Method and system for analyzing and detecting form of heavy metal in soil remediation process

Abstract

The invention relates to the technical field of soil remediation detection, and discloses a heavy metal form analysis and detection method and system in a soil remediation process. According to the method, a soil sample to be detected is collected and pretreated to obtain homogenized soil particles, the homogenized soil particles are placed in a continuous extraction device, different chemical reagents are added according to a preset program, and the dissolution concentration of heavy metal ions is monitored in real time to generate a dynamic dissolution curve. Dividing five form intervals of an exchangeable state, a carbonate binding state and the like based on a curve, and calculating form proportion distribution according to the dissolution peak area of each interval. A heavy metal form migration model is established, proportion distribution is matched with a soil remediation stage parameter time sequence, each form conversion path is output, the remediation agent adding amount and the reaction time are adjusted accordingly, adjustment parameters are fed back to a remediation equipment control terminal, and the soil remediation efficiency and accuracy can be effectively improved; and reliable technical support is provided for soil remediation process management and control.

Description

Technical Field

[0001] This invention relates to the field of soil remediation testing technology, specifically to a method and system for analyzing and detecting heavy metal speciation in the soil remediation process. Background Technology

[0002] Heavy metal pollution control is a common task in current soil remediation operations. The forms in which heavy metals exist in soil directly affect their mobility and bioavailability, and different forms of heavy metals have significantly different impacts on the ecological environment and human health. Existing methods for analyzing and detecting heavy metal forms during soil remediation mostly employ an offline sampling followed by laboratory analysis. This method requires collecting samples at the soil remediation site, transporting them to the laboratory, undergoing a series of complex pretreatment procedures, and then using detection instruments for speciation analysis. The entire process is time-consuming and makes it difficult to reflect the dynamic changes in heavy metal forms during soil remediation in real time. This results in remediation operators being unable to promptly grasp the transformation patterns of heavy metal forms during the remediation process, and can only adjust remediation parameters based on experience or pre-set plans, easily leading to problems such as unreasonable dosage of remediation agents and improper control of reaction time. Existing analytical methods for classifying heavy metal speciation often suffer from inconsistent classification standards and ambiguous ranges. Some methods can only distinguish a few heavy metal speciations, failing to comprehensively cover the five key speciations: exchangeable, carbonate-bound, iron-manganese oxide-bound, organic matter-bound, and residual. This results in an incomplete and inaccurate assessment of heavy metal migration characteristics in soil. Furthermore, existing methods lack effective means to correlate heavy metal speciation analysis results with soil remediation stage parameters, making it difficult to establish a direct link between heavy metal speciation changes and remediation operations. This prevents dynamic adjustments to remediation strategies based on real-time changes in heavy metal speciation, potentially leading to low remediation efficiency, or even substandard or over-remediation, increasing the cost and time of soil remediation. In actual soil remediation projects, the physicochemical properties of the soil change at different remediation stages, and the forms of heavy metals also change accordingly. Existing detection methods cannot capture these dynamic changes in real time, leaving the remediation process in a relatively passive state and hindering precise and efficient soil remediation operations. Furthermore, some detection methods are cumbersome in data processing and analysis, require high levels of professional skills from operators, and have complex operating procedures, which is not conducive to their widespread application in actual remediation projects and cannot meet the needs of large-scale soil remediation operations for rapid and accurate detection of heavy metal forms. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for heavy metal speciation analysis and detection in soil remediation processes, so as to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides a method for heavy metal speciation analysis and detection in soil remediation processes, the method comprising: Soil samples were collected and pretreated to obtain homogenized soil particles. The homogenized soil particles are placed in a continuous extraction device, and different chemical reagents are added sequentially according to a preset program; The concentration of heavy metal ions leached in the continuous extraction device is monitored in real time to generate a dynamic leaching curve; Based on the dynamic leaching curve, heavy metals are divided into five state ranges: exchangeable state, carbonate-bound state, iron-manganese oxide-bound state, organic matter-bound state, and residue state. The proportion of heavy metals in each form is calculated based on the dissolution peak area corresponding to each form range. A heavy metal speciation migration model was established, and the aforementioned proportion distribution was time-series matched with soil remediation stage parameters. The transformation paths of each form of heavy metal during the repair process are output through the heavy metal morphology migration model. Adjust the dosage of the repair agent and the reaction time parameters according to the described conversion pathway; The adjusted parameters are fed back to the control terminal of the soil remediation equipment.

[0005] Preferably, the preprocessing includes: The collected soil samples were air-dried under constant temperature conditions until the moisture content was less than 5%. The air-dried soil was mechanically sieved using a nylon screen to retain particles with a diameter of less than two millimeters. The sieved particles were divided into three samples using a quartering method to obtain homogenized soil particles.

[0006] Preferably, the preset program includes: In the first stage, magnesium chloride solution was used as the extractant, and the mixture was shaken at 25 degrees Celsius for two hours. In the second stage, the pH was adjusted to five using sodium acetate buffer solution, and the mixture was shaken at the same temperature for four hours. The third stage involves reacting a hydroxylamine hydrochloride solution in an 85-degree Celsius water bath for six hours. The fourth stage involves heating the solution of hydrogen peroxide and nitric acid to 95 degrees Celsius in two stages. The fifth stage involves microwave digestion using a mixture of hydrofluoric acid and perchloric acid.

[0007] Preferably, the generation of the dynamic dissolution curve includes: The concentration of heavy metals in the extract was collected at a frequency of once per minute using inductively coupled plasma mass spectrometry. Perform three repeated measurements on the concentration data at each sampling point and take the arithmetic mean; The arithmetic mean was fitted to a Gaussian distribution curve over time, and the dissolution time points corresponding to each peak were marked.

[0008] Preferably, the division of the morphological intervals includes: The Gaussian distribution curve is identified as having a commutative state if the first peak occurs in the interval between 0 and 30 minutes. The second peak occurring within the 30-minute to 3-hour timeframe was identified as a carbonate-bound state. The broad peaks appearing in the three to eight hour interval were identified as iron-manganese oxide bound states. The multi-peak superposition region appearing in the eight to twenty hour interval was identified as an organic matter bound state; The total amount of heavy metals detected during the digestion stage is subtracted from the sum of the first four forms to obtain the residue state.

[0009] Preferably, the establishment of the heavy metal morphological migration model includes: Construct a three-dimensional matrix with the number of repair days as the horizontal axis and the proportion of each form as the vertical axis; Mark the time nodes of the repair agent application events in the three-dimensional matrix; The correlation coefficients between the proportion of each morphology and the amount of repair agent added were calculated using partial least squares method. The dominant migration path is defined based on the correlation coefficient being greater than 0.8.

[0010] Preferably, the generation of the conversion path includes: When the dominant migration pathway shows a conversion from carbonate-bound to exchangeable state, it is determined to be a form release caused by a decrease in pH value; When the bound states of iron and manganese oxides and organic matter decrease simultaneously, it is determined to be desorption caused by an increase in redox potential; The determination result is cross-validated with the real-time operating data of the repair equipment.

[0011] Preferably, the adjustment of the dosage of the repair agent includes: When the proportion of exchangeable states exceeds the threshold of 30%, the amount of passivating agent added is increased by 20%. When the proportion of residue is less than 5% for three consecutive days, extend the running time of the electric repair equipment by one hour. The cycle is dynamically updated and adjusted based on the morphological transformation rate in the transformation path.

[0012] Preferably, the parameter feedback of the control terminal includes: The instruction to increase the amount of passivating agent added is converted into a pulse delivery control signal; The instruction to extend the running time of the electric repair equipment is superimposed on the original timing program of the equipment. The original timing program of the equipment refers to the timing program that comes with the equipment at the factory or is set in the early stage of debugging to control the daily basic running time of the electric repair equipment. The monitoring cycle of the dynamic dissolution curve is restarted after the feedback parameters take effect.

[0013] Preferably, the present invention also includes a heavy metal speciation analysis and detection system for soil remediation processes, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor, when executing the computer program, implements the steps of the aforementioned method for heavy metal speciation analysis and detection in soil remediation processes.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This method for heavy metal speciation analysis in soil remediation involves collecting and pre-treating soil samples to obtain homogenized soil particles. This ensures the consistency and representativeness of subsequent samples, avoiding deviations in test results caused by sample inhomogeneity and making subsequent heavy metal speciation analysis more reliable. The homogenized soil particles are placed in a continuous extraction device, and different chemical reagents are added sequentially according to a preset program, enabling continuous extraction of heavy metals from the soil. This eliminates the need for multiple sample transfers, reducing sample loss and contamination during processing. Furthermore, the application of the preset program ensures the standardization and stability of the extraction process, allowing different batches and different operators to follow a unified standard, thus improving the repeatability of the detection process. Real-time monitoring of the dissolution concentration of heavy metal ions in the continuous extraction device and the generation of dynamic dissolution curves enable real-time capture of changes in heavy metal dissolution concentration during soil remediation. This provides a clear view of the dissolution patterns of heavy metals at different extraction stages, overcoming the limitations of traditional offline detection methods that cannot acquire data in real time. It allows remediation operators to understand the dynamic trends of heavy metal speciation, providing real-time data references for subsequent speciation classification and parameter adjustments. Based on the dynamic dissolution curves, five speciation ranges of heavy metals are defined, clearly distinguishing between exchangeable, carbonate-bound, iron-manganese oxide-bound, organic matter-bound, and residual states. This comprehensive coverage of the main speciations of heavy metals in soil makes the analysis of heavy metal speciation more comprehensive and allows for a more accurate understanding of the existence and migration characteristics of heavy metals in soil. The proportion of each heavy metal form was calculated based on the leaching peak area corresponding to each form range. This quantitative approach presents the proportion of different heavy metal forms in the soil, allowing operators to clearly understand the relative content of each form and providing concrete data support for subsequent analysis of heavy metal migration and transformation. A heavy metal form migration model was established, and the proportion distribution was time-series matched with soil remediation stage parameters. This achieved an organic combination of heavy metal form changes and remediation stages, clearly demonstrating the changing patterns of heavy metal forms at different remediation stages. This clarifies the impact of remediation operations on heavy metal forms and provides a basis for adjusting subsequent remediation strategies. By using a heavy metal speciation migration model to output the transformation pathways of various heavy metal speciations during the remediation process, the transformation direction and process of heavy metals can be intuitively presented. This allows operators to clearly understand how heavy metals transform from one speciation to another and what remediation factors influence this transformation. Adjusting the dosage of remediation agents and reaction time parameters based on the transformation pathways enables precise adjustments to be made according to the transformation status of heavy metal speciation, avoiding blind adjustments to remediation parameters and making the remediation operation more targeted. This effectively promotes the transformation of heavy metals to speciations with less environmental harm.

[0015] The adjusted parameters are fed back to the control terminal of the soil remediation equipment, realizing real-time linkage between the test results and equipment control. This eliminates the need for manual adjustment of equipment parameters, reduces human error, and improves the timeliness and accuracy of parameter adjustments. It ensures that the remediation equipment can operate according to optimal parameters, thereby promoting the orderly and efficient implementation of soil remediation operations. This meets the needs of precise and efficient remediation in actual soil remediation projects. At the same time, the method has a standardized operation process and does not require overly complicated operation steps, making it easy to promote and apply in soil remediation projects of different scales. Attached Figure Description

[0016] Figure 1 This is a schematic diagram illustrating the working principle of the heavy metal speciation analysis and detection method in the soil remediation process described in this invention. Figure 2 A schematic diagram illustrating the working principle of soil sample pretreatment; Figure 3 A schematic diagram illustrating the working principle of dynamic dissolution curve generation. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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.

[0018] Please see Figure 1 This invention provides a method and system for heavy metal speciation analysis in soil remediation processes. The method includes continuous extraction technology combined with real-time monitoring to dynamically track heavy metal speciation transformation and optimize remediation process parameters accordingly. The specific implementation of this method includes sample pretreatment, continuous extraction, real-time monitoring, speciation classification, model establishment, and parameter feedback. After collecting soil samples, pretreatment is first performed to obtain homogenized soil particles, ensuring the representativeness and repeatability of the samples. The homogenized soil particles are placed in a dedicated continuous extraction device, and extraction reagents with different chemical properties are added sequentially according to a preset time sequence. The reagent selection is designed for different binding speciations of heavy metals. During the extraction process, the concentration changes of heavy metal ions in the solution are monitored in real time using a high-precision sensor, generating a time-concentration dynamic curve. Based on the characteristic peaks of the dynamic dissolution curve, five speciation ranges of heavy metals are identified and classified, including exchangeable, carbonate-bound, iron-manganese oxide-bound, organic-bound, and residual states. The proportion of each speciation is calculated by integrating the corresponding peak area. Subsequently, a heavy metal speciation migration model was established, matching speciation percentage data with time-series parameters during the remediation process (such as remediation agent dosage time and equipment operating status) to analyze the transformation paths between speciations. The model output guides remediation operations, such as adjusting the passivating agent dosage or extending the reaction time, and automatically feeds the optimized parameters back to the control terminal of the remediation equipment to achieve closed-loop control. The entire method emphasizes real-time performance and adaptability, effectively improving remediation efficiency and reducing the risk of secondary pollution.

[0019] Example 1: See Figure 2 The pretreatment process was carried out in a constant-temperature laboratory, with the ambient temperature maintained within the range of 20±2 degrees Celsius. This was to minimize the potential interference of temperature fluctuations on the stability of heavy metal speciation in the soil. The collected raw soil samples were evenly spread in clean enamel dishes, with the thickness strictly controlled to less than 2 cm to ensure uniform moisture evaporation. The samples were placed in a well-ventilated environment without direct sunlight for natural air drying. During this period, the samples were turned over every 12 hours using a plastic shovel to ensure uniform drying throughout the soil. The drying process was determined by gravimetric analysis: two consecutive weighings were performed at 24-hour intervals. A change in sample mass of less than 0.1% was considered sufficient to indicate that the moisture content had dropped below 5%, meeting the predetermined standard.

[0020] After air-drying, the soil samples underwent mechanical sieving. A nylon screen was used, its chemical inertness effectively preventing metal contamination. A 2mm mesh size was selected, and the sieving was performed on a standard vibrating sieve machine, which oscillated at a fixed frequency of 200 times per minute for 10 minutes. After oscillation, impurities such as stones and plant roots remaining on the screen were discarded, and fine particles smaller than 2mm that passed through the screen were collected for subsequent experiments. This step effectively removed heterogeneous components and improved sample homogeneity. After obtaining the sieved particles, the classic quartering method was used to further ensure sample representativeness. Specifically, all sieved soil particles were piled into a cone, then gently flattened from the apex to the bottom with a pressure plate, creating a uniform circular layer. A cross-shaped sample divider was then used to precisely divide the layer into four equal parts, combining the two diagonally opposite parts and discarding the remaining two. The merged samples were mixed again, and the process of stacking, flattening, dividing, and sampling was repeated three times. After three quartering processes, the soil particles obtained were highly homogeneous, with minimal spatial differences in their chemical composition, meeting the stringent requirements for sample homogenization in continuous extraction experiments.

[0021] The pre-programmed continuous extraction procedure was executed in an automated continuous extraction unit equipped with multi-channel reagent addition and precise temperature control. The extraction procedure consisted of five clearly defined stages, each with strictly set reagent, temperature, and time parameters. The first stage aimed to extract exchangeable heavy metals, using a 1.0 mol / L magnesium chloride solution as the extractant, with the soil sample to extractant mass-to-volume ratio controlled at 1:20. The mixed suspension was transferred to a thermostatic shaker and continuously shaken at 200 rpm for 2 hours at a constant temperature of 25°C. The relatively mild conditions in this stage aimed to desorb heavy metal ions adsorbed on the surface of soil particles by electrostatic attraction. The second stage targeted carbonate-bound heavy metals, replacing the extractant with a 1.0 mol / L sodium acetate buffer solution adjusted to pH 5.0 with acetic acid. This stage was also conducted at 25°C, but the shaking time was extended to 4 hours. The slightly acidic environment simulated the dissolution process of carbonates under natural conditions when the pH slightly decreased, thereby releasing the bound heavy metals. The third stage aims to dissociate the bound state of iron-manganese oxides. This stage uses a 0.1 mol / L hydroxylamine hydrochloride solution as the extractant. The reaction is carried out in a constant-temperature water bath at 85°C with continuous magnetic stirring, and the reaction time is set to 6 hours. The combined effect of the high temperature and reducing environment effectively releases the heavy metals encapsulated or co-precipitated in the iron-manganese oxides.

[0022] The fourth stage treats heavy metals bound to organic matter using a mixture of hydrogen peroxide and concentrated nitric acid at a volume ratio of 2:1. The operation consists of two steps: first, a portion of the mixture is added at room temperature and reacted for one hour to initially oxidize the organic matter; then, the remaining mixture is added, and the mixture is heated to 95 degrees Celsius and reacted for another two hours to completely decompose the organic matter through strong oxidation, releasing the fixed heavy metals. The fifth stage targets residual heavy metals present in the crystal lattice of primary or secondary minerals. This stage uses a mixture of hydrofluoric acid and perchloric acid at a volume ratio of 3:1, conducted in a high-pressure, sealed microwave digestion vessel. The digestion program is set to heat to 180 degrees Celsius and maintain for 30 minutes. Utilizing the strong corrosiveness of the mixed acid and the high temperature and pressure conditions, stable minerals such as silicates are completely decomposed, allowing for the determination of the total heavy metal content. After each extraction stage, the reaction mixture is centrifuged at high speed to achieve solid-liquid separation. The supernatant is collected for testing, and the residual solid precipitate is washed with an appropriate amount of deionized water and centrifuged before entering the next extraction stage. This ensures the specificity and completeness of the extraction of each form.

[0023] Example 2: See Figure 3 The acquisition of dynamic dissolution curves relies on a high-temporal-resolution online monitoring system. This system uses an inductively coupled plasma mass spectrometer (ICP-MS) as its core detection unit, which possesses extremely high sensitivity and the ability to simultaneously detect multiple elements. After the extract flows out of the continuous extraction device, it is introduced through a precisely designed flow injection analysis module. This module is equipped with a quantitative sample loop and a continuously flowing carrier, which can quantitatively deliver the test droplets into the ICP-MS nebulizer at a stable flow rate of 1 mL per minute. The instrument method has been optimized, and the optimal signal-to-noise ratio parameters have been set for the target heavy metal elements (such as lead, cadmium, chromium, copper, and zinc). The data acquisition frequency is fixed at once per minute, that is, at the end of each minute in each extraction stage, the instrument automatically completes a spectral scan and records the characteristic ion intensity values ​​of each element. To minimize analytical errors, measurements at each time point are not single readings, but rather three consecutive, repeated acquisitions automatically executed by the instrument's internal control software. The three ion intensity values ​​are first converted into corresponding mass concentration values ​​based on a pre-established multi-point calibration curve. Then, the arithmetic mean of these three concentration values ​​is immediately calculated, and this average is used as the final heavy metal concentration value for that specific minute. This workflow, from sample introduction and signal acquisition to data preprocessing, runs throughout the entire continuous extraction process, starting from the zero point without added extractant and continuing until the completion of the fifth stage of microwave digestion and detection. This generates a complete time-series concentration dataset that clearly depicts the dynamic trajectory of heavy metal leaching under the influence of different chemical reagents.

[0024] After obtaining the raw time-concentration data sequence, mathematical processing is required to generate a smooth, dynamic dissolution curve suitable for peak analysis. Data processing is performed using specialized chemometrics software. Operators plot all data points on a scatter plot, using time points as independent variables (x-axis) and the corresponding average heavy metal concentration as the dependent variable (y-axis). Due to the kinetic characteristics of the extraction process, the data points typically exhibit a peak-shaped trend of first rising and then falling, but some random fluctuations may exist. To reveal its inherent patterns, a nonlinear curve fitting algorithm based on the Gaussian function is used to fit the scatter plot. The Gaussian function can well describe the natural distribution of concentrations in many physicochemical processes. The software iteratively calculates and adjusts the parameters of the Gaussian function (such as peak height, peak position, and peak width) to minimize the sum of squared residuals between the fitted curve and all measured data points. A goodness of fit (R²) greater than 0.99 is typically required to confirm that the curve highly reproduces the actual dissolution process. After successful fitting, a continuous and smooth dynamic dissolution curve is generated.

[0025] The software uses numerical calculation methods to calculate the first derivative of the fitted curve. The points where the first derivative equals zero correspond to the peak points (maximum points) on the curve. The x-axis value corresponding to each peak point represents the time elapsed before the leaching concentration of that heavy metal form reaches its maximum. These key time points are precisely marked and recorded as important criteria for dividing the form intervals. The peak height reflects the maximum leaching concentration of that heavy metal form, while the peak width reflects, to some extent, the dissolution rate and the heterogeneity of binding strength. The division of form intervals follows the generation of the dynamic leaching curve, and its core lies in establishing a scientific correspondence between the characteristic peaks appearing in different time intervals on the curve and the specific chemical forms of heavy metals in the soil. The division is mainly based on the chemical extraction characteristics of each form and its dissolution order on the time axis. Exchangeable heavy metals, adsorbed onto soil particle surfaces only through weak electrostatic attraction, are most easily and rapidly desorbed upon contact with neutral salt extractants (such as magnesium chloride solution). Therefore, they appear as the first significant peak on the dynamic dissolution curve. This peak is typically very sharp, indicating a rapid dissolution rate, and its occurrence is defined within 0 to 30 minutes after the start of extraction. Following this, carbonate-bound heavy metals dissolve. These heavy metals are associated with carbonate precipitates or co-precipitates in the soil. When the pH of the extraction system slightly decreases to a weakly acidic level (such as in a sodium acetate buffer environment), the carbonates begin to dissolve, releasing the metals. This is represented by the second peak on the curve, typically occurring between 30 minutes and 3 hours later. The peak shape is slightly wider than that of the exchangeable form, indicating that its release requires a certain reaction time.

[0026] Heavy metals bound to iron and manganese oxides refer to those encapsulated within or within the lattice of amorphous iron and manganese oxides. Their release requires stronger reducing conditions and longer reaction times (such as the action of hydroxylamine hydrochloride solution at higher temperatures). Therefore, on the dynamic dissolution curve, it exhibits a broad peak shape appearing in the 3-8 hour range. The peak may not be particularly prominent, but its duration is long, forming a distinct "bulge" region, reflecting the slow and complex nature of the oxide reduction and dissolution process. The release of heavy metals bound to organic matter depends on the oxidative decomposition of organic matter (such as the action of hydrogen peroxide / nitric acid mixture). Due to the complex composition of soil organic matter and the varying degrees of oxidative difficulty, the release of heavy metals is often not synchronous. Therefore, on the curve, corresponding to the 8-20 hour range, a complex peak group formed by the superposition of multiple small peaks or a broad peak with an asymmetric "shoulder" is frequently observed, reflecting the stepwise release characteristics of organically bound heavy metals. For residual heavy metals, which exist in the stable lattice of primary or secondary minerals such as silicates, they are not extracted by the reagents in the first four steps of a conventional continuous extraction process. They are only completely released in the final strong acid digestion stage (hydrofluoric acid / perchloric acid mixture, microwave digestion). Therefore, the residual state does not directly appear as an independent dissolution peak on the dynamic dissolution curve. The content of these residual heavy metals needs to be determined indirectly by measuring the total amount of heavy metals in the solution after the fifth stage of digestion, and then subtracting the first four form ranges (i.e., exchangeable state, carbonate-bound state, iron-manganese oxide-bound state, and organic matter-bound state) and integrating the sum of the heavy metal contents calculated from the corresponding dissolution peak areas. The quantification of each form is achieved by integrating the peak area of ​​the dynamic dissolution curve under its corresponding range. The area value is automatically calculated by software. Finally, the content of all forms is normalized to obtain the percentage distribution of each form of heavy metal in the soil.

[0027] Example 3: The starting point for model construction is the systematic organization of data. Its core structure is a three-dimensional data matrix, whose abstract space is defined by three dimensions. The first dimension is the timeline of the remediation process, marked in days, starting from day zero of the remediation project and continuing until the current analysis day, with each day as an independent time layer. The second dimension is the daily percentage data of five heavy metal forms (exchangeable, carbonate-bound, iron-manganese oxide-bound, organic matter-bound, and residual forms). These percentage data are derived from continuous extraction and speciation analysis of soil samples collected daily from representative sites at the remediation site. The third dimension is a set of operational parameters related to the remediation operation, including but not limited to the specific daily dosage of various remediation agents (such as passivators, oxidants, nutrients, etc.) (usually in kilograms per hectare), real-time pH value of the soil medium, redox potential (Eh value), soil moisture, and other environmental indicators. Each remediation day corresponds to a slice in the matrix, which contains the percentage of all forms and the values ​​of all operational parameters for that day.

[0028] In the three-dimensional matrix, key remediation events need to be specifically marked with their time points, such as the date and time when a specific type and dosage of remediation agent was added, or the time when the electric remediation equipment was started or the rinsing system was activated. These events are marked as specific markers in the matrix, typically associated with the base data in the form of event identifiers and timestamps. After completing the construction of the data matrix and event labeling, the next step is to analyze the quantitative relationship between the proportion of different heavy metal forms and the remediation operations. Here, partial least squares (PLS) regression is used for modeling. PLS regression is particularly suitable for handling situations where there is multicollinearity among independent variables (such as remediation agent dosage, pH value, and other operating parameters) and the sample size (number of days) may be less than the number of variables. The analysis process is performed separately for each heavy metal form. For example, the "proportion of exchangeable forms" is used as the dependent variable Y, and the daily "dosage of remediation agent A", "dosage of remediation agent B", "pH value", and "Eh value" are used as the independent variable matrix X. The PLS algorithm extracts latent variables (principal components) from X and Y, maximizing the explanation of Y's variation while taking into account its correlation with X, thereby establishing a linear regression model between X and Y.

[0029] For a well-established PLS model, it is necessary to evaluate the significance of the influence of each independent variable (operating condition parameter) on the dependent variable (morphological proportion). This is typically achieved by calculating the variable importance projection (VIP) value. The VIP value quantifies the contribution of each independent variable to explaining the variation in the dependent variable, and its calculation is based on the principal components extracted from the PLS model. The larger the VIP value of a given independent variable, the more important it is in explaining changes in morphological proportion. The formula for calculating the VIP value can be expressed as:

[0030] in: Representing the The variable importance projection (VIP) values ​​of each independent variable. This represents the total number of principal components extracted by the PLS regression model. It is the index in the summation formula, representing the index of the currently being calculated. Principal components, Representing the The sum of squares of the variation in the dependent variable (proportion of heavy metal forms) that each principal component can explain. Representing the The independent variable at the th independent variable in the th ... The square of the loadings on each principal component reflects the contribution weight of that independent variable to that principal component.

[0031] To simplify model output and focus on key issues, a threshold value (VIP) is set, for example, 1.0. When the VIP value of a condition parameter is greater than or equal to 1.0, the parameter is generally considered to have a significant impact on morphological changes. After the model runs, it may output, for example, a VIP value of 1.85 for "pH value" relative to "exchangeable state percentage," and a VIP value of 1.72 for "passivating agent dosage" relative to the same morphology. Both exceed the threshold, therefore their relationship with exchangeable states is identified as a key correlation. After identifying important variables, the model further examines the partial correlation coefficients between these important variables and morphological percentages. This coefficient reflects the strength and direction of the linear association between the two after excluding the influence of other variables. A threshold value for the correlation coefficient is set, for example, 0.8. When the absolute value of the partial correlation coefficient between an important condition parameter and a certain morphological percentage is greater than 0.8, the relationship is defined as a "dominant migration path." For example, the analysis results might show a significant negative correlation between "pH value" and "exchangeable state ratio" (partial correlation coefficient < -0.8), while "passivator dosage" and "exchangeable state ratio" also show a significant negative correlation (partial correlation coefficient < -0.8). In this case, "pH decrease → exchangeable state increase" and "passivator dosage → exchangeable state decrease" are identified as two dominant migration pathways acting on the same state but in opposite directions.

[0032] Based on the identified dominant migration pathways and the sign of their correlation coefficients, the model can provide qualitative interpretations. For example, when "pH value" and "exchangeable state ratio" show a significant negative correlation (i.e., a decrease in pH is accompanied by an increase in exchangeable state), the model can infer the chemical process that "soil acidification leads to the release of some bound heavy metals into more reactive exchangeable states." When "redox potential Eh" and "iron-manganese oxide bound state ratio" show a significant negative correlation (i.e., an increase in Eh is accompanied by a decrease in the ratio of this form), the model can infer that "enhanced oxidation conditions promote the dissolution of some iron-manganese oxides, releasing the heavy metals they bind." These inferences connect abstract mathematical correlations with specific soil chemical principles. Ultimately, the model's output is a semi-quantitative description of heavy metal speciation during remediation, systematically demonstrating which operational parameters are the main driving factors and which speciations they drive in which direction. This output can be represented as a directed network graph, where nodes are various speciation or operational parameters, edges represent dominant pathways with significant correlations, and VIP values, partial correlation coefficients, and inferred transformation mechanisms can be labeled on the edges.

[0033] Example 4: The generation of the transformation pathway is not a simple data mapping process, but a closed-loop analysis based on soil chemistry principles to infer and empirically verify the data relationships. The following example illustrates this with a specific data sequence. Assume that during a cycle of stabilization-based remediation of cadmium-contaminated soil, a heavy metal speciation migration model, using partial least squares analysis of data from the past seven days, identifies a dominant migration pathway: a strong positive correlation (correlation coefficient > 0.8) exists between the proportion of carbonate-bound cadmium and soil pH; that is, the proportion of carbonate-bound cadmium decreases as pH decreases. The model also detects a corresponding increase in the proportion of exchangeable cadmium during this process. Based on this statistical phenomenon, the system initially determines the existence of a migration pathway of "conversion from carbonate-bound to exchangeable cadmium," and speculates that its driving force is "pH decrease leading to carbonate dissolution and heavy metal release." However, simple statistical correlation is insufficient to confirm the chemical mechanism; cross-validation with real-time sensor data from the remediation site is necessary.

[0034] The soil pH sensors deployed on-site transmitted data hourly. The system retrieved pH records for the corresponding time period for speciation analysis. For example, it was found that the soil pH value decreased from an initial 7.2 to 6.5 and fluctuated at this low level the day before the proportion of carbonate-bound cadmium began to decline significantly, due to the infiltration of acidic precipitation or a slight overdose of an acidification remediation agent. This independent pH monitoring data highly matched the triggering condition (pH decrease) inferred by the model, thus strongly supporting the determination that "pH decrease is the main cause of this speciation transformation." See Table 1, which presents a summary of simulated time-series data used to support this determination process.

[0035] Table 1: Summary of Time-Series Data on Cadmium Speciation and Key Operating Parameters During the Restoration Process

[0036] The data in the table clearly shows that from day 2 to day 4, the soil pH decreased from 7.1 to 6.5. Simultaneously, the proportion of exchangeable cadmium increased from 13.1% to 18.9%, while the proportion of carbonate-bound cadmium decreased from 34.8% to 28.7%. This inverse trend is clearly visible, and the changes mainly occurred between these two forms among the first four forms, while the proportions of other forms remained relatively stable. This is consistent with the characteristics of the "carbonate-bound to exchangeable" pathway. Another common transformation pathway involves changes in redox conditions. The model may identify another dominant pathway: there is a negative correlation between the proportion of cadmium bound to iron and manganese oxides and the redox potential (Eh) (correlation coefficient < -0.8). The proportion of cadmium bound to organic matter also shows a similar negative correlation trend with Eh, meaning that both proportions decrease synchronously as Eh increases.

[0037] The system identifies this phenomenon as "desorption caused by the synergistic effect of the reducing dissolution of iron and manganese oxides and the oxidative decomposition of organic matter due to the increased redox potential." To verify this pathway, the system retrieves historical data from the on-site redox potential sensor. Assuming the data records show that on the 10th day of remediation, a slow-release oxygen agent was added to activate local microorganisms, causing the soil Eh to gradually increase from around 150mV and stabilize above 400mV in the following days. Speciation analysis data for the corresponding time period reveals that it was precisely after Eh exceeded 300mV that the proportions of bound iron and manganese oxides and bound organic matter began to decrease continuously and synchronously. The Eh sensor data and speciation show a high degree of consistency in both timing and trend. The generation of the transformation pathway requires further cross-validation with the real-time operating data of the remediation equipment to eliminate other interfering factors or confirm synergistic effects.

[0038] In the aforementioned case of pH decrease leading to carbonate-bound transformation, the system simultaneously queries irrigation records and mixing equipment operation logs. If the data shows a significant increase in soil mixing intensity during the pH decrease, the system needs further analysis to determine whether the pH decrease alone dominated the transformation, or whether intense mechanical mixing accelerated the contact and reaction of acidified soil particles, jointly promoting the transformation rate. Through multivariate time-series comparisons, the contribution of each operational parameter to the morphological transformation can be more accurately defined. Finally, effective transformation pathways, confirmed through cross-validation, are transformed into a directed network primitive. Each pathway contains several key elements: the starting morphological node, the ending morphological node (or intermediate morphology), the main operating parameters driving the pathway (such as pH, Eh), the pathway strength (based on correlation coefficients or magnitude of change), and the time window in which the pathway takes effect. This empirically validated pathway network dynamically depicts the evolution of the geochemical behavior of heavy metals during remediation.

[0039] Example 5: Assume a site using stabilization technology to remediate cadmium-contaminated soil, where the system is continuously monitoring. Daily speciation data shows that the exchangeable cadmium content in the soil has been steadily increasing over the past three days, reaching 28%, 31%, and 33% respectively. The system's preset threshold is that when the exchangeable cadmium content exceeds 30%, the risk of active migration of heavy metals in the soil is considered to be increased, requiring intervention. Therefore, upon entering the data for the third day, the system immediately triggers adjustment logic, generating an instruction to increase the dosage of the currently used phosphate passivating agent by 20%. The increase is determined based on a historical data model, which indicates that under these soil conditions, a 20% increase can typically suppress the exchangeable cadmium content back below the threshold while avoiding excessive dosage that could lead to wasted costs or secondary pollution. The instruction clearly specifies the type of remediation agent, the percentage increase, and the specific dosage.

[0040] Another adjustment scenario focuses on the final effect of remediation, namely the efficiency of heavy metal conversion to a stable residual state. The system monitored the changes in the proportion of residual cadmium over the past three days: 18.5% on the first day, 18.7% on the second day, and 18.8% on the third day. Although the trend was upward, the cumulative increase over the three consecutive days was less than 0.5%, far below the system's threshold condition of "a continuous increase of less than 5% for three consecutive days is considered a remediation stagnation." Once this condition was triggered, the system generated another instruction: extend the daily operating time of the electric remediation equipment by 1 hour. Extending the operating time aims to promote the conversion of deeper or more stable heavy metals to soluble or exchangeable states by enhancing the electric field, thereby providing substrates for subsequent stabilization reactions. The adjustment instruction also clearly specifies the equipment and the specific time unit for the extension. The key advanced function lies in the dynamic updating of the adjustment cycle, which gives the system a certain degree of self-adaptability. The adjustment cycle is not fixed but dynamically calculated based on the rate of morphological transformation. The system periodically (e.g., every three days) calculates the conversion rates of recent major morphologies (e.g., the conversion from exchangeable to carbonate-bound, or from iron-manganese oxide-bound to residue). The conversion rates are calculated using linear regression analysis of the percentage data from the most recent five time points to obtain the slope value. A baseline conversion rate V0 is set within the system. If the calculated current actual conversion rate V is greater than 1.5 times V0, it indicates a vigorous repair reaction and rapid morphological change. The system will automatically shorten the adjustment cycle, for example, changing from weekly assessments to daily assessments, to respond more responsively to changes. Conversely, if V is less than 0.5 times V0, it indicates a slow process, and the adjustment cycle can be extended to once every two weeks to avoid unnecessary frequent disturbances.

[0041] The generated adjustment instructions need to be accurately converted into signals that the control terminal can recognize and execute. For instructions like "increase the passivating agent dosage by 20%", the control software encodes it into a set of pulse control signals. This set of signals contains two key parameters: pulse frequency and pulse width. The pulse frequency corresponds to the baseline flow rate of the remediation equipment (such as a metering pump), while the pulse width corresponds to the required increase in dosage. After the instruction is issued, the metering pump controller receives these pulse signals and drives the actuator to deliver the passivating agent at a faster pace or for a longer operating time, ensuring the precise implementation of the incremental dosage. This pulsed delivery method helps to evenly distribute the agent in the soil. For instructions like "extend the operating time of the electric remediation equipment by 1 hour", the system integrates it into the equipment's original control program using an instruction overlay method. The operation of the remediation equipment is usually controlled by a programmable logic controller (PLC) according to a preset schedule. The system does not directly modify the core timing program but instead overlays a delay instruction on top of it. This instruction takes effect at the equipment's original shutdown time, forcing the equipment to continue running for 60 minutes and automatically shutting down after the extension period expires. This approach reduces the impact on the stability of the original program and enables seamless runtime expansion. Once all feedback parameters take effect, the system does not stop working but immediately restarts a new monitoring cycle. This means that in the next sampling cycle after the command is issued, the system will collect soil samples again, perform a new round of continuous extraction and speciation analysis, generate new dynamic leaching curves, and reassess the proportion of each speciation. This new assessment result will be used to determine whether the previous adjustment command was effective.

[0042] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0043] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for analyzing and detecting heavy metal speciation in soil remediation processes, characterized in that, Includes the following steps: Soil samples were collected and pretreated to obtain homogenized soil particles. The homogenized soil particles are placed in a continuous extraction device, and different chemical reagents are added sequentially according to a preset program; The concentration of heavy metal ions leached in the continuous extraction device is monitored in real time to generate a dynamic leaching curve; Based on the dynamic leaching curve, heavy metals are divided into five state ranges: exchangeable state, carbonate-bound state, iron-manganese oxide-bound state, organic matter-bound state, and residue state. The proportion of heavy metals in each form is calculated based on the dissolution peak area corresponding to each form range. A heavy metal speciation migration model was established, and the aforementioned proportion distribution was time-series matched with soil remediation stage parameters. The transformation paths of each form of heavy metal during the repair process are output through the heavy metal morphology migration model. Adjust the dosage of the repair agent and the reaction time parameters according to the described conversion pathway; The adjusted parameters are fed back to the control terminal of the soil remediation equipment.

2. The method for heavy metal speciation analysis and detection in soil remediation process according to claim 1, characterized in that, The preprocessing includes: The collected soil samples were air-dried under constant temperature conditions until the moisture content was less than 5%. The air-dried soil was mechanically sieved using a nylon screen to retain particles with a diameter of less than two millimeters. The sieved particles were divided into three samples using a quartering method to obtain homogenized soil particles.

3. The method for heavy metal speciation analysis and detection in soil remediation process according to claim 1, characterized in that, The preset program includes: In the first stage, magnesium chloride solution was used as the extractant, and the mixture was shaken at 25 degrees Celsius for two hours. In the second stage, the pH was adjusted to five using sodium acetate buffer solution, and the mixture was shaken at the same temperature for four hours. The third stage involves reacting a hydroxylamine hydrochloride solution in an 85-degree Celsius water bath for six hours. The fourth stage involves heating the solution of hydrogen peroxide and nitric acid to 95 degrees Celsius in two stages. The fifth stage involves microwave digestion using a mixture of hydrofluoric acid and perchloric acid.

4. The method for heavy metal speciation analysis and detection in soil remediation process according to claim 1, characterized in that, The generation of the dynamic dissolution curve includes: The concentration of heavy metals in the extract was collected at a frequency of once per minute using inductively coupled plasma mass spectrometry. Perform three repeated measurements on the concentration data at each sampling point and take the arithmetic mean; The arithmetic mean was fitted to a Gaussian distribution curve over time, and the dissolution time points corresponding to each peak were marked.

5. The method for heavy metal speciation analysis and detection in soil remediation process according to claim 4, characterized in that, The division of the morphological intervals includes: The Gaussian distribution curve is identified as having a commutative state if the first peak occurs in the interval between 0 and 30 minutes. The second peak occurring within the 30-minute to 3-hour timeframe was identified as a carbonate-bound state. The broad peaks appearing in the three to eight hour interval were identified as iron-manganese oxide bound states. The multi-peak superposition region appearing in the eight to twenty hour interval was identified as an organic matter bound state; The total amount of heavy metals detected during the digestion stage is subtracted from the sum of the first four forms to obtain the residue state.

6. The method for heavy metal speciation analysis and detection in soil remediation process according to claim 1, characterized in that, The establishment of the heavy metal morphological migration model includes: Construct a three-dimensional matrix with the number of repair days as the horizontal axis and the proportion of each form as the vertical axis; Mark the time nodes of the repair agent application events in the three-dimensional matrix; The correlation coefficients between the proportion of each morphology and the amount of repair agent added were calculated using partial least squares method. The dominant migration path is defined based on the correlation coefficient being greater than 0.

8.

7. The method for heavy metal speciation analysis and detection in soil remediation process according to claim 6, characterized in that, The generation of the conversion path includes: When the dominant migration pathway shows a conversion from carbonate-bound to exchangeable state, it is determined to be a form release caused by a decrease in pH value; When the bound states of iron and manganese oxides and organic matter decrease simultaneously, it is determined to be desorption caused by an increase in redox potential; The determination result is cross-validated with the real-time operating data of the repair equipment.

8. The method for heavy metal speciation analysis and detection in soil remediation process according to claim 1, characterized in that, The adjustment of the dosage of the repair agent includes: When the proportion of exchangeable states exceeds the threshold of 30%, the amount of passivating agent added is increased by 20%. When the proportion of residue is less than 5% for three consecutive days, extend the running time of the electric repair equipment by one hour. The cycle is dynamically updated and adjusted based on the morphological transformation rate in the transformation path.

9. The method for heavy metal speciation analysis and detection in soil remediation process according to claim 8, characterized in that, The parameter feedback from the control terminal includes: The instruction to increase the amount of passivating agent added is converted into a pulse delivery control signal; The instruction to extend the running time of the electric repair equipment is superimposed on the original timing program of the equipment. The original timing program of the equipment refers to the timing program that comes with the equipment at the factory or is set in the early stage of debugging to control the daily basic running time of the electric repair equipment. The monitoring cycle of the dynamic dissolution curve is restarted after the feedback parameters take effect.

10. A heavy metal speciation analysis and detection system for soil remediation processes, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for heavy metal speciation analysis and detection in soil remediation process as described in any one of claims 1 to 9.

Images