In-situ direct measurement device and method for carbon and nitrogen solute fluxes of groundwater

Abstract

This invention discloses a method and system for directly measuring carbon and nitrogen transport fluxes in groundwater. The method involves deploying integrated microfluidic serpentine Dean flow sorting chips in each layer of the monitoring well. After groundwater samples are filtered through a membrane and driven at a constant velocity, they enter a curved microchannel with progressively decreasing radii of curvature. Under the combined action of Dean vortex secondary flow and inertial lift within the channel, carbon and nitrogen components are physically sorted to different lateral positions on the channel cross-section due to their molecular weight differences. The carbon and nitrogen transport fluxes are then determined by the laterally arranged C… 4 The D-electrode array directly outputs a voltage signal proportional to the mass flux of each component in differential mode; the downhole rotary valve group realizes multi-layer automatic polling measurement, and the ground ARM controller uniformly schedules and remotely transmits data via a wireless module. This invention transforms groundwater carbon and nitrogen flux from "estimated quantity" to "measured quantity," avoiding the interference of competing ions and membrane scaling problems of chemically sensitive membranes at the principle level, and is suitable for long-term continuous monitoring of agricultural non-point source pollution groundwater.

Description

Technical Field

[0001] This invention belongs to the field of geological environment monitoring technology, specifically relating to a device and method for in-situ direct measurement of the transport flux of dissolved carbon and nitrogen components in groundwater using the principle of fluid dynamics sorting. Background Technology

[0002] In hydrogeological studies of groundwater pollution, obtaining the vertical transport fluxes of dissolved carbon (mainly dissolved inorganic carbon DIC and dissolved organic carbon DOC) and dissolved nitrogen (mainly nitrate nitrogen and ammonium nitrogen) in aquifers at different depths is fundamental for quantifying agricultural non-point source pollution loads and establishing watershed water quality models. The accuracy of flux data directly affects the design of pollution source apportionment and control schemes.

[0003] For a long time, the standard practice in the engineering field for obtaining this data has been the indirect route of "stratified sampling + laboratory analysis + Darcy's law estimation". This involves first obtaining water samples from each stratum through stratified sampling wells, determining the concentration of the target ions, then obtaining the hydraulic conductivity and hydraulic gradient of the corresponding stratum through hydrogeological tests, and finally estimating the transport flux of a certain component at that stratum by multiplying "concentration × Darcy velocity × cross-sectional area". While this method is mature, it suffers from structural flaws in its error structure: concentration analysis error, hydraulic parameter estimation error, and cross-sectional area measurement error. These three independent error sources are amplified by the superposition of each other during the multiplication operation. More importantly, the measured concentration value is the net result of the combined effects of solute migration input and in-situ biochemical reactions (such as denitrification and organic nitrogen mineralization), and the indirect estimation method cannot distinguish the individual contributions of these two processes.

[0004] To address the challenge of instantaneous flow velocity measurement, US Patent 9404783B2 discloses a passive flux meter scheme. Its principle involves placing an adsorbent matrix containing a tracer downhole for a period of time, and then analyzing the mass of the residual tracer and adsorbed contaminants after recovery to estimate the average flux during that period. This scheme avoids instantaneous flow velocity measurement, but correspondingly loses the ability to capture short-duration, high-flux events (such as contaminant pulses within hours after a heavy rain).

[0005] Regarding detection methods, US patent 10564122B1 discloses an in-situ ion sensor based on microchip electrophoresis. This scheme applies a high-voltage electric field to a microfluidic chip, utilizing the difference in ion mobility within the electric field to achieve separation, and then uses a capacitively coupled non-contact conductivity detector for quantification. This scheme compresses the single analysis time to several minutes, significantly improving in-situ detection efficiency. However, its measurement endpoint remains the ion concentration at a specific time point, not changing the fundamental principle that "flux is an estimated quantity." Furthermore, this scheme and other existing in-situ ion detection technologies, especially those involving chemically sensitive interfaces such as ion-selective membranes and ion-exchange membranes, face risks of signal drift and failure due to membrane fouling, competing ion interference, and biofilm contamination during long-term deployment. Our long-term in-situ tests conducted in shallow groundwater monitoring wells in the North China Plain show that, without special protection, the measurement deviation of ion-sensitive electrodes generally exceeds the acceptable calibration compensation range after 1 to 2 weeks of installation.

[0006] In summary, there is an urgent need in this field for a technical solution that can shift the measurement endpoint from "concentration" to "flux," thereby avoiding the problem of multi-source error superposition in principle and bypassing the core weak link of chemically sensitive membranes, so as to support long-term continuous monitoring of groundwater solute transport. Summary of the Invention

[0007] One objective of this invention is to provide a method for in-situ direct measurement of groundwater carbon and nitrogen fluxes without relying on chemical separation membranes. A second objective is to provide a measurement device for implementing this method. This invention combines a microfluidic inertial flow physical sorting mechanism with a multi-channel non-contact conductivity detection array, enabling the direct conversion of solute flux signals into readable electrical signals.

[0008] Specifically, the measurement method of the present invention includes the following steps: a) deploying microfluidic sorting measurement units at at least two depth layers vertically distributed within the monitoring well; b) after filtering the groundwater sample through a pre-filter membrane, smoothly driving it into the main sorting channel of the microfluidic chip at a calibrated flow rate maintained in the range of 100 to 500 microliters per minute; c) the main sorting channel is a serpentine tortuous microchannel with at least two sections of decreasing radius of curvature. When the water sample flows through this channel, the dissolved carbon and nitrogen components are separated due to differences in molecular weight under the combined action of Dean vortex secondary flow and inertial lift. The effective hydrated ions have different radii and migrate to different lateral equilibrium positions on the channel cross section, thus achieving physical spatial sorting; d) In the straight detection zone downstream of the main sorting channel, at least three capacitively coupled non-contact conductivity detection electrode pairs are fixedly arranged along the channel cross section direction. The at least three electrode pairs are respectively aligned with the lateral positions corresponding to the carbon component enrichment zone, nitrate nitrogen enrichment zone, and ammonium nitrogen enrichment zone after sorting; e) Under the condition that the calibrated flow rate is kept constant, the voltage signal output by each capacitively coupled non-contact conductivity detection electrode pair is collected. The voltage signal is directly proportional to the mass flux of the corresponding component.

[0009] Based on the above method, this application also provides a specific inertial sorting microchannel design: the main sorting channel has a rectangular cross-section with a width of 180 to 220 micrometers and a height of 40 to 60 micrometers, and the total length of the channel is between 100 and 150 millimeters. This channel consists of at least three series-connected curved sections with progressively decreasing radii of curvature. The radius of curvature of each curved section can be selected sequentially as: 450 to 550 micrometers, 250 to 350 micrometers, and 120 to 180 micrometers. The design intention of decreasing curvature is to gradually increase the Dean number, allowing ions to undergo a progressively stronger inertial focusing and sorting process in the channel, reaching a stable lateral equilibrium position before entering the detection zone.

[0010] To address the complex chemical background of groundwater in engineering practice, this method further includes an in-situ matrix calibration step: using the in-situ groundwater from the aquifer of the monitoring well as the base fluid, after filtration and sterilization via a 0.45-micron filter membrane, a series of standard flux solutions containing known carbon and nitrogen component concentrations are prepared; according to a preset cycle or instruction, the standard solutions are sequentially injected into a microfluidic chip, the responses of each detection electrode pair are recorded, and the standard working curve and sensitivity correction coefficient are established and updated. The advantage of this approach is that the chemical matrix of the calibration solution is consistent with the actual water sample, effectively compensating for slight sorting position shifts that may be caused by background differences such as temperature, ionic strength, and coexisting components.

[0011] To prevent physical blockage and biofouling during long-term downhole deployment, this method integrates a three-tiered synergistic maintenance approach: First, a 0.45-micron PTFE membrane is used as the pre-filter, with a controlled micro-electromagnetic oscillator mounted above it. This oscillator does not operate continuously during measurement; instead, it generates high-frequency micro-amplitude vibrations lasting several seconds at preset time intervals (e.g., every 15 minutes) or triggered based on a pressure difference threshold across the membrane, to peel off the still loose filter cake layer from the membrane surface. The selection of amplitude and frequency must balance deposit removal efficiency with potential disturbance to the internal flow field of the chip; the recommended parameter range is a vibration frequency of 50 to 200 Hz and an amplitude of 5 to 20 microns.

[0012] In the second stage, the inner wall of the microfluidic chip's channel is pre-treated with octadecyltrichlorosilane (OTS) solution to form a hydrophobic self-assembled monolayer, which inhibits the non-specific adsorption of dissolved organic matter on the channel wall.

[0013] In the third stage, within the multi-layered monitoring well, each layer's measurement unit is connected to a surface-embedded ARM controller via a downhole rotary valve assembly driven by a stepper motor and a CAN bus. The controller switches the power supply and signal channels of each layer unit in a "polling" mode according to a program, enabling a single surface system to perform time-sharing measurements and long-term data acquisition across multiple depth layers.

[0014] The present invention also provides a measuring device for implementing the above method. The device includes a downhole stratified measurement subsystem comprising at least two stratified measurement units that can be deployed at different preset depths within the monitoring well; each stratified measurement unit integrates: a pre-filter membrane, an electromagnetic oscillator, a micro-piezoelectric pump, a microfluidic serpentine Dean flow sorting chip, and at least three pairs of capacitively coupled non-contact conductivity detection electrodes arranged in the chip's detection area. The main sorting channel inside the microfluidic serpentine Dean flow sorting chip consists of at least two serpentine bends with decreasing radii of curvature connected in series. Each bend is designed to sort carbon and nitrogen components of different molecular weights to different lateral equilibrium positions on the channel cross-section under a constant injection flow rate. The wellhead control and communication subsystem includes an embedded processor and a wireless communication module. This subsystem is connected to each layer measurement unit via a wired bus to transmit flux data back to the host computer and receive control commands.

[0015] The substantial contribution of this invention compared to the prior art, through the above-described solution, lies in: Firstly, it transforms flux from a "computational quantity" to a "measured quantity." Regardless of advancements in detection methods, current technologies consistently estimate flux as the product of concentration and flow rate. This approach leverages the principle that, under constant-rate injection conditions, the number of ions in the detection zone after physical sorting is directly proportional to the injection rate, thus directly mapping the transport flux of carbon and nitrogen components to... The detection signal from the electrode eliminates the superposition of multiple source errors caused by indirect calculation through multiplication from the perspective of measurement principle.

[0016] Secondly, it bypasses the chemically sensitive membrane at the principle level. Component separation relies entirely on physical parameters such as channel geometry, flow rate, and the hydration radius of the analyte ions. Within the common chemical conditions of groundwater (conductivity 500 to 5000 microsiemens per centimeter, calcium ion concentration 5 to 500 milligrams per liter, pH between 6.5 and 8.5), the types and concentrations of coexisting ions in the water do not affect the mechanical mechanism of the Dean flow field within the sorting channel. Therefore, the detection signal will not experience irreversible attenuation due to competing ions or scaling issues. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the overall system architecture of the measuring device of the present invention.

[0018] Figure 2 This is a schematic diagram of the structure of a microfluidic sorting chip within a single layered measurement unit.

[0019] Figure 3 yes Figure 2 A magnified schematic diagram of the horizontal arrangement of the non-contact conductivity electrode array in the detection zone.

[0020] Figure 4 This is a circuit diagram of the wellhead control and communication subsystem.

[0021] Figure 5 This is a schematic diagram of the working state of the filter membrane self-cleaning component.

[0022] Figure 6 This is a schematic diagram of the in-situ matrix calibration process.

[0023] Figure 7 This is a system timing diagram of the measuring device of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0025] Before describing the various embodiments, a unified explanation of several key terms and basic measurement principles in this invention will be provided to facilitate subsequent understanding. Unless otherwise specified, the term "carbon component" mentioned in this invention refers to dissolved inorganic carbon (DIC, mainly composed of...) The terms "existing form" and "dissolved organic carbon (DOC)" refer to both nitrate nitrogen (nitrate nitrogen) and dissolved organic carbon (DOC). ) and ammonium nitrogen ( These four types of substances are the most concerning indicators in monitoring groundwater pollution from agricultural non-point source pollution. The "microfluidic sorting measurement unit" mentioned in this invention refers to an integrated, sealed device that combines filtration, fluid drive, physical sorting, and electrical signal detection functions; it is a fundamental functional node constituting the downhole stratified measurement system of this invention. The "Dean vortex" described in this invention refers to two counter-rotating secondary flow vortices generated in a curved microchannel due to centrifugal force acting on a cross-section perpendicular to the main flow direction. These vortices exert force on particles in the fluid (target ions in this invention), pushing them to a specific lateral position within the channel cross-section, and this position is related to the particle's size (or molecular weight). This is the basic principle of "physical spatial sorting." Below, based on the accompanying drawings, the various components of this invention and their working principles are described in detail.

[0026] The overall system architecture is as follows Figure 1 As shown, the entire system mainly consists of two parts: a downhole stratified measurement subsystem deployed underground and a wellhead control and data communication subsystem located on the surface. The downhole part includes 2 to 4 stratified measurement units arranged vertically along a monitoring well (three are shown in the figure as an example: 100a, 100b, and 100c). Each measurement unit is connected to the embedded ARM controller 200 at the wellhead via a CAN bus 210 to achieve bidirectional data interaction. On the one hand, the ARM controller 200 controls a rotary valve group driven by a stepper motor (not shown in the figure) according to a preset timing sequence to switch the power supply and signal paths of different stratified measurement units, realizing "polling" measurement; on the other hand, the ARM controller 200 remotely transmits the collected and processed carbon and nitrogen flux data of each stratum to the cloud data platform via a 4G module or a Beidou short message module 220.

[0027] Figure 2 This is a structural diagram of the "microfluidic sorting chip," the core component within a single layered measurement unit. Figure 3 yes Figure 2 An enlarged diagram of the detection area shows... The horizontal arrangement of the electrode array is the core difference between this invention and all existing groundwater monitoring technologies. Within each layered measurement unit, a microfluidic sorting chip 110 is encapsulated. This chip enables the physical process of converting carbon and nitrogen fluxes from "calculated values" to "measured values," and its structure and principle are explained in detail below.

[0028] The microfluidic sorting chip 110 uses a PDMS (polydimethylsiloxane)-glass composite structure as its substrate and measures approximately 25mm × 15mm × 3mm. A functional microchannel network is etched inside the chip.

[0029] The groundwater sample first passes through a PTFE (polytetrafluoroethylene) pre-filter membrane 120 with a pore size of 0.45 μm to remove suspended particles such as silt before entering the chip inlet. Then, a miniature piezoelectric pump 130 smoothly drives the water sample into the main sorting channel 140 at a constant flow rate of 100 to 500 μL / min.

[0030] The main sorting channel 140 is a serpentine flow channel that we determined after repeated verification. Its cross-section is a rectangle with a width of 200μm and a height of 50μm, and a total length of approximately 120mm. This serpentine flow channel is not a simple bend with a constant radius of curvature, but is composed of three functional segments connected in series with progressively decreasing radii of curvature: the first segment 141 has a bending radius of approximately 500μm and contains 10 bends; the second segment 142 has a bending radius that decreases to approximately 300μm and contains 8 bends; and the third segment 143 has a bending radius that further decreases to approximately 150μm and contains 6 bends.

[0031] The underlying logic of this design is as follows: with a fixed channel size and flow velocity, a decrease in the bend radius R leads to an increase in the Dean number (a dimensionless parameter characterizing the Dean vortex intensity in a bend). In the 500 μm bend, the Dean vortex intensity is relatively weak, around 0.8. At this point, the inertial lift initially affects the hydrated ions, but the separation between different ions is not yet significant. In the 150 μm bend, the Dean number increases to approximately 3.2, and the Dean vortex intensity significantly strengthens. Under the combined and repeated action of the gradually increasing Dean vortex drag force and the inertial lift pointing towards the channel wall, ions of different radii are continuously "driven" towards their respective hydrodynamic equilibrium positions. Ultimately, at the straight detection zone 144 at the outlet of the third bend section 143, for example, ions with a molecular weight of approximately 61... and a molecular weight of around 62 Its lateral equilibrium position is stably offset by a distance of approximately 8 to 12 μm.

[0032] This difference of a few micrometers is the basis for the physical sorting capability of this invention. To detect this extremely small spatial sorting effect, this invention does not employ any complex chemical or optical detectors, but instead directly uses three independent... Electrode pairs ( Figure 3 Electrode pairs 151, 152, and 153 are arranged laterally on the glass substrate of detection area 144, aligned with the corresponding component enrichment areas of linear detection area 150: electrode pair 151 is aligned with the carbon component enrichment area, 152 with the nitrate nitrogen enrichment area, and 153 with the ammonium nitrogen enrichment area. Each electrode pair consists of an excitation electrode and a pickup electrode, and its operating frequency is adjustable. When the piezoelectric pump maintains a constant flow rate, the number of ions flowing through the high-concentration enrichment area directly below each electrode pair is directly proportional to the rate at which that component is delivered into the chip, which is its "mass flux". Therefore, the voltage signal output by these three electrode pairs in differential mode (differential amplifier 160) is directly proportional to the flux of each component, without needing to multiply by the flow rate or cross-sectional area for calculation. Flux here becomes a directly readable electrical signal.

[0033] Figure 4 This is a circuit diagram of the surface wellhead control and communication subsystem. Figure 5 This is a schematic diagram illustrating the working principle of the self-cleaning component of the filter membrane in a single layered measurement unit. The microfluidic chip described above works well in the laboratory, but to enable it to operate continuously and stably for more than a year in a monitoring well more than ten meters underground, two serious challenges need to be addressed: first, how to switch measurements between different depth layers; and second, how to prevent the chip inlet from being blocked by silt, colloids, and microorganisms in the groundwater.

[0034] This embodiment solves the measurement switching problem in the following way: The ARM controller 200 at the wellhead runs a configurable timed scheduling program. When the preset time point is reached, the ARM controller 200 sends a switching command to the downhole stepper motor and rotary valve group via the CAN bus, turning on the power and signal channels of the next measurement unit 100 to be measured, while turning off the channel of the current unit, thereby realizing "timed polling" measurement at multiple depths.

[0035] For the more challenging problem of clogging, the "three-level defense" strategy of this invention constitutes a synergistic protection system. The first level is a 0.45μm PTFE filter membrane 120. This pore size was chosen based on our experience testing in various types of groundwater (with sediment content ranging from almost 0 to 500 mg / L) in multiple regions. Any smaller, like an overly fine net, would be easily clogged by instantaneous fluctuations in suspended solids concentration, causing the pressure difference across the membrane to increase dramatically within hours. Any larger, while extending the membrane's lifespan, would allow more and finer particles to pass through. Although these particles might temporarily pass through the microchannels, they could slowly deposit in low-velocity bends, forming potential sources of long-term clogging. 0.45μm is an empirical value that strikes a balance between "preventing instantaneous clogging" and "preventing long-term sedimentation."

[0036] The second level of defense is a miniature electromagnetic vibrator 170 attached above the filter membrane 120. It is directly controlled by the ARM controller 200 and activated according to a specific cycle. Our strategy is not to apply vibration continuously, as this would waste energy and potentially accelerate mechanical fatigue. Instead, we trigger a pulsating high-frequency micro-amplitude vibration lasting about 10 seconds at a preset time interval, such as every 15 minutes, with the amplitude controlled between 5 and 20 μm. This intermittent rapping method can effectively peel off and disperse the loose filter cake layer that has just begun to form on the filter membrane surface, preventing it from being compacted and turning into a dense filter cake that obstructs water flow.

[0037] The third level of protection is located inside the chip. Before encapsulation, the inner walls of the microfluidic sorting channel 140 are impregnated with an OTS (octadecyltrichlorosilane) solution to form a nanoscale hydrophobic self-assembled monolayer. Deep groundwater often contains dissolved organic matter such as humic acid, which is much smaller than 0.45 μm and can penetrate filter membranes. Without treatment, these organic substances will slowly adhere to the PDMS channel walls, gradually altering the wall surface charge and the effective cross-sectional size of the channel, interfering with the highly sensitive Dean vortex fine flow field. The OTS hydrophobic coating acts like a non-stick varnish on a wall, greatly suppressing the non-specific adsorption of organic matter, thus ensuring the long-term stability of the channel wall properties.

[0038] These three layers of protection—filter membrane interception, intermittent rapping stripping, and channel wall hydrophobicity—each have their own roles in time and space, yet they work together to form the engineering foundation for ensuring that the downhole system can operate unattended for extended periods.

[0039] like Figure 6 As shown, all measuring devices require calibration, especially field devices operating in environments with natural fluctuations in temperature and water chemistry. This invention employs a "field matrix spiking" method to maintain measurement traceability, the steps of which are as follows: Figure 6 As shown.

[0040] First, a standard flux solution needs to be prepared. Unlike conventional sensors that are calibrated in laboratory pure water or simple salt solutions, this invention emphasizes using the raw aquifer water from the monitoring well as the base solution for pretreatment: filtration through a 0.45 μm filter membrane followed by high-temperature sterilization to ensure the elimination of biological and physical interferences. Then, DIC and... The standard was prepared into a series of standards with known concentrations of low, medium, and high.

[0041] During monthly calibration, the operator connects an external standard solution bag to the calibration port. The calibration program in the ARM controller 200 takes over control, sequentially injecting the three concentration gradients of standard solution into the chip at the normal measurement flow rate. The program automatically records and stores the values ​​at each concentration gradient. Figure 3 Middle Group 3 The response curves and peak heights of the electrode pair.

[0042] Unlike conventional two- or three-point calibration, the calibration of this invention not only calculates the sensitivity coefficient and updates the firmware of the ARM controller 200, but also compares the current calibration coefficient with historical data. If the sensitivity drift of a certain electrode exceeds a preset threshold, such as a deviation of more than 30%, the system will mark this event as "requiring attention" and automatically generate a maintenance warning on the cloud platform, reminding managers to check the measurement unit at that level—which may require manual cleaning of the filter membrane, replacement of the chip, or investigation of other causes. This "diagnostic-in-calibration" approach changes periodic maintenance from time-driven to state-driven, further improving the long-term operational reliability and availability of the system.

[0043] like Figure 1As shown, in a typical winter wheat-summer maize rotation agricultural non-point source pollution monitoring area in the North China Plain, the system of this invention was deployed in three monitoring wells with depths between 15 and 20 meters. Based on the regional hydrogeological profile, stratified measurement units were installed in each well at three depths: 3m (shallow), 9m (medium), and 14m (deep). The system operates at a frequency of polling one depth per hour, obtaining approximately eight sets of effective flux data per depth per day. Data from a complete hydrological year, spanning from March to February of the following year, reveals two crucial pollution migration processes that were previously unseen in monthly sampling data: First, during several heavy rainfall events in late June and mid-July each year, the nitrate nitrogen flux in the mid-layer (9m) channel experiences a pulse-like increase of an order of magnitude within approximately 4 to 8 hours after the onset of rainfall, lasting for 1 to 2 days before declining as the rainfall ends; second, after basal and topdressing fertilizers are applied in early April and early July, both carbon and nitrogen fluxes in the shallow layer (3m) exhibit a broad peak lasting approximately 7 to 10 days. These temporal resolutions and process details directly demonstrate the invention's ability to capture peak fluxes from short-duration events, which are precisely the main contributors to the total annual solute transport.

[0044] In a large irrigation area of ​​the Yellow River Basin, a monitoring section was set up along the direction of water retreat. One monitoring well was selected at the upstream, middle, and downstream of the section, and the system of this invention was deployed at two depths of 4m and 10m for each well. The flux data of the system was accessed through a 4G network to the digital twin hydraulic model of the irrigation area once per hour, serving as real-time boundary condition verification data for the solute transport module. During the two concentrated irrigation periods of the year, "spring irrigation" and "winter irrigation," the system recorded clear deep (10m) carbon and nitrogen flux change curves that were highly synchronized with the irrigation process. This high temporal resolution data significantly improved the efficiency of model calibration, solving the previous problem of "inaccurate model parameter calibration" caused by the order-of-magnitude mismatch between data sampling points and model time steps (monthly data versus sub-hourly model step size).

[0045] In the above design, an easily overlooked detail is the volume of the pre-filter membrane and the inner wall of the flow channel. During the design process, we controlled the ratio of the filter membrane area to the total volume of the downstream flow channel within a specific range. This ensures that even under abnormal operating conditions where the piezoelectric pump experiences minute flow rate fluctuations, the fluid buffer volume in the chamber behind the filter membrane is sufficient to absorb these pulsations, preventing the flow rate changes from being directly transmitted to the Dean flow sorting area within the chip. This, in turn, guarantees the stability of the sorting effect (ion equilibrium position). Furthermore, the frequency and amplitude of the oscillator were also matched and tested, aiming to effectively shear the filter membrane while avoiding the induction of disruptive secondary flow fields within the chip due to excessive vibration energy.

[0046] Regarding the detection of ammonium nitrogen NH4⁺, NH4⁺ is positively charged, while HCO3⁻ and NO3⁻ are negatively charged. This does not affect the fundamental physical mechanism of Dean vortex sorting—Dean vortex sorting relies on particle size (the hydrated ionic radius related to molecular weight), not charge. However, since different charges imply differences in the structure of the hydrated layer, its effective hydrated ionic radius will differ from that of anions with the same charge. Therefore, in the chip flow channel, the lateral enrichment sites of NH4⁺ will naturally not coincide with those of anions. This provides a basis for using the third group of C... 4 Electrode D creates the conditions for its specific detection. It should be noted that, because the difference in hydrated ionic radii between NH4⁺ and NO3⁻ is greater than that between NO3⁻ and HCO3⁻, its sorting effect is actually superior, with a larger lateral offset. 4 The D electrode is more advantageous for signal pickup and resolution.

[0047] Another aspect is the timing of the cleaning system's operation. As mentioned earlier, the oscillator on the filter membrane surface operates intermittently. In fact, the software control logic also includes a "smart trigger" backup mode—when the ARM controller calculates the water pressure difference across the filter membrane based on changes in the piezoelectric pump's operating current and determines that it reaches a certain critical value (approximately 0.5 to 1.0 bar in maintenance mode), an additional cleaning cycle will be temporarily inserted. This is equivalent to adding a layer of "conditional maintenance" safety redundancy on top of "timed maintenance."

[0048] It should be noted that the above embodiments and accompanying drawings are merely illustrative examples of the core principles and key structures of the present invention, "A Direct Measurement Method and System for Groundwater Carbon and Nitrogen Transport Flux." The accompanying drawings are simplified schematic diagrams, intended to clearly illustrate the structural, process, or data flow relationships related to the innovative points of the technical solution, and are not intended to limit the complete form of the actual product. This specification focuses on the innovative technical means necessary to achieve the invention's objectives and solve the technical problems. While auxiliary or commonly known details such as downhole waterproof connectors, CAN bus communication protocol stacks, piezoelectric pump drive circuits, conventional signal filtering algorithms, power management circuits, and standard screws / clips and other fasteners, which can be implemented by those skilled in the art without creative effort, are not described in detail, they should all be understood as naturally included in the specific implementation of this invention and fall within the protection and implementation scope of this technical solution.

Claims

1. A method for in-situ direct measurement of groundwater carbon and nitrogen transport flux, characterized in that, Includes the following steps: (a) Deploy microfluidic sorting measurement units at at least two depth levels vertically distributed within the monitoring well; (b) After the groundwater sample is filtered through a pre-filter membrane, it is smoothly driven into the main sorting channel of the microfluidic chip at a calibrated flow rate maintained in the range of 100 to 500 microliters per minute. (c) The main sorting channel is a serpentine microchannel with at least two segments of decreasing curvature radius. When the water sample flows through the channel, the dissolved carbon and nitrogen components in it migrate to different lateral equilibrium positions on the channel cross section due to the different effective hydrated ion radii caused by the difference in molecular weight under the combined action of Dean vortex secondary flow and inertial lift, thus achieving physical spatial sorting. (d) In the straight detection zone downstream of the main sorting channel, at least three capacitively coupled non-contact conductivity detection electrode pairs are fixedly arranged along the cross-sectional direction of the channel. The at least three electrode pairs are respectively aligned with the lateral positions corresponding to the carbon component enrichment zone, nitrate nitrogen enrichment zone and ammonium nitrogen enrichment zone after sorting. (e) Under the condition that the calibrated flow rate is kept constant, the voltage signal output by each of the capacitively coupled non-contact conductivity detection electrode pairs is collected. The voltage signal is directly proportional to the mass flux of the corresponding component.

2. The method according to claim 1, characterized in that, The main sorting channel has a rectangular cross-section with a width of 180 to 220 micrometers and a height of 40 to 60 micrometers. The total length of the channel is between 100 and 150 millimeters. The at least two segments with decreasing radii of curvature are specifically: the first curved segment has a radius of curvature of 450 to 550 micrometers, the second curved segment has a radius of curvature of 250 to 350 micrometers, and the third curved segment has a radius of curvature of 120 to 180 micrometers.

3. The method according to claim 1, characterized in that, The method also includes an in-situ calibration step for the microfluidic sorting measurement unit: using in-situ groundwater taken from the monitoring well as the base liquid, after filtration and deactivation, at least three standard flux solutions with known carbon and nitrogen component concentrations are prepared; each standard solution is injected into the microfluidic sorting measurement unit sequentially at the calibration flow rate, and the output response of each capacitively coupled non-contact conductivity detection electrode pair is recorded, thereby establishing and updating a standard curve or correction coefficient for converting voltage signals into mass flux.

4. The method according to claim 1, characterized in that, The method also includes a self-cleaning step for the pre-filter membrane: an electromagnetic oscillator is fixed above the pre-filter membrane in a sealed shell submerged in groundwater. The electromagnetic oscillator is periodically triggered at preset time intervals or when the water pressure difference across the pre-filter membrane reaches a preset threshold, so as to generate vibrations with a frequency between 50 and 200 Hz and an amplitude between 5 and 20 micrometers to peel off the deposits on the surface of the filter membrane.

5. The method according to claim 4, characterized in that, The inner wall surface of the main sorting channel of the microfluidic chip is pre-impregnated with an octadecyltrichlorosilane solution to form a hydrophobic self-assembled monolayer on the inner wall of the channel.

6. The method according to claim 1, characterized in that, It also includes the step of time-division polling measurement of microfluidic sorting measurement units at multiple depth levels: a wellhead embedded controller controls a downhole rotary valve group driven by a stepper motor through a bus interface according to a preset timing sequence, and connects or disconnects the power supply and signal path of each depth measurement unit one by one, so that multiple levels share the same set of wellhead control and communication units.

7. An in-situ direct measurement device for groundwater carbon and nitrogen transport flux, used to implement the method according to any one of claims 1 to 6, characterized in that, include: At least two stratified measurement units that can be deployed at different preset depths within the monitoring well; each stratified measurement unit integrates: a pre-filter membrane, an electromagnetic oscillator, a micro piezoelectric pump, a microfluidic snake-shaped Dean flow sorting chip, and at least three pairs of capacitively coupled non-contact conductivity detection electrodes arranged in the chip's detection area; The main sorting channel inside the microfluidic serpentine Dean flow sorting chip is composed of at least two serpentine curved sections with decreasing radii of curvature connected in series. Each curved section is designed to sort carbon and nitrogen components of different molecular weights to different lateral equilibrium positions on the channel cross section under a constant injection flow rate. The wellhead control and communication subsystem includes an embedded processor and a wireless communication module. This subsystem is connected to each layer measurement unit via a wired bus to transmit flux data back to the host computer and receive control commands.

8. The apparatus according to claim 7, characterized in that, The microfluidic serpentine Dean flow sorting chip is made of PDMS-glass composite substrate, and its internal main sorting channel has a rectangular cross-section with a width of 180 to 220 micrometers and a height of 40 to 60 micrometers, and a total channel length of 100 to 150 millimeters; the at least two curved sections with decreasing radii of curvature specifically include three serpentine curved sections with radii of curvature of 500 micrometers, 300 micrometers and 150 micrometers respectively.

9. The apparatus according to claim 7, characterized in that, It also includes an in-situ calibration subsystem located at the wellhead, comprising a standardized interface for connecting an external standard flux solution bag, a controlled switching valve, and a calibration control program stored in the embedded processor. When the embedded processor responds to a calibration command, it takes over control of the switching valve and the micro piezoelectric pump to sequentially introduce at least three standard solutions of known concentrations into the selected stratified measurement unit, automatically completing sensitivity calibration and coefficient updates.

10. The apparatus according to claim 7, characterized in that, It also includes a downhole polling and switching subsystem, which includes a stepper motor controlled by the embedded processor and a multi-channel rotary valve group. The rotary valve group is connected in series between the wired bus and each layer measurement unit. The embedded processor controls the rotation angle of the stepper motor to activate the measurement units of the target layer one by one according to a preset time sequence, thereby realizing time-division measurement of multiple depth layers.

Images