Preparation method and application of gold and silver nano cubic particle solution

Abstract

The invention discloses a preparation method and application of a gold and silver nano cubic particle solution, and the method comprises the following steps: 1) mixing a gold nanorod solution and a hexadecyl trimethyl ammonium chloride solution with the concentration of 0.06-0.10 M according to the volume ratio of 1: (10-15), and slightly and uniformly mixing to obtain a solution A; (2) a silver nitrate solution with the concentration being 0.008-0.012 M and an ascorbic acid solution with the concentration being 0.08-0.12 M are sequentially added into the solution A, the adding amount of the silver nitrate solution and the adding amount of the ascorbic acid solution are (0.15-0.25): 1 and (0.6-0.8): 1 of the volume of the gold nanorod solution respectively, the mixture is stirred to be uniform, and a solution B is obtained; 3) heating the solution B for 2.5-3.5 hours under a 55-65 DEG C water bath condition to form a solution C; and 4) centrifuging the solution C for 8-12 minutes at the rotating speed of 5500-6500 rpm, discarding the supernate, retaining the precipitate, and re-dispersing with deionized water to finally obtain the gold and silver nano cubic particle solution serving as the SERS substrate. The method is used for detecting the organic toxins. The invention provides a new technical solution for rapid and high-sensitivity detection of multiple toxins in food and feed.

Description

Technical Field

[0001] The present invention belongs to the field of detection technology, and in particular relates to a preparation method of a gold-silver nanocubic particle solution and an application thereof. Background Art

[0002] In recent years, gold and silver nanostructures have garnered widespread attention in the field of chemical detection due to their excellent surface plasmon resonance properties and strong surface-enhanced Raman scattering (SERS) effect. Among them, gold and silver nanocubic particles, due to their unique rectangular structure and controllable surface chemistry, have become a highly effective SERS substrate material. These nanoparticles are typically prepared via a seed growth method, which specifically involves the following steps: first preparing a gold nanoseed solution, then introducing a silver source into a reaction system containing a surfactant and a reducing agent to achieve surface silver coating on the gold nanoseeds, ultimately forming a gold and silver nanocubic structure.

[0003] Existing technologies have made certain progress in the preparation of gold and silver nanocubic particles, but the following deficiencies and challenges still exist: Insufficient control of particle morphology and size: The morphology and size of gold and silver nanocubic particles play a decisive role in their SERS performance. However, due to the lack of precise control of reaction conditions (such as temperature, time and reagent ratio) in existing preparation methods, it is easy to cause uneven particle size distribution and inconsistent morphology, thereby reducing the stability and repeatability of the SERS signal. Complex substrate preparation process: Current technology requires strict control of reaction steps and achieves specific recognition function through surface modification, but this increases the complexity of substrate preparation and has a high risk of introducing impurities, further affecting the sensitivity and stability of detection. Insufficient multi-target analysis capabilities: Although gold and silver nanocubic particles show good performance in single target detection, there is little research on the simultaneous detection of multiple targets in complex samples (such as mixtures of multiple toxins). Existing methods make it difficult to achieve simultaneous identification and quantitative analysis of multiple toxins (such as aflatoxin B1, ochratoxin A and zearalenone). High preparation costs and limited applicability: Existing methods for preparing gold and silver nanoparticles typically rely on high-purity reagents and expensive equipment, resulting in complex processes and low yields, limiting their widespread adoption in large-scale, low-cost applications. Furthermore, traditional spectral analysis methods have limited ability to rapidly process large amounts of sample data, making them difficult to meet practical application requirements. Summary of the Invention

[0004] This invention provides a method for preparing a solution of cubic gold and silver nanoparticles and its application, aiming to address the shortcomings of existing multi-toxin detection methods in sensitivity, specificity, and stability. By optimizing the preparation process of the nanoparticle substrate and integrating deep learning with intelligent analysis of spectral data, this system provides a novel technical solution for the rapid, highly sensitive, and high-precision detection of multi-toxins in food and feed.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for preparing a gold-silver nanocubic particle solution comprises the following steps:

[0007] 1) Mix the gold nanorod solution with a 0.06-0.10 M hexadecyltrimethylammonium chloride solution in a volume ratio of 1:(10-15) and gently mix to obtain solution A;

[0008] 2) adding a 0.008-0.012 M silver nitrate solution and a 0.08-0.12 M ascorbic acid solution to solution A in a ratio of (0.15-0.25):1 and (0.6-0.8):1 to the volume of the gold nanorod solution, respectively, and stirring to obtain solution B;

[0009] 3) Heat solution B in a water bath at 55-65°C for 2.5-3.5 hours to form solution C;

[0010] 4) Centrifuging Solution C at a rotation speed of 5500-6500 rpm for 8-12 minutes, discarding the supernatant, retaining the precipitate, and redispersing it with deionized water to finally obtain a solution of gold and silver nanocubic particles serving as a SERS substrate.

[0011] A further improvement of the present invention is that the gold-silver nanocubic particles are 70-100 nm in length and 25-40 nm in width, and are generally rectangular in shape with an angular structure. The inner core is a gold nanorod, and the outer surface is evenly coated with a layer of silver, thus forming a gold-silver composite nanostructure.

[0012] A further improvement of the present invention is that the gold nanorods are uniform in shape, with a length of 60-90 nm and a width of 15-30 nm.

[0013] A further improvement of the present invention is that the gold nanorods are synthesized by the following method:

[0014] First, prepare the gold seed solution:

[0015] Subsequently, gold nanorods were synthesized by the seed growth method using a gold seed solution.

[0016] A further improvement of the present invention is that the gold seed solution is prepared, comprising:

[0017] 1) Dissolve cetyltrimethylammonium bromide and deionized water in a mass-to-volume ratio of (0.03-0.04):1, then add tetrachloroauric acid solution in a volume ratio of (1-2):100, and shake to mix thoroughly to obtain an initial solution A;

[0018] 2) dissolving sodium borohydride in refrigerated deionized water at a mass-to-volume ratio of (0.0002-0.0003):1 to obtain solution B;

[0019] 3) Under high-speed stirring, slowly add solution B dropwise to solution A at a volume ratio of 1:(8-12) and continue stirring for 2-3 minutes to obtain solution C;

[0020] 4) Heat solution C in a water bath at 25-35° C. for 20-40 minutes to obtain a gold seed solution.

[0021] A further improvement of the present invention is that the gold nanorods are synthesized by a seed growth method using a gold seed solution, comprising:

[0022] 1) Dissolve CTAB and sodium oleate in deionized water at a mass ratio of (5-6):1 (0.02-0.03):1, and stir in a 45-55°C water bath until completely dissolved to obtain Solution A;

[0023] 2) Place Solution A in a water bath at 25-35°C with magnetic stirring at 1200-1600 rpm. Add silver nitrate solution at a volume ratio of (1-2):20. Stir thoroughly and let stand for 10-20 minutes to obtain Solution B.

[0024] 3) Add 50 mM tetrachloroauric acid solution and deionized water to Solution B in a volume ratio of 1:(40-60), and stir at 1200-1600 rpm in a water bath at 25-35°C for 60-120 minutes to obtain Solution C;

[0025] 4) Add concentrated hydrochloric acid to Solution C in a volume ratio of (1-2):100, and react for 1-3 minutes to obtain Solution D;

[0026] 5) Add 0.1 M ascorbic acid solution dropwise to solution D at a volume ratio of (1-2):200 and mix thoroughly to obtain solution E;

[0027] 6) adding the gold seed solution to solution E at a volume ratio of (1-2):100, mixing, and allowing to stand in a water bath at 25-30° C. for 10-14 hours to generate a gold nanorod solution;

[0028] 7) Centrifuge the gold nanorod solution twice, each time for 8-12 minutes at a speed of 6500-7500 rpm, discard the supernatant, retain the precipitate and redisperse it in deionized water to obtain a gold nanorod solution.

[0029] A further improvement of the present invention is that the silver nitrate solution is prepared by mixing silver nitrate and deionized water in a mass volume ratio of (0.0006-0.0008):1.

[0030] An application of a gold-silver nanometer cubic particle solution is used for detecting organic toxins, and the gold-silver nanometer cubic particle solution is prepared by adopting the preparation method.

[0031] A further improvement of the present invention is that it specifically comprises the following steps:

[0032] 1) Substrate Preparation: 10 μL of the gold and silver nanocubic particle solution was pipetted onto a clean silicon wafer surface and then dried at room temperature to form a uniformly distributed nanostructured substrate.

[0033] 2) Sample preparation: Organic toxin solutions of varying concentrations are added dropwise onto the dried substrate surface. After the toxin solutions dry naturally, their molecules adsorb onto the surface of the gold and silver nanocubic particles, forming an effective SERS active area.

[0034] 3) Spectral measurement: Spectral measurements of the samples were performed using a Raman spectrometer. The specific measurement conditions were as follows: excitation wavelength: 633 nm; laser power: 0.49 mW; integration time: 10 s. The measurement was repeated three times for each sample to improve signal stability and accuracy.

[0035] 4) Data analysis: By analyzing the obtained Raman spectral characteristic peaks, qualitative and quantitative detection of organic toxins at different concentrations can be achieved.

[0036] A further improvement of this invention is that it leverages the surface plasmon resonance properties of gold and silver nanocubic particles to significantly enhance detection sensitivity, making it particularly suitable for efficient analysis of trace toxins. Furthermore, by incorporating deep learning technology into the intelligent processing of spectral data, the accuracy and reliability of trace toxin detection are further enhanced.

[0037] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0038] The gold and silver nano-rectangular particle substrate prepared by the present invention can realize ultra-trace detection of three mycotoxins, AFB-1, OTA and ZEN, through its sharp structural characteristics and local electromagnetic field enhancement effect, combined with the high enhancement performance of silver materials. The detection limit meets the requirements of food safety regulations and meets the needs of practical applications. At the same time, the neural network model realizes efficient qualitative and quantitative analysis of complex samples through in-depth mining of the association between standard spectra and mixed spectral features, with excellent performance. This method adopts SERS technology, which does not require the modification of additional specific recognition molecules or complex sample pre-treatment processes, can quickly complete the detection operation, and is suitable for high-throughput analysis. In addition, the neural network model can simultaneously process single sample, double sample and multi-sample mixed spectra, and has the ability to simultaneously identify and quantify multi-target toxins. In summary, the present invention has the advantages of high sensitivity, fast efficiency, simple operation and strong applicability, and provides an efficient and reliable technical solution for food safety monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 SEM image of gold nanorods.

[0041] Figure 2 This is the SEM image of gold and silver nano-rectangular particles.

[0042] Figure 3 This is the absorption spectrum of the gold and silver nano-rectangular particle substrate.

[0043] Figure 4 Raman spectra of AFB-1, OTA and ZEN.

[0044] Figure 5 SERS spectra of AFB-1, OTA and ZEN at different concentrations.

[0045] Figure 6 Schematic diagram of the neural network structure of the present invention.

[0046] Figure 7 This is the neural network training curve of the present invention.

[0047] Figure 8 This is the SEM image of the substrate synthesized in Example 1.

[0048] Figure 9 This is the SEM image of the substrate synthesized in Example 2.

[0049] Figure 10 This is the SEM image of the substrate synthesized in Example 3. DETAILED DESCRIPTION

[0050] Hereinafter, only certain exemplary embodiments are briefly described. As will be appreciated by those skilled in the art, the described embodiments may be modified in various ways without departing from the spirit or scope of the present invention. Therefore, the drawings and description are to be considered as illustrative in nature and not restrictive.

[0051] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0052] It should also be understood that the terms used in the present specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0053] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0054] The accompanying drawings illustrate various schematic diagrams of structures according to embodiments disclosed herein. These figures are not drawn to scale; for clarity, some details are exaggerated and some details may be omitted. The shapes of the various regions and layers shown in the figures, as well as their relative sizes and positional relationships, are merely exemplary and may deviate in practice due to manufacturing tolerances or technical limitations. Those skilled in the art may design regions / layers with different shapes, sizes, and relative positions as needed.

[0055] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0056] The present invention provides a method for preparing a solution of gold-silver nanocubic particles. The preparation process is divided into three steps: preparation of a gold seed solution, synthesis of gold nanorods, and preparation of gold-silver nanocubic particles. The specific steps are as follows:

[0057] First, prepare the gold seed solution:

[0058] 1) Dissolve 0.3644 g of cetyltrimethylammonium bromide (CTAB) in 10 mL of deionized water. Add 50 μL of 50 mM HAuCl4 and gently shake to mix thoroughly to obtain initial solution A.

[0059] 2) Dissolve 0.0023 g of sodium borohydride (NaBH4) in 10 mL of refrigerated deionized water to obtain solution B.

[0060] 3) Under high-speed stirring, slowly add 1 mL of solution B to 10 mL of solution A and continue stirring for 2 minutes to obtain solution C.

[0061] 4) Heat solution C in a 30° C. water bath for 30 minutes to obtain a gold seed solution.

[0062] Subsequently, gold nanorods were synthesized by the seed growth method using the gold seed solution:

[0063] 1) 1.4 g CTAB and 0.2468 g sodium oleate (C 17 H 33 CO2Na) was dissolved in 50 mL of deionized water and stirred in a 50°C water bath until completely dissolved to obtain solution A.

[0064] 2) Solution A was placed in a 30°C water bath with magnetic stirring at 1400 rpm. At the same time, 4.8 mL of a solution prepared by dissolving 0.0068 g of silver nitrate (AgNO3) in 10 mL of deionized water was added. The mixture was stirred and allowed to stand for 15 minutes to obtain Solution B.

[0065] 3) Add 1 mL of 50 mM tetrachloroauric acid solution and 50 mL of deionized water to Solution B, and stir at 1400 rpm in a 30°C water bath for 90 minutes to obtain Solution C.

[0066] 4) Add 0.34 mL of concentrated hydrochloric acid (HCl) to solution C and react for 2 minutes to obtain solution D.

[0067] 5) Add 250 μL of 0.1 M ascorbic acid (AA) dropwise to solution D and mix thoroughly to obtain solution E.

[0068] 6) Add 160 μL of the gold seed solution prepared above to Solution E, mix well, and let stand in a 28° C. water bath for 12 hours to generate a gold nanorod solution.

[0069] 7) Centrifuge the gold nanorod solution twice (10 minutes each time, 7000 rpm), discard the supernatant, retain the precipitate and redisperse it in deionized water to obtain a purified gold nanorod solution.

[0070] Figure 1 The scanning electron microscope (SEM) image of the gold nanorods synthesized in the above steps; Figure 1 It can be seen that the synthesized gold nanorods have regular morphology and their length is less than 100 nm.

[0071] The process of synthesizing gold and silver nanoparticles by chemical reduction using gold nanorods is as follows:

[0072] 1) Add 300 μL of the gold nanorod solution to 4 mL of a 0.08 M cetyltrimethylammonium chloride (CTAC) solution and mix gently to obtain solution A.

[0073] 2) Add 55 μL of 0.01 M silver nitrate solution and 200 μL of 0.1 M ascorbic acid solution to solution A and stir well to obtain solution B.

[0074] 3) Heat solution B in a 60°C water bath for 3 hours to form solution C.

[0075] 4) Centrifuge Solution C for 10 minutes (6000 rpm), discard the supernatant, retain the precipitate, and redisperse it in deionized water to obtain a solution of gold and silver nanocubic particles serving as a SERS substrate.

[0076] The technical solution of the present invention is described in detail below with reference to specific embodiments and drawings, and this embodiment does not limit the present invention.

[0077] Example 1:

[0078] 1) The gold nanorod solution was mixed with a 0.06 M cetyltrimethylammonium chloride (CTAC) solution at a volume ratio of 1:10 and gently mixed to obtain solution A.

[0079] 2) Silver nitrate solution (concentration 0.008 M) and ascorbic acid solution (concentration 0.08 M) were sequentially added to solution A in a ratio of 0.15:1 and 0.6:1 to the volume of the gold nanorod solution, respectively, and stirred to obtain solution B.

[0080] 3) Solution B was heated in a 55°C water bath for 2.5 hours to form solution C.

[0081] 4) Centrifuging Solution C at 5500 rpm for 8 minutes, discarding the supernatant, retaining the precipitate, and redispersing it with deionized water to obtain a gold and silver nanocubic particle solution as a SERS substrate.

[0082] Example 2:

[0083] 1) The gold nanorod solution was mixed with a 0.10 M cetyltrimethylammonium chloride (CTAC) solution at a volume ratio of 1:15 and gently mixed to obtain solution A.

[0084] 2) Silver nitrate solution (concentration 0.012 M) and ascorbic acid solution (concentration 0.12 M) were sequentially added to solution A in a ratio of 0.25:1 and 0.8:1 to the volume of the gold nanorod solution, respectively, and stirred to obtain solution B.

[0085] 3) Solution B was heated in a water bath at 65°C for 3.5 hours to form solution C.

[0086] 4) Centrifuge Solution C at 6500 rpm for 12 minutes, discard the supernatant, retain the precipitate, and redisperse it with deionized water to obtain a gold and silver nanocubic particle solution as a SERS substrate.

[0087] Example 3:

[0088] 1) The gold nanorod solution was mixed with a 0.08 M cetyltrimethylammonium chloride (CTAC) solution at a volume ratio of 1:12.5 and gently mixed to obtain solution A.

[0089] 2) Silver nitrate solution (concentration 0.01 M) and ascorbic acid solution (concentration 0.10 M) were sequentially added to solution A in a ratio of 0.20:1 and 0.7:1 to the volume of the gold nanorod solution, respectively, and stirred to obtain solution B.

[0090] 3) Heat solution B in a 60°C water bath for 3 hours to form solution C.

[0091] 4) Centrifuge Solution C at 6000 rpm for 10 minutes, discard the supernatant, retain the precipitate, and redisperse it with deionized water to obtain a gold and silver nanocubic particle solution as a SERS substrate.

[0092] Figure 8 The synthetic substrate of Example 1 is Figure 9 The synthetic substrate for Example 2, Figure 10 The synthetic substrate of Example 3. As can be seen from the figure, Example 1 ( Figure 8 ) The obtained substrate has fewer nanoparticles with regular morphology, showing a certain degree of synthetic heterogeneity; Example 2 ( Figure 9) has a significant increase in the number of substrates with regular morphology, but the presence of larger particles has a certain impact on the uniformity of the substrate; in contrast, Example 3 ( Figure 10 ) almost all substrates synthesized in Example 3 exhibited regular morphology, uniform particle size, and the highest yield, demonstrating excellent morphology control. Therefore, the ratio and synthesis conditions used in Example 3 were most suitable, enabling efficient preparation of substrates with regular morphology.

[0093] Figure 2 This is an SEM image of synthesized gold and silver nanoparticles; it can be observed that these rectangular particles have sharp right-angle features. When the particles are aggregated together, these right angles and the gaps between the particles can form "hot spots" with strong local electromagnetic field enhancement. The hot spots are due to the strong electromagnetic field concentration effect generated by the localized surface plasmon resonance (LSPR) near these structures, which significantly enhances the SERS signal. In addition, the plasmon resonance frequency of silver is closer to the visible light range than that of gold, and its free electron density is higher, so it can produce a stronger local electromagnetic field enhancement effect, which makes the silver-based particles have higher detection sensitivity in SERS applications. Therefore, this gold and silver nanoparticle substrate has the dual advantages of structure and material, providing an ideal hot spot distribution and signal enhancement effect for efficient SERS detection.

[0094] Figure 3 The absorption spectrum of the gold and silver nano-rectangular particle substrate is displayed, with a measurement range of 300-1100nm, in order to fully characterize the optical properties of the substrate in the ultraviolet, visible and near-infrared bands. The results show that the strongest absorption peak of the substrate is located at 620-660nm, which just covers the excitation light wavelength of 633nm. This matching relationship significantly improves the SERS detection performance, because the position of the absorption peak reflects the surface plasmon resonance (SPR) peak of the substrate. When the excitation light wavelength is close to the SPR peak, the enhancement effect of the local electromagnetic field on the substrate surface is most significant, thereby greatly improving the intensity of the Raman scattering signal. The specific position of the SPR peak is determined by the size, shape and composition of the nanoparticles, and the peak of the absorption spectrum is the optical manifestation of the SPR peak. Therefore, the good matching of the substrate absorption peak and the excitation light wavelength not only verifies the rationality of the SPR peak position, but also fully demonstrates the advantages of the substrate in high sensitivity and signal enhancement capabilities in SERS detection.

[0095] Figure 4The Raman spectra of AFB-1, OTA and ZEN are shown. These spectra contain a series of characteristic peaks, each of which corresponds to a specific chemical bond vibration, as shown in Table 1. By analyzing these characteristic peaks, it can be found that each organic toxin has a unique vibration mode, and the uniqueness of these vibration peaks provides a theoretical basis for distinguishing the three toxins. The independence and non-overlapping nature of the characteristic peaks enable SERS technology to achieve accurate identification and quantitative analysis of multiple toxins in mixed samples based on these unique Raman signals. Therefore, Figure 4 The spectra in the image not only intuitively reflect the molecular characteristics of each toxin, but also demonstrate the feasibility and efficiency of using SERS technology to identify toxins.

[0096] Table 1 Raman shifts of characteristic peaks of different organic toxins and their corresponding chemical bonds

[0097]

[0098]

[0099] Figure 5 The use of synthesized gold and silver nanoparticles as a substrate was demonstrated to measure the SERS spectra of three organic toxins (AFB-1, OTA and ZEN) at different concentrations. During the measurement, 10 μL of the gold and silver nanoparticles substrate solution was taken with a pipette and dropped on a clean silicon wafer. After drying, organic toxin solutions of different concentrations were dropped on the substrate surface, and Raman spectra were measured after it was naturally dried. The measurement conditions were an excitation light wavelength of 633 nm, a laser power of 0.49 mW, an integration time of 10 s, and three integration times for each sample to improve the stability and accuracy of the signal. The experimental results show that the substrate has high sensitivity, and the minimum detection limit for AFB-1 reached 10 -9 M / L, corresponding to 0.3 μg / kg; while for OTA and ZEN, the minimum detection limits were 10 -8 The SERS substrate was used for ultratrace detection of mycotoxins, with a detection limit of 4 μg / kg and 3 μg / kg, respectively. These detection limits are significantly lower than those stipulated by many food safety regulations, demonstrating the high efficiency and practicality of this SERS substrate for ultratrace detection of mycotoxins, providing an important basis for achieving rapid and sensitive food safety testing.

[0100] In order to achieve efficient detection and quantitative analysis of three organic toxins (AFB-1, OTA and ZEN), the experiment first collected the standard Raman spectrum of each toxin as a reference for distinguishing the toxin characteristics. At the same time, the SERS spectral data of single samples, double samples and triple samples of these three toxins at different concentrations were measured to provide a complete data set for subsequent neural network training and prediction. The neural network structure is as follows Figure 6As shown in the figure, the model architecture design uses the SERS spectrum of the mixture and the standard Raman spectrum of the organic toxin as input data. These two data sets are then passed through two one-dimensional convolutional layers for feature extraction. Each convolutional layer uses 32 kernels with a kernel size of 2 to capture local characteristic patterns in the spectrum. Reinforced Luminance (ReLU) activation functions are used to achieve nonlinear mapping. The convolutional output undergoes dimensionality reduction using a max-pooling operation (pooling factor of 3) to improve the compactness of feature representation and computational efficiency. The features extracted by the convolutional layer are then fed into a single-layer long short-term memory (LSTM) network with 50 neurons to exploit temporal correlations and dynamic features in the spectral data. The output of the LSTM layer is two high-dimensional vectors, representing the global characteristics of the standard spectrum and the mixture spectrum, respectively. These two vectors are concatenated and fed into a fully connected layer with 1024 neurons to further fuse and extract feature associations between the sample and standard spectra. In the output stage, the model employs two independent output layers to simultaneously perform qualitative and quantitative analysis. The first output layer contains one neuron, which is used to predict whether the mixed sample contains the target toxin, and makes a qualitative judgment through the probability value; the second output layer also contains one neuron, which is used to predict the specific concentration value of the target toxin, thereby realizing quantitative analysis. For the two output tasks, the model uses the binary cross entropy loss function and the mean square error loss function for optimization respectively. Since qualitative analysis is more important than quantitative analysis in practical applications, the loss function weights are set to 0.75 for the binary cross entropy and 0.25 for the mean square error to prioritize the accuracy of qualitative predictions. The training process uses the Adam optimizer (Adaptive Moment Estimation) for parameter updates, and the learning rate is set to 0.0001 to ensure the stability and convergence efficiency of the training. Through the above model architecture and training strategy, the feature correlation between the standard Raman spectrum and the mixed spectrum is fully utilized, and the experiment achieves accurate identification and concentration prediction of three organic toxins.

[0101] Figure 7 The changing trends of the accuracy and loss function of the neural network model during 200 rounds of training are shown. After the training is completed, the constructed neural network model is used to predict the SERS spectrum of the mixture. The accuracy of the model's qualitative analysis reached 0.954, indicating that it has high reliability in identifying the presence or absence of target organic toxins in the mixture. The mean squared error (MSE) of the quantitative analysis results is 0.3, indicating that the model has good accuracy in predicting the concentration of organic toxins. The above results verify the effectiveness and practicality of the neural network model for efficient qualitative and quantitative detection of multiple organic toxins in complex samples.

[0102] The invention points for pre-protection of the present invention are:

[0103] 1. Preparation method of gold and silver nano rectangular particle substrate;

[0104] 2. Gold and silver nanoparticle SERS substrate;

[0105] 3. A neural network model integrating SERS spectral feature extraction;

[0106] 4. Rapid and efficient detection system based on gold and silver nanoparticle substrate and neural network.

[0107] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, from all points of view, the embodiments should be regarded as illustrative and non-restrictive. The scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes that fall within the meaning and range of equivalents of the claims are included in the present invention. Any reference signs in the claims should not be construed as limiting the claim to which they relate.

[0108] In addition, it should be understood that although this specification describes the embodiments, not every embodiment contains only one independent technical solution. This description is for clarity only. Those skilled in the art should consider the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art. The above content is only for the purpose of illustrating the technical concept of the present invention and cannot be used to limit the scope of protection of the present invention. Any changes made based on the technical solution in accordance with the technical concept proposed by the present invention fall within the scope of protection of the claims of the present invention.

Claims

1. A method for preparing a gold-silver nanocubic particle solution, characterized in that: The following steps are involved: 1) Mix the gold nanorod solution with a 0.06-0.10 M hexadecyltrimethylammonium chloride solution in a volume ratio of 1:(10-15) and gently mix to obtain solution A; 2) adding a 0.008-0.012 M silver nitrate solution and a 0.08-0.12 M ascorbic acid solution to solution A in a ratio of (0.15-0.25):1 and (0.6-0.8):1 to the volume of the gold nanorod solution, respectively, and stirring to obtain solution B; 3) Heat solution B in a water bath at 55-65°C for 2.5-3.5 hours to form solution C; 4) Centrifuging Solution C at a rotation speed of 5500-6500 rpm for 8-12 minutes, discarding the supernatant, retaining the precipitate, and redispersing it with deionized water to finally obtain a solution of gold and silver nanocubic particles serving as a SERS substrate.

2. The method for preparing a gold-silver nanocubic particle solution according to claim 1, characterized in that: The length of the gold-silver nanocubic particles is 70-100nm, the width is 25-40nm, and the overall shape is a cuboid with an angular structure. The internal core is a gold nanorod, and the outside is evenly wrapped with a layer of silver, thus forming a gold-silver composite nanostructure.

3. The method for preparing a gold-silver nanocubic particle solution according to claim 1, characterized in that: The gold nanorods are uniform in shape, with a length of 60-90nm and a width of 15-30nm.

4. The method for preparing a gold-silver nanocubic particle solution according to claim 3, characterized in that: Gold nanorods were synthesized by the following method: First, prepare the gold seed solution: Subsequently, gold nanorods were synthesized by the seed growth method using a gold seed solution.

5. The method for preparing a gold-silver nanocubic particle solution according to claim 4, characterized in that: Prepare the gold seed solution, including: 1) Dissolve cetyltrimethylammonium bromide and deionized water in a mass-to-volume ratio of (0.03-0.04):1, then add tetrachloroauric acid solution in a volume ratio of (1-2):100, and shake to mix thoroughly to obtain an initial solution A; 2) dissolving sodium borohydride in refrigerated deionized water at a mass-to-volume ratio of (0.0002-0.0003):1 to obtain solution B; 3) Under high-speed stirring, slowly add solution B dropwise to solution A at a volume ratio of 1:(8-12) and continue stirring for 2-3 minutes to obtain solution C; 4) Heat solution C in a water bath at 25-35° C. for 20-40 minutes to obtain a gold seed solution.

6. The method for preparing a gold-silver nanocubic particle solution according to claim 5, characterized in that: Gold nanorods were synthesized by a seed growth method using a gold seed solution, including: 1) Dissolve CTAB and sodium oleate in deionized water at a mass ratio of (5-6):1 (0.02-0.03):1, and stir in a 45-55°C water bath until completely dissolved to obtain Solution A; 2) Place Solution A in a water bath at 25-35°C with magnetic stirring at 1200-1600 rpm. Add silver nitrate solution at a volume ratio of (1-2):

20. Stir thoroughly and let stand for 10-20 minutes to obtain Solution B. 3) Add 50 mM tetrachloroauric acid solution and deionized water to Solution B in a volume ratio of 1:(40-60), and stir at 1200-1600 rpm in a water bath at 25-35°C for 60-120 minutes to obtain Solution C; 4) Add concentrated hydrochloric acid to Solution C in a volume ratio of (1-2):100, and react for 1-3 minutes to obtain Solution D; 5) Add 0.1 M ascorbic acid solution dropwise to solution D at a volume ratio of (1-2):200 and mix thoroughly to obtain solution E; 6) adding the gold seed solution to solution E at a volume ratio of (1-2):100, mixing, and allowing to stand in a water bath at 25-30° C. for 10-14 hours to generate a gold nanorod solution; 7) Centrifuge the gold nanorod solution twice, each time for 8-12 minutes at a speed of 6500-7500 rpm, discard the supernatant, retain the precipitate and redisperse it in deionized water to obtain a gold nanorod solution.

7. The method for preparing a gold-silver nanocubic particle solution according to claim 6, characterized in that: The silver nitrate solution was prepared by mixing silver nitrate and deionized water in a mass-to-volume ratio of (0.0006-0.0008):

1.

8. An application of a gold-silver nanocubic particle solution, characterized in that: The invention is used for detecting organic toxins, and the gold-silver nanocubic particle solution is prepared by the preparation method according to any one of claims 1 to 7.

9. The use of a gold-silver nanocubic particle solution according to claim 8, characterized in that: The specific steps include: 1) Substrate Preparation: 10 μL of the gold and silver nanocubic particle solution was pipetted onto a clean silicon wafer surface and then dried at room temperature to form a uniformly distributed nanostructured substrate. 2) Sample preparation: Organic toxin solutions of varying concentrations are added dropwise onto the dried substrate surface. After the toxin solutions dry naturally, their molecules adsorb onto the surface of the gold and silver nanocubic particles, forming an effective SERS active area. 3) Spectral measurement: Spectral measurements of the samples were performed using a Raman spectrometer. The specific measurement conditions were as follows: excitation wavelength: 633 nm; laser power: 0.49 mW; integration time: 10 s. The measurement was repeated three times for each sample to improve signal stability and accuracy. 4) Data analysis: By analyzing the obtained Raman spectral characteristic peaks, qualitative and quantitative detection of organic toxins at different concentrations can be achieved.

10. The use of a gold-silver nanocubic particle solution according to claim 9, characterized in that: Since the surface plasmon resonance properties of gold and silver nanocubic particles can enhance detection sensitivity, they are suitable for the analysis of trace toxins.

Images