substrate, and a detection area is arranged on the substrate, the detection area being provided with quantum sheets and / or quantum dots, the quantum sheets and / or quantum dots being formed by dropping quantum sheet solution and / or quantum dot solution.

[0064] The quantum sheets are selected from at least one of CdSe quantum sheets, MoS2 quantum sheets or perovskite quantum sheets, for example, CdSe quantum sheets or CsPbBr3 quantum sheets.

[0065] The quantum dots are selected from at least one of CdSe quantum dots or perovskite quantum dots, for example, CdSe quantum dots.

[0066] The quantum sheets and / or quantum dots can be combined with the to-be-detected pollutants, and the fluorescence intensity of the quantum sheets and / or quantum dots after combination is the detection signal, and the fluorescence spectrum of the quantum sheets and / or quantum dots changes after being combined with the to-be-detected pollutants.

[0067] The signal acquisition assembly is a fluorescence collector for shooting color picture information of the quantum sheets and / or quantum dots, and in the embodiment, a mobile phone with a visible light multi-spectrum imaging device is selected for field use.

[0068] The signal processing assembly is used for image processing of the color picture information to obtain image data information, and according to the image data information, the concentration of the halogenated hydrocarbon pollutants is obtained.

[0069] The signal processing assembly comprises an inversion model, the inversion model being a quadratic nonlinear inversion model established by using a machine learning algorithm based on existing image data information, and the quadratic nonlinear inversion model can obtain the corresponding concentration of the halogenated hydrocarbon pollutants according to the image data information.

[0070] Specifically, the establishment of the quadratic nonlinear inversion model comprises the following steps:

[0071] 1) Pre-experiment part:

[0072] CdSe quantum sheet solutions with concentrations of 6.25 mg / ml, 13.5 mg / ml and 26.5 mg / ml are configured, and appropriate CdSe quantum sheet solution is added dropwise into a cuvette, and the transmission spectrum and the fluorescence emission spectrum of the CdSe quantum sheet solution in the cuvette are obtained by using an ultraviolet-visible spectrophotometer and a fluorescence spectrometer respectively, and the fluorescence intensity peak position of the quantum sheet solution is measured.

[0073] CdSe quantum sheet solutions with a concentration of 13.5 mg / ml are added dropwise to six different areas on an alumina substrate, and the added amounts are 0 μL, 10 μL, 20 μL, 30 μL, 40 μL and 50 μL respectively, and the transmission spectrum and the fluorescence emission spectrum of the CdSe quantum sheet solution with different added amounts on the alumina substrate are obtained by using an ultraviolet-visible spectrophotometer and a fluorescence spectrometer respectively.

[0074] Select 1-1-1-2 tetrachloroethane halogenated hydrocarbon (1-1-1-2 tetrachloroethane is a more common halogenated hydrocarbon, and other halogenated hydrocarbons are expected to have similar characteristics in the case of good response in 1-1-1-2 tetrachloroethane detection) to perform the following experiments: 50 μL of the corresponding concentration of CdSe quantum sheet solution was added to 6 different regions of the test paper, and the concentration of the CdSe quantum sheet solution was 13.5 mg / ml. Take pictures of each test paper under an ultraviolet light source, and take 10 pictures of each test paper. Then add pure 1-1-1-2 tetrachloroethane solution to the second to sixth regions, with the amount of each being 10 μL, 15 μL, 20 μL, 25 μL, and 30 μL, respectively. No halogenated hydrocarbon is added to the first region. Observe the changes in the fluorescence of the quantum sheets after the halogenated hydrocarbon is combined on the corresponding test paper by taking pictures, and select halogenated hydrocarbon species that respond well to the fluorescence quenching effect of CdSe quantum sheets and have similar response rules.

[0075] Table 1 Halogenated hydrocarbons and quantum sheet solutions used in the experiment

[0076] Region 1 Region 2 Region 3 Region 4 Region 5 Region 6 CdSe quantum plate amount 50 μL 50 μL 50 μL 50 μL 50 μL 50 μL 1-1-1-2 tetrachloroethane Not dropped 10 μL 15 μL 20 μL 25 μL 30 μL

[0077] 2) The screening method is as follows:

[0078] Referring to Figure 1 , take pictures of the test paper after adding the halogenated hydrocarbon as the original image. The original image is sequentially pre-processed, Otsu binarization processed (adaptive threshold segmentation method based on maximum inter-class variance), and multiplied. The RGB feature information of the picture after multiplication is extracted, and the average value of the RGB three channels of the picture is calculated.

[0079] The pre-processing includes Gaussian filtering of the image data (to reduce random noise) and extraction of the gray channel. The Otsu binarization processing is used to remove the background information in the pre-processed image data to obtain the target information. The multiplication processing is used to filter the invalid data area in the target information to obtain the valid data picture.

[0080] According to the fluorescence intensity of the quantum sheet at different wavebands measured by the steady-state / transient fluorescence spectrometer, the fluorescence intensity peak is concentrated in the green waveband (as shown in Figure 3 ), and the contribution of the blue waveband and the green waveband to the fluorescence intensity can be ignored. Therefore, after calculating the average value of the RGB three channels, only the G channel data average value is further processed and modeled. Specifically, the G average value is normalized to the effective range of [50, 250] by cropping to further weaken the influence of noise signals, highlight the relative changes of target signals, and improve the calculation accuracy.

[0081] Based on the fluorescence quenching effect, with the normalized channel data average value x as the independent variable, the CdSe quantum sheet drop amount or the halogenated hydrocarbon drop amount as the dependent variable, quadratic non-linear regression modeling is carried out, and a quadratic non-linear inversion model is fitted to be:

[0082] Y = ax 2 + bx + c

[0083] Wherein, y is the drop amount, a, b and c are the quadratic term coefficient, the first order term coefficient and the constant term respectively.

[0084] All the normalized channel data average values are taken as experimental data, and are divided into a test set and a training set according to a ratio of 3:7; the training set is used to fit a quadratic non-linear inversion model by machine learning (a polynomial regression algorithm), and the test set is used to test the prediction accuracy of the inversion model, by calculating the determination coefficient r 2 of the inversion model and the root mean square error of the predicted drop amount, so as to determine the accuracy of the model.

[0085] Referring to Figure 4 , it is found that the determination coefficient r 2 of the quadratic non-linear regression inversion model obtained by the application is 0.987, and the model prediction value is close to the actual value, so that the quadratic non-linear regression inversion model obtained by the application has high prediction accuracy.

[0086] Example 2

[0087] By using the above method, CdSe quantum sheet solutions with concentrations of 6.25 mg / ml, 13.5 mg / ml and 26.5 mg / ml are configured,

[0088] 1.2 dichloroethane is selected to perform the following experiment: 50 μL of CdSe quantum sheet solution with a corresponding concentration is added to each of 6 different regions on a test paper, and 10 pictures are taken for each test paper under an ultraviolet light source; then 10 μL, 15 μL, 20 μL, 25 μL and 30 μL of pure 1.2 dichloroethane solution are added to the second to sixth regions, respectively, and no halogenated hydrocarbon is added to the first region; the fluorescence change of the quantum sheet after being combined with the halogenated hydrocarbon on the corresponding test paper is observed by taking pictures, and the halogenated hydrocarbon species with excellent response to the fluorescence quenching effect of the CdSe quantum sheet and similar response rules are screened out.

[0089] Referring to Figure 5 , it is found that the determination coefficient r 2 of the quadratic non-linear regression inversion model obtained by the application is 0.987, and the model prediction value is close to the actual value, so that the quadratic non-linear regression inversion model obtained by the application has high prediction accuracy.

[0090] Referring to Figure 6 Fig. 3 shows a quadratic non-linear regression inversion model obtained by processing the corresponding images after the quantum plate array formed by the CdSe quantum plate solution with a concentration of 13.5 mg / ml reacts with 1.2 dichloroethane, the model determination coefficient r 2 is 0.91, and the model predicted value is close to the actual value.

[0091] Referring to Figure 7 Fig. 4 shows a quadratic non-linear regression inversion model obtained by processing the corresponding images after the quantum plate array formed by the CdSe quantum plate solution with a concentration of 26.5 mg / ml reacts with 1.2 dichloroethane, the model determination coefficient r 2 is 0.48, at this time, the quantum plate concentration is too high, and the predicted value is far from the actual value.

[0092] Example 3

[0093] The CdSe quantum dot solution with a concentration of 13.5 mg / ml is configured.

[0094] 1.2 dichloropropane is selected for the following experiment: 50 μL of the CdSe quantum dot solution with a concentration of 13.5 mg / ml is added to 6 different regions on a test paper, 10 pictures are taken for each test paper under an ultraviolet light source; then 2 μL, 4 μL, 6 μL, 8 μL, and 10 μL of pure 1.2 dichloropropane solution is added to the second to sixth regions, respectively, and no halogenated hydrocarbon is added to the first region, the fluorescence change of the quantum dots after combining with the halogenated hydrocarbon on the corresponding test paper is observed by taking pictures, and the halogenated hydrocarbon species with excellent response to the fluorescence quenching effect of the CdSe quantum dots and similar response rules are screened out.

[0095] Referring to Figure 8 Fig. 5 shows the fluorescence spectrum change of the quantum dot array formed by the CdSe quantum dot solution with a concentration of 13.5 mg / ml under different 1.2 dichloropropane solution drop amounts, according to the fluorescence intensity of the quantum dots at different wave bands measured by a steady-state / transient fluorescence spectrometer, the fluorescence intensity peak is concentrated in the red wave band (such as Figure 8 shown).

[0096] Referring to Figure 9 Fig. 6 shows that the CdSe quantum dot solution with a concentration of 13.5 mg / ml is fixedly added to the test paper in an amount of 50 uL, and pure 1.2 dichloropropane is added in an amount of 0 uL, 2 μL, 4 μL, 6 μL, 8 μL, and 10 μL, respectively, and the corresponding fluorescence spectrum presents fluorescence quenching phenomenon with the increase of the drop amount.

[0097] Referring to Figure 10The figure shows the quadratic non-linear regression inversion model obtained by processing the corresponding images after the quantum dot array formed by the CdSe quantum dot solution with a concentration of 13.5 mg / ml reacts with 1.2 dichloropropane, at this time the quadratic term coefficient is 0, the R channel data is selected for processing, the regression inversion model is: y = -0.43 * R + 17.54, the model determination coefficient r 2 is 0.92, and the model predicted value is close to the actual value.

[0098] When CdSe quantum dots are used for detection, when the concentration of CdSe quantum dots is greater than 20 mg / ml, the model determination coefficient r 2 is less than 0.50; when the concentration of CdSe quantum dots is less than 4 mg / ml, the model determination coefficient r 2 is less than 0.70.

[0099] Existing comparison methods

[0100] Wang et al. combined spectral analysis model, handheld FTIR instrument, soil erosion model (RUSLE) and hydrological model (Hydrus), based on the differences of different carbon chains in the 4000-400 cm -1 band infrared spectrum, studied the migration path of soil petroleum hydrocarbon in contaminated sites, and improved the efficiency of contaminated site assessment. However, the application of this method in practice is difficult to exclude the interference of other components in the soil (Spectrochim Acta A Mol Biomol Spectrosc. 2019, 207: 183-188.); Chakraborty et al. first developed a system integrating X-ray fluorescence and near-infrared spectroscopy, which realized rapid screening of large-area soil petroleum hydrocarbon pollution through two-dimensional quantitative measurement: on the one hand, VisNIR spectrum (Visible and Near-Infrared Spectroscopy, covering visible light (Vis, 400-700 nm) and near-infrared (NIR, 700-2500 nm) band reflection or transmission spectrum technology) and PSR (partial least squares regression) were used to predict the initial value of petroleum hydrocarbon, on the other hand, PXRF (portable X-ray fluorescence spectrometer) element data and RF (Random Forest, random forest) were used to predict the residual of petroleum hydrocarbon, and the initial prediction result was corrected. Although this method improves the efficiency of field detection, the model determination coefficient (R 2The results of the study by Correa et al. (Sci Total Environ. 2019, 649:1224-1236) showed that the accuracy of the method was 0.64, which was still not ideal, and lacked the ability to analyze spatial distribution (Sci Total Environ. 2015, 514:399-408.). Correa et al. developed a fusion of visible, near-infrared and shortwave infrared band (wavelength range 350-2500 nm) reflectance spectroscopy and imaging spectroscopy technology, which uses a portable spectrometer and an aerial imaging system to obtain spectral images of contaminated areas, accurately identifies contaminated soil based on the spectral absorption characteristics of the material, and then realizes systematic and seamless monitoring of land oil facilities and pipelines. Although this technology has achieved remarkable results in real-time large-area pollution identification, it still has limitations in quantitative detection (Sci Total Environ. 2019, 649:1224-1236).

[0101] The above describes the specific embodiments of the present application by way of examples. However, the protection scope of the present application is not limited to the above exemplary embodiments. Any modifications, equivalent replacements, improvements, etc. made by those skilled in the art within the spirit and principles of the present application shall be included in the protection scope of the claims of the present application.

Claims

1. A sensing system for the rapid detection of halocarbon contaminants, characterized in that, The sensing system comprises a detection component, a signal collection component and a signal processing component, wherein the detection component is used to combine with the to-be-detected pollutant to form a detection signal, the signal collection component is used to collect the detection signal and send it to the signal processing component for processing to obtain the concentration and type information of the to-be-detected pollutant; The detection component comprises a substrate, and a detection area is arranged on the substrate, wherein the detection area is provided with quantum sheets and / or quantum dots, the quantum sheets and / or quantum dots can combine with the to-be-detected pollutant, and the fluorescence spectrum of the quantum sheets and / or quantum dots changes after combining with the to-be-detected pollutant; The signal collection component is used to shoot picture information of the quantum sheets and / or quantum dots on the detection component and send it to the signal processing component; The signal processing component is used to perform image processing on the picture information to obtain image data information, and obtain the concentration of the halogenated hydrocarbon pollutant according to the image data information.

2. The sensor system for rapid detection of halocarbon contaminants of claim 1, wherein, The quantum sheets and / or quantum dots are formed by dropping quantum sheet solution and / or quantum dot solution on the substrate.

3. The sensor system for rapid detection of halocarbon contaminants of claim 2, wherein, The quantum sheets are selected from at least one of CdSe quantum sheets, MoS2 quantum sheets or perovskite quantum sheets, and the concentration of the quantum sheet solution is 5-20 mg / ml. Preferably, the quantum dots are selected from CdSe quantum dots or perovskite quantum dots, and the concentration of the quantum dot solution is 4-20 mg / ml.

4. The sensor system for rapid detection of halocarbon contaminants of claim 1, wherein, The sensing system further comprises an ultraviolet light source, which is used to provide ultraviolet light to excite the quantum sheets and / or quantum dots to emit light; The signal collection component is selected from a device capable of multispectral imaging of visible light, ultraviolet light or near-infrared light.

5. The sensor system for rapid detection of halocarbon contaminants according to any one of claims 1-4, wherein, The processing component comprises a constructed inversion model, which is a quadratic nonlinear inversion model established based on existing image data information by using a machine learning algorithm, and the quadratic nonlinear inversion model can obtain the corresponding concentration of the halogenated hydrocarbon pollutant according to the image data information.

6. The sensor system for rapid detection of halocarbon contaminants according to any one of claims 1-4, wherein, The processing component comprises a data preprocessing module and a calculation module, wherein the data preprocessing module is used to preprocess the image data information and send it to the calculation module, and the calculation module is used to calculate the concentration of the halogenated hydrocarbon pollutant according to the preprocessed image data information.

7. The sensor system for rapid detection of halocarbon contaminants of claim 6, wherein, The processing component further comprises an output module, which is used to output the concentration of the halogenated hydrocarbon pollutant calculated by the calculation module. Preferably, the data preprocessing module comprises a preprocessing unit, a binarization unit, a multiplication processing unit, an RGB average value calculation unit and a normalization unit which sequentially perform data transmission; The preprocessing unit is used to filter the image data information; The binarization unit is an Otsu binarization processing unit, which is used to separate the preprocessed data information from the background to obtain binarized data; The multiplication processing unit is used to multiply the binarized data with a color image to obtain enhanced data; The RGB average value calculation unit is used to extract the RGB feature information of the enhanced data subjected to the multiplication processing, calculate the average values of the RGB three channels of the picture respectively, and take the average value of the channel corresponding to the fluorescence spectrum peak value of the quantum dot / sheet; The normalization unit is used to normalize the average value of the corresponding channel data to an effective range by cropping.

8. The sensor system for rapid detection of halocarbon contaminants of claim 5, wherein, The constructing the inversion model comprises the following steps: Based on the fluorescence quenching effect, the normalized channel data average value is used as the independent variable, and the CdSe quantum dot / drop addition amount or halogenated hydrocarbon drop addition amount is used as the dependent variable, and a quadratic nonlinear regression modeling is performed, and a quadratic nonlinear inversion model is fitted to be: y = ax 2 + bx + c Wherein, y is the drop addition amount of CdSe quantum sheet or halogenated hydrocarbon, a, b, c are respectively the quadratic term coefficient, the first order term coefficient and the constant term. Preferably, the quadratic non-linear inversion model further comprises an accuracy prediction mechanism, which is realized by taking all the normalized channel data mean values as experimental data, dividing them into a test set and a training set in a ratio of 3:7, using the training set to fit the quadratic non-linear inversion model through machine learning, using the test set to test the prediction accuracy of the inversion model, and calculating the coefficient of determination r 2 and the root mean square of the predicted drop amounts.

9. The sensor system for rapid detection of halogenated hydrocarbon contaminants of claim 6, wherein, the determination coefficient r of the inversion model 2 greater than or equal to 0.85, preferably the determination coefficient r of the inversion model 2 greater than or equal to 0.

90.

10. A method for detecting the concentration of a halocarbon contaminant using the sensing system of any one of claims 1-9, wherein, The method comprises the following steps: The halogenated hydrocarbon pollutant to be measured is added to the surface of the quantum sheet and / or quantum dot, the signal acquisition component acquires the image information of the quantum sheet and / or quantum dot and sends it to the signal processing component for processing, and the concentration of the halogenated hydrocarbon pollutant to be measured is obtained.

Images