Nitrogen-protected atmospheric particulate matter iron speciation semi-automatic analysis device and method

Abstract

This invention discloses a semi-automatic device and method for analyzing the iron speciation of atmospheric particulate matter under nitrogen protection, belonging to the field of environmental monitoring and analytical chemistry technology, specifically addressing the analysis of Fe speciation in atmospheric particulate matter extracts. 2+ The device addresses technical challenges such as easy oxidation, large manual operation errors, and low analytical throughput. It is a continuous analysis system operating under nitrogen protection, comprising a sample extraction module, an automated sample introduction module, a reaction chamber, a detection module, and an automated cleaning module connected sequentially via PTFE tubing. A nitrogen protection system covers the entire process to maintain an oxygen-free environment, and a control system coordinates the operation of each module and manages the data. This device enables automated analysis from sample extraction to detection and cleaning, and is used for atmospheric particulate matter Fe. 2+ and Fe 3+ High-precision, high-throughput semi-automatic analysis provides a reliable technical means for studying the source migration and oxidation potential of atmospheric particulate matter.

Description

Technical Field

[0001] This invention belongs to the field of environmental monitoring and analytical chemistry, specifically relating to the extraction of iron (Fe) from atmospheric particulate matter under nitrogen protection. 2+ / Fe 3+ A semi-automatic colorimetric analysis device and method. Background Technology

[0002] Iron form (Fe) 2+ / Fe 3+ Accurate determination of Fe(II) is a key indicator for studying the sources, migration and transformation patterns of atmospheric particulate matter and its environmental oxidation potential. Currently, analysis in this field mainly relies on the classic Ferrozine colorimetric method, the principle of which is Fe(II)... 2+ The Fe reacts specifically with Ferrozine reagent to form a stable purple-red complex, which is then quantified spectrophotometrically. While this method is widely used due to its high sensitivity and ease of operation, it faces specific challenges in practical applications, particularly when processing atmospheric particulate matter extracts: Fe 2+ Easily oxidized by air, with cumbersome operating procedures that can introduce errors, and difficulty in achieving high-throughput standardized analysis, resulting in poor accuracy and repeatability of measurement results.

[0003] First, traditional manual analytical methods have inherent limitations. During open sample processing and measurement, dissolved Fe... 2+ It is easily oxidized to Fe by oxygen in the air. 3+ This leads to Fe 2+ The concentration was underestimated while Fe 3+ The concentration was overestimated, ultimately causing Fe 2+ / Fe 3+ The ratios are severely distorted, resulting in poor accuracy and repeatability of the measurement results. This problem is particularly prominent in atmospheric particulate matter samples with low iron concentrations.

[0004] Secondly, the adaptability of existing automation technologies is insufficient. Although some semi-automatic or online iron ion analysis systems have been developed, such as the online iron ion analyzer from Endress+Hauser or the microfluidic chip based on Ferrozine by Geißler et al., these devices have demonstrated the feasibility of automated iron ion detection. However, their initial design was primarily geared towards seawater or industrial water samples with relatively high iron concentrations and relatively stable matrices. Their flow path design and operating environment have not been optimized for the sensitive characteristics of low concentrations and high oxidizability of atmospheric particulate matter extracts. Specifically, existing systems generally lack effective oxygen isolation measures and nitrogen protection systems covering the entire sample pretreatment and analysis process. They also fail to fully integrate functions such as automated sample introduction, high-precision quantification, and programmed cleaning, thus failing to meet the urgent needs of atmospheric environmental research for high-throughput and high-stability analysis.

[0005] Furthermore, it should be specifically pointed out that the atmospheric particulate matter oxidation potential (OP) measuring device disclosed in the authorized patent CN118190841A cannot solve the aforementioned technical problems in iron speciation analysis. The OP measuring device aims to assess the overall oxidative toxicity of particulate matter, and its detection target is the combined effect of multiple active components. Its measurement principle (such as the DTT method) is fundamentally different from the chemical system and iron valence state analysis (Ferrozine colorimetric method). More importantly, the OP measurement does not require and does not possess a full-process inert gas protection function; its open analytical environment is precisely what leads to the formation of Fe... 2+ The main cause of oxidation. Existing OP detection technologies differ from iron speciation analysis in principle, measurement system, and oxygen barrier design. They cannot be directly applied to the accurate determination of iron valence state, nor can effective oxygen barrier be achieved through simple structural improvements.

[0006] In summary, there is an urgent need in this field for an innovative technical solution that can deeply integrate end-to-end nitrogen protection with a highly automated technical module specifically designed for iron valence state analysis. This solution would be specifically designed to overcome the core pain points in the analysis of iron speciation in atmospheric particulate matter, such as oxidation interference, human error, and low throughput, in order to achieve accurate, rapid, and stable determination of this type of sample. Summary of the Invention

[0007] One objective of this invention is to collaboratively address three core challenges in the analysis of iron speciation in atmospheric particulate matter: 1) Fe 2+ 1) Easily oxidized by air, leading to distorted results; 2) Errors introduced by manual operation result in poor repeatability; 3) Manual analysis is inefficient and has insufficient throughput.

[0008] To achieve these objectives, the present invention provides a semi-automatic iron speciation analysis device, which constitutes a continuous analysis system operating under nitrogen protection. The device includes: a sample extraction module for receiving and processing atmospheric particulate matter samples and outputting an iron-containing extract; and an automatic sample introduction module connected to the output of the sample extraction module via a first flow path for receiving the iron-containing extract and quantitatively introducing Ferrozine chromogenic reagent, buffer solution, and a reagent for removing Fe from the iron-containing extract. 3+ Reduced to Fe 2+ The reducing agent is used to form the test mixture; the reaction chamber, which is connected to the output of the automatic sample injection module through a second flow path, is used to receive the test mixture and incubate it under constant temperature conditions to complete the Fe... 2+ The colorimetric reaction with Fe 3+ The reduction reaction forms Fe 2+ -Ferrozine complex; Detection module, connected to the output of the reaction chamber via a third flow path, is used for real-time optical detection of the post-reaction liquid, specifically determining Fe.2+ - Absorbance of the Ferrozine complex at 562 nm wavelength; Automatic cleaning module, which is connected to the automatic sample injection module, reaction chamber and detection module through a multi-way valve system, is used to programmatically flush the flow path through these modules after detection to prepare for the next analysis; Nitrogen protection system, which is connected to and covers the entire flow path system from the sample extraction module to the detection module through a gas path, is used to perform vacuum-nitrogen replacement of the flow path system before analysis and maintain a positive pressure nitrogen atmosphere during analysis to isolate oxygen interference; Control system, which is signal connected to the sample extraction module, automatic sample injection module, reaction chamber, detection module, automatic cleaning module and nitrogen protection system, is used to coordinate commands, execute preset analysis procedures and collect, process and store detection data.

[0009] To address the three core challenges mentioned above, this solution introduces a nitrogen protection system. This system connects to and covers the entire flow path from the sample extraction module to the detection module via a gas path, performing vacuum-nitrogen purging before analysis and maintaining a continuous, slightly positive pressure nitrogen atmosphere during the analysis. This method ensures that the unstable Fe... 2+ It provides a completely inert environment. Vacuum-nitrogen replacement ensures an oxygen-free initial state; continuous slight positive pressure actively prevents the infiltration of external oxygen, fundamentally inhibiting Fe... 2+ The oxidation of particulate matter ensures the authenticity and accuracy of the measurement results. An automatic sample introduction module, an automatic cleaning module, and a control system are set up to automate and program steps such as sample addition, mixing, and cleaning, eliminating human error and improving repeatability and throughput. The six functional modules are connected sequentially through flow paths and uniformly scheduled by a central control unit, forming a continuous, closed, airtight, coherent, and cyclical automated process, achieving efficient and standardized operation from sample input to data output. While existing automated equipment (such as industrial online iron analyzers) has achieved process automation, its design does not address the low concentration and highly oxidizable characteristics of atmospheric particulate matter extracts. They typically lack full-process nitrogen protection, only covering certain local stages (such as reagent bottles) with nitrogen, or operating entirely in open atmosphere, failing to effectively prevent Fe... 2+ It is oxidized during the long process of transport and reaction.

[0010] Furthermore, the oxidation potential (OP) measuring device disclosed in the authorized patent CN118190841A aims to address the problem of how to comprehensively assess the overall toxicity potential of particulate matter; while this invention directly addresses a specific and long-standing pain point in environmental analytical chemistry: how to accurately and stably determine the valence state of chemically unstable specific elements (such as Fe) in the complex, low-concentration, and easily oxidized matrix of atmospheric particulate matter. 2+These are two completely different technical problems. To address the specific problems mentioned above, this invention constructs a closed-loop nitrogen protection system covering the entire process from sample extraction to the detection endpoint. This system actively protects unstable Fe by combining vacuum-nitrogen replacement with continuous micro-positive pressure maintenance. 2+ This creates an inert analytical environment. In contrast, OP measurements do not require and typically lack such global inert atmosphere control. Furthermore, this invention integrates a priori Fe assay designed specifically for iron valence state analysis. 2+ This invention employs a two-step automated analysis process for total iron determination followed by reduction, and is equipped with a high-precision reducing agent addition module. This represents a fundamental difference in technical approach compared to the single-path, continuous monitoring mode based on probe molecule consumption commonly used in OP determination. The final output of this invention is specific chemical speciation concentration data (Fe). 2+ Fe 3+ The concentration and ratio of OP (topic toxicity) are directly used for the analysis of environmental chemical processes, source identification, and biogeochemical cycle research. The OP measurement device outputs a macroscopic indicator (OP value) characterizing the overall toxicity effect, primarily used in the field of environmental health and risk assessment.

[0011] Preferably, the automatic sample introduction module includes a programmable 12-way valve and a 6-way valve; the flow path interface of the 12-way valve is configured to connect to the output end of the sample extraction module, the cleaning solution inlet, the standard solution bottle, the buffer solution bottle, the reducing agent bottle, the input end of the reaction chamber, and the flow cell inlet of the detection module; the flow path interface of the 6-way valve is configured to connect to the nitrogen transfer container, the cleaning solution inlet, the reaction chamber, and the Ferrozine reagent bottle.

[0012] Preferably, the autosampler module further includes at least two syringe pumps, including a 10 mL syringe pump and a 100 µL syringe pump, for achieving high-precision quantitative transfer of samples and reagents.

[0013] Preferably, the reaction chamber includes a polytetrafluoroethylene reaction tube placed in a thermostatic jacket, the thermostatic conditions of which are controlled at 37 ± 0.5℃, and the incubation time of the sample is 12 minutes.

[0014] Preferably, the detection module includes a quartz glass flow colorimeter with an optical path of 10 mm, and a detection system consisting of an LED light source with a center wavelength of 562 nm and a silicon photodiode detector, with an absorbance detection linear range of 0 to 2 AU.

[0015] Preferably, the reducing agent is a 4.2 M hydroxylamine hydrochloride solution or a 5 mM ascorbic acid solution, thereby reducing Fe... 3+ Effective and stable reduction to Fe 2+To accurately determine the total soluble iron concentration. Furthermore, a 4.2 M hydroxylamine hydrochloride solution is preferably used as the reducing agent.

[0016] Preferably, the performance indicators of the semi-automatic atmospheric particulate iron speciation analyzer meet the following conditions: the concentration range of the online working curve is 0.01 to 2.5 mg·L⁻¹. -1 Fe 2+ And the coefficient of determination R 2 Greater than 0.999; the method detection limit for atmospheric particulate matter extract is 0.022 ng·m -3 .

[0017] Preferably, the performance indicators of the semi-automatic analyzer for iron speciation of atmospheric particulate matter meet the following conditions: the average recovery rate of the spiked atmospheric particulate matter extract is 97.1%, and the relative standard deviation (RSD) of 10 consecutive measurements is 2.2%.

[0018] Preferably, the nitrogen protection system further includes a dissolved oxygen monitoring subsystem, comprising at least one miniature dissolved oxygen sensor located at the inlet of the reaction chamber, and a feedback control unit electrically connected to the dissolved oxygen sensor and the nitrogen flow control valve of the nitrogen protection system. The feedback control unit is configured to dynamically adjust the nitrogen supply flow rate of the nitrogen protection system based on the dissolved oxygen concentration data monitored in real time by the dissolved oxygen sensor. Specifically, when the dissolved oxygen concentration is higher than a first preset threshold, the nitrogen supply flow rate is increased to perform enhanced purging; when the dissolved oxygen concentration is lower than a second preset threshold, the nitrogen supply flow rate is reduced to a maintenance level. This intelligent feedback adjustment mechanism further reduces the risk of oxygen intrusion, ensuring that the system maintains an inert, oxygen-free environment during long-term continuous operation.

[0019] Furthermore, the dissolved oxygen monitoring subsystem also includes a sensor self-calibration module connected to the feedback control unit. During the system initialization phase before each analysis cycle, this module automatically performs the following operations: injects a standard oxygen-saturated solution of known concentration into the flow path; compares the readings of the miniature dissolved oxygen sensor with the standard value, automatically performs software calibration when the deviation exceeds the allowable range, and records the calibration log; diagnoses the sensor status by analyzing the sensor's response curves in zero-oxygen and standard oxygen environments, and issues an early warning when sensor performance degrades; the feedback control unit employs an adaptive predictive control algorithm based on historical data, configured to: establish a dynamic model of the system's background oxygen permeability based on the dissolved oxygen concentration change trends recorded in multiple consecutive analysis cycles; in subsequent analyses, predict potential oxygen concentration fluctuations at key nodes based on this dynamic model, and adjust the nitrogen flow rate in advance of the sensor threshold triggering conditions to achieve proactive suppression of oxidation risks.

[0020] Preferably, the feedback control unit is connected to the control system and is further configured to: during the transport process after the automatic sample module quantitatively extracts the sample but before mixing it with reagents, the dissolved oxygen sensor monitors the instantaneous change curve of the dissolved oxygen concentration during the transport of the pure sample solution in a high-frequency sampling mode; the feedback control unit analyzes the characteristics of the change curve in real time, including the rising / falling / stable trend and the rate of change; based on the result of this characteristic analysis, a dynamic compensation factor K is applied to the nitrogen flow rate adjustment output by the prediction model for the subsequent reaction stage: if the sample solution itself causes a rapid decrease in oxygen concentration, K is set to <1 to appropriately weaken the predetermined enhanced purging intensity and avoid excessive nitrogen; if the sample solution causes an abnormal increase or violent fluctuation in oxygen concentration, K is set to >1 to enhance the predetermined nitrogen protection intensity; if the oxygen concentration is stable, K is set to =1 to maintain the original predictive control output. This approach addresses the problem of predictive models being insensitive to sample-specific perturbations. By detecting the oxygen behavior characteristics of samples in advance and adjusting subsequent steps accordingly, it adapts to transient perturbations caused by different samples. This makes the analytical method more reliable when dealing with atmospheric particulate matter samples from a wide range of sources and with complex compositions, significantly improving the method's universality and robustness.

[0021] This invention provides a semi-automatic iron speciation analysis method using any of the described semi-automatic atmospheric particulate iron speciation analysis devices, comprising the following steps:

[0022] S1: System initialization, performing vacuum-nitrogen replacement of the entire flow path through the nitrogen protection system, and executing system self-test and baseline stabilization; S2: Automatically transferring blank solution, standard solution, or sample solution according to the program via the autosampler module, mixing it with Ferrozine chromogenic reagent and buffer, and delivering it to the reaction chamber; S3: After incubation in the reaction chamber, delivering it to the detection module, measuring the absorbance at a wavelength of 562 nm to obtain Fe 2+ Concentration; S4: Add a reducing agent through the automatic sample injection module to reduce the Fe concentration in the sample. 3+ Reduced to Fe 2+ Repeat steps S2 and S3 to determine the total soluble iron concentration, and calculate the Fe concentration using the difference method. 3+ Concentration; S5: The programmed cleaning process is executed by the automatic cleaning module, while the control system stores the analysis data and prepares for the next analysis cycle.

[0023] Existing analytical methods (such as the standard Ferrozine method) involve discrete manual procedures or lack a thorough oxygen-barrier initiation step (vacuum-nitrogen replacement) in automated equipment, and do not integrate oxygen barrier treatment throughout the entire extraction-injection-reaction-detection process. The accuracy of their reduction-subtraction steps is significantly compromised by oxidation issues. This method comprises a workflow with five core steps: system oxygen barrier initialization, automated colorimetric development, specificity detection, reduction-subtraction, and automated cleaning. It uses vacuum-nitrogen replacement as a necessary starting point and integrates inert atmosphere maintenance, automated fluid control, subtraction calculation logic, and programmed cleaning into an inseparable and coherent workflow.

[0024] Preferably, step S3 further includes an automatic dilution sub-step: when the absorbance is measured by the detection module, it is determined in real time whether the value exceeds the preset upper limit threshold of the linear range; if it does not exceed the threshold, subsequent steps proceed normally; if it does exceed the threshold, the control system automatically performs the following operations: pausing the current analysis sequence, recording the sample number and the information indicating that the sample exceeds the limit; controlling the automatic injection module and the automatic cleaning module to automatically and quantitatively dilute the original sample with diluent to prepare a test solution with a preset dilution factor; re-inserting the diluted sample into the analysis queue for priority measurement; and finally, the control system automatically calculates and stores the final result based on the dilution factor. The concentration range of samples in real-world environments can be very wide. Traditionally, manual judgment and intervention are required to address sample exceedance issues. This solution integrates real-time monitoring, automatic judgment, intelligent dilution, and re-injection into a fully automated analysis process, achieving self-adjustment and adaptation. This avoids situations where a single sample exceeding the limit renders the entire analysis sequence invalid or interrupted, saving reagents and time. Users only need to prepare sufficient original samples before analysis, eliminating the need for pre-judgment and manual dilution, greatly simplifying the operation process.

[0025] The present invention has at least the following beneficial effects:

[0026] 1. This invention significantly inhibits Fe 2+ Oxidation significantly improves detection accuracy. Unlike traditional manual methods or automated equipment that only provides partial oxygen isolation, this invention establishes and maintains a complete inert environment throughout the entire flow path from sample extraction to final detection through pre-analytical vacuum-nitrogen replacement and continuous micro-positive pressure coverage during the process. This fundamental measure completely blocks Fe... 2+ The main pathway for Fe to come into contact with oxygen in the air, thereby directly converting Fe 2+ The detection rate improved by 8.11% to 13.71%, and enabled subsequent calculations of Fe using the difference method. 3+ The concentration is also more accurate and reliable.

[0027] 2. This invention exhibits comprehensive analytical performance characterized by high accuracy, high precision, and high sensitivity. This is the result of the synergistic effect of nitrogen protection, automated fluid control, and optimization of key chemical parameters. While ensuring that the true values ​​of the sample are not interfered with by oxidation, the system achieves nano-level precise metering and mixing of samples and reagents through a programmable multi-port valve and a high-precision syringe pump (e.g., a combination of 10 mL and 100 µL pumps), completely eliminating volumetric errors introduced by manual operation. Furthermore, the optimal reducing agent (4.2 M hydroxylamine hydrochloride, Example 3) selected through system screening and optimized reaction conditions (37°C, 12 minutes, Examples 1 and 3) ensures the high efficiency and stability of the colorimetric and reduction reactions. It is the interconnectedness of these technical aspects that collectively achieves a recovery rate of up to 97.1%, a precision of 2.2% (RSD, n=10, Example 2), and a range of 0.01–2.5 mg·L⁻¹. -1 Within the range R 2 Excellent linearity >0.999 and as low as 0.022 ng·m -3 The method detection limit (Examples 1 and 4).

[0028] 3. This invention achieves high-throughput batch analysis and excellent repeatability, thanks to its fully automated workflow integration and intelligent cleaning function. The automated sample introduction-reaction-detection-cleaning closed-loop process, uniformly scheduled by the control software, not only increases the analytical throughput to 12 samples per hour (Example 4), but also standardizes each analytical step through programmed operation, fundamentally ensuring the consistency of batch sample determinations. The accompanying automated cleaning module performs programmed rinsing after each analysis, effectively controlling the residual rate to below 0.5% (Example 4), eliminating cross-contamination between samples. This is a key guarantee for the system's long-term stable operation and high repeatability.

[0029] 4. The modular design of this invention endows it with excellent practicality and deployability. The system is clearly divided into six functional modules: sample extraction, automatic sample introduction, reaction, detection, cleaning, and nitrogen protection. These modules are connected through standardized interfaces (such as PTFE tubing and unified electrical control), resulting in a compact structure and convenient maintenance. This design not only reduces the difficulty and cost of daily operation and maintenance but also allows it to overcome the limitations of traditional laboratory instruments, making it suitable for a wider range of applications, including fieldwork.

[0030] 5. Conventional oxygen-free measures are usually limited to localized or intermittent nitrogen purging, or only briefly inerting samples in a sealed container. This invention innovatively achieves continuous nitrogen protection throughout the entire analytical flow path, removing initial oxygen through vacuum-nitrogen replacement and maintaining a slightly positive pressure nitrogen cover during the analysis. In contrast, traditional methods cannot provide such a continuous inert environment throughout the entire analytical process. Therefore, this invention establishes a fully oxygen-free analytical system that is significantly different from conventional inert protection methods.

[0031] 6. Maintaining a completely airtight and inert environment throughout the entire flow path in a continuous flow analysis system is extremely challenging, requiring overcoming difficulties such as multiple connection seals, valve switching status changes, and oxygen permeation during long-term operation. This invention successfully solves the difficulties in achieving complete flow path inert protection through a precise sealed flow path design and continuous micro-positive pressure nitrogen compensation, supplemented by real-time monitoring and feedback control. From sample extraction, reagent addition, reaction to detection, the entire process is conducted in an oxygen-free state, a bottleneck that existing technologies have failed to overcome.

[0032] 7. Thanks to the synergistic effect of nitrogen protection and automated control throughout the entire process, the device of this invention achieves technical results exceeding expectations. Compared with conditions without nitrogen protection, Fe... 2+ The detection signal was significantly improved (by approximately 8%–14%), and the method detection limit was reduced to 0.022 ng·m -3 The analytical precision was improved to 2.2% (and even reached 1.8% after optimization, Example 6). These data are far superior to traditional methods, demonstrating the unexpected improvement in accuracy, sensitivity, and repeatability of the present invention.

[0033] 8. Simple vacuum degassing or nitrogen purging under open conditions is not equivalent to the full-process nitrogen protection solution provided by this invention. Vacuum treatment can only temporarily remove initial oxygen, and nitrogen purging is difficult to completely isolate air in an open system. This invention combines vacuum-nitrogen replacement with continuous micro-positive pressure coverage and implements it in an airtight flow path, preventing oxygen from intruding. This holistic solution is fundamentally different from simple vacuum or nitrogen purging measures, ensuring Fe... 2+ It is not affected by oxidation throughout the entire analytical process, which is something that traditional oxygen isolation methods cannot achieve.

[0034] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the first structure of the semi-automatic atmospheric particulate iron speciation analysis device of the present invention; wherein, the nitrogen protection system is connected to the entire flow path system through the gas path;

[0036] Figure 2 This is a schematic diagram of the second structure of the semi-automatic analysis device for iron speciation of atmospheric particulate matter according to the present invention; wherein, the automatic sample injection module uses a 10 mL syringe pump A and a 100 µL syringe pump B to quantitatively transfer samples and reagents.

[0037] Figure 3 This is a schematic diagram of the third structure of the semi-automatic atmospheric particulate iron speciation analysis device of the present invention; wherein, the automatic cleaning module is connected to the flow path of the module to be cleaned through a multi-way valve;

[0038] Figure 4 This is a general structural and flow diagram of the semi-automatic atmospheric particulate iron speciation analysis device described in this invention.

[0039] Figure 5 Flow diagram of the semi-automatic atmospheric particulate iron speciation analysis device of the present invention;

[0040] Figure 6 This is a physical overall view of the semi-automatic atmospheric particulate iron speciation analysis device described in this invention.

[0041] Figure 7 This is an example of an online operating curve for the semi-automatic atmospheric particulate iron speciation analysis device of the present invention.

[0042] Figure 8 The results of the comparison of reducing agents in the semi-automatic analysis device for iron speciation of atmospheric particulate matter of the present invention are shown.

[0043] Figure 9 The results of nitrogen / non-nitrogen conditions in the semi-automatic atmospheric particulate iron speciation analyzer of the present invention are shown.

[0044] Figure 10 The values ​​represent the instantaneous dissolved oxygen peaks during sample injection under different control modes in Example 6; where Const N2 represents the constant nitrogen flow mode; Fedback represents the real-time feedback control mode; and Predictive represents the predictive feedback control mode.

[0045] Figure 11 Fe under different control modes in Example 6 2+ The relative standard deviation (RSD) of the measurements; where Const N2 is the constant nitrogen flow mode; Feedback is the real-time feedback control mode; and Predictive is the predictive feedback control mode.

[0046] Figure 12 The spike recovery range for continuous operation in predictive feedback control mode for 72 hours in Example 6;

[0047] Figure 13In Example 7, the instantaneous dissolved oxygen curves of samples A, B, and C after 3 seconds and the corresponding compensation factor K;

[0048] Figure 14 The spiked recoveries of samples A, B, and C with sample specificity compensation enabled / disabled in Example 7 are shown, where the error bar represents RSD and n=6. Detailed Implementation

[0049] The present invention will be further described in detail below with reference to examples, so that those skilled in the art can implement it based on the description.

[0050] It should be understood that terms such as “having,” “comprising,” and “including” as used herein do not exclude the presence or addition of one or more other elements or combinations thereof.

[0051] I. Overall Structure and Working Principle

[0052] As attached Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 and Figure 6 As shown, the semi-automatic analysis device for iron speciation of atmospheric particulate matter under nitrogen protection of the present invention constructs a modular, fully automated, and continuously analytical system protected by nitrogen throughout the entire process, both spatially and temporally. This system integrates the originally discrete and easily interfered manual operation into a closed, coherent, and cyclical automated process through precise hardware connections and intelligent software control. Its basic working principle is as follows: Under the program control of the control system, atmospheric particulate matter samples are extracted in a nitrogen environment in the sample extraction module. The resulting iron-containing extract is quantitatively extracted by the automatic sample injection module and precisely mixed with reagents such as Ferrozine colorimetric reagent and buffer solution to form the test solution. The test solution is transported to the reaction chamber to complete the colorimetric and reduction reactions, and then enters the UV-Vis detection module for real-time, online detection. After detection, the automatic cleaning module thoroughly flushes the system flow path, preparing for the next analysis. The nitrogen protection system runs through and covers all the above modules, ensuring that the entire process from sample preparation to the detection endpoint is in an oxygen-free environment. The control system uniformly schedules the execution and data management of the entire analysis process.

[0053] II. Modular Components and Connections

[0054] The device consists of the following six functional modules and a control system connected in sequence:

[0055] 1. Sample Extraction Module. This module is the starting point of the analysis, used to extract water-soluble iron from an atmospheric particulate filter membrane under air-isolated conditions. Its main components include: a sealable extraction container (usually a glass bottle), an ultrasonic water bath, a 0.22 µm filter unit, and a peristaltic pump. The extraction container's lid integrates a nitrogen inlet and a vacuum port for performing vacuum-nitrogen purging operations; a drain port at the bottom of the container connects to the input of the autosampler module via a polytetrafluoroethylene (PTFE) tube for delivering the extracted sample solution.

[0056] 2. Automated Sample Introduction Module. This module is responsible for the precise introduction and path switching of all liquids (samples, standards, reagents, and cleaning solutions). Its core components are a programmable 12-way valve and a 6-way valve, equipped with two syringe pumps for precise quantitative delivery (Pump A: 10 mL, for large volumes of samples / cleaning solutions; Pump B: 100 µL, for small volumes of reagents). The 12-way valve's flow path interface is configured to connect to: the sample extraction module's inlet, various cleaning solution inlets (deionized water, isopropanol, waste liquid), standard solution bottles, reagent bottles (acetic acid buffer, hydroxylamine hydrochloride, ascorbic acid), the reaction chamber inlet, and the flow cell inlet of the UV-Vis detection module. The 6-way valve's flow path interface is configured to connect to: a nitrogen transfer container, the cleaning solution inlet, the reaction chamber, and the Ferrozine reagent bottle. This module acts as a central node, receiving the sample extraction module upstream and, downstream, using valve position switching, delivering the mixed liquid to the reaction chamber.

[0057] 3. Reaction Chamber. This module provides a stable and controllable environment for the chemical reaction. Its main component is a PTFE reaction tube (e.g., approximately 1.5 cm in inner diameter and 12 cm in height) housed in a precision thermostatic heating mantle, with the temperature constantly controlled at 37 ± 0.5 °C. The sample is mixed with reagents here and incubated for approximately 12 minutes to complete the Fe reaction. 2+ The colorimetric reaction with Ferrozine was achieved by adding 4.2 M hydroxylamine hydrochloride (HA) or 5 mM ascorbic acid (AA) to Fe 3+ Reduced to Fe 2+ Fe 2+ -Ferrozine complex, used for the determination of total soluble iron. The inlet of this module is connected to the output of the autosampler module to receive the reaction mixture; its outlet is connected to the inlet of the UV-Vis detection module.

[0058] 4. UV-Vis Detection Module. This module is used for quantitative analysis of the reaction products from the reaction module. Its core is a quartz glass flow cell with an optical path of 10 mm and a volume of approximately 0.5 mL. The optical system includes an LED light source with an emission center wavelength of 562 nm (and Fe... 2+(Most absorption peak matching of the Ferrozine complex) and a silicon photodiode detector. This module features a light-shielding structure, an absorbance linear response range of 0-2 AU, and the detection data is transmitted to the control system in real time. The module's inlet directly receives the post-reaction liquid from the reaction chamber; its outlet connects to the automatic cleaning module and the waste liquid flow path.

[0059] 5. Automatic Cleaning Module. This module prevents sample cross-contamination and ensures the long-term stability of the system. It includes a 2 L cleaning solution reservoir (containing deionized water or isopropanol), a 2 L waste solution bottle, and a multi-way valve system consisting of several solenoid valves controlling the cleaning path. After analysis, the control system automatically flushes the tubing and cells flowing through the automatic sample introduction module, reaction chamber, and UV-Vis detection module according to a preset program (e.g., rinsing with deionized water for 5 minutes, cyclically twice). Isopropanol is mainly used to clean the colorimetric cell to remove organic residues. This module is connected to the flow paths of the modules requiring cleaning via the multi-way valve system and directs all waste liquid into a unified waste solution bottle.

[0060] 6. Nitrogen Protection System. This module is crucial for ensuring analytical accuracy, establishing and maintaining an inert environment throughout the system. It consists of a high-purity nitrogen cylinder, a vacuum pump, a multi-way distribution valve, and sealed piping and interfaces covering all modules. Its operation is divided into two phases: Before analysis, the system undergoes vacuum-nitrogen purging; after evacuation, nitrogen is introduced, repeated several times to completely eliminate oxygen; during analysis, nitrogen is continuously introduced into the system to maintain a slight positive pressure, actively isolating it from air.

[0061] The system is global and comprehensive. It is connected to key components such as the air inlet of the sample extraction module, the nitrogen transfer container of the automatic sample injection module, and the flow path of the reaction chamber through gas paths, forming an invisible nitrogen protective shield.

[0062] 7. Control System. This system, implemented with programmable software and a computer interface, serves as the command center of the apparatus. It is electrically connected to the valves and pumps of the automatic sample injection module, the temperature controller of the reaction chamber, the signal output terminals of the detection module, and the solenoid valves of the automatic cleaning module. Through software programming, it coordinates the start-up, shutdown, timing, and parameters of all modules, executing a complete analytical cycle from sample injection, reaction, detection, and cleaning, and automatically completing data acquisition and calibration curve plotting (requiring R...). 2 >0.999), result calculation and storage.

[0063] III. Work Process and Steps

[0064] Based on the above-mentioned device, its semi-automatic analysis method is performed according to the following sequential steps:

[0065] S1: System Initialization and Oxygen Barrier Preparation. Start the system; the control system triggers the nitrogen protection system to perform vacuum-nitrogen purging of the entire flow path. Simultaneously, the system performs a self-test and stabilizes the baseline of the detection module. S2: Automatic Sample Injection and Fe... 2+ Colorimetric determination. The control system instructs the valves and pumps of the automatic sample injection module to sequentially and quantitatively aspirate blank solution, standard solution, or the sample solution from the sample extraction module into the system, and mix it with Ferrozine colorimetric reagent and buffer. The mixture is then transported to the reaction chamber and incubated at 37°C for 12 minutes to form Fe. 2+ -Ferrozine complex. Subsequently, the chromogenic solution was fed into a UV-Vis detection module, and the absorbance was measured at 562 nm to obtain Fe... 2+ Concentration values. S3: Total iron determination and Fe 3+ Calculation. After completing Fe... 2+ After the measurement, the system is not cleaned. Instead, the autosampler module restarts, quantitatively adding a reducing agent (4.2 M HA or 5 mM AA) to the same sample solution. The mixture is returned to the reaction chamber, removing the original Fe from the sample. 3+ Reduced to Fe 2+ The solution was then incubated again for color development. This colorimetric solution was then sent to a detection module to measure the absorbance, yielding the total soluble iron concentration. (Fe) 3+ The concentration is calculated by subtracting the difference between the two measurements (Fe). 3+ =Total iron - Fe 2+ S4: Automatic System Cleaning and Data Loop. After the measurement is completed, the control system immediately starts the automatic cleaning module to thoroughly flush the system flow path according to the preset program. Simultaneously, all data from this analysis are automatically recorded, processed, and stored. After cleaning, the system automatically prepares to trigger the next analysis cycle.

[0066] Example 1: Analysis of system establishment, calibration curve plotting, and system adaptability verification

[0067] This embodiment aims to explain in detail how to assemble and operate the semi-automatic iron speciation analysis device described in this invention, complete the initial setup and calibration of the system, verify its basic analytical performance (linearity, stability and automation level), and lay the foundation for subsequent actual sample analysis.

[0068] 1. System Components and Parameter Settings

[0069] As attached Figure 1-6 As shown, the semi-automatic iron morphology analysis device was assembled strictly according to the connection relationships described in the instruction manual. The system assembly and key parameter settings are as follows:

[0070] Module Connection: All functional modules, namely sample extraction module, automatic sample injection module, reaction chamber, UV-Vis detection module, automatic cleaning module and nitrogen protection system, are physically connected through polytetrafluoroethylene (PTFE) pipes with an inner diameter of 1.5 mm, strictly following the sequence of sample extraction → automatic sample injection → reaction chamber → detection → automatic cleaning → waste liquid.

[0071] Flow path and control: The core of the automatic sample injection module consists of a programmable 12-way valve and a 6-way valve, which are connected to the control system via a computer interface. Fluid power is provided by one peristaltic pump and two syringe pumps (Pump A: 10 mL; Pump B: 100 µL).

[0072] Reaction conditions: The temperature of the heating mantle in the reaction chamber was set and controlled at 37.0±0.5℃. The incubation time of the sample and reagent in the reaction chamber was set to 12 minutes in the system control software.

[0073] Detection parameters: The wavelength of the light source in the UV-Vis detection module is fixed at 562 nm. The flow-through colorimetric cell is a quartz cell with an optical path of 10 mm. The absorbance acquisition frequency is set to 1 Hz in the software, and the average value is read during the stable plateau period of each sample detection.

[0074] Fluid parameters: Sample / standard transfer flow rate: Peristaltic pump speed set to 25 rpm, corresponding to a flow rate of approximately 3.0 mL / min.

[0075] Reagent addition flow rate: The flow rate of reagents (Ferrozine, buffer, reducing agent) delivered by syringe pump B (100 µL) is set to 50 µL / s.

[0076] System cleaning and waste discharge flow rate: The flow rate of the cleaning solution propelled by the syringe pump A (10 mL) is set to 5.0 mL / min.

[0077] 2. Calibration Curve Plotting Steps and Results

[0078] Preparation of standard solutions: Accurately prepare solutions with concentrations of 0.01, 0.05, 0.1, 0.5, 1.0, 2.0, and 2.5 mg·L⁻¹. -1 Fe 2+ A series of standard working solutions (the solvent is ultrapure water, containing 0.1% HCl to stabilize ions).

[0079] System Start-up: Turn on the apparatus and first run the nitrogen protection system. Perform three cycles of vacuuming (to -0.08 MPa) and purging with high-purity nitrogen (to 0.05 MPa) throughout the entire flow path to completely replace the air in the system. Then maintain a slight positive nitrogen pressure (0.02 MPa) throughout the entire analytical process.

[0080] Automated measurement: Initiate the calibration program in the control software. The system automatically executes the following procedures according to the preset valve position sequence and pump stroke:

[0081] Sample introduction: The 12-way valve is switched to the sampling position, and the peristaltic pump delivers a specific concentration of Fe. 2+ The standard solution was drawn into a quantitative loop (1.5 mL).

[0082] Delivery and mixing: After switching the valve position, syringe pump B pushes 300 µL of ammonium acetate buffer (1 M, pH 6.8) and 150 µL of Ferrozine reagent (5 mM) into the flow path, mixes with the standard solution, and then delivers it to the reaction chamber.

[0083] Color development and detection: After the mixture was incubated in a reaction chamber at 37°C for 12 minutes, it was transferred to the detection cell, and the absorbance was measured at 562 nm and automatically recorded.

[0084] Cleaning: After each standard point measurement, the system automatically calls the cleaning program, using syringe pump A to push deionized water at a flow rate of 5.0 mL / min to rinse the flow path for 5 minutes (2 cycles) to ensure no residue.

[0085] Data processing: The control system collects the absorbance values ​​corresponding to all standard concentration points and automatically performs linear regression fitting.

[0086] 3. Results and Conclusions

[0087] Calibration curve: as attached Figure 7 As shown, the Fe was automatically drawn by the system. 2+ The calibration curve is in the range of 0.01–2.5 mg·L⁻¹. -1 Excellent linearity was observed within the concentration range. The linear regression equation is y = 0.3754x + 0.0021, where y is absorbance and x is Fe. 2+ Concentration (mg·L) -1 Coefficient of determination R 2 = 0.9993.

[0088] System performance verification: linearity and range R 2 The value >0.999 demonstrates excellent linearity of the system within this range, meeting the requirements for quantitative analysis. After 8 hours of continuous operation, the baseline drift was less than 0.001 AU / h, indicating extreme stability of the system under nitrogen protection and isothermal conditions. From sample injection, reaction, detection to cleaning, the entire process requires no manual intervention and automatically generates high-precision calibration curves, verifying the integrated and automated technical features and effectiveness of the device.

[0089] Example 2: Precision, accuracy and recovery verification

[0090] This embodiment aims to quantitatively evaluate the accuracy (characterized by recovery rate) and precision (characterized by relative standard deviation RSD) of the semi-automatic iron speciation analyzer in actual analysis of atmospheric particulate matter samples through a standard addition recovery experiment, and to verify the key role of its automatic sample introduction and cleaning functions in ensuring the consistency of batch sample analysis.

[0091] 1. Experimental Materials and Methods

[0092] Base sample: Atmospheric PM2.5 collected from environmental monitoring points 2.5 The filter membrane sample was ultrasonically extracted with 20 mL of ultrapure water for 30 minutes under nitrogen protection in the sample extraction module, and then filtered through a 0.22 µm aqueous filter to obtain a blank matrix extract.

[0093] Spiked Samples: Accurately measure 5.0 mL of the above blank matrix extract into multiple sample vials, and precisely add Fe to each vial. 2+ The standard solution was prepared into spiked samples at three concentration levels: low, medium, and high, with theoretical spiking concentrations of 0.1 mg·L⁻¹, respectively. -1 0.5 mg·L -1 and 1.0 mg·L -1 Ten parallel samples were prepared for each concentration level (n=10).

[0094] System configuration: Use the exact same apparatus configuration as in Example 1 (see system parameters in Example 1) to ensure that the entire analysis process is carried out under full nitrogen protection.

[0095] 2. Experimental Procedure

[0096] In the control software, set up a batch analysis sequence, including the 30 spiked samples (3 concentrations × 10 replicates), calibration curve points, and blank samples. Start the fully automated analysis program; the system will then perform the following operations in sequence:

[0097] Automatic sample injection: Under software control, the twelve-way valve of the automatic sample injection module precisely switches to the designated sample vial position, and the syringe pump A (10 mL) quantitatively extracts 1.5 mL of sample into the flow path.

[0098] Fe 2+ Determination: The system automatically adds Ferrozine chromogenic reagent and buffer according to a preset program. After reacting at 37℃ for 12 minutes, Fe is measured at a wavelength of 562 nm. 2+ Absorbance.

[0099] Data recording: The absorbance values ​​obtained by the detection module are automatically recorded, and the control system software program calculates and stores Fe in real time according to the calibration curve established in Example 1. 2+ The measured concentration.

[0100] Automatic cleaning: After each sample is measured, the automatic cleaning module is immediately activated to perform a standard 5-minute × 2 cycle of deionized water rinsing procedure to ensure no cross-contamination in the flow path.

[0101] 3. Results and Calculations

[0102] Recovery rate calculation (accuracy): Recovery rate (%) is calculated using the following formula: Recovery(%) = (C measured -C background ) / C spiked ×100%. Where, C measured To determine the concentration of the spiked sample, C background The background concentration of the blank matrix extract (which was determined to be negligible), C spiked The theoretical spiking concentration.

[0103] Calculation results: Three spiking levels (0.1 mg·L⁻¹) -1 0.5 mg·L -1 1.0 mg·L -1 The average recovery rate was 97.1%.

[0104] Precision was expressed as relative standard deviation (RSD, %), calculated for 10 parallel samples at each spiking level. The average RSD for the three concentration levels was 2.2%.

[0105] The results of this embodiment demonstrate that the semi-automatic analysis device can effectively detect Fe in atmospheric particulate matter extracts. 2+ The assay exhibits excellent accuracy (average recovery rate of 97.1%) and high precision (average RSD of 2.2%), fully meeting the technical requirements for environmental sample analysis.

[0106] The results validate the effectiveness of the automated sample introduction and automated cleaning modules. Automated sample introduction eliminates volumetric errors caused by manual operation through precise pump and valve control; automated cleaning completely avoids cross-contamination between samples. The synergistic operation of these two modules is a key technical means to ensure the consistency and high repeatability of batch sample determinations. Furthermore, in this embodiment, the entire process, from sample sequence import, standard addition, analysis execution to data recording and calculation, is automatically completed by software control, fully demonstrating the automation and intelligence level of the device.

[0107] Example 3: Comparison of reducing agents and optimization of chemical parameters

[0108] This embodiment aims to determine the method for processing Fe through systematic comparison and optimization experiments. 3+ Reduced to Fe 2+The optimal type, concentration, and reaction conditions (temperature and time) of reducing agent for total soluble iron were determined to ensure the accuracy and stability of total iron determination and to verify the technical advantages of the preferred scheme.

[0109] 1. Experimental Materials and Methods

[0110] Reagents and Samples:

[0111] Reducing agents: hydroxylamine hydrochloride (HA) and ascorbic acid (AA).

[0112] Test sample: Prepare Fe of known concentration 3+ Standard solution (1.0 mg·L) -1 ) and actual atmospheric PM 2.5 Extract.

[0113] Using the exact same apparatus as in Example 1, all measurements were performed under nitrogen protection throughout the process to avoid Fe during reduction. 2+ It was re-oxidized.

[0114] 2. Experimental Procedures and Results

[0115] 2.1 Screening of Reducing Agent Types and Concentrations

[0116] Single-factor screening experiment: HA solutions (1.0, 2.0, 3.0, 4.0, 4.2, 5.0 M) and AA solutions (1.0, 2.0, 5.0, 7.5, 10.0 mM) of different concentrations were prepared. Using the aforementioned apparatus, Fe was treated with different concentrations of reducing agent under fixed reaction conditions (37℃, 12 minutes). 3+ The total iron recovery rate was determined by comparing the standard solution with the actual sample.

[0117] Results and Analysis: See attached. Figure 8 As shown, the vertical axis represents the recovery rate, and the horizontal axis represents different reducing agents. For the atmospheric particulate matter extract matrix, 4.2 M HA exhibits the best overall performance, especially for Fe... 3+ The average recovery rate reached 83.2%, and the working curve showed good linearity (R0). 2 >0.998). The reduction effect of 5 mM AA was acceptable, with an average recovery rate of 78.5%, but it was poor in the low concentration range (<0.1 mg·L⁻¹). -1 The recovery rate and linear stability of HA were slightly inferior to those of AA. Excessively high concentrations (e.g., HA > 4.5 M, AA > 8 mM) may cause absorbance reading drift due to reagent background interference, while excessively low concentrations result in incomplete reduction. Furthermore, the experiment also investigated the effectiveness of high-concentration solutions of HA and AA (both 10% by mass) as reducing agents. The results showed that in the atmospheric particulate matter extract matrix, such high-concentration reducing agents led to Fe...3+ The recovery rates decreased significantly, with 50.0% for 10% HA and 75.5% for 10% AA.

[0118] Therefore, 4.2 M HA was determined to be the optimal reducing agent, and 5 mM AA was determined to be an effective alternative.

[0119] 2.2 Optimization of reaction temperature and time

[0120] Optimization experiment: Using 4.2 M HA as the reducing agent, the reaction chamber temperature (30, 33, 35, 37, 40℃) and incubation time (5, 8, 10, 12, 15 minutes) were varied to optimize Fe... 3+ The recovery rate and absorbance signal stability of the standard solution were used as indicators for optimization.

[0121] The results showed that at 37℃, Fe 3+ The reduction rate and completeness are optimally balanced, resulting in the highest recovery rate and stable signal. The reaction is slow below 35°C, and may cause Fe2+ degradation above 40°C. 2+ The ferrozine complex undergoes slight decomposition. The reduction reaction is complete at 12 minutes, with the absorbance signal plateauing and stabilizing. Incubation times shorter than 10 minutes indicate incomplete reduction, while incubation times longer than 15 minutes show no benefit and reduce analytical throughput.

[0122] Therefore, 37°C and 12 minutes were determined to be the optimal combination of temperature and time for the reduction reaction.

[0123] This embodiment, through systematic experimental screening, determined that hydroxylamine hydrochloride (HA) is a superior reducing agent to ascorbic acid (AA), with an optimal working concentration of 4.2 M. This preferred method ensures that the Fe content in the atmospheric particulate matter extract is within acceptable limits. 3+ The reduction was rapid and complete, resulting in accurate total iron determination and validating the technical characteristics of the optimized reducing agent. Simultaneously, the optimal temperature for the reduction reaction was determined to be 37°C, and the optimal incubation time to be 12 minutes. This combination of parameters provides crucial assurance for the reliability and reproducibility of the entire analytical method, ensuring the accurate determination of Fe. 3+ The complete reduction and avoidance of interference from excess reagents support the constraints on isothermal reaction conditions.

[0124] Example 4: Evaluation of Nitrogen Protection Effect and Comprehensive System Performance Test

[0125] This embodiment aims to quantitatively evaluate the effectiveness of nitrogen protection systems in preventing Fe2+ gas leaks. 2+ The device plays a key role in oxidation and improving the accuracy of the measurement. The comprehensive performance indicators of the device are systematically measured, including the method detection limit, analytical throughput and long-term operational stability, in order to fully verify its reliability in practical applications.

[0126] 1. Experimental Materials and Methods

[0127] Sample preparation: Preparation of Fe 2+ Standard solution (1.0 mg·L) -1 Prepare for actual atmospheric PM2.5 levels. 2.5 Extract sample.

[0128] The system configuration and comparison settings are as follows:

[0129] Nitrogen protection conditions: Nitrogen protection is activated throughout the entire system process, and standard vacuum-nitrogen replacement is performed while maintaining a slight positive pressure.

[0130] Non-nitrogen conditions: Turn off the nitrogen protection system and do not perform nitrogen protection.

[0131] The remaining device parameters are exactly the same as in Example 1.

[0132] 2. Experimental Procedures and Results

[0133] 2.1 Verification of Nitrogen Protection Effect

[0134] Comparative experiment: Under nitrogen protection and without nitrogen protection, using the same apparatus, the above Fe was tested respectively. 2+ The standard solution and the actual sample were measured repeatedly (n=6).

[0135] Results and Analysis: As shown in Appendix 9, under non-nitrogen conditions, due to Fe... 2+ Oxidized by oxygen in the air, its measured value is systematically lower than that measured under nitrogen protection. Calculations show that Fe... 2+ The specific deviation range of the measurement results is between 8.11% and 13.71%, and the deviation of actual samples is usually greater than that of standard solutions. Therefore, the nitrogen protection system can effectively isolate oxygen and prevent Fe from being released. 2+ Oxidation improved the accuracy of the determination by more than 8.11%, verifying the effectiveness of the vacuum-nitrogen replacement technique in improving detection.

[0136] 2.2 Method Detection Limit and Throughput Determination

[0137] Limit of Detection (MDL): Following the method recommended by the International Union of Pure and Applied Chemistry (IUPAC), 11 consecutive measurements of the blank matrix solution were performed, and the standard deviation (σ) of the absorbance was calculated. The method limit of detection (MDL) was calculated using the formula MDL = 3.143 × σ × C / A (where C is the concentration of the low-concentration standard, and A is its average absorbance), and converted to atmospheric concentration. The results showed that the limit of detection (MDL) of this method was 0.022 ng·m⁻¹. -3 (Based on standard sampling volume).

[0138] System throughput determination: The continuous operation unit processed a batch containing 12 samples, and the total time from the injection of the first sample to the completion of data output for the last sample was recorded. Results showed that the system successfully completed the full analysis of all 12 samples (containing Fe). 2+ The total time for the determination of total iron and cleaning is approximately 60 minutes, meaning the system throughput can reach 12 samples per hour.

[0139] 2.3 Long-term operational stability and automatic cleaning effect

[0140] Test method: The apparatus was run continuously for 8 hours, processing over 90 samples. A quality control sample of known concentration was inserted every 10 samples, and baseline noise was monitored. After the run, a high-concentration sample (2.0 mg·L⁻¹) was used. -1 After the test, a blank solution was immediately injected, and its absorbance was measured to assess the residue.

[0141] Results and Analysis: The RSD of all quality control sample measurements was <2.5%, indicating that the system remained stable during long-term operation. Residual testing showed that the concentration corresponding to the increase in absorbance of the blank solution was <0.01 mg·L⁻¹. -1 The calculated residual rate is <0.5%. Therefore, the automatic cleaning module can effectively control cross-contamination, with a residual rate of less than 0.5%, ensuring the stability and data reliability of the system during long-term operation at high throughput (12 samples / hour).

[0142] This embodiment confirms that the nitrogen protection system is the core guarantee for accurate measurement in this device, effectively preventing Fe... 2+ Oxidation improves detection accuracy by 8.11%-13.71%. The device exhibits high sensitivity (detection limit 0.022 ng·m⁻²). -3 High throughput and high efficiency (12 samples / hour). The automatic cleaning module ensures the stability and reliability of the system in batch analysis (residual rate <0.5%), which is the foundation for achieving the aforementioned high throughput and high precision.

[0143] Example 5: Actual Atmospheric PM 2.5 Sample application verification

[0144] This embodiment aims to verify the reliability, stability, and overall workflow effectiveness of the semi-automatic iron speciation analyzer in practical applications by analyzing real-world environmental samples, and to evaluate the accuracy and uncertainty of its measurement results.

[0145] 1. Experimental Materials and Methods

[0146] Sample source: Atmospheric PM2.5 collected from urban environmental monitoring stations 2.5 Filter membrane sample (the sampling membrane is a quartz fiber filter membrane, and the sampling volume is a standard cubic meter).

[0147] Pretreatment: Using the sample extraction module of this device, place 1 / 8 of a filter membrane in a sealed extraction container and add 20.0 mL of ultrapure water. Turn on the nitrogen protection system and continuously purge the upper space of the extraction container of the sample extraction module with high-purity nitrogen (flow rate 0.5 L / min) for 5 minutes to replace the air. Then, ultrasonically extract for 30 minutes under nitrogen cover. After filtering through the 0.22 µm filter head built into the container, the extract is directly introduced into the analytical flow path.

[0148] System configuration: The device configuration and optimized parameters are exactly the same as those in Example 1, including a reaction temperature of 37°C, an incubation time of 12 minutes, and 4.2 M hydroxylamine hydrochloride as a reducing agent.

[0149] 2. Experimental Procedure

[0150] Set a complete analysis sequence in the control system software program to automatically execute the following process on the same extract:

[0151] Fe 2+ Determination: 1.5 mL of the extract was quantitatively extracted using the automated sampler, mixed with Ferrozine chromogenic reagent and buffer, and after incubation, the absorbance was measured at 562 nm to obtain water-soluble Fe. 2+ concentration.

[0152] Total soluble iron determination: Fe 2+ After measurement, without cleaning, 150 µL of 4.2 M hydroxylamine hydrochloride solution is automatically added to the same in-orbit sample, followed by another reaction incubation to remove all Fe from the sample. 3+ After reduction, the absorbance of total iron was measured to obtain the total soluble iron concentration.

[0153] Fe 3+ Calculation: The control system automatically calculates Fe according to the formula. 3+ = Total soluble iron - Fe 2+ Calculate and record water-soluble Fe 3+ concentration.

[0154] Automatic cleaning: After the measurement is completed, the standard automatic cleaning procedure is executed.

[0155] Each sample was measured in triplicate, and a standard reference material (NIST 1648a urban particulate matter) was analyzed simultaneously for quality control.

[0156] 3. Results and Discussion

[0157] Measurement results: Actual PM 2.5 Water-soluble Fe was measured in the sample. 2+ The concentration was (0.15 ± 0.01) μg / m³. 3 Water-soluble Fe3+ The concentration was (0.38 ± 0.02) μg / m³. 3 The total water-soluble iron concentration was (0.53 ± 0.02) μg / m³. 3 Fe 2+ The / Fetotal ratio is approximately 0.28. The determination results of the standard reference material are within the uncertainty range of its certified value.

[0158] Uncertainty Assessment: A comprehensive relative uncertainty assessment of the measurement results was performed, taking into account factors such as calibration curves, instrument readings, injection volume, and sample homogeneity. The calculations showed that the comprehensive relative uncertainty of this method for actual samples was <5%, indicating that the measurement results have high reliability.

[0159] System operating status: Throughout the entire actual sample analysis process, as shown in the attached... Figure 6 As shown, the device has a compact structure, reasonable module connections, stable operation, and convenient maintenance. From the start of sample extraction to the end of data output, the entire process is seamlessly automated with nitrogen protection, successfully verifying the feasibility and efficiency of the complete workflow of extraction → nitrogen protection → automatic sample injection → reaction → detection → automatic cleaning → data output.

[0160] This embodiment successfully applies the semi-automatic analysis device to real atmospheric PM2.5. 2.5 The sample yielded reliable Fe 2+ Fe 3+ The concentration and ratio data, with a comprehensive relative uncertainty of less than 5%, fully demonstrate the accuracy and reliability of the device in practical applications. This embodiment comprehensively demonstrates and verifies the entire process of the technical solution of this invention, from sample pretreatment to final result calculation, showcasing the synergistic advantages of nitrogen protection, automated control, and modular integration. It solves the long-standing problems of oxidation interference, cumbersome operation, and poor repeatability in the analysis of iron speciation in atmospheric particulate matter.

[0161] Example 6: Intelligent Nitrogen Protection and Predictive Anti-oxidation Control

[0162] This embodiment aims to verify the feasibility and effectiveness of the intelligent nitrogen protection scheme. Specifically, it will evaluate the comprehensive performance of real-time dissolved oxygen monitoring and feedback control, combined with sensor self-calibration and adaptive predictive control algorithms, in improving the system's reliability against oxygen interference and achieving proactive protection.

[0163] 1. System Configuration and Optimization

[0164] Based on the device described in Example 1, the nitrogen protection system is upgraded as follows:

[0165] Hardware Upgrade: A miniature fluorescence quenching dissolved oxygen sensor (model: FireSting-O2, PyroScience) with a detection limit of 1 ppb and a response time of <5 seconds is integrated at the entrance of the reaction chamber. The dissolved oxygen sensor is connected to the control system, and a feedback control unit and a sensor self-calibration and status diagnosis module are added to the control system.

[0166] The control strategy is set as follows: the feedback control logic sets the first preset threshold to 10 ppb (triggering enhanced purging) and the second preset threshold to 2 ppb (switching to sustain mode). The base nitrogen flow rate is 0.5 L / min, and the enhanced purging flow rate is 1.0 L / min.

[0167] Configure an adaptive predictive control algorithm: Predictive control is implemented using a linear autoregressive exogenous model based on historical data, specifically including:

[0168] Model construction: The model uses the dissolved oxygen concentration (O2) at future time (t+T) t+T Expressed as a linear combination of dissolved oxygen observations and valve switching events over a past period: O2 t+T =a1·O2 t +a2·O2 t-1 +b·ValveEvent t +c, where T is the prediction step size (set to 80ms). O2 t and O2 t-1 These represent the dissolved oxygen concentration values ​​actually measured by the sensor at the current time t and the previous time t-1, respectively, serving as the historical inputs to the model. ValveEvent t The variable 't' represents whether a valve switching event occurred at time t. This is an exogenous input variable, typically represented by "0" and "1" in the model. For example, "0" represents no event, and "1" represents a valve switching command being issued. This variable is used to capture the instantaneous disturbance of the control action to the system state. 'a1', 'a2', and 'b' are the core parameters (weighting coefficients) of the model. 'a1' and 'a2' represent the degree of influence of historical dissolved oxygen concentration on future predicted values. They quantify the inertia or autocorrelation characteristics of the system's oxygen concentration. 'b' represents the contribution of the valve switching event to the predicted future oxygen concentration. A positive 'b' value indicates that the valve event typically introduces oxygen, leading to an increase in concentration. 'c' represents the constant term (intercept) of the model. It represents the background oxygen concentration level of the system, or the baseline value of the combined influence of all other unconsidered factors in the model. This model is an empirical model based on data fitting. The coefficients 'a1' and 'a2' are dimensionless weights, the coefficient 'b' has the same dimension as dissolved oxygen concentration, and the constant term 'c' also has the dimension of concentration.

[0169] Parameter determination: During the initial installation of the system, dissolved oxygen time series data and valve event markers are collected by performing 20-30 normal analysis operations. The optimal model parameter set [a1,a2,b,c] is automatically fitted using the least squares system identification method and then fixed to the control unit.

[0170] Workflow: When a valve switching command is detected, the algorithm immediately predicts the oxygen concentration for the next 80ms based on the current and historical dissolved oxygen readings. If the predicted value exceeds the preventive threshold (8 ppb), the nitrogen flow rate is pre-increased by 30% before the valve actually operates.

[0171] Self-calibration process: During system initialization, zero-oxygen solution (prepared with sodium dithionite) and air-saturated standard solution are automatically injected to complete the two-point calibration of the sensor and the diagnosis of the response curve.

[0172] 2. Experimental Procedures and Results

[0173] 2.1 Intelligent Nitrogen Protection Effect Test

[0174] Method: Prepare low concentration Fe 2+ Standard solution (0.05 mg·L) -1 The measurements were performed in three modes: (A) constant nitrogen flow mode; (B) real-time feedback control mode; and (C) predictive feedback control mode (with adaptive predictive control algorithm enabled). Each group was measured 10 times.

[0175] The results are as follows Figure 10 and 11 As shown:

[0176] Mode A: The peak dissolved oxygen concentration at the moment of injection reached 15 ppb, and the RSD of the measurement result was 3.5%.

[0177] Mode B: Feedback control suppresses the peak dissolved oxygen concentration to <5 ppb, reducing the measured RSD to 2.5%.

[0178] Model C: The prediction algorithm further suppressed the peak dissolved oxygen concentration to <3 ppb, significantly reducing the measured RSD to 1.8%, and Fe 2+ The average measured value was about 4.5% higher than that of the A mode.

[0179] 2.2 Long-term stability verification of the system

[0180] The system operated continuously for 72 hours in predictive feedback control mode, processing over 500 samples. The system automatically performed sensor self-calibration after analyzing every 50 samples.

[0181] The self-calibration module successfully detected a slight drift in the sensor signal during operation and automatically completed software correction. The predictive feedback control mode remained stable throughout the test, without any prediction inaccuracies. Figure 12 As shown, the system treats Fe 2+ The recovery rate remained within a narrow range of 96.5%-98.0%.

[0182] This embodiment achieves a leap from constant-flow oxygen isolation to on-demand responsive oxygen isolation by introducing real-time dissolved oxygen monitoring and feedback control. The adaptive predictive control algorithm, through model learning of the system's dynamic characteristics, achieves a transition from passive response to active prediction and intervention, effectively controlling low-concentration Fe... 2+ The measurement precision was further optimized to 1.8%. The sensor self-calibration and condition diagnosis module ensures the long-term reliability of the monitoring data.

[0183] Example 7: Testing of Samples with Different Redox Properties

[0184] This embodiment evaluates how to improve the accuracy and robustness of analysis by dynamically optimizing nitrogen protection strategies through real-time detection of the oxygen behavior of actual atmospheric particulate matter samples with different redox properties.

[0185] 1. Experimental Materials and Methods

[0186] Prepare the following types of samples:

[0187] Category A (reducing) samples: PM2.5 samples collected from areas significantly affected by biomass burning. 2.5 The filter membrane is expected to extract a large amount of reducing organic matter.

[0188] Category B (Neutral / Inert) Samples: PM2.5 collected from a clean control site in the suburbs. 2.5 Filter membrane.

[0189] Category C (potentially catalytic) samples: collected from PM2.5 near industrial areas. 2.5 The filter membrane is expected to contain catalytic components such as transition metals.

[0190] All samples were extracted under nitrogen protection according to the method described in Example 5.

[0191] System configuration: The sample-specific oxygen behavior adaptive compensation algorithm has been further embedded in the program of the feedback control unit in Example 6.

[0192] 2. Experimental Procedures and Results

[0193] 2.1 Monitoring and Compensation Validation of Sample-Specific Oxygen Behavior

[0194] Methods: Samples of categories A, B, and C were measured separately. In each analytical cycle, as the sample was drawn by the autosampler and flowed through the dissolved oxygen sensor at the inlet of the reaction chamber before being mixed with any reagents, the sensor recorded the instantaneous change in dissolved oxygen concentration over approximately 3 seconds at a frequency of 10 Hz. The system automatically analyzed the characteristics (trend and slope) of this change curve and determined the dynamic compensation factor K accordingly.

[0195] The results are as follows Figure 13 As shown:

[0196] Type A (reducing) samples: During the pure sample delivery stage, they exhibit a rapid decreasing trend (slope < -0.5 ppb / s). The system classifies them as "oxygen-consuming" samples and automatically sets the compensation factor K = 0.7. This means that even if the prediction model suggests strengthening purging due to system background, the final nitrogen flow rate adjustment will be reduced by 30%, avoiding unnecessary nitrogen waste and potential disturbances to the reaction system.

[0197] Type B (neutral) samples: Dissolved oxygen concentration remained stable (absolute slope < 0.1 ppb / s). The system was set with compensation factor K = 1.0, fully conforming to the original output of the predictive feedback control mode in Example 6.

[0198] Category C (potentially catalytic) samples: Dissolved oxygen concentration exhibits abnormal fluctuations or a slight increase (slope > 0.2 ppb / s). The system classifies these samples as "oxygen risk" samples, automatically sets a compensation factor K = 1.4, and increases the predetermined nitrogen-enhanced purging intensity by 40%, providing stronger suppression of potential rapid oxidation reactions.

[0199] 2.2 Evaluation of Compensation Effectiveness

[0200] Methods: For samples of types A, B, and C, low concentrations of Fe were measured in two modes: with compensation function (experimental group) and with compensation function disabled (control group, predictive feedback control mode only in Example 6). 2+ (0.05 mg·L) -1 Spike recovery and repeatability (n=6).

[0201] The results are as follows Figure 14 The results show that for Class A (reducing) samples, the average recovery was 101.5% in the uncompensated mode with an RSD of 3.8%, and 98.9% in the compensated mode with an RSD of 2.1%. For Class B (neutral) samples, the average recovery was 98.2% in the uncompensated mode with an RSD of 2.0%, and 98.0% in the compensated mode with an RSD of 1.9%. For Class C (catalytic) samples, the average recovery was 94.5% in the uncompensated mode with an RSD of 4.5%, and 96.8% in the compensated mode with an RSD of 2.5%.

[0202] For Class A samples, the recovery rate was higher and the RSD was larger in the uncompensated mode, possibly because excessive nitrogen purging slightly affected the reaction kinetics or signal stability. After enabling compensation, the recovery rate was closer to the true value, and the precision was significantly improved. For Class B samples, the difference between the two modes was not significant, indicating that the compensation algorithm had little interference with normal samples. For Class C samples, the recovery rate was significantly lower and the precision was poor in the uncompensated mode, indicating the presence of oxidation interference. After enabling compensation, oxidation was effectively suppressed through enhanced protection, and both the recovery rate and precision were significantly improved. Therefore, by introducing sample-specific oxygen behavior adaptive compensation, this invention successfully solved the problem of the predictive control model being insensitive to oxygen behavior perturbations introduced by the sample itself.

[0203] Example 8: Automatic Determination of Sample Concentration Range

[0204] This embodiment evaluates whether the method can automatically identify exceeding the limit, perform precise dilution, and complete accurate measurement when encountering high-concentration samples, thereby improving the automation and reliability of the method when dealing with actual samples with large concentration range fluctuations.

[0205] 1. Experimental Materials and Methods

[0206] Sample preparation: Ultra-high concentration Fe 2+ Standard solution: Prepared with a concentration of 8.0 mg·L⁻¹ -1 Fe 2+ The standard solution had a concentration much higher than the upper limit of the calibration curve established in Example 1 (2.5 mg·L⁻¹). -1 The actual high-concentration samples were selected from PM2.5 samples collected near industrial sources with known high iron content. 2.5 The filter membrane sample was extracted according to the method in Example 5. The upper limit threshold of the linear range was preset in the control software to correspond to an absorbance of 2.0 AU, and the automatic dilution factor was set to 5 times, with ultrapure water as the diluent.

[0207] 2. Experimental Procedures and Results

[0208] Method and steps: The above 8.0 mg·L -1 Fe 2+ The standard solution is placed as an unknown sample in the automated sample introduction sequence. After the system executes steps S1 and S2 according to the normal procedure, detection is performed in step S3.

[0209] The detection module detected that the absorbance value of the sample momentarily exceeded 2.0 AU. The control system immediately triggered the automatic range judgment logic, determining that the sample concentration exceeded the limit. The system then paused the current analysis sequence and recorded the sample tag number, absorbance value, and the fact that it exceeded the linear range, triggering the automatic dilution process. Simultaneously, the system interface provided a clear prompt message, but no manual intervention was required; subsequent steps were executed automatically.

[0210] 2.2 Validation of Automated Dilution and Re-testing

[0211] Method: The system performed automatic dilution according to a preset procedure. The syringe pump A of the automatic sample injection module precisely drew 1.0 mL of the original ultra-high concentration sample. Subsequently, 4.0 mL of ultrapure water was drawn from the diluent bottle and mixed with the sample in the flow path, achieving a 5-fold precise quantitative dilution to obtain a theoretical concentration of approximately 1.6 mg·L⁻¹. -1 The sample solution to be analyzed. The diluted sample is automatically inserted at the front of the analysis queue, and the complete analysis starting from S2 is performed first (to avoid residue, S1 is skipped but a nitrogen environment is maintained).

[0212] The diluted sample measured an absorbance of 0.598 AU, falling within the ideal linear range of the calibration curve. The control system automatically calculated the concentration based on a dilution factor of 5, ultimately reporting a concentration of 8.05 mg·L⁻¹. -1 The concentration of this ultra-high concentration standard solution (8.0 mg·L⁻¹) was compared with that of the standard solution. -1 Compared to the previous method, the relative error was 0.63%, demonstrating the accuracy of the automatic dilution and calculation.

[0213] 2.3 Effectiveness Evaluation

[0214] Prepare a batch of sequences containing 10 samples, with 3 of the aforementioned high-concentration samples randomly inserted. Analyze the entire sequence using both the automatic dilution function and manual processing (requiring manual detection of exceeding limits, interruption of the program, manual dilution, and reloading of samples). Record the total time and the number of manual operations required.

[0215] The manual processing mode requires 95 minutes and at least 3 manual operations (intervention, dilution, and reloading). The automatic dilution mode requires 65 minutes and requires no manual operations. It is evident that by introducing automatic range judgment and dilution functions, this invention successfully solves the efficiency bottleneck and data reliability problems caused by sample concentrations exceeding the preset range in automated analysis processes. It not only ensures the accuracy of high-concentration sample determination results but also significantly improves overall analytical efficiency.

[0216] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. It can be applied to various fields suitable for the present invention. Further modifications can be readily implemented by those skilled in the art.

Claims

1. A device for semi-automatic analysis of iron forms of atmospheric particulate matter under nitrogen protection, characterized in that, The device constitutes a continuous analysis system operating under nitrogen protection, comprising: a sample extraction module for receiving and processing atmospheric particulate matter samples and outputting iron-containing extraction solution; an autosampler module connected to the output of the sample extraction module through a first flow path for receiving the iron-containing extract and for quantitative introduction of Ferrozine color reagent, buffer, and a reducing agent for reduction of Fe 3+ to Fe 2+ in the iron-containing extract to form a mixture to be measured; a reaction chamber connected to the output end of the automatic sampling module through the second flow path, for receiving the mixed solution to be tested and incubating under constant temperature conditions to complete the color development reaction of Fe 2+ and the reduction reaction of Fe 3+ ; a detection module connected to the output end of the reaction chamber through a third flow path, for real-time optical detection of the reacted liquid, specific determination of Fe 2+ absorbance of the Ferrozine complex at a wavelength of 562 nm; an automatic cleaning module for flushing the flow paths through the automatic sampling module, the reaction chamber and the detection module after detection is completed to prepare for the next analysis; a nitrogen protection system connected by gas paths and covering the entire flow path system from the sample extraction module to the detection module, for vacuum-nitrogen replacement of the flow path system before analysis is performed, vacuum pumping and then filling with nitrogen for several times to completely remove oxygen; and continuously introducing nitrogen into the system to maintain a positive pressure nitrogen atmosphere during analysis to isolate oxygen interference; a control system connected in signal with the sample extraction module, the automatic sampling module, the reaction chamber, the detection module, the automatic cleaning module and the nitrogen protection system, for unified coordination of instructions, execution of preset analysis processes and collection, processing and storage of detection data; The nitrogen protection system further comprises a dissolved oxygen monitoring subsystem comprising at least one micro-dissolved oxygen sensor arranged at the inlet of the reaction chamber and a feedback control unit electrically connected with the dissolved oxygen sensor and the nitrogen flow control valve of the nitrogen protection system; the feedback control unit is configured to dynamically adjust the nitrogen supply flow rate of the nitrogen protection system according to the dissolved oxygen concentration data monitored by the dissolved oxygen sensor in real time; specifically, when the dissolved oxygen concentration is higher than a first preset threshold of 10 ppb, the nitrogen supply flow rate is controlled to be increased to perform a strengthened purging flow rate of 1.0 L / min, and when the dissolved oxygen concentration is lower than a second preset threshold of 2 ppb, the nitrogen supply flow rate is controlled to be reduced to a maintenance level flow rate of 0.5 L / min; The feedback control unit employs an adaptive predictive control algorithm based on historical data, which is configured to: establish a dynamic model of the system background oxygen permeation according to the recorded dissolved oxygen concentration variation trend in a plurality of consecutive analysis periods, the dynamic model being configured to predict the dissolved oxygen concentration O2 t+T at a future time (t+T) based on the dissolved oxygen concentration O2 t+T =a1·O2 t +a2·O2 t-1 +b·ValveEvent t +c, where the prediction step T=80 ms, O2 t and O2 t-1 represent the actual measured concentration values at the current time t and the previous time t-1 by the dissolved oxygen sensor, as historical inputs to the model, ValveEvent t represents whether a valve switching event has occurred at time t, a1, a2, b are core parameters of the model, wherein the coefficients a1, a2 are dimensionless weights, the coefficient b has the same dimension as the dissolved oxygen concentration; c represents a constant term of the model, which is the background oxygen concentration level of the system, or the reference value of the combined influence of all other factors not explicitly considered in the model, and also has the same dimension as the concentration dimension; in subsequent analysis, the potential oxygen concentration fluctuation at the key node is predicted based on the dynamic model, and the nitrogen flow rate is adjusted in advance before the sensor threshold trigger condition.

2. The apparatus for semi-automatic analysis of iron speciation of atmospheric particulate matter according to claim 1, characterized in that, The automatic sampling module comprises a programmable twelve-way valve and a six-way valve; the flow path interface of the twelve-way valve is configured to be connected to the output end of the sample extraction module, the cleaning liquid inlet, the standard solution bottle, the buffer bottle, the reducing agent bottle, the input end of the reaction chamber and the flow cell inlet of the detection module; the flow path interface of the six-way valve is configured to be connected to the nitrogen transmission container, the cleaning liquid inlet, the reaction chamber and the Ferrozine reagent bottle.

3. The apparatus for semi-automatic analysis of iron speciation of atmospheric particulate matter according to claim 2, characterized in that, The automatic sampling module further comprises at least two syringe pumps, including a 10 mL syringe pump and a 100 µL syringe pump, for realizing quantitative transfer of samples and reagents.

4. The apparatus for semi-automatic analysis of iron speciation of atmospheric particulate matter according to claim 1, wherein, The reaction chamber comprises a polytetrafluoroethylene reaction tube placed in a constant temperature jacket, the constant temperature condition of which is controlled to be 37±0.5℃, and the incubation time of the sample is 12 minutes.

5. The apparatus for semi-automatic analysis of iron speciation in atmospheric particulate matter according to claim 1, characterized in that, The detection module comprises a quartz glass flow-through cuvette with an optical path of 10 mm, and a detection system composed of an LED light source with a central wavelength of 562 nm and a silicon photodiode detector, with an absorbance detection linear range of 0 to 2 AU.

6. The apparatus for semi-automatic analysis of iron speciation in atmospheric particulate matter according to claim 1, characterized in that, The reducing agent is a 4.2M hydroxylamine hydrochloride solution or a 5 mM ascorbic acid solution.

7. The apparatus for semi-automatic analysis of iron speciation in atmospheric particulate matter according to claim 1, characterized in that, Three cycles of vacuum pumping to -0.08 MPa and filling with pure nitrogen to 0.05 MPa are performed on the entire flow path to completely replace the air in the system; then a nitrogen micro-positive pressure of 0.02 MPa is maintained throughout the analysis process.

8. The apparatus for semi-automatic analysis of iron speciation of atmospheric particulate matter according to claim 7, characterized in that, The dissolved oxygen monitoring subsystem further comprises a sensor self-calibration module connected to the feedback control unit, which automatically performs the following operations during the system initialization phase before each analysis cycle: inject a standard oxygen-saturated solution with known concentration into the flow path; compare the reading of the micro-dissolved oxygen sensor with the standard value, and automatically perform software calibration when the deviation exceeds the allowed range, and record the calibration log; diagnose the sensor state by analyzing the response curve of the sensor in zero-oxygen and standard-oxygen environments, and issue a warning when the sensor performance degrades; When the valve switching instruction is detected, the oxygen concentration in the next 80 ms is predicted based on the current and historical dissolved oxygen readings, and if the predicted value exceeds the preventive threshold of 8 ppb, the nitrogen flow rate is pre-increased by 30% before the actual action of the valve.

9. The apparatus for semi-automatic analysis of iron speciation of atmospheric particulate matter according to claim 8, characterized in that, The feedback control unit is connected to the control system and is further configured to monitor the instantaneous change curve of the dissolved oxygen concentration during the transportation of the pure sample liquid in the high-frequency sampling mode after the automatic sampling module quantitatively extracts the sample, but before the reagent mixing process; The feedback control unit analyzes the characteristics of the change curve in real time, including the rising / falling / stable trend and the change rate; based on the analysis results of the characteristics, a dynamic compensation factor K is applied to the nitrogen flow rate adjustment amount output by the prediction model in the subsequent reaction stage: if the oxygen concentration is rapidly reduced by the sample liquid itself, the slope is <-0.5 ppb / s, K<1 is set, the predetermined intensive purging intensity is appropriately weakened, and the nitrogen flow rate adjustment amount is weakened by 30%, avoiding excessive nitrogen; if the oxygen concentration is abnormally increased or fluctuates violently by the sample liquid, K>1 is set, the predetermined nitrogen intensive purging intensity is increased by 40%, and the predetermined nitrogen protection intensity is enhanced; If the oxygen concentration is stable, the absolute value of the slope is <0.1 ppb / s, K=1 is set, and the original prediction control output is maintained.

10. A method for semi-automatic analysis of iron speciation using the apparatus for semi-automatic analysis of iron speciation according to any one of claims 1 to 9, characterized in that, The method comprises the following steps: S1: system initialization, vacuum-nitrogen replacement of the entire flow path by the nitrogen protection system, and system self-checking and baseline stability detection; S2: automatically transport blank solution, standard solution or sample solution according to the program by the automatic sampling module, and mix with Ferrozine color developing agent and buffer solution, and transport to the reaction chamber; S3: After incubation in the reaction chamber, transport to the detection module, determination of the absorbance at a wavelength of 562 nm, resulting in Fe 2+ concentration; S4: Add reducing agent by automatic sample injection module, reduce Fe in sample to Fe 3+ S2: Add acid to sample to dissolve Fe 2+ S3: Add reducing agent to sample to reduce Fe to Fe 3+ S2: Add acid to sample to dissolve Fe S3: Add reducing agent to sample to reduce Fe to Fe S5: perform the programmed cleaning process by the automatic cleaning module, and store the analysis data by the control system and prepare for the next analysis cycle.

Images