Physical virtual mixed volatile organic compound sensing array based on cascade structure

Abstract

The invention discloses a cascade structure-based physical virtual hybrid volatile organic compound sensing array, which is formed by cascading a plurality of quartz crystal microbalance (QCM) devices, and realizes the acquisition of output signals of the plurality of QCM devices in the same data measurement process. And the four QCM devices are respectively coated with different metal organic framework (MOF) films. According to the invention, the mixed physical virtual sensor array is constructed, the advantages of the sensor array and the virtual sensor array are combined, parameters such as resistance, reactance and impedance angle can be calculated through the Smith diagram, characteristic fingerprints can be formed for each gas, and the method can be used for selective and quantitative detection of various VOCs under different concentrations.

Description

Technical Field

[0001] The invention relates to a gas sensing device, in particular to a physical virtual hybrid volatile organic compound sensing array based on a cascade structure. Background Art

[0002] The hazards of gas pollution include harm to human body, harm to climate, harm to plants, harm to industry and agriculture, etc. In order to achieve the goal of gas pollution control, the basis is to detect the existence and concentration of various gases in the environment, which has promoted the continuous development of gas detection technology.

[0003] According to the different signal response mechanisms of sensitive elements, gas sensors can be divided into: electrical gas sensors, optical gas sensors, thermal gas sensors, and mass gas sensors. Electrical gas sensors have a simple structure and mature manufacturing process, but poor stability and difficulty in reading data; optical gas sensors have high detection accuracy, high sensitivity, wide detection range, high signal-to-noise ratio, good real-time performance, but poor anti-interference ability and are susceptible to optical pollution. Thermal gas sensors have high stability and are less affected by temperature and humidity, but can only measure combustible gases.

[0004] The mass-type gas sensor represented by QCM is based on the mass effect of the sensitive element and is made by using the relationship between the oscillator parameters (frequency, amplitude, wave velocity, etc.) of a specific piezoelectric crystal and the change of its surface mass. Usually, a thin film material sensitive to a specific gas is coated on the surface of the piezoelectric crystal to form a sensitive element, and the concentration information of the gas is detected by detecting the change in the oscillator parameters caused by the change in the mass of the adsorbed gas.

[0005] Driven by gas detection technology, the demand for gas classification and mixed gas detection is also gradually increasing. Therefore, physical sensor arrays (MSA) and virtual sensor arrays (VSA) have emerged. MSA consists of multiple gas sensors coated with different sensitive materials, each of which can produce a different single output response to a specific gas. The performance of MSA depends on the number of sensor units and the chemical diversity of sensitive materials. The increase in sensor units makes the electronic nose system more complicated. In addition, the existing MSA requires each sensor unit to be measured sequentially, which increases the difficulty and workload of data acquisition. In order to circumvent this problem, VSA was proposed to achieve the classification and identification of different VOC gases by using only a single physical sensor, signal collection and data processing methods. Although the existing VSA method effectively solves the problem of the complexity of the electronic nose system, since a single physical sensor can only be coated with one sensitive material, its performance will be limited when facing more complex VOC classification problems. Summary of the invention

[0006] In order to solve the problems in the prior art, the present invention proposes a physical virtual hybrid volatile organic compound sensor array based on a cascade structure. The present invention simplifies the measurement process and complex circuits, and meets the requirements in gas classification and mixed gas detection.

[0007] The technical solution adopted by the present invention is as follows:

[0008] A physical-virtual hybrid volatile organic compound sensor array based on a cascade structure includes a plurality of QCM devices connected in parallel. The QCM devices include a quartz crystal plate and electrodes arranged on both sides of the quartz crystal plate. Different MOF material films are coated on the surfaces of the plurality of QCM devices.

[0009] As a preferred solution of the present invention, the resonance frequencies of the multiple QCM devices are all between 9.97 MHz and 10.01 MHz, and the difference between the echo dissipation peak values ​​corresponding to the resonance frequencies of every two QCM devices is greater than 3 dB.

[0010] As a preferred solution of the present invention, both ends of the plurality of QCM devices connected in parallel are also connected to a measuring instrument for reading the Smith chart of the QCM devices; the measuring instrument is preferably a network analyzer.

[0011] The present invention also provides a detection method, which is implemented based on the above-mentioned physical virtual hybrid volatile organic compound sensor array based on the cascade structure, and the method includes the following steps:

[0012] A plurality of volatile organic compounds are prepared, and each volatile organic compound gas is detected by using a sensor array, and the gas concentration is gradually increased, and the Smith chart read by the network analyzer at each concentration is recorded; based on the Smith chart, multiple parameters of each QCM device under different gas environments are calculated, and the parameters include echo dissipation, resistance, reactance, impedance angle, impedance value, and resonance frequency; a plurality of parameters corresponding to the current volatile organic compound are combined as the gas characteristic fingerprint of the volatile organic compound; the gas characteristic fingerprint of each volatile organic compound constitutes a gas characteristic fingerprint library;

[0013] The gas to be tested is detected by using a sensor array, and the gas characteristic fingerprint of the gas to be tested is calculated based on the Smith chart read by the network analyzer;

[0014] The gas characteristic fingerprint of the gas to be tested is compared with the gas characteristic fingerprint library to obtain the compound type and gas concentration of the gas to be tested.

[0015] Compared with the prior art, the present invention has the following beneficial effects:

[0016] The present invention realizes the acquisition of output signals of four QCM devices in the same data measurement process through cascade structure design, breaking through the complex data acquisition process required for the construction of sensor arrays and virtual sensor arrays by existing sensors; and through the hybrid physical virtual sensor array technology, high-dimensional feature vector construction is realized, combining the chemical diversity advantages of physical sensor arrays (MSA) and the advantages of virtual sensor arrays (VSA) that focus on the interaction between gas molecules and adsorption materials, which can be used for gas selective and quantitative detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a circuit structure diagram of the sensor array of the present invention;

[0018] Figure 2 It is a schematic diagram of the experimental device layout of the present invention;

[0019] Figure 3 is a SEM characterization image of the adsorbed material on the device of the present invention;

[0020] Figure 4 is a graph of the Smith chart used in the present invention and the calculated multi-dimensional parameters;

[0021] Figure 5 This is a test diagram of the response performance of the device of the present invention to volatile organic compounds;

[0022] Figure 6 This is a comparison chart of the gas classification performance of MSA, VSA, and M-VSA. DETAILED DESCRIPTION

[0023] The present invention proposes a physical-virtual hybrid volatile organic compound sensor array based on a cascade structure, which will be further described below in conjunction with the accompanying drawings and embodiments.

[0024] like Figure 1A physical virtual hybrid volatile organic compound sensor array based on a cascade structure is shown, and the sensor array includes four cascaded quartz crystal microbalance (Quartz Crystal Microbalance, QCM) devices. The surfaces of the four QCM devices are coated with a metal organic framework (Metal Organic Frameworks-Polydimethylsiloxane, MOF) material film. The QCM device is mainly a circular quartz crystal plate with a diameter of 14mm, and electrodes are provided on both sides of the quartz crystal plate. The MOF material film coating position is specifically a circle with a diameter of 1cm on the geometric center of the single-side surface of the QCM device with electrodes. The four QCM devices are connected in parallel through electrodes, and the two ends of the parallel QCM devices can be connected to measuring instruments such as network analyzers. The resonant frequencies of the QCM devices are all between 9.97MHz-10.01MHz, but they are different from each other. The difference in echo dissipation corresponding to the resonant frequency between each two QCM devices is greater than 3dB. The MOF material films coated on the surfaces of the four QCM devices are MIL-101(Cr), UIO-66, ZIF-8 and PCN-222(H) respectively.

[0025] Figure 3 The SEM characterization images of the MOF material films coated on four QCM devices are shown. The MOF material films present crystalline and tubular morphologies at the microscopic level, which are conducive to the physical adsorption of volatile organic compounds (such as methanol, ethanol, isopropanol, etc.).

[0026] The method of coating the MOF material film on the QCM device is as follows:

[0027] 1) Plasma cleaning the surface of the QCM device with electrodes for 10 min to make the surface hydrophilic;

[0028] 2) Prepare a thermal release tape with an aperture of 10 mm as a mask and stick it on the surface of the QCM device;

[0029] 3) Prepare a MOF suspension with a concentration of 0.2% w / v and stir evenly;

[0030] 4) Apply a total of 120ul of MOF suspension on the surface of the QCM device 6 times every 7 minutes to ensure that the ethanol in the previous MOF suspension has evaporated completely to form a MOF film;

[0031] 5) Remove the thermal release tape.

[0032] In order to more clearly express the detection process of the above-mentioned volatile organic compound sensor, this embodiment also provides a detection method of the volatile organic compound sensor, which is specifically:

[0033] In order to construct a gas characteristic fingerprint library of various volatile organic compounds (VOCs), it is necessary to observe the sensor array under the influence of various volatile organic compounds through measuring instruments, read the Smith chart of the QCM device and convert it into the gas characteristic fingerprint of the corresponding volatile organic compounds. In this embodiment, the specific steps are as follows:

[0034] A network analyzer is used as a measuring instrument, and the input and output ports of the network analyzer are connected to the electrodes at both ends of the QCM device, such as Figure 2 As shown, for a variety of volatile organic compounds, the sensor array experimental space is filled with volatile organic compound gas for gas detection, and the gas concentration is gradually increased from 10ppm to 100ppm at intervals of 10ppm, and then from 100ppm to 1000ppm at intervals of 100ppm; at each concentration, the Smith charts of the four QCM devices read by the network analyzer are recorded, as shown in FIG. Figure 4 (a) Based on the Smith chart, the parameters of the four QCM devices in different gas environments, such as echo dissipation S, resistance R, reactance X, impedance angle A, impedance value Z, and resonance frequency f corresponding to the echo dissipation peak, are calculated, as shown in Figure 2. Figure 4 In part (b), these parameters will change due to the adsorption of gas molecules; these parameters are combined as the gas characteristic fingerprints of the corresponding volatile organic compounds; gas detection and gas characteristic fingerprints are constructed for a variety of volatile organic compounds to obtain a gas characteristic fingerprint library.

[0035] Using a volatile organic compound sensor array to detect the gas to be detected, and obtaining a gas characteristic fingerprint of the gas to be detected as an output signal of the volatile organic compound sensor;

[0036] The output signal is compared with the characteristic fingerprint library of volatile organic compounds to obtain the compound type of the gas to be tested and the detection results of the gas concentration.

[0037] like Figure 5 Part (a) shows the dynamic response curve of the first QCM device (i.e., MIL101-coated QCM) in the sensing array in the concentration range of 10-40ppm (10ppm interval). It can be seen that the resonant frequency of the sensor decreases with the increase of VOC concentration. The higher the concentration, the lower the frequency. Figure 5 Part (b) shows the three-cycle repeatability curve of the first QCM device in a low-concentration acetone atmosphere of 10 ppm, indicating that the repeatability of the sensor is reliable. Figure 5 Part (c) of Figure 1 analyzes the response / recovery time of the first QCM device, and the response time is 24s and the recovery time is 111s. The stable signal of the sensor at each concentration is a function of various concentrations of volatile organic compound vapors, such as Figure 5 In addition, in order to study the long-term stability in practical applications, the sensor response to various volatile organic compounds of the same concentration (10 ppm) was tested within 14 days, as shown in part (d). Figure 5 As shown in part (d), at the same concentration, the response of the sensor within 14 days fluctuated slightly within a very small coefficient of variation (defined as the ratio of the standard deviation to the mean), and the coefficient of variation was 1.76%, indicating that the sensor has good long-term stability.

[0038] Figure 6 The graphs show the linear discriminant analysis of 8 VOC gases based on the physical sensor array (MSA), virtual sensor array (VSA) and the hybrid physical virtual sensor array (M-VSA) of the present invention and the confusion matrix of SVM analysis under leave-one-out validation. Figure 6 Parts (a), (b), and (c) are linear discriminant analysis diagrams of MSA, VSA, and M-VSA for 8 VOC gases, respectively. Parts (e), (f), and (g) are confusion matrix diagrams of SVM analysis of MSA, VSA, and M-VSA for 8 VOC gases under leave-one-out validation. The 8 VOC gases tested are acetone, methanol, isopropanol, methanol, acetonitrile, tetrahydrofuran, toluene, and chloroform (AC, MeOH, IPA, EtOH, ACN, THF, TOL, and CF). In this comparative test, the MSA is composed of four independent QCM devices, and the four QCM devices and their coating materials are the same as the QCM device used in the present invention, and the output signal is the resonant frequency of each QCM device; the VSA detects gas through a single QCM device, and the QCM device and its coating material are the same as the first QCM device used in the present invention, and the output signal is the return loss, resistance, reactance, impedance angle, impedance value, and resonant frequency calculated based on the Smith chart of the single QCM device.

[0039] It can be seen that the classification accuracy of MSA for alcohols is only 78.95%, and the classification accuracy for substances other than alcohols is 85.25%. The classification accuracy of VSA for alcohols is 89.47%, and the classification accuracy for substances other than alcohols is 91.58%. This shows that MSA can have a better classification effect on different types of VOCs due to its rich chemical diversity, while VSA can have a better classification effect on the same type of substances due to its use of the viscoelasticity and other properties of the sensitive film, and pays more attention to the interaction between the film and the VOC molecules. However, due to the limitation of chemical diversity, the classification effect of multiple types of substances is unsatisfactory. M-VSA can combine the advantages of both and show excellent recognition ability for both the same type of substances and multiple types of substances.

[0040] In summary, the physical-virtual hybrid volatile organic compound sensor array based on a cascade structure designed in the present invention realizes the acquisition of the output signals of four QCM devices in the same data measurement process through the cascade structure design, thereby reducing the workload and complexity of data acquisition; the Smith chart analysis combining four QCM devices and a single QCM device realizes the hybrid physical-virtual sensor array technology constructed based on high-dimensional data, which can be applied to gas classification and quantitative detection, and provides important progress in realizing artificial olfaction.

[0041] The above-mentioned embodiments only express several implementation methods of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. For ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A physical-virtual hybrid volatile organic compound sensor array based on a cascade structure, characterized in that: The invention comprises a plurality of QCM devices connected in parallel, wherein the QCM devices comprise a quartz crystal plate and electrodes arranged on both sides of the quartz crystal plate, and the surfaces of the plurality of QCM devices are coated with different MOF material films.

2. The physical-virtual hybrid volatile organic compound sensor array based on a cascade structure according to claim 1, characterized in that: The MOF materials used in the MOF material film include MIL-101 (Cr), UIO-66, ZIF-8 and PCN-222 (H).

3. The physical-virtual hybrid volatile organic compound sensor array based on cascade structure according to claim 1, characterized in that: The MOF material film coating position is specifically a circle with a geometric center diameter of 1 cm on a single side surface of the QCM device.

4. The physical-virtual hybrid volatile organic compound sensor array based on cascade structure according to claim 1, characterized in that: The resonance frequencies of the multiple QCM devices are all between 9.97 MHz and 10.01 MHz, and the difference between the echo dissipation peak values ​​corresponding to the resonance frequencies of every two QCM devices is greater than 3 dB.

5. The physical-virtual hybrid volatile organic compound sensor array based on cascade structure according to claim 1, characterized in that: Both ends of the multiple QCM devices connected in parallel are also connected to a measuring instrument for reading the Smith chart of the QCM devices.

6. The physical-virtual hybrid volatile organic compound sensor array based on cascade structure according to claim 5, characterized in that: The measuring instrument is a network analyzer.

7. A detection method of a physical-virtual hybrid volatile organic compound sensor array based on a cascade structure as claimed in claim 6, characterized in that: The following steps are involved: A plurality of volatile organic compounds are prepared, and each volatile organic compound gas is detected by using a sensor array, and the gas concentration is gradually increased, and the Smith chart read by the network analyzer at each concentration is recorded; based on the Smith chart, multiple parameters of each QCM device under different gas environments are calculated, and the parameters include echo dissipation, resistance, reactance, impedance angle, impedance value, and resonance frequency; a plurality of parameters corresponding to the current volatile organic compound are combined as the gas characteristic fingerprint of the volatile organic compound; the gas characteristic fingerprint of each volatile organic compound constitutes a gas characteristic fingerprint library; The gas to be tested is detected by using a sensor array, and the gas characteristic fingerprint of the gas to be tested is calculated based on the Smith chart read by the network analyzer; The gas characteristic fingerprint of the gas to be tested is compared with the gas characteristic fingerprint library to obtain the compound type and gas concentration of the gas to be tested.

Images