A method for dynamically monitoring CO2 dissolution and diffusion based on microfluidic fluorescence imaging
By constructing a closed-loop measurement system with pH=7 as the baseline under high temperature and high pressure, and combining fluorescence image processing and CO2 dissolution equilibrium relationship, the problem of real-time and in-situ monitoring of CO2 dissolution and diffusion process was solved, achieving high-precision CO2 concentration field inversion and ensuring the stability and accuracy of the observation results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WUHAN UNIV
- Filing Date
- 2026-05-18
- Publication Date
- 2026-06-12
AI Technical Summary
Existing technologies struggle to achieve real-time, in-situ, and visual monitoring of the CO2 dissolution and diffusion process under high temperature and high pressure conditions. Furthermore, fluorescence signals are easily affected by environmental disturbances, leading to unstable measurements and an inability to accurately reflect real chemical changes.
A closed-loop measurement system based on pH=7 was constructed. By normalizing fluorescence images and using binarization masking technology, combined with the solubility equilibrium of CO2 in water, a high-resolution quantitative conversion from fluorescence intensity to CO2 concentration was achieved, eliminating the influence of environmental interference.
Stable, reliable, and quantitative observation of the CO2 dissolution and diffusion process under high temperature and high pressure conditions has been achieved, providing monitoring results with high spatiotemporal resolution and providing reliable experimental basis for the CO2 geological storage process.
Smart Images
Figure CN122193014A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of CO2 dissolution and diffusion dynamic monitoring methods. Background Technology
[0002] CO2 geological storage, as a key technology for mitigating the greenhouse effect, relies heavily on the occurrence and migration behavior of CO2 in deep reservoirs for its safety and long-term stability. At the pore scale, the dissolution and diffusion processes of CO2 directly affect the long-term stability and storage efficiency of the storage system. Therefore, accurately monitoring the behavior of CO2 during the storage process is crucial for assessing storage effectiveness and ensuring environmental safety.
[0003] However, real-time observation of CO2 dissolution and diffusion processes at the pore scale still faces significant challenges. Traditional intermittent sampling and offline analysis methods have low spatial resolution and disrupt the original spatial distribution of fluids within porous media, making it impossible to capture dynamic processes. Although fluorescence imaging-based monitoring techniques have the potential for in-situ and visualization, they face two major bottlenecks when applied to simulate high-temperature and high-pressure environments that mimic real geological conditions: First, the physicochemical properties of the solution under high-temperature and high-pressure conditions, such as dissociation equilibrium and changes in ionic strength, cause drift in the calibration relationship between fluorescence signals and pH / CO2 concentration, resulting in poor measurement stability. Second, microscopic experimental images are easily interfered with by fluorescence background from non-target areas such as the chip substrate and impurities, and fluctuations in illumination during experiments can lead to inaccurate fluorescence intensity, making it difficult to directly extract reliable signals reflecting real chemical changes from the original images.
[0004] Therefore, developing a method that can fundamentally overcome the interference of high temperature and high pressure environments and achieve stable, accurate, and quantitative inversion from raw fluorescence images to CO2 concentration fields is of urgent scientific need and important engineering value. Summary of the Invention
[0005] To address the aforementioned technical challenges, this invention provides a dynamic monitoring method for CO2 dissolution and diffusion based on microfluidic fluorescence imaging. The core of this invention lies in constructing a closed-loop measurement system with pH=7 as a unified and stable benchmark. This system not only enables real-time, in-situ, and visualized observation of the CO2 dissolution and diffusion process under simulated geological conditions, but also fundamentally solves the measurement inaccuracy problem caused by environmental disturbances through an innovative data processing workflow. This ensures high precision, high spatiotemporal resolution, and high robustness of the observation results, enabling real-time, in-situ observation and quantitative characterization of the CO2 dissolution and diffusion process at the pore scale under high temperature and high pressure environments. This provides reliable experimental evidence for fundamental research and technological optimization of CO2 geological storage processes. Specifically, this invention is achieved through the following technologies.
[0006] This invention provides a method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging, comprising the following steps:
[0007] Based on several sodium fluorescein solutions with different pH values, the original average fluorescence intensity of the corresponding fluorescence images was obtained. The original average fluorescence intensity was normalized to obtain the relative fluorescence intensity, and a pH-relative fluorescence intensity calibration curve was established by fitting.
[0008] CO2 percolation experiments were conducted in a microfluidic chip, and fluorescence images were acquired at predetermined time intervals to form a fluorescence image sequence.
[0009] The fluorescence image with the best contrast between the microchannel and the background is selected from the fluorescence image sequence, and then binarized and denoised to obtain a binarized mask image of the microchannel covering the microfluidic chip.
[0010] Based on the binarized mask image, the relative fluorescence intensity of each fluorescence image in the fluorescence image sequence is calculated to form a relative fluorescence intensity sequence;
[0011] Using the pH-relative fluorescence intensity calibration curve, the relative fluorescence intensity sequence was converted into a pH field evolution dataset;
[0012] Based on the pH field evolution dataset and the CO2 dissolution equilibrium relationship in water, the spatiotemporal distribution of CO2 concentration at the pore scale of the microfluidic chip was calculated.
[0013] Furthermore, the method for normalization and fitting to establish the pH-relative fluorescence intensity calibration curve is as follows:
[0014] The average fluorescence intensity I of the fluorescence image corresponding to the sodium fluorescein solution at pH 7.0 was selected. 7,calib As a normalization benchmark, according to Formula I rel,calib =I raw,calib / I 7,calib Calculate the relative fluorescence intensity I of the fluorescence images corresponding to sodium fluorescein solutions at different pH values. rel,calib I raw,calib The original average fluorescence intensity of the fluorescence images corresponding to sodium fluorescein solutions at different pH values;
[0015] Using the measured pH value of each of the sodium fluorescein solutions as the independent variable, the corresponding relative fluorescence intensity I rel,calib Using the variable as the dependent variable, a nonlinear regression fitting was performed to obtain the candidate pH-relative fluorescence intensity calibration curve;
[0016] The parameters of the candidate pH-relative fluorescence intensity calibration curves were optimized to obtain pH-relative fluorescence intensity calibration curves that meet the requirements.
[0017] Furthermore, a four-parameter Logistic function was used for nonlinear regression analysis and fitting.
[0018] Furthermore, the method for optimizing the parameters of the candidate pH-relative fluorescence intensity calibration curve is as follows: the parameters are optimized using the least squares method, and the fitting quality is evaluated by calculating the coefficient of determination, mean absolute error, and root mean square error.
[0019] Furthermore, the mathematical expression for the pH-relative fluorescence intensity calibration curve is:
[0020] ;
[0021] Where I is the relative fluorescence intensity, a and d represent the lower and upper limits of fluorescence intensity, respectively, b is the Hill coefficient, and c is the pH value corresponding to the inflection point of the curve.
[0022] Furthermore, during the CO2 infiltration test, the predetermined time interval for acquiring the fluorescence images is 30 s / frame.
[0023] Furthermore, the method for selecting the fluorescence image with the best contrast between the microchannel and the background from the fluorescence image sequence, performing binarization and noise reduction processing, and obtaining the binarized mask image covering the microchannel of the microfluidic chip is as follows:
[0024] Several fluorescence images were selected from the fluorescence image sequence when the background solution filled the microchannels, and the fluorescence image with the best contrast between the microchannels and the background was selected.
[0025] The Otsu automatic thresholding method based on grayscale histogram was used to binarize the fluorescence image and segment the microchannels in the fluorescence image.
[0026] A morphological closing operation method of first dilation and then erosion is used to eliminate noise, smooth the region boundaries of the microchannels, and generate a binary mask image that accurately covers all the microchannels.
[0027] Further, based on the binarized mask image, the relative fluorescence intensity of each fluorescence image in the fluorescence image sequence is calculated to form a relative fluorescence intensity sequence (that is, the binarized mask image covering the microchannels of the microfluidic chip is applied to the entire fluorescence image sequence). The method is as follows:
[0028] A binarized mask image covering the microchannels of the microfluidic chip is applied to the entire fluorescence image sequence;
[0029] Under calibration test conditions, the fluorescence image of the initial frame was selected when the entire microchannel was filled with sodium fluorescein solution at pH=7.0 at the initial moment of the CO2 percolation test, and the average fluorescence intensity I within the masked area was obtained.7,exp ;
[0030] The average absolute fluorescence intensity I of sodium fluorescein solution at pH 7.0 7,calib As a reference standard, according to the formula k=I 7,calib / I 7,exp The global fluorescence intensity correction coefficient k was calculated.
[0031] According to Formula I corr,exp =k·I raw,exp The corrected average fluorescence intensity I was calculated. corr,exp I raw,exp The original average fluorescence intensity within each fluorescence image mask region in the fluorescence image sequence during the CO2 infiltration test;
[0032] According to Formula I rel,exp =I corr,exp / I 7,calib The corrected average fluorescence intensity I corr,exp Converted to relative fluorescence intensity I based on pH=7.0 rel,exp This forms a corresponding relative fluorescence intensity sequence.
[0033] Furthermore, the method for converting the relative fluorescence intensity sequence into a pH field evolution dataset using the aforementioned pH-relative fluorescence intensity calibration curve is as follows:
[0034] Each corrected relative fluorescence intensity (I) in the relative fluorescence intensity sequence rel,exp Substitute the values into the pH-relative fluorescence intensity calibration curve that meets the requirements, calculate the corresponding pH values, and form a pH field evolution dataset.
[0035] Furthermore, based on the pH field evolution dataset and the CO2 dissolution equilibrium relationship in water, the method for calculating the spatiotemporal distribution of CO2 concentration at the pore scale of the microfluidic chip is as follows:
[0036] Using the pH field evolution dataset, the dissolved CO2 concentration [CO2(aq)] corresponding to each pixel is calculated according to the following formula;
[0037] ;
[0038] ;
[0039] ;
[0040] Where K1 and K2 are the first and second dissociation constants, respectively, K w Let H be the ion product constant of water, [H] + (aq)] Dissolved H for each pixel+ concentration.
[0041] In the above method, this invention first establishes a pH-relative fluorescence intensity calibration curve with pH=7 fluorescence intensity as the normalization benchmark, fundamentally avoiding the disturbance of the calibration relationship by environmental factors. Secondly, before inverting the time-series fluorescence images obtained from the CO2 seepage experiment, it innovatively introduces two preprocessing steps: first, it accurately locates the microchannel reaction area using image segmentation technology (mask extraction) to eliminate background interference; second, it normalizes and corrects the fluorescence intensity of the target area using the same pH=7 benchmark to eliminate light intensity fluctuations in time-series acquisition and obtain reliable relative fluorescence intensity data. Finally, combined with the CO2 dissolution equilibrium model, it achieves high-resolution, quantitative inversion from standardized optical signals to the spatiotemporal distribution of CO2 concentration.
[0042] Compared with the prior art, the advantages of the present invention are:
[0043] 1. This invention establishes a highly robust anti-interference in-situ measurement system: by creating a relative fluorescence intensity calibration curve with pH=7 as a constant internal reference, and by normalizing the fluorescence intensity of the CO2 seepage test fluorescence image to the same standard, the system drift effect of high temperature and high pressure environment, light source fluctuation and solution physicochemical property changes on fluorescence measurement signal is fundamentally eliminated, ensuring the long-term stability and comparability of quantitative data throughout the observation process, and realizing stable and reliable in-situ observation under real geological temperature and pressure conditions.
[0044] 2. Achieved precise and automated quantitative conversion from raw images to concentration fields: Through innovative "mask extraction" technology, the microchannel reaction region is accurately segmented and locked, effectively eliminating signal interference from chip background and non-target areas. Combined with the "fixed coefficient correction and normalization based on initial pH=7 state calibration" algorithm, the time-series fluorescence image sequence is transformed into standardized relative fluorescence intensity data. This preprocessing workflow, combined with a high-precision fitting calibration model, achieves fully automated, high-fidelity, and high spatial resolution inversion from complex raw optical signals to pH fields and CO2 concentration spatiotemporal distribution fields.
[0045] 3. Combining realistic physical simulation with broad applicability: The microfluidic chip used is flexibly designed, and its microchannels can simulate various geological structures, from simple geometric channels to pore networks replicated based on real core scanning data. It is also compatible with various substrate materials such as glass, PDMS and real rocks. While ensuring high temperature and high pressure resistance, it greatly improves the fidelity of physical simulation in the experiment, enabling the method to adapt to the study of CO2 dissolution and diffusion behavior under different reservoir conditions.
[0046] 4. Provides a highly integrated standard analysis workflow: From image acquisition, mask extraction, fluorescence intensity correction, calibration and inversion to concentration calculation, a standardized, automated, and non-destructive complete analysis workflow is formed. This workflow avoids manual intervention and sample damage in traditional methods, significantly improving experimental efficiency and the consistency of results, and providing a powerful quantitative analysis tool for studying the multi-physicochemical coupling processes in CO2 geological storage. Attached Figure Description
[0047] Figure 1 This is a flowchart illustrating the dynamic monitoring method for CO2 dissolution and diffusion provided by the present invention.
[0048] Figure 2 The images shown are fluorescence images of the sodium fluorescein test solution at different pH values in step one of the examples.
[0049] Figure 3 This is the pH-relative fluorescence intensity calibration curve obtained in step two of the examples, which meets the requirements.
[0050] Figure 4 The image shown is a binarized mask image of the microchannel region obtained in step three of the embodiment.
[0051] Figure 5 The fluorescence intensity correction and normalization curves for step four in the embodiment are shown.
[0052] Figure 6 This is a spatiotemporal distribution map of CO2 concentration obtained in step five of the embodiment. Detailed Implementation
[0053] The technical solution of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0054] The CO2 dissolution and diffusion dynamic monitoring method provided by this invention uses a microchannel structure that is a simulated geological structure or a real geological structure to simulate the geological reservoir environment. These structures include a single linear channel, a serpentine parallel channel, a two-dimensional pore network, and a three-dimensional porous medium structure replicated based on real core scanning data or a real pore fracture channel formed by encapsulating rock thin sheets / particles.
[0055] Optionally, the microfluidic chip is made of an artificial substrate material or a real rock material; wherein the artificial substrate material includes glass, polydimethylsiloxane, polymethyl methacrylate and cyclic olefin copolymers, and the real rock material is various encapsulated rock flakes or rock particles.
[0056] In the CO2 dissolution and diffusion dynamic monitoring method provided by the present invention, image acquisition is performed by a fluorescence microscope system, which includes a scientific-grade camera and an excitation light source.
[0057] Optionally, the light source is a mercury lamp, used in conjunction with a fluorescent filter with an excitation wavelength of 450-490 nm and an emission wavelength of >500 nm.
[0058] Example
[0059] The CO2 dissolution and diffusion dynamic monitoring method based on microfluidic fluorescence imaging provided in this embodiment has the following process: Figure 1 As shown, it includes the following steps.
[0060] Step 1: Establish pH-relative fluorescence intensity calibration curve
[0061] This step aims to establish a quantitative relationship between solution pH and fluorescence intensity, providing a basis for conversion in subsequent CO2 percolation experiments.
[0062] 1. Prepare a 10 mM sodium fluorescein working solution
[0063] Accurately weigh 198 mg of sodium fluorescein solid (purity ≥95%) using an analytical balance, dissolve it in ultrapure water, transfer it to a 50 mL brown volumetric flask, dilute to the mark, and shake well before use.
[0064] 2. Prepare a series of sodium fluorescein test solutions with different pH values.
[0065] Take 12 50 mL test tubes and number them 1-12. Accurately add 10.0 mL of Britton-Robinson (BR) buffer stock solution (pH=3) to each test tube, and then add 0.1 M NaOH standard solution and 0.4 mL of the above-mentioned 10 mM sodium fluorescein working solution sequentially, following the solution preparation method in Table 1 below. Finally, bring the total volume to 25.0 mL with ultrapure water, tighten the cap, and shake thoroughly to mix.
[0066] Table 1. Preparation of a series of fluorescence test solutions with different pH values
[0067]
[0068] Note: The NaOH volume in Table 1 above is an estimated value based on theory. Because there are differences between different batches of BR mother liquor, the final "measured pH" of each sample is the basis for the pH-relative fluorescence intensity calibration curve.
[0069] 3. Measure pH
[0070] A series of prepared sodium fluorescein test solutions with different pH values were placed in a constant temperature water bath at 60℃ (±0.5℃) and preheated for at least 30 minutes to ensure temperature equilibrium.
[0071] Subsequently, using a precision pH meter calibrated at the same temperature with standard buffer solutions of pH 4.01, 6.86, and 9.18, the measured pH value of each solution was sequentially measured and recorded. This measured pH value was used as the accurate x-axis of the pH-relative fluorescence intensity calibration curve.
[0072] 4. Capture fluorescence images
[0073] (1) Inject each constant temperature solution into a special fluorescence cuvette and fix it on the stage of the fluorescence microscope. Fluorescence imaging is performed using an upright fluorescence microscope system equipped with a scientific-grade CCD camera. The excitation source is a mercury lamp, and the excitation light is provided by a blue excitation block (excitation wavelength range 450-490 nm, beam splitter cutoff wavelength 495 nm, emission wavelength > 500 nm).
[0074] During image acquisition, the microscope objective lens focal length was fixed at 80 mm, the CCD camera exposure time was set to 50 ms, and the intensity was 1; fluorescence images were captured under a constant temperature of 60℃, such as... Figure 2 As shown.
[0075] (2) Use image processing programs to batch calculate and extract the original average fluorescence intensity I of each fluorescence image. raw,calib Image processing programs can be either existing, known programs or custom-written programs using the Python platform.
[0076] 5. Normalization of average fluorescence intensity and fitting of pH-relative fluorescence intensity calibration curve.
[0077] (1) The average fluorescence intensity I of the fluorescence image of the sodium fluorescein solution with pH=7.0. 7,calib As a normalization benchmark, the relative fluorescence intensity I of the fluorescence images corresponding to each sodium fluorescein solution was calculated. rel,calib =I raw,calib / I 7,calib .
[0078] (2) Using the measured pH value of each sodium fluorescein solution as the independent variable and the relative fluorescence intensity as the dependent variable, nonlinear regression analysis was performed. Existing nonlinear regression curve fitting software can be used for the analysis, or a nonlinear regression curve fitting program can be written based on the Python platform.
[0079] The nonlinear regression curve fitting program in this embodiment is based on the SciPy optimization library of the Python platform. It uses the four-parameter Logistic function (S-shaped function) to establish a quantitative relationship model between pH and relative fluorescence intensity. Its mathematical expression is as follows:
[0080]
[0081] Where I is the relative fluorescence intensity, a and d represent the lower and upper limits of fluorescence intensity, respectively, b is the Hill coefficient, and c is the pH value corresponding to the inflection point of the curve.
[0082] The fitting process uses the least squares method to optimize parameters, and the coefficient of determination (R²) is calculated. 2 A comprehensive evaluation of the fit quality is performed. The specific calculation method is as follows:
[0083]
[0084] in, To measure the pH value, To predict pH values for the model, R represents the average measured pH value, where n is the total number of samples. 2 A value ≥0.98 indicates that the fitted curve can explain most of the variation in the experimental data and has a good fit.
[0085] like Figure 3 As shown, its goodness of fit R 2 A value >0.98 was obtained, yielding a high-precision, invertible pH-relative fluorescence intensity calibration curve, providing a reliable basis for the subsequent accurate reconstruction of the pH field.
[0086] Step 2: Conduct CO2 micro-permeation experiments and image acquisition
[0087] This embodiment uses a microfluidic chip based on a basalt core sheet with a thickness of 500 μm. After cutting, polishing and plasma cleaning, the chip is encapsulated in a high-temperature and high-pressure sapphire glass with pre-set inlet / outlet microchannels to form a parallel fracture microchannel with a cross-section of 0.5 mm × 1 mm and a length of 10 mm.
[0088] The chip was placed on the stage of a fluorescence microscope. First, a solution containing 0.2 mM sodium fluorescein (initial pH=7.0) was introduced. Then, CO2 gas was injected at a rate of 10 μL / min under an outlet back pressure of 10 MPa.
[0089] During the experiment, a fluorescence microscope system was used to acquire fluorescence images in real time at a frequency of 30 s / frame, forming a fluorescence image sequence to fully record the dynamic pH change process caused by CO2 dissolution.
[0090] Step 3: Preprocessing of fluorescence image sequences
[0091] This step is the core of the present invention to achieve stable and quantitative inversion, aiming to extract reliable relative fluorescence intensity signals that reflect chemical changes from the original image.
[0092] 1. Generate a binarized mask image
[0093] (1) In the early stage of CO2 percolation test, select several fluorescent images in the fluorescent image sequence when the background sodium fluorescein solution fills the microchannel, and screen out the fluorescent image with the best contrast between the microchannel (fluorescent area) and the background (non-fluorescent area).
[0094] Using the OpenCV image processing library on the Python platform, the Otsu automatic thresholding method based on grayscale histogram was employed to binarize the fluorescence image and preliminarily segment the microchannel region in the fluorescence image.
[0095] (2) Apply morphological closing operation (dilation followed by erosion) to eliminate minute noise, smooth the microchannel boundary, and finally generate a binary mask image that accurately covers all microchannel regions.
[0096] like Figure 4 The binary mask image shown has white areas representing microchannels and black areas representing the background.
[0097] This binarized mask image is saved and applied to the entire fluorescence image sequence to ensure that only pixels within the microchannels are analyzed.
[0098] 2. Fluorescence intensity normalization correction based on pH=7.0 benchmark
[0099] This step aims to eliminate overall fluorescence intensity variations in the fluorescence image caused by factors such as light source fluctuations and camera drift during the experiment, and to normalize the corrected fluorescence intensity to a relative fluorescence intensity based on the relative fluorescence intensity at pH=7.0. The specific procedure is as follows:
[0100] (1) Obtain the reference fluorescence intensity of the initial fluorescence image
[0101] Under calibration experimental conditions, the original fluorescence intensity I within the masked region of each fluorescence image in the fluorescence image sequence was extracted. raw,exp ;
[0102] The fluorescence image (usually the first frame) was acquired at the initial time of the experiment (t=0), when the entire microchannel was filled with sodium fluorescein solution at an initial pH of 7.0. The average fluorescence intensity of this frame within the masked region (i.e., the entire microchannel) was calculated and denoted as I. 7,exp .
[0103] I7,exp This represents the average fluorescence intensity actually measured under the initial conditions of this experiment at pH=7.0.
[0104] (2) Determine the reference standard for average fluorescence intensity
[0105] The average absolute fluorescence intensity of the sodium fluorescein solution with pH=7.0 measured under the calibration test conditions is denoted as I. 7,calib This serves as a theoretical benchmark for the average fluorescence intensity at pH 7.0.
[0106] (3) Calculate the global fluorescence intensity correction coefficient
[0107] According to the formula k=I 7,calib / I 7,exp A fixed global fluorescence intensity correction coefficient k is calculated.
[0108] k is used to set the initial strength benchmark I for this CO2 seepage test. 7,exp Corrected to the fluorescence intensity reference I of the calibration test. 7,calib level.
[0109] (4) Fluorescence intensity correction was performed on all test images.
[0110] According to Formula I corr,exp =k·I raw,exp The original fluorescence intensity I within each fluorescence image mask region in the entire fluorescence image sequence. raw,exp Multiply by the global fluorescence intensity correction factor k to obtain the corrected average fluorescence intensity I. corr,exp .
[0111] (5) Fluorescence intensity normalization to relative fluorescence intensity
[0112] To match and invert the established pH-relative fluorescence intensity calibration curve, the corrected absolute fluorescence intensity I needs to be... corr,exp Converted to relative fluorescence intensity I relative to pH=7.0 baseline rel,exp The calculation formula is I rel,exp =I corr,exp / I 7,calib .
[0113] After this normalization process, the relative fluorescence intensity I of all pixels within the entire microchannel in the initial frame (t=0) is... rel,exp It is always equal to 1, precisely corresponding to the reference point of pH=7.0 in the pH-relative fluorescence intensity calibration curve.
[0114] like Figure 5As shown, after correction and normalization, the relative fluorescence intensity within the microchannel at the initial experimental moment (t=0) stabilized at 1, precisely corresponding to the pH=7.0 baseline state. With continuous CO2 injection and dissolution diffusion, the pH of the solution within the microchannel gradually decreased, and the relative fluorescence intensity showed a regular decrease, exhibiting a continuous, stable, and unaffected overall trend. This indicates that the fluorescence intensity correction and normalization method employed in this invention can effectively eliminate interference from light source fluctuations and system biases, providing reliable relative fluorescence intensity data for the subsequent accurate inversion of the pH and CO2 concentration fields.
[0115] Step 4: Convert to obtain pH field evolution dataset
[0116] The relative fluorescence intensity image sequence obtained in the above steps, which has been masked and normalized, is substituted into the high-precision pH-relative fluorescence intensity calibration curve established in step one.
[0117] For each pixel within the masked area, based on its corrected absolute fluorescence intensity (I... rel,exp The corresponding pH value is calculated using the inverse function of the pH-relative fluorescence intensity calibration curve.
[0118] This process is performed automatically frame by frame and pixel by pixel, ultimately generating a high-resolution pH field evolution dataset that corresponds spatiotemporally to the original image sequence.
[0119] Step 5: Calculate the spatiotemporal distribution of CO2 concentration
[0120] Based on the pH field obtained in step four above, the spatiotemporal distribution of CO2 concentration is calculated using the CO2 dissolution equilibrium relationship in water, such as... Figure 6 As shown. The specific theoretical basis is as follows:
[0121] CO2 first undergoes physical dissolution upon contact with water:
[0122] ;
[0123] Dissolved CO2 reacts with water to produce bicarbonate ions (HCO3-). 3- ) and release hydrogen ions (H + This process can be represented as:
[0124] ;
[0125] Since carbonic acid (H2CO3) exists in solution for a very short time and dissociates rapidly, its concentration is negligible. Therefore, the above reaction is regarded as the first equivalent dissociation step.
[0126] The generated HCO 3- Further dissociation is possible:
[0127] ;
[0128] The dissociation constants for the two dissociation steps described above are defined as follows:
[0129] ;
[0130] ;
[0131] K1 and K2 are the first and second dissociation constants, respectively, and their values are temperature-dependent and can be obtained by consulting thermodynamic databases.
[0132] Based on the principle of solution electroneutrality and neglecting the minor effects of other ions, the concentration of dissolved CO2 ([CO2(aq)]) and the concentration of hydrogen ions ([H2)]) can be derived. + The relationship between ])
[0133] ;
[0134] Among them, K w is the ion product constant of water.
[0135] Since pH = -log[H] + After obtaining the pH field evolution dataset as described in step four, it can be obtained through the above-mentioned [CO2(aq)] and [H + The relationship between the two pixels is used to directly calculate the dissolved CO2 concentration for each pixel, thereby generating a spatiotemporal distribution map of CO2 at the pore scale of the microfluidic chip, achieving precise quantification of the CO2 dissolution and diffusion process.
[0136] The above detailed embodiments describe the implementation of the present invention; however, the present invention is not limited to the specific details described in the above embodiments. Within the scope of the claims and technical concept of the present invention, various simple modifications and changes can be made to the technical solution of the present invention, and these simple modifications all fall within the protection scope of the present invention.
Claims
1. A method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging, characterized in that, Includes the following steps: Based on several sodium fluorescein solutions with different pH values, the original average fluorescence intensity of the corresponding fluorescence images was obtained. The original average fluorescence intensity was normalized to obtain the relative fluorescence intensity, and a pH-relative fluorescence intensity calibration curve was established by fitting. CO2 percolation experiments were conducted in a microfluidic chip, and fluorescence images were acquired at predetermined time intervals to form a fluorescence image sequence. The fluorescence image with the best contrast between the microchannel and the background is selected from the fluorescence image sequence, and then binarized and denoised to obtain a binarized mask image of the microchannel covering the microfluidic chip. Based on the binarized mask image, the relative fluorescence intensity of each fluorescence image in the fluorescence image sequence is calculated to form a relative fluorescence intensity sequence; Using the pH-relative fluorescence intensity calibration curve, the relative fluorescence intensity sequence was converted into a pH field evolution dataset; Based on the pH field evolution dataset and the CO2 dissolution equilibrium relationship in water, the spatiotemporal distribution of CO2 concentration at the pore scale of the microfluidic chip was calculated.
2. The method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging according to claim 1, characterized in that, The method for normalization and fitting to establish the pH-relative fluorescence intensity calibration curve is as follows: The average fluorescence intensity I of the fluorescence image corresponding to the sodium fluorescein solution at pH 7.0 was selected. 7,calib As a normalization benchmark, according to Formula I rel,calib =I raw,calib / I 7,calib Calculate the relative fluorescence intensity I of the fluorescence images corresponding to sodium fluorescein solutions at different pH values. rel,calib I raw,calib The original average fluorescence intensity of the fluorescence images corresponding to sodium fluorescein solutions at different pH values; Using the measured pH value of each of the sodium fluorescein solutions as the independent variable, the corresponding relative fluorescence intensity I rel,calib Using the variable as the dependent variable, a nonlinear regression fitting was performed to obtain the candidate pH-relative fluorescence intensity calibration curve; The parameters of the candidate pH-relative fluorescence intensity calibration curves were optimized to obtain pH-relative fluorescence intensity calibration curves that meet the requirements.
3. The method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging according to claim 2, characterized in that, A four-parameter Logistic function was used for nonlinear regression analysis and fitting.
4. The method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging according to claim 2, characterized in that, The method for optimizing the parameters of the candidate pH-relative fluorescence intensity calibration curve is as follows: the parameters are optimized using the least squares method, and the fitting quality is evaluated by calculating the coefficient of determination.
5. The method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging according to claim 2, characterized in that, The mathematical expression for the pH-relative fluorescence intensity calibration curve is: ； Where I is the relative fluorescence intensity, a and d represent the lower and upper limits of fluorescence intensity, respectively, b is the Hill coefficient, and c is the pH value corresponding to the inflection point of the curve.
6. The method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging according to claim 1, characterized in that, During the CO2 seepage test, the predetermined time interval is 30 s / frame.
7. The method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging according to claim 1, characterized in that, The method for selecting the fluorescence image with the best contrast between the microchannel and the background from the fluorescence image sequence, performing binarization and noise reduction processing, and obtaining the binarized mask image covering the microchannel of the microfluidic chip is as follows: Several fluorescence images were selected from the fluorescence image sequence when the background solution filled the microchannels, and the fluorescence image with the best contrast between the microchannels and the background was selected. The Otsu automatic thresholding method based on grayscale histogram was used to binarize the fluorescence image and segment the microchannels in the fluorescence image. A morphological closing operation method of first dilation and then erosion is used to eliminate noise, smooth the region boundaries of the microchannels, and generate a binary mask image that accurately covers all the microchannels.
8. The method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging according to claim 1, characterized in that, Based on the binarized mask image, the relative fluorescence intensity of each fluorescence image in the fluorescence image sequence is calculated, and the method for forming the relative fluorescence intensity sequence is as follows: Under calibration test conditions, the fluorescence image of the initial frame was selected when the entire microchannel was filled with sodium fluorescein solution at pH=7.0 at the initial moment of the CO2 percolation test, and the average fluorescence intensity I within the masked area was obtained. 7,exp ; The average absolute fluorescence intensity I of sodium fluorescein solution at pH 7.0 7,calib As a reference standard, according to the formula k=I 7,calib / I 7,exp The global fluorescence intensity correction coefficient k was calculated. According to Formula I corr,exp =k·I raw,exp The corrected average fluorescence intensity I was calculated. corr,exp I raw,exp The original average fluorescence intensity within each fluorescence image mask region in the fluorescence image sequence during the CO2 infiltration test; According to Formula I rel,exp =I corr,exp / I 7,calib The corrected average fluorescence intensity I corr,exp Converted to relative fluorescence intensity I based on pH=7.0 rel,exp This forms a corresponding relative fluorescence intensity sequence.
9. The method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging according to claim 1, characterized in that, The method for converting the relative fluorescence intensity sequence into a pH field evolution dataset using the aforementioned pH-relative fluorescence intensity calibration curve is as follows: Each corrected relative fluorescence intensity in the relative fluorescence intensity sequence is substituted into a pH-relative fluorescence intensity calibration curve that meets the requirements, and the corresponding pH value is calculated to form a pH field evolution dataset.
10. The method for dynamic monitoring of CO2 dissolution and diffusion based on microfluidic fluorescence imaging according to claim 1, characterized in that, Based on the pH field evolution dataset and the CO2 dissolution equilibrium relationship in water, the method for calculating the spatiotemporal distribution of CO2 concentration at the pore scale of the microfluidic chip is as follows: Using the pH field evolution dataset, the dissolved CO2 concentration [CO2(aq)] corresponding to each pixel is calculated according to the following formula; ； ； ； Where K1 and K2 are the first and second dissociation constants, respectively, K w Let H be the ion product constant of water, [H] + (aq)] Dissolved H for each pixel + concentration.