A near-infrared flow-based upconversion nanoparticle dispersity detection method and device

Abstract

The application discloses a kind of upconversion nanoparticle dispersity detection method and device based on near-infrared flow, belong to nanometer material fluorescence characterization and optical detection field.980nm near-infrared continuous laser collimation is expanded, and micro excitation area is formed in 10 μm inner diameter quartz microtube by high NA objective lens focusing;Constant-speed injector pushes the sample to be measured to flow through the region, and excitation fluorescence is collected by the same objective lens, and single-photon counting sequence is obtained by SPAD multichannel acquisition after dichroic mirror, optical filter spectrometer. Data processing module extracts single-particle fluorescence burst multidimensional feature by relying on physical model, and gets agglomeration rate and dispersity index by cluster analysis. Device includes excitation light generation, microfluidic sampling, light collection spectrometer, single-photon detection module. The application can be liquid in-situ high-throughput real-time detection, avoid drying sample artifact, realize nanometer particle dispersion state high-precision multidimensional quantitative evaluation.

Description

Technical Field

[0001] This invention relates to the field of fluorescence characterization and optical detection of nanomaterials, specifically to a method and apparatus for detecting the dispersibility of upconversion nanoparticles based on near-infrared flow cytometry. Background Technology

[0002] Upconversion nanoparticles (UCNPs), as a novel type of luminescent nanomaterial, have wide applications in fields such as bioimaging, anti-counterfeiting, and single-molecule detection. In practical applications, the dispersibility of nanoparticles in the liquid phase directly determines the stability of their luminescent properties and the effectiveness of their applications.

[0003] Currently, commonly used methods for detecting the dispersibility of nanoparticles include dynamic light scattering (DLS) and transmission electron microscopy (TEM). However, DLS is too sensitive to the scattered light from large aggregates, easily masking the true signal of small particles, and it is difficult to combine the fluorescence and luminescence characteristics of the particles themselves for multi-dimensional evaluation. TEM, on the other hand, requires sample drying and cannot reflect the true flow state and aggregation of particles in solution. Although traditional flow cytometry can achieve high-throughput detection in liquid phase, its excitation spot and flow chamber size are usually designed for micrometer-sized cells, resulting in an excessively large detection area and high background noise, making it difficult to meet the requirements for single-particle-level sensitivity detection of upconversion particles at the tens of nanometer scale.

[0004] To overcome the dual bottlenecks of spatial resolution and sensitivity in traditional optical characterization, there is an urgent need to introduce a detection method with extremely small detection volume and single-photon-level capture capability. An ideal technical solution should be able to extract extremely weak fluorescence pulses from single particles in situ and in real-time within a realistic liquid microfluidic environment, thereby simultaneously achieving high-precision, multi-dimensional quantification of the physical hydrodynamic dimensions and multi-channel optical properties of nanoparticles.

[0005] Therefore, it is necessary to provide a method and apparatus for detecting the dispersibility of upconversion nanoparticles based on near-infrared cytometry to solve the above-mentioned technical problems. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to overcome the above-mentioned technical defects and provide a method and apparatus for detecting the dispersion of upconversion nanoparticles based on near-infrared cytometry, which can achieve high throughput and real-time detection of upconversion nanoparticle dispersion in a liquid environment.

[0007] To solve the above-mentioned technical problems, the technical solution provided by the present invention is: a method for detecting the dispersibility of upconversion nanoparticles based on near-infrared flow cytometry, comprising the following steps:

[0008] S1: A 980nm near-infrared continuous laser beam is input into the optical shaping component for collimation and beam expansion to obtain an excitation beam with uniform energy distribution.

[0009] S2: The excitation beam shaped by S1 is input into the high numerical aperture objective lens and focused inside the flow cytometry microtube with an inner diameter of 10 micrometers, forming an excitation region of extremely small volume at the center of the flow cytometry microtube.

[0010] S3: Using a constant-speed injector, the sample solution containing upconversion nanoparticles is propelled through the excitation region in the flow microtube at a constant speed. The sample particles are excited by a 980nm laser to generate a fluorescence signal.

[0011] S4: The fluorescence signal generated by the excitation is collected and returned by the same objective lens, separated by a dichroic mirror for the main optical path, and the fluorescence signals of different wavelengths are split into different detection channels by a combination of filters.

[0012] S5: Fluorescence signals from different channels are received by the corresponding single-photon avalanche diodes and converted into single-photon counting sequences. The digital signals are input into the data processing module, and the time series data of single-particle fluorescence bursts are analyzed by the algorithm pipeline combined with the physical model to calculate the aggregation rate and dispersion of the sample.

[0013] Furthermore, the specific steps in S5 for the algorithm to analyze sample dispersion include:

[0014] Step 1: Signal preprocessing and denoising: Poisson denoising and smoothing is performed on the photon counting time series output by the single-photon avalanche diode, and an adaptive background threshold is set in combination with the pure solvent background measurement value.

[0015] Step 2: Dynamic peak finding based on long lifetime characteristics: In view of the asymmetric signal tailing caused by the long lifetime characteristics of rare earth upconversion luminescence, a dynamic threshold determination method is adopted to extract continuous pulse signals above the background threshold as effective particle events.

[0016] Step 3: Multidimensional physical feature extraction: Based on fluid dynamics and Gaussian beam distribution theory, a physical model is constructed to extract the feature parameters of each burst pulse, including at least: the peak area reflecting the number of luminous centers of a single particle, and the pulse width reflecting the hydraulic diameter and residence time of the particle as it passes through the excitation region.

[0017] Step 4: Cluster Analysis and Dispersion Quantification Output: The extracted multidimensional feature data is input into the clustering algorithm for scatter projection to distinguish between the core main group and the discrete group, and finally outputs the dispersion evaluation index.

[0018] Furthermore, the third step of multidimensional physical feature extraction also includes multi-channel ratio analysis: calculating the integral area ratio of luminescence channels in different wavelength bands;

[0019] By utilizing the phenomenon that the cross-relaxation effect caused by aggregation changes the spectral proportions, we can achieve cross-verification of sample dispersibility and eliminate interference from impurities.

[0020] Furthermore, the formula for calculating the dispersion evaluation index DI in S5 is as follows:

[0021]

[0022] This represents the number of single-particle events obtained from cluster analysis. The table shows the total number of events detected.

[0023] The higher the DI value, the better the sample dispersion and the lower the aggregation rate.

[0024] The lower the DI value, the worse the sample dispersion and the higher the aggregation rate.

[0025] Furthermore, when the sample to be tested is a multi-component mixed upconversion nanoparticle sample, different types of nanoparticles are distinguished by using signals from different emission bands acquired through multiple channels, and the dispersibility assessment of each component is completed separately.

[0026] Furthermore, the upconversion nanoparticles to be tested are water-soluble after ligand replacement treatment, making them suitable for microfluidic liquid phase detection environments.

[0027] A near-infrared flow cytometry-based upconversion nanoparticle dispersion detection device, applied to the above detection method, includes an excitation light generation module, a microfluidic sample introduction module, a light collection and dispersion module, and a single-photon detection module;

[0028] The excitation light generation module includes a 980nm near-infrared continuous laser and a collimating beam expander.

[0029] The microfluidic injection module includes a constant-speed injection pump and a quartz flow microtube with an inner diameter of 10 micrometers. The flow microtube is used to provide a microchannel for single-particle-level queued flow.

[0030] The light collection and beam splitting module includes a confocal high numerical aperture objective lens, a first high-reflection, low-transmission dichroic mirror, a second high-reflection, low-transmission dichroic mirror, and a third high-reflection, low-transmission dichroic mirror, as well as a first narrowband filter, a second narrowband filter, and a third narrowband filter.

[0031] The single-photon detection module includes at least three sets of single-photon avalanche diodes and a data acquisition and processing motherboard.

[0032] Furthermore, the single-photon detection module includes a first single-photon photodetector, a second single-photon photodetector, and a third single-photon photodetector;

[0033] Furthermore, the laser beam output by the excitation light generation module, after being collimated, expanded, and filtered by a collimating beam expander and a filter, fills the entrance pupil of the high numerical aperture objective lens, forming a Gaussian spot close to the diffraction limit inside the flow microtube.

[0034] Furthermore, the inner diameter of the flow microtube is 10 μm. Relying on hydrodynamic constraints, the upconversion nanoparticles are arranged in a single row as much as possible to flow through the excitation region, thereby realizing in-situ excitation and detection in the liquid phase.

[0035] The first, second, and third high-reflection, low-transmission dichroic mirrors in the light collection and splitting module split the fluorescence signal according to the wavelength. The split fluorescence signal passes sequentially through the first, second, and third focusing lenses set behind each filter and is then input into the corresponding single-photon avalanche diode, realizing synchronous acquisition of multi-band fluorescence.

[0036] With the above structure, the present invention has the following advantages: 1. The present invention utilizes a 10-micrometer inner diameter microtube in conjunction with an objective lens for focusing, which greatly compresses the excitation volume. Combined with the ultra-high sensitivity of a single-photon avalanche diode (SPAD), it achieves fluorescence capture of a single upconversion nanoparticle, overcoming the limitation of insufficient sensitivity of traditional flow cytometry at the nanoscale.

[0037] 2. This invention uses constant-speed microfluidic drive to enable in-situ detection of nanoparticles in a liquid-phase suspension state, avoiding the agglomeration artifacts caused by traditional electron microscopy drying sample preparation, and truly reflecting the actual dispersion state of the material.

[0038] 3. This invention designs a multi-channel spectroscopic detection optical path. The algorithm can not only rely on the physical width of the photon pulse to determine agglomeration, but also combine the fluorescence intensity ratio of multiple emission channels (such as the green / red ratio) to assist in verification, so as to achieve higher precision multi-parameter detection and classification of samples with overlapping spectra or complex composition. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the structure of the device of the present invention.

[0040] Figure 2 This is a comparison diagram of the fluorescence burst signal characteristics of discrete single particles and aggregates in this invention;

[0041] Figure 3 This is a flowchart illustrating the flow cytometry algorithm for detecting the single-particle dispersion of upconversion nanoparticles in this invention.

[0042] Figure 4 This is a schematic diagram of two-dimensional scatter clustering analysis of fluorescence quenching characteristics in this invention;

[0043] The following labels appear in the diagram: 1. Near-infrared continuous laser; 2. Collimating beam expander; 3. Filter; 4. First high-reflection, low-transmission dichroic mirror; 5. High numerical aperture objective lens; 6. Flow cytometer; 7. Second high-reflection, low-transmission dichroic mirror; 8. First narrowband filter; 9. First focusing lens; 10. First single-photon photodetector; 11. Third high-reflection, low-transmission dichroic mirror; 12. Second narrowband filter; 13. Second focusing lens; 14. Second single-photon photodetector; 15. Third narrowband filter; 16. Third focusing lens; 17. Third single-photon photodetector. Detailed Implementation

[0044] The present invention will now be described in further detail with reference to the accompanying drawings.

[0045] Combined with appendix Figure 1-4 A near-infrared flow cytometry-based upconversion nanoparticle dispersion detection device includes an excitation light generation module, a microfluidic sample introduction module, a light collection and dispersion module, and a single-photon detection module.

[0046] The excitation light generation module includes a near-infrared continuous laser 1, a collimating beam expander 2, and a filter 3;

[0047] The microfluidic sample delivery module includes a flow microtube 6;

[0048] The light collection and beam splitting module includes a first high-reflection, low-transmission dichroic mirror 4, a second high-reflection, low-transmission dichroic mirror 7, a third high-reflection, low-transmission dichroic mirror 11, a high numerical aperture objective lens 5, a first narrowband filter 8, a second narrowband filter 12, and a third narrowband filter 15.

[0049] The single-photon detection module includes a first focusing lens 9, a second focusing lens 13, a third focusing lens 16, a first single-photon photodetector 10, a second single-photon photodetector 14, and a third single-photon photodetector 17.

[0050] Among them, the excitation light generation module is used to generate a continuous near-infrared steady-state laser beam as the excitation light. The near-infrared continuous laser 1 emits a stable laser beam. The collimating beam expander 2 is a biconvex lens. A pinhole filter is set in the collimated beam. After the laser beam is shaped by the collimating beam expander 2, other wavelengths of laser beam are filtered out by the filter 3. The filtered and shaped laser beam fills the entrance pupil of the high numerical aperture objective lens to ensure the focusing quality of the beam.

[0051] Near-infrared laser 1 generates a continuous near-infrared steady-state laser beam output. The laser beam is collimated, expanded, and filtered by collimating and expanding lens 2 and filter 3.

[0052] The infrared laser beam, after being shaped and expanded, passes through the first high-reflection, low-transmission dichroic mirror 4 and fills the entrance pupil of the high numerical aperture objective lens 5. After being focused by the objective lens 5, it forms a near-diffraction-limited Gaussian spot in the flow microtube 6, which excites the sample particles flowing through it.

[0053] By using a high numerical aperture objective lens, the size of the excitation region can be greatly reduced, enabling high-quality and high-precision single-particle detection.

[0054] The fluorescence generated by the sample after excitation by a Gaussian spot is collected by objective lens 5 and separated into fluorescence signals by a first high-reflection, low-transmission dichroic mirror 4, a second high-reflection, low-transmission dichroic mirror 7, and a third high-reflection, low-transmission dichroic mirror 11. The fluorescence signals are then passed through a first narrowband filter 8, a first focusing lens 9, a second narrowband filter 12, a second focusing lens 13, a third narrowband filter 15, and a third focusing lens 16, and received by a first single-photon photodetector 10, a second single-photon photodetector 14, and a third single-photon photodetector 17, achieving preliminary screening of fluorescence signals with different emission wavelengths. After extracting the "pulse width" and "area" features of thousands of single-particle events, the system generates a two-dimensional scatter plot and uses a clustering algorithm to automatically identify the main group (monolithic particles) and the anomalous group (aggregates). Furthermore, by calculating the integral area ratio of different wavelengths (e.g., the green / red light channels in the visible light region), impurity interference can be further eliminated, and cross-validation of the sample can be achieved based on the phenomenon of changes in spectral proportions caused by aggregation-induced cross-relaxation effects. Finally, the data processing module automatically outputs the percentage of "monodisperse particle count / total effective event count," i.e., the dispersibility evaluation index, enabling quantitative and online monitoring of the dispersion state of nanomaterials in suspension.

[0055] A method for detecting the dispersibility of upconversion nanoparticles based on near-infrared flow cytometry, comprising the following steps:

[0056] Step 1: Input a 980nm near-infrared continuous laser beam into the optical shaping component for collimation and beam expansion, input it into the high numerical aperture objective lens, and focus it inside a quartz microtube with an inner diameter of 10 micrometers to form an excitation region of extremely small volume at the center of the microtube.

[0057] Step 2: The sample particles flow through the Gaussian detection area, and the resulting fluorescence is dispersed by a dichroic mirror according to the wavelength difference.

[0058] Step 3: The fluorescence signals from different channels are received by the corresponding single-photon avalanche diodes and converted into single-photon counting sequences. The digital signals are input into the data processing module, and the time series data of single-particle fluorescence bursting is analyzed by the algorithm pipeline combined with the physical model to calculate the aggregation rate and dispersion of the sample.

[0059] The specific steps of the algorithm for analyzing sample dispersion in this invention include:

[0060] Step 1, Signal preprocessing and denoising: The photon counting time series output by the single-photon avalanche diode is subjected to Poisson denoising and smoothing, and an adaptive background threshold is set in combination with the pure solvent background measurement value.

[0061] The second step is dynamic peak finding based on long lifetime characteristics: In view of the asymmetric signal tailing caused by the long lifetime characteristics of rare earth upconversion luminescence, a dynamic threshold determination method is adopted to extract continuous pulse signals above the background threshold as effective particle events.

[0062] Step 3: Multidimensional physical feature extraction: Based on fluid dynamics and Gaussian beam distribution theory, a physical model is constructed to extract the feature parameters of each burst pulse, including at least: the peak area reflecting the number of luminous centers of a single particle, and the pulse width reflecting the hydraulic diameter and residence time of the particle as it passes through the excitation region.

[0063] Step 4: Cluster Analysis and Dispersion Quantification Output: The extracted multidimensional feature data is input into the clustering algorithm for scatter projection to distinguish between the core main group and the discrete group, and finally outputs the dispersion evaluation index.

[0064] The multi-dimensional physical feature extraction in the algorithm of this invention also includes multi-channel ratio analysis: calculating the integral area ratio of emission channels in different bands; utilizing the phenomenon that the cross-relaxation effect caused by aggregation will change the spectral ratio, to achieve cross-verification of sample dispersion and eliminate impurity interference.

[0065] The present invention calculates the aggregation rate and dispersion evaluation index using an algorithm as follows:

[0066]

[0067] in:

[0068] This represents the number of single-particle events obtained through cluster analysis during detection; The DI index indicates the total number of events detected; a higher DI index indicates better dispersibility and lower aggregation rate of the batch of samples; a lower DI index indicates poorer dispersibility and higher aggregation rate of the batch of samples.

[0069] The nanoparticles detected in this invention can be either quantum dots or rare-earth upconversion nanoparticles, or any fluorescent dye. Rare-earth upconversion nanoparticles can be detected by reacting with NOB... The reaction removes the oleic acid ligands on the particle surface, allowing the rare earth upconversion nanoparticles to dissolve in water for application, thus meeting the requirements of flow cytometry experiments.

[0070] A method for detecting the dispersibility of upconversion nanoparticles based on near-infrared flow cytometry includes the following two detection methods:

[0071] S1. During single-sample detection, a near-infrared laser, after passing through a collimating beam expander 2 and a filter 3, is focused by an objective lens 5 to form a Gaussian excitation region, which is then focused into a flow cytometer 6. The upconversion nanoparticles to be tested sequentially pass through the Gaussian excitation region. The fluorescence of the single sample enters a single-photon detector, and is then processed by a flow cytometer... Figure 3 The algorithm shown processes the fluorescence signal.

[0072] S2. When detecting multi-component mixed samples, the system further introduces three-channel fluorescence feature differentiation on top of Gaussian excitation. Specifically, since upconversion nanoparticles with different dopants exhibit spectral differences, the mixed sample is separated by analyzing the fluorescence of the three channels. This method can effectively detect the dispersibility of mixed samples, enabling fine classification and multi-parameter analysis of complex nanoparticle samples. Example 1

[0073] Based on the upconversion nanoparticle dispersibility detection method of near-infrared flow cytometry in the specific embodiment, this example demonstrates the single-particle dispersibility detection data of a single sample.

[0074] Erbium-doped NaYF4:Yb / Er (20 / 2%) upconversion nanoparticles were excited using a continuous near-infrared excitation beam. In this example, a wavelength of 980 nm was selected. Under the excitation of a 980 nm near-infrared laser with a certain power, the Time-Trace collected by the single-photon avalanche diode was used to collect the fluorescence burst signal of all target areas in a 60-second acquisition time. At the same time, the background signal with large wavelength differences could be perfectly separated by the dichroic mirror, and the extremely weak single-particle fluorescence pulses could be extracted.

[0075] This example demonstrates the detection of single-particle dispersion data using near-infrared flow cytometry. A 980nm near-infrared laser beam is expanded by a collimating beam expander 2 and focused by a high numerical aperture objective lens 5 to form a very small volume excitation region, thereby enabling the detection of single particles in a convective microtube 6. Simultaneously, the fluorescence signal passes through a first high-reflection, low-transmission dichroic mirror 4, a second high-reflection, low-transmission dichroic mirror 7, and a third high-reflection, low-transmission dichroic mirror 11, as well as a first narrowband filter 8, a second narrowband filter 12, and a third narrowband filter 15, and is collected by a first single-photon photodetector 10, a second single-photon photodetector 14, and a third single-photon photodetector 17, achieving the detection of fluorescence signals with single-particle-level sensitivity.

[0076] This embodiment provides a near-infrared flow cytometry upconversion nanoparticle dispersion detection device. The structure of this device is described in [reference needed]. Figure 1 It includes an excitation light generation module, a microfluidic sample introduction module, a light collection and dispersion module, and a single-photon detection module.

[0077] The excitation light generation module includes a near-infrared continuous laser 1, a collimating beam expander 2, and a filter 3. The infrared continuous laser 1 emits a stable laser beam. After being shaped by the collimating beam expander 2, the laser beam passes through the filter 3 to filter out other wavelengths of laser light. The filtered and shaped laser beam fills the entrance pupil of the high numerical aperture objective lens 5, ensuring the focusing quality of the beam, thereby obtaining the Gaussian excitation spot required for a very small detection volume.

[0078] The microfluidic sample introduction module includes a constant-speed injection pump and a flow microtube 6 with an inner diameter of 10 micrometers. The constant-speed injection pump propels the sample solution, and the 10-micrometer flow microtube 6 is used to form a hydrodynamic constraint, forcing the upconversion nanoparticles in the solution to flow through the excitation region focused by the objective lens in a single-row arrangement to the greatest extent possible, so as to achieve in-situ excitation detection.

[0079] The light collection and dispersion module includes a confocal high numerical aperture objective lens 5, a first high-reflection, low-transmission dichroic mirror 4, a second high-reflection, low-transmission dichroic mirror 7, and a third high-reflection, low-transmission dichroic mirror 11, as well as a first narrowband filter 8, a second narrowband filter 12, and a third narrowband filter 15. Excited fluorescence is collected through the same objective lens, and then the fluorescence signals of different detection channels are separated by wavelength through the dichroic mirrors, thereby achieving precise dispersion of the light path.

[0080] The single-photon detection module includes a focusing lens and a single-photon avalanche photodetector. After the excitation fluorescence is separated by a dichroic mirror, the fluorescence signal is passed through narrowband filters and the focusing lens of the corresponding channels, and then received by the single-photon avalanche photodetector, realizing the signal conversion of the extremely weak fluorescence burst of a single particle. According to... Figure 3 The algorithm flow shown illustrates how, in subsequent data processing, the data acquisition and processing motherboard performs Poisson denoising and dynamic peak finding on the photon counting sequence, as follows: Figure 2 Based on the physical model, pulse width and peak area features are extracted, and as follows: Figure 4 Cluster analysis was performed on a two-dimensional scatter plot, as shown. All detection signals were demodulated and analyzed by the data acquisition and control system to finally obtain the monodisperse particle ratio information and quantitative evaluation parameters of dispersibility of the sample. Example 2

[0081] Based on the upconversion nanoparticle dispersibility detection method of near-infrared flow cytometry in a specific embodiment, this example demonstrates the spectral separation and dispersibility detection data of multi-component mixed samples.

[0082] NaYF4:Yb / Er (20 / 2%) nanoparticles and NaYF4:Yb / Tm (18 / 2%) mixed nanoparticles were synthesized using a continuous near-infrared excitation beam, with a wavelength of 980 nm selected in this example. Under 980 nm near-infrared laser excitation at a certain power, the two types of nanoparticles exhibited different emission bands due to the different doping ions (Er mainly emits green / red light, while Tm mainly emits blue / near-infrared light). Although they were mixed and flowed in a microfluidic system, their emission channels responded differently; for example... Figure 4 The results of two-dimensional scatter cluster analysis of extracted particle features are presented. Through multi-channel spectral comparison and feature clustering methods, different luminescent components in mixed samples can be effectively distinguished, and the independent dispersion of Er and Tm samples can be evaluated respectively.

[0083] The present invention and its embodiments have been described above. This description is not restrictive, and the actual structure is not limited thereto. In conclusion, if those skilled in the art, inspired by this description, design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the scope of protection of this invention.

Claims

1. A method for detecting the dispersibility of upconversion nanoparticles based on near-infrared cytometry, characterized in that, Includes the following steps: S1: A 980nm near-infrared continuous laser beam is input into the optical shaping component for collimation and beam expansion to obtain an excitation beam with uniform energy distribution. S2: The excitation beam after S1 shaping is input into the high numerical aperture objective lens (5) and focused inside the flow microtube (6) with an inner diameter of 10 micrometers, forming an excitation region of extremely small volume in the center of the flow microtube (6); S3: Using a constant-speed injector, the sample solution containing upconversion nanoparticles is pushed through the excitation region in the flow microtube (6) at a constant speed. The sample particles are excited by a 980nm laser to generate a fluorescence signal. S4: The fluorescence signal generated by the excitation is collected and returned by the same objective lens, separated by a dichroic mirror for the main optical path, and the fluorescence signals of different wavelengths are split into different detection channels by a combination of filters. S5: Fluorescence signals from different channels are received by the corresponding single-photon avalanche diodes and converted into single-photon counting sequences. The digital signals are input into the data processing module, and the time series data of single-particle fluorescence bursts are analyzed by the algorithm pipeline combined with the physical model to calculate the aggregation rate and dispersion of the sample.

2. The method for detecting the dispersibility of upconversion nanoparticles based on near-infrared cytometry according to claim 1, characterized in that, The specific steps in S5 for analyzing sample dispersion include: Step 1: Signal preprocessing and denoising: The photon counting time series output by the single-photon avalanche diode is subjected to Poisson denoising and smoothing, and an adaptive background threshold is set in combination with the pure solvent background measurement value. Step 2: Dynamic peak finding based on long lifetime characteristics: In view of the asymmetric signal tailing caused by the long lifetime characteristics of rare earth upconversion luminescence, a dynamic threshold determination method is adopted to extract continuous pulse signals above the background threshold as effective particle events. Step 3: Multidimensional physical feature extraction: Based on fluid dynamics and Gaussian beam distribution theory, a physical model is constructed to extract the feature parameters of each burst pulse, including at least: the peak area reflecting the number of luminous centers of a single particle, and the pulse width reflecting the hydraulic diameter and residence time of the particle as it passes through the excitation region. Step 4: Cluster Analysis and Dispersion Quantification Output: The extracted multidimensional feature data is input into the clustering algorithm for scatter projection to distinguish between the core main group and the discrete group, and finally outputs the dispersion evaluation index.

3. The method for detecting the dispersibility of upconversion nanoparticles based on near-infrared cytometry according to claim 2, characterized in that: The third step of multidimensional physical feature extraction also includes multi-channel ratio analysis: calculating the integral area ratio of luminescence channels in different wavelength bands; By utilizing the phenomenon that the cross-relaxation effect caused by aggregation changes the spectral proportions, we can achieve cross-verification of sample dispersibility and eliminate interference from impurities.

4. The method for detecting the dispersibility of upconversion nanoparticles based on near-infrared cytometry according to claim 2, characterized in that: The formula for calculating the dispersion evaluation index DI in S5 is as follows: This represents the number of single-particle events obtained from cluster analysis. The table shows the total number of events detected. The higher the DI value, the better the sample dispersion and the lower the aggregation rate. The lower the DI value, the worse the sample dispersion and the higher the aggregation rate.

5. The method for detecting the dispersibility of upconversion nanoparticles based on near-infrared cytometry according to claim 1, characterized in that: When the sample to be tested is a multi-component mixed upconversion nanoparticle sample, different types of nanoparticles are distinguished by using signals from different emission bands acquired through multiple channels, and the dispersibility of each component is evaluated separately.

6. The method for detecting the dispersibility of upconversion nanoparticles based on near-infrared cytometry according to claim 1, characterized in that: The upconversion nanoparticles to be tested are water-soluble after ligand replacement treatment, making them suitable for microfluidic liquid phase detection environments.

7. A near-infrared flow cytometry-based upconversion nanoparticle dispersibility detection device, applied to the detection method according to any one of claims 1-6, characterized in that: It includes an excitation light generation module, a microfluidic sample introduction module, a light collection and dispersion module, and a single-photon detection module; The excitation light generation module includes a 980nm near-infrared continuous laser (1) and a collimating beam expander (2). The microfluidic injection module includes a constant-speed injection pump and a quartz flow microtube (6) with an inner diameter of 10 micrometers. The flow microtube (6) is used to provide a microchannel for single-particle-level queued flow. The light collection and beam splitting module includes a confocal high numerical aperture objective lens (5), a first high-reflection low-transmission dichroic mirror (4), a second high-reflection low-transmission dichroic mirror (7) and a third high-reflection low-transmission dichroic mirror (11), as well as a first narrowband filter (8), a second narrowband filter (12) and a third narrowband filter (15). The single-photon detection module includes at least three sets of single-photon avalanche diodes and a data acquisition and processing motherboard.

8. The detection device according to claim 7, characterized in that: The single-photon detection module includes a first single-photon photodetector (10), a second single-photon photodetector (14), and a third single-photon photodetector (17).

9. The detection device according to claim 7, characterized in that: The laser beam output by the excitation light generation module is collimated, expanded and filtered by the collimating beam expander (2) and the filter (3), and then fills the entrance pupil of the high numerical aperture objective lens (5), forming a Gaussian spot close to the diffraction limit inside the flow microtube (6).

10. The detection device according to claim 7, characterized in that: The flow microtube (6) has an inner diameter of 10 μm. By relying on fluid dynamics constraints, the upconversion nanoparticles are arranged in a single row as much as possible and flow through the excitation region to achieve in-situ excitation detection in the liquid phase. The first high-reflection, low-transmission dichroic mirror (4), the second high-reflection, low-transmission dichroic mirror (7), and the third high-reflection, low-transmission dichroic mirror (11) in the light collection and splitting module split the fluorescence signal according to the wavelength. The split fluorescence signal passes through the first focusing lens (9), the second focusing lens (13), and the third focusing lens (16) set behind each filter in sequence and is then input into the single-photon avalanche diode of the corresponding channel to realize multi-band fluorescence synchronous acquisition.

Images