Double-layer piezoelectric sensor and application and method of double-layer piezoelectric sensor in characterization of beta-phase polyvinylidene fluoride membrane adsorptive pollution

Abstract

The invention discloses a double-layer piezoelectric sensor and an application and method of the double-layer piezoelectric sensor in representing beta-phase polyvinylidene fluoride membrane adsorptive pollution, a piezoelectric beta-phase polyvinylidene fluoride membrane is a porous membrane formed by a blending system of polyvinylidene fluoride and ionic liquid and has a beta-phase crystal structure and a piezoelectric response characteristic, and the piezoelectric beta-phase polyvinylidene fluoride membrane is prepared by spin-coating a PVDF solution on a chip. The double-layer piezoelectric sensor containing the beta-phase polyvinylidene fluoride membrane is obtained, and QCM-D detection is carried out to characterize the adsorptive pollution condition of the beta-phase polyvinylidene fluoride membrane. Adsorption and desorption behaviors of organic matters with different dissolvability and mixtures thereof on the beta-phase PVDF membrane layer are represented, the adsorption result is matched with the hole blocking resistance in the actual membrane pollution process, and the hole blocking resistance is calculated by monitoring the fifth overtone of the piezoelectric quartz sensor. And the quality of soluble organic matters adsorbed on the PVDF surface and the property of an adsorption layer can be evaluated by using a frequency change value and a dissipation value, and a scheme is provided for researching the pollution behavior of the beta-phase polyvinylidene fluoride membrane.

Description

Technical Field

[0001] The present invention relates to the technical field of membrane fouling detection, and in particular to a bilayer piezoelectric sensor and its application and method for characterizing the adsorption fouling of β-phase polyvinylidene fluoride membranes. Background Art

[0002] Membrane fouling is an inevitable problem in membrane separation applications and an important limiting factor. Dissolved organic matter (DOM) is considered the main factor causing fouling in the ultrafiltration membrane process for water treatment. Membrane fouling caused by DOM mainly stems from pore blockage, concentration polarization, and the formation of gels or cake layers. Recently, researchers have classified membrane fouling into two categories: "adsorptive fouling" and "filter fouling". Adsorptive fouling is considered to originate from the intrinsic molecular interactions between pollutants and membrane materials, and this interaction can occur without filtration, i.e., without any hydrodynamic effects. Filter fouling, on the other hand, is caused by the dragging, deposition, and enrichment of pollutants in a pressure-driven membrane process under hydrodynamic conditions and is mainly affected by the internal hydrodynamics of the membrane module, the membrane pore structure, and other physical factors. Adsorptive fouling occurs in the initial stage of filtration and is caused by the strong interaction between pollutant molecules and membrane materials. This type of fouling is difficult to remove by physical cleaning methods (such as water washing) and is usually regarded as irreversible fouling.

[0003] The quartz crystal microbalance dissipation characterization technique (QCM-D) provides a valuable tool for the detailed study of adsorptive fouling. QCM-D is a piezoelectric mass sensor based on an angle-tuned (AT-cut) quartz crystal. The sensor element has a sandwich structure with metal electrodes coated on the upper and lower surfaces of the quartz, enabling it to resonate when an alternating current signal is applied. In QCM-D, the piezoelectric quartz exhibits shear vibration in the d31 mode, where "d" represents the piezoelectric coefficient. When the mass on the oscillator surface changes, the vibration resonance frequency of the piezoelectric quartz oscillator also changes accordingly. QCM-D can detect mass changes as low as a few nanograms per square centimeter. Previous QCM-D studies have explored the effects of factors such as membrane materials, membrane surface functions, the ionic strength of the feed solution, and the type of feed solution on membrane fouling behavior. Generally, the change in resonance frequency Δf is considered to be proportional to the amount of pollutants adsorbed on the membrane coating. The change in the dissipation factor ΔD represents the rate at which the piezoelectric quartz oscillation decays to zero and is an effective parameter for characterizing the structural properties of the adsorption layer. The dissipation value per unit mass ΔD / Δf is particularly useful for evaluating the structural properties of the adsorption layer because there is an inherent correlation between ΔD and Δf. The attachment of pollutants on polymer films measured by QCM-D is comparable to membrane fouling, and this conclusion is confirmed by comparing ultrafiltration membrane fouling in polymer films and membranes with QCM-D measurements using the same feed (process water from thermomechanical pulping) and the same material (hydrophilized polysulfone).

[0004] However, the correlation between the membrane fouling process and the adsorption-desorption results obtained by QCM-D has not been established. Generally, the flux decline during the ultrafiltration process is the combined result of concentration polarization, pore blockage, and cake layer formation. Since adsorptive fouling is usually considered irreversible, it may not be reasonable to directly compare the flux decline with the decrease in the adsorption frequency of QCM-D.

[0005] References: 1. G. Rudolph, A. Hermansson, A.-S. Jonsson, F. Lipnizki, In-situreal-time investigations on adsorptive membrane fouling by thermomechanicalpulping process water with quartz crystal microbalance with dissipationmonitoring (QCM-D), Separation and Purification Technology, 254 (2021)117578. 2. R. Miao, L. Wang, N. Mi, Z. Gao, T. Liu, Y. Lv, X. Wang, X. Meng,Y. Yang, Enhancement and mitigation mechanisms of protein fouling ofultrafiltration membranes under different ionic strengths, EnvironmentalScience&Technology, 49 (2015) 6574-6580. 3. A. Dolatshahi-Pirouz, K. Rechendorff, M.B. Hovgaard, M. Foss, J.Chevallier, F. Besenbacher, Bovine serum albumin adsorption on nano-roughplatinum surfaces studied by QCM-D, Colloids and Surfaces B-Biointerfaces, 66(2008) 53-59. 4、A. Tsortos, G. Papadakis, E. Gizeli, Shear acoustic wave biosensor for detecting DNA intrinsic viscosity and conformation: A study with QCM-D, Biosensors&Bioelectronics, 24 (2008) 836-841. Summary of the Invention

[0006] To solve the deficiencies of the prior art, the object of the present invention is to prepare a double-layer piezoelectric sensor, on the surface of the quartz crystal of which a β-phase polyvinylidene fluoride film is provided, which can be used for QCM-D analysis to evaluate the quality and stiffness changes of the dissolved organic matter layer adsorbed on the β-phase polyvinylidene fluoride film.

[0007] To achieve the above object, the present invention adopts the following technical solutions: A double-layer piezoelectric sensor, comprising a quartz crystal, on the surface of which a β-phase polyvinylidene fluoride film is provided.

[0008] Preferably, the preparation method of the aforementioned β-phase polyvinylidene fluoride film comprises the following specific steps: S1. Dissolve polyvinylidene fluoride, a pore-forming agent and an ionic liquid in a polar organic solvent to form a uniform casting solution; S2. Degas the casting solution and coat it on the quartz crystal to form a primary film; S3. Subject the primary film to phase separation, rinse it, and obtain a piezoelectric β-phase polyvinylidene fluoride film by a phase inversion process.

[0009] Preferably, the method of forming the primary film of the aforementioned casting solution on the quartz crystal adopts one of spin coating, drop coating or evaporation method. In the spin coating method, the rotation speed of the spin coater is 1000~4000 rpm, the humidity is 10%~30%, and the temperature is 22~25°C.

[0010] Preferably, the aforementioned ionic liquid is [EMIM][HSO 4 , [EMIM][Cl], [EMIM][Br] or [EMIM][BF 4 of one kind.

[0011] Preferably, the mass ratio of the aforementioned polyvinylidene fluoride, pore-forming agent, ionic liquid and polar organic solvent is 9~10:1~3:0.1~5:30~40; the pore-forming agent is polyvinylpyrrolidone K30 or polyethylene glycol 400; the polar organic solvent is one or more of N,N-dimethylformamide, N,N-dimethylacetamide, tetrahydrofuran or triethyl phosphate.

[0012] Preferably, the aforementioned phase inversion process is a non-solvent induced phase separation process or a process combining vapor induced phase separation and non-solvent induced phase separation.

[0013] Application of a double-layer piezoelectric sensor in characterizing the adsorption pollution of a β-phase polyvinylidene fluoride membrane.

[0014] A method for characterizing the adsorption pollution of a β-phase polyvinylidene fluoride membrane by a double-layer piezoelectric sensor, comprising the following specific steps: (1) Fix the quartz crystal described in claim 6 in a flow cell, pump in ultrapure water, and establish a baseline; (3) Then pump in the pollutant solution until adsorption equilibrium is reached. (4) Then pump in ultrapure water for rinsing; (5) Monitor the frequency drop value and dissipation factor value of the QCM-D device to obtain the result of the adsorption pollution of the β-phase polyvinylidene fluoride membrane; if the frequency drop value is large, it is determined that there is a large pore blockage resistance or irreversible pollution in the β-phase polyvinylidene fluoride membrane; if the dissipation factor value is high, it is determined that the viscoelasticity of the β-phase polyvinylidene fluoride membrane is large.

[0015] Preferably, the aforementioned pollutant solution is a solution of one or a mixture of dextran, humic acid, and bovine serum albumin at 30 - 70 mg / L. The time for pumping in the pollutant solution is 50 - 70 min, the temperature is 22 - 25 °C, and the feed liquid flow rate is 70 - 90 μL / min; the time for the double-layer piezoelectric sensor to contact pure water is 15 - 25 min; the rinsing time is 30 - 50 min.

[0016] The advantages of the present invention are as follows: (1) The present invention uses a water-soluble ionic liquid to prepare a PVDF ultrafiltration membrane with a piezoelectric phase, which exhibits an obvious β-phase composition and shows obvious piezoelectric response, hysteresis loop, and characteristic "butterfly-shaped" amplitude loop; (2) The present invention successfully prepares a PVDF-coated sensor and can control its polymorphism, showing good repeatability; (3) The present invention uses QCM-D to characterize the adsorption - desorption of pollutants by the β-phase PVDF membrane, and the adsorption result is consistent with the pore blockage resistance, indicating that it has broad application prospects in analyzing the irreversible pollution behavior of the β-phase PVDF membrane. Description of the Drawings

[0017] Figure 1 : a, [EMIM]HSO 4Mechanism for promoting the formation of β-phase PVDF in the coagulation bath; b, pore size distribution of VN1, VN2, VN3, and VN4 membranes; c, rejection rate of bovine serum albumin by the VN4 membrane; d, infrared spectrogram of the membrane; e, X-ray diffraction pattern of the membrane; f, differential scanning calorimetry curve of the membrane; Figure 2 are SEM morphology diagrams of the top surface, cross-section, and bottom surface of the β-phase polyvinylidene fluoride membrane; Figure 3 : a, amplitude butterfly curve of the piezoelectric polyvinylidene fluoride membrane; b, phase lag loop of the piezoelectric polyvinylidene fluoride membrane; c, fouling curves of the β-phase PVDF membrane when contacting hyaluronic acid, dextran, bovine serum albumin, and their mixture solutions during the cycling process described in the experimental section; d, pore blockage resistance of the β-phase PVDF membrane after one fouling cycle; Experimental conditions: feed solution concentration 100 mg / L, dextran mixture (molecular weights 40000:70000:100000:450000 - 650000, ratio 1:1:1:1), temperature 25 °C, pressure constant at 1 bar; Figure 4 is a schematic diagram for preparing β-phase PVDF crystals by spin coating and α-quartz; Figure 5 : a, two-dimensional height atomic force microscopy (AFM) images, phase images, and three-dimensional height AFM images of the gold sensor substrate, and corresponding images of the spin-coated β-phase polyvinylidene fluoride (PVDF) layer; b, frequency values of the spin-coated β-phase sensor at different harmonics; c, dissipation values of the spin-coated β-phase sensor at different harmonics; d, frequency reduction values of different sensors at the 5th harmonic by injecting water instead of air; Figure 6 : a, variation of frequency with time to characterize the adsorption of dissolved organic matter on the PVDF-coated sensor; b, variation of dissipation with time to characterize the adsorption of dissolved organic matter on the PVDF-coated sensor; c, variation of ΔD / Δf (ratio of dissipation change to frequency change) with time to characterize the adsorption of dissolved organic matter on the PVDF-coated sensor; d, pore blockage resistance of the β-phase PVDF membrane after one fouling cycle; e, final stable frequency reduction value of the β-PVDF sensor after fouling and rinsing (one adsorption-desorption cycle). f, comparison between the pore blockage resistance and the stable frequency reduction value; Experimental conditions: solution concentration 50 mg / L, test temperature 24 °C, flow rate 80 μL / min. Detailed implementation manners

[0018] The present invention will be specifically introduced below in conjunction with the accompanying drawings and specific embodiments.

[0019] A preparation method of a piezoelectric β-phase polyvinylidene fluoride membrane includes the following specific steps: S1. Dissolve polyvinylidene fluoride (PVDF) and polyvinylpyrrolidone K30 (PVP K30) in N,N-dimethylacetamide (DMAc), and stir at a mechanical stirring speed of 300 rpm at 30 °C for 16 hours to prepare a homogeneous casting solution.

[0020] S2. Then place the casting solution in an oven at 30 °C for degassing. Use a coating knife with a gap height of 250 μm to coat the PVDF solution on a plexiglass substrate at a controlled speed to form a nascent membrane. S3. After a set atmospheric exposure time, immerse the nascent membrane in a 30% (v / v) aqueous DMAc solution at 30 °C for phase separation. Then thoroughly rinse the membrane with deionized water to remove residual solvents and ionic liquids.

[0021] S4. Two phase inversion methods can be used to prepare the membrane: (1) directly immerse it in a coagulation bath for non-solvent induced phase separation (NIPS); (2) first expose it to ambient humidity (20%) for vapor induced phase separation (VIPS), and then immerse it in a coagulation bath for NIPS (referred to as V-NIPS).

[0022] Adjust the addition amounts of each raw material and the preparation method, and design Examples 1-6 and Comparative Examples 1-3 respectively. The detailed information on the composition and preparation method of the casting solution is shown in Table 1.

[0023] Table 1 Composition and preparation details of the casting solution

[0024] The preparation steps of the double-layer piezoelectric sensor are as follows: Prepare the PVDF coating in a controlled environment with a temperature of 20 °C and a humidity below 20%. First, dissolve PVDF 6010 in DMAc by magnetic stirring and let it stand for several hours to obtain a PVDF solution with a mass fraction of 2%. Install a new gold (Au) sensor on a spin coater. Carefully transfer 50 μL of the PVDF solution to the original sensor using a pipette to avoid forming bubbles. The spin coating program is set to accelerate from 0 to 1000 rpm with an acceleration of 5000 rpm / s, then hold at 1000 rpm for 20 seconds, and then increase to 4000 rpm and hold for 40 seconds. This process helps the evaporation of volatile solvents and forms a uniform PVDF coating on the Au sensor. The prepared sensor is stored in a dedicated sample box for subsequent use.

[0025] Performance detection experiment: 1. Detection method (1) Filtration and pollution experiment The filtration and fouling experiments were conducted using a self-built cross-flow filtration system, which consisted of a pump circulation system, a pressure control unit, a cross-flow ultrafiltration module, and an on-line mass monitoring system. The ultrafiltration cell had a capacity of 75 mL and an effective filtration area of 15 cm². A gear pump circulated the feed solution through the membrane module and returned the retentate to the feed tank. The operating conditions included a pressure of 1 bar, a flow rate of 600 mL / min, and maintaining the feed solution temperature at 25 °C using a water bath. Fresh pollutant solutions were prepared for each fouling experiment to ensure consistency.

[0026] The filtration resistance was determined by Darcy's law combined with the series resistance model. The total membrane fouling resistance (Rt) (m -1 ) included the contributions of the membrane intrinsic resistance (Rm) (m -1 ), the concentration polarization resistance (Rp) (m -1 ), the pore blocking resistance (Rb) (m -1 ), and the cake layer resistance (Rc) (m -1 ).

[0027]

[0028] Where Js is the flux (L·m -2 ·h -1 , or LMH), Δp is the transmembrane pressure (Pa), and μ is the liquid viscosity (Pa·s). Before the fouling experiment, the membrane intrinsic resistance (R 0 ) was calculated by measuring the flux of pure water through the fresh membrane (J m ). Then the feed fouling solution was filtered, and the steady-state flux (J 1 ) was recorded to calculate the total resistance. After that, the device was depressurized for 1 minute and then restarted to measure the flux (J 2 ), at which point the concentration polarization resistance (R p ) was eliminated. Finally, after removing the cake layer from the PVDF membrane, the flux of pure water (J 3 ) was measured. After this process, only the pore blocking resistance (R b ) and the membrane intrinsic resistance (R m ) remained.

[0029] The fouling resistance components were calculated using the following equations:

[0030] (2) The QCM-D testing procedure was as follows: The adsorption and desorption behaviors of dissolved organic compounds on the surface of polyvinylidene fluoride (PVDF) coatings were studied using a dissipative quartz crystal microbalance. All sensors were fabricated using a consistent manufacturing process. The QCM-D sensors had a fundamental resonance frequency of 5 MHz and an underwater mass sensitivity of 0.5 ng / cm² and were used as the substrates for the PVDF coatings. Before each experiment, the sensors were rinsed with ultrapure water and dried with nitrogen. To ensure coating uniformity and experimental reproducibility, the sensors were used only once.

[0031] For all QCM-D tests, the temperature was maintained at 24 °C, and the fluid flow was controlled to ensure laminar flow conditions, with the flow rate set to 80 μL / min using a peristaltic pump. First, a new PVDF-coated Au sensor was placed in the flow module of the QCM-D device, and the dissipation and frequency of the original sensor were recorded, measuring different overtones. Then, ultrapure water was pumped into the module to establish a baseline while allowing the sensor to stress equilibrate (temperature equilibration, O-ring stress relaxation, etc.). The flow of ultrapure water was recorded for 20 minutes to ensure a stable baseline, and then the ultrapure water was replaced with the contaminant solution. The contaminant solution was pumped into the module for 1 h to reach adsorption equilibrium, and then rinsed with ultrapure water for 40 minutes to remove any loosely bound contaminant molecules. The fifth overtone of the piezoelectric quartz sensor was monitored, and the frequency and dissipation values were used to evaluate the mass of the dissolved organic matter adsorbed on the PVDF surface and the properties of the adsorption layer. The average value of five recording points during the stabilization period was used to calculate the stable decrease value of the frequency. A new contaminant solution was prepared for each test.

[0032] When contaminant molecules deposit or adsorb on the sensor, the resonance frequency of the piezoelectric sensor decreases. The decrease or increase in the adsorbed mass due to the deposition and adsorption of contaminants on the PVDF surface can be calculated using the Sauerbrey relationship:

[0033] where Δm is the adsorbed mass, n is the overtone number, and C is the crystal constant (17.7 ng·Hz -1 ·cm -2 ).

[0034] (3) Characterization methods The morphology of the PVDF membrane was observed using a cold field emission scanning electron microscope at different magnifications. For membranes with pore sizes larger than 100 nm, the pore size distribution was measured using the gas-liquid displacement method. For membranes with pore sizes smaller than 100 nm, the liquid-liquid displacement method was adopted. The crystallinity of the PVDF membrane was analyzed using an X-ray diffractometer with a copper (Cu) target. The parameters were set as follows: acceleration voltage 40 kV, tube current 15 mA, scanning speed 5° / min, and angular range 5° to 60°. The functional groups present in the PVDF membrane were detected using a Fourier transform infrared spectrometer with an attenuated total reflection (ATR) mode. The scanning time was 32 min, and the resolution was 4 cm -1 . The thermogram of the PVDF membrane was characterized using a differential scanning calorimeter. DSC was also used to calculate the crystallinity, and the sample test temperature range was 20~200 °C.

[0035] The morphology of the sample was observed using an atomic force microscope in tapping mode. A sharp silicon AFM probe (RTESPA-300) with a spring constant of 40 ± 5 N / m was used for imaging. Piezoelectric response force microscopy (PFM) tests were performed using another atomic force microscope. The probe model was ARROW-EFM-20, and both the probe side and the detector side were coated with platinum / iridium (Pt / Ir). The amplitude and phase of the sample were obtained under a DC bias voltage of -80~80 V. The applied voltage frequency was 0.2 Hz, and the duration of a single pulse was 0.05 s.

[0036]

[0037] The rejection performance of the membrane was evaluated by measuring the bovine serum albumin (BSA) content in the permeate and feed solutions using formula (8). Based on the absorbance at 278 nm, a UV-visible spectrophotometer was used to determine the BSA concentration. A Zetasizer Nano instrument was used to measure the hydrodynamic diameter of the model contaminants. A total organic carbon (TOC) analyzer was used to determine the TOC concentration in the contaminant solution.

[0038] 2. Detection Results (1) Preparation and Characteristics of Piezoelectric Phase PVDF Membrane As Figure 1 shown in (a) of 4It is water-soluble and can be easily removed during the phase inversion and washing steps, which is beneficial for the preparation of pure PVDF membranes. Specifically, membranes VN1 and VN2 were prepared from 18 wt% PVDF, while VN3 and VN4 were prepared from 20 wt% PVDF. The slight difference between VN1 and VN2 (or VN3 and VN4) stems from the addition of 0.1 g of ionic liquid (IL) to the casting solution. Notably, the most probable pore sizes and pore size distributions of VN1 and VN2 (or VN3 and VN4) are similar, with VN1 and VN2 being 45 nm and 44 nm respectively, and VN3 and VN4 being 29 nm and 30.5 nm respectively, as shown in Figure 1 of (b). Figure 1 In (c) of, the BSA rejection rates of VN3 and VN4 over the filtration time further confirm the similarity of the pore sizes.

[0039] Figure 1 of (d) shows the infrared spectra of the prepared membranes. The bands at 763 cm -1 , 796 cm -1 and 976 cm -1 correspond to the characteristic peaks of α-phase PVDF, while the bands at 840 cm -1 and 1279 cm -1 are attributed to β-phase PVDF. The peaks at 880 cm -1 , 1071 cm -1 , 1194 cm -1 and 1396 cm -1 are related to amorphous PVDF. Compared with VN1 and VN3 without [EMIM][HSO 4 , the peaks of membranes VN2 and VN4 at 763 cm -1 , 796 cm -1 , 976 cm -1 and 1383 cm -1 decrease, while the peaks at 840 cm -1 and 1279 cm -1 increase, indicating that [EMIM][HSO 4 inhibits the formation of α-phase PVDF and promotes the formation of β-phase PVDF. The absence of additional absorption bands near 827 cm -1 , 1158 cm -1 and 1570 cm -1 indicates the presence of [EMIM][HSO 4 . In fact, all samples, including N5 prepared using a casting solution containing 9% IL, did not show any detectable [EMIM][HSO 4Peaks. FTIR generally cannot provide information about the polymorphism in the bulk of the membrane structure. Due to the numerous irregular gas-solid interfaces in the porous structure, a large amount of photon scattering occurs, especially at higher wavenumbers, resulting in a significant slope and a low signal-to-noise ratio. Therefore, the practicality of the traditional transmission mode is limited, and alternative techniques such as X-ray diffraction (XRD) are used to supplement the FTIR characterization.

[0040] The XRD spectra of the membranes are as Figure 1 shown in (e). Membranes VN2 and VN4 are mainly in the β-phase, as indicated by the distinct peak at 20.26°. In contrast, the main peaks of VN1 and VN3 PVDF membranes are broader and shifted to 19.9°, indicating the presence of the (110) crystal plane of α-phase PVDF. In addition, the weak peaks at 18.3° and 26.6° suggest the incorporation of α-phase PVDF into the membrane matrix. Specifically, VN3 shows an obvious α-phase, while VN4 exhibits a unique β-phase. It is worth noting that the melting points of all samples are similar, approximately 170 °C, as Figure 1 shown in (f).

[0041] The prevalence of β-phase PVDF after adding IL can be mainly explained by the ion-dipole interaction between the CF2 groups of PVDF and the cationic part of IL ([EMIM]), as Figure 1 shown in (a). Incorporating cations (rather than anions) can induce the polar phase of PVDF. The interaction between the charged ions and the dipole moments at the water-membrane intimate interface arranges the polymer chains along the interface. This structure improves the stacking of CH 2 -CF 2 dipoles and leaves [EMIM]HSO 4 residual in the coagulation due to its high water solubility. Overall, it is feasible to prepare β-phase PVDF membranes without additives using [EMIM]HSO 4 as an additive in combination with the traditional phase inversion method.

[0042] Figure 2 The results show that all membranes exhibit a typical asymmetric liquid-liquid phase separation structure, characterized by the presence of semi-finger-like pores and sponge-like pores in the membrane cross-section. The top surface of the membrane is rich in dense nanoscale pores, serving as the separation layer. Membranes prepared with a higher polymer concentration show higher mechanical strength and smaller pore sizes, which can enhance anti-fouling properties. β-phase PVDF can not only be used as a membrane material but also as a piezoelectric sensor material, with a lower adhesion interaction with dissolved organic matter (DOM) pollutants. Therefore, VN4 containing β-phase PVDF was selected for further fouling studies.

[0043] (2) Piezoelectric response of the membrane and its fouling behavior towards dissolved organic matter Figure 3Panels (a) and (b) show the piezoelectric response amplitude loop and phase lag loop of the VN4 (β-phase) membrane. The lag loop in the phase and the classical "butterfly shape" in the amplitude loop indicate the presence of well-defined polarization. The phase saturates at ±80 V, and the phase flip is close to 160°. The oriented CH 2 / CF 2 dipoles form self-polarized nanocrystals within the membrane matrix. To evaluate the fouling behavior of the membrane, humic acid (HA), dextran (DEX), bovine serum albumin (BSA), and their mixtures were used as organic pollutants for testing. A self-made crossflow system was used to record the membrane flux throughout the cycle to study the interaction between the piezoelectric PVDF surface and the pollutants. Figure 3 Panel (c) shows the fouling curve of the VN4 (β-phase) membrane.

[0044] The initial flux of the new membrane was approximately 390 LMH and gradually decreased after introducing the pollutant solution, with the flux decreasing the fastest in the presence of bovine serum albumin. The entire cycle included a 30-minute water permeation baseline, followed by 60 minutes of introducing DOM pollutants to reach a stable flux, and then 20 minutes of replacement with pure water. After that, a 1-minute decompression was performed to remove concentration polarization, during which the water flux was recorded for 20 minutes. After removing the cake layer, the water flux was obtained for another 10 minutes. After removing the concentration polarization and the pollutant cake layer, the stable fluxes of the β-phase membranes were measured. For the membranes fouled by humic acid, dextran, bovine serum albumin, and the mixture, the stable fluxes were 324 LMH, 300 LMH, 162 LMH, and 271 LMH, respectively. This indicates that the resistance order of PVDF to DOM is: humic acid > dextran > mixture > bovine serum albumin.

[0045] To further analyze the membrane fouling mechanism, a resistance analysis of the fouling process was carried out by combining Darcy's law and the series resistance model ( Figure 3 Panel (d)). It can be observed that when the membrane is fouled by humic acid and dextran, the fouling resistance mainly comes from concentration polarization. In contrast, the resistance generated by bovine serum albumin and the mixture pollutants is mainly due to the formation of a cake layer, indicating that in the presence of bovine serum albumin, a dense cake layer is easily formed on the membrane surface, resulting in a rapid decline in flux. In addition, the pore blockage resistance caused by bovine serum albumin is significant, with a value of 5.4 × 10 11 m -1, which is four times the pore plugging resistance caused by dextran and ten times the resistance caused by humic acid. As mentioned above, in the resistance series model, the pore plugging resistance is considered irreversible because it cannot be eliminated by water washing throughout the cycle. In contrast, concentration polarization and cake layer fouling are considered reversible fouling and can be alleviated by mechanical hydraulic cleaning of the membrane. The pore plugging resistance can be used as an indicator of the level of interaction between the pollutant and the membrane material, indicating that the binding strength trend between the membrane material and the pollutant is: bovine serum albumin > mixture > dextran > humic acid.

[0046] (3) Design and properties of bilayer piezoelectric sensors See Figure 4 , the spin coating method produces a stretching effect on the PVDF chains due to the outward action of the centrifugal force, which is beneficial to the formation of β-phase PVDF. The higher the rotation speed during the spin coating process, the greater the stretching effect on the PVDF chains, resulting in a higher β-phase content in the final PVDF membrane. In the present invention, a rotation speed of 4000 rpm was adopted, and it is expected that this speed is sufficient to induce the formation of the β-phase. In addition, humidity control during the coating deposition process is crucial because water vapor in a high-humidity environment will lead to the formation of a porous and rough membrane structure. Therefore, a relative humidity of 20% and a temperature of 20 °C were maintained in the present invention.

[0047] Images of the PVDF coating were obtained by atomic force microscopy (AFM) ( Figure 5 (a)), but due to the very thin PVDF layer (only a few nanometers) and the interference of the gold substrate, infrared analysis could not be performed. The AFM images showed that the surface of the original gold sensor was smooth and there were spherical structures corresponding to β-phase PVDF spherulites. There were significant differences between the phase diagrams and three-dimensional images of the original sensor and the spin-coated sensor, confirming the formation of a uniform PVDF layer on the substrate. This process resulted in a bilayer piezoelectric sensor composed of α-quartz and β-phase PVDF. The surface roughness (Ra) values of the original sensor (scanning sizes of 5×5 μm and 20×20 μm respectively) were measured to be 0.66 nm and 1.06 nm, while the surface roughness values of the PVDF-coated sensor were 6.57 nm and 9.84 nm respectively. These measurement results indicate that the membrane surface is very smooth, with a root mean square value below 10 nm, suggesting a dense and compact structure. The process of transforming β-phase PVDF into α-phase PVDF by melt crystallization has been well established. In the present invention, α-phase PVDF was obtained by heating the spin-coated β-phase PVDF layer to 200 °C and holding for 10 minutes, and then naturally cooling. The resulting radial lamellae are characteristic of microcrystalline α-phase PVDF.

[0048] (4) QCM-D analysis and irreversible fouling situation As Figure 6As shown in (a), after introducing the pollutant solution, a frequency decrease was immediately observed, indicating that DOM was adsorbed onto the PVDF layer. It should be noted that the adsorption and desorption of pollutants is a dynamic process involving conformational reconstruction until equilibrium is reached. Similar to membrane fouling, the adsorption of dissolved organic matter can be classified into two categories: reversible and irreversible. In the present invention, pollutants that can be washed off by water are considered to have undergone reversible adsorption. After gentle water flushing (80 μL / min for 40 minutes), the degrees of frequency decrease for HA, DEX, and the mixed solution were different, while the frequency of BSA did not change, indicating that BSA could not be removed by flushing, suggesting a strong binding between BSA and the PVDF layer. In addition, the frequency decrease of BSA was significantly higher than that of other pollutants, and it steadily decreased by 85 Hz after flushing, indicating a relatively high Sauerbrey adsorption mass. In contrast, the frequency decreases for HA, DEX, and the mixed solution were 5 Hz, 18 Hz, and 25 Hz, respectively.

[0049] Figure 6 (b) shows the dissipation curves of the piezoelectric PVDF-coated sensor when exposed to different pollutants. The dissipation values of BSA and HA were relatively low, close to the X-axis. The dissipation curves of DEX and the mixed solution were similar.

[0050] BSA caused a significant frequency decrease, but its dissipation value remained very low, and there was no obvious change in dissipation throughout the test. As mentioned above, a higher dissipation value (D) indicates a greater viscoelasticity of the adsorption layer. This indicates that a dense and tightly bound BSA layer was formed on the PVDF-coated sensor. The relatively low dissipation value of humic acid may be attributed to its relatively low Sauerbrey adsorption mass. In addition, the ΔD / Δf ratio was calculated to further analyze the characteristics of the adsorption layer, as Figure 6 shown in (c). A lower │ΔD / Δf│ value indicates the formation of a dense and compact layer, while a higher value indicates the formation of a "soft" and open structure with higher dissipation. Therefore, the adsorption of HA on the PVDF layer was the loosest, the adsorption of BSA was the tightest, and DEX was in between, which also indicates that PVDF was more easily fouled by BSA.

[0051] Figure 6 (d) and (e) show the pore blocking resistance during the membrane filtration process and the final frequency shift value of the QCM-D test, respectively. As Figure 6 shown in (f), there was a strong correlation between these trends, indicating that both the pore blocking resistance and the frequency decrease value in QCM-D reflect the interaction between pollutants and the membrane material. These measurements can be used to compare and predict the degree of irreversible membrane fouling, and for the first time in the present invention, a numerical correlation was established between the membrane fouling process and the QCM-D results.

[0052] The basic principles, main features and advantages of the present invention have been shown and described above. Those skilled in the art should understand that the above embodiments do not limit the present invention in any form, and all technical solutions obtained by means of equivalent replacement or equivalent transformation fall within the protection scope of the present invention.

Claims

1. A double-layer piezoelectric sensor, comprising a quartz crystal, characterized in that: The surface of the quartz crystal is provided with a β-phase polyvinylidene fluoride film.

2. A double-layer piezoelectric sensor according to claim 1, characterized in that: The preparation method of the β-phase polyvinylidene fluoride film comprises the following specific steps: S1, dissolving polyvinylidene fluoride, a porogen and an ionic liquid in a polar organic solvent to form a uniform film casting solution; S2, degassing the casting solution and coating it on a quartz crystal to form a primary film; S3, phase separation of the primary film, after washing, a phase inversion process is used to obtain a piezoelectric β-phase polyvinylidene fluoride film.

3. A double-layer piezoelectric sensor according to claim 2, characterized in that: The method for forming a primary film on a quartz crystal using the casting liquid adopts one of spin coating, drop coating or evaporation. In the spin coating method, the rotation speed of the spin coater is 1000-4000 rpm, the humidity is 10%-30%, and the temperature is 22-25°C.

4. A double-layer piezoelectric sensor according to claim 2, characterized in that: The ionic liquid is one of [EMIM][HSO4], [EMIM][Cl], [EMIM][Br] or [EMIM][BF4].

5. A double-layer piezoelectric sensor according to claim 2, characterized in that: The mass ratio of the polyvinylidene fluoride, porogen, ionic liquid and polar organic solvent is 9-10:1-3:0.1-5:30-40; the porogen is polyvinyl pyrrolidone K30 or polyethylene glycol 400; the polar organic solvent is one or more of N,N-dimethylformamide, N,N-dimethylacetamide, tetrahydrofuran or triethyl phosphate.

6. A double-layer piezoelectric sensor according to claim 2, characterized in that: The phase inversion process is a non-solvent induced phase separation process or a process combining steam induced phase separation and non-solvent induced phase separation.

7. Use of the double-layer piezoelectric sensor described in claim 1 in characterizing adsorptive contamination of β-phase polyvinylidene fluoride membranes.

8. The method for characterizing adsorptive contamination of β-phase polyvinylidene fluoride membrane using a double-layer piezoelectric sensor as claimed in claim 1, characterized in that: The specific steps include: (1) Fixing the quartz crystal described in claim 1 in a flow cell, pumping ultrapure water into the flow cell, and establishing a baseline; (3) Then the pollutant solution is pumped in to reach adsorption equilibrium. (4) Pump in ultrapure water for flushing; (5) Monitor the frequency drop and dissipation factor of the QCM-D device to obtain the results of the adsorption of pollutants by the β-phase polyvinylidene fluoride membrane; If the frequency drop value is large, it is judged that the β-phase polyvinylidene fluoride membrane has a large pore blocking resistance or irreversible pollution; If the dissipation factor value is high, it is judged that the viscoelasticity of the β-phase polyvinylidene fluoride film is large.

9. The method according to claim 8, characterized in that The pollutant solution is a 30-70 mg / L solution of one of dextran, humic acid, and bovine serum albumin or a mixture thereof. The time for pumping the pollutant solution is 50-70 min, the temperature is 22-25° C., and the flow rate of the feed liquid is 70-90 μL / min.

Images