Model and method for predicting toxicity of multi-metal compound
By establishing a biotoxicity prediction model based on the composition of multi-metal compound elements, the problem of difficulty in evaluating the toxicity of multi-metal compounds in the prior art is solved, and the rapid and accurate prediction of the toxicity of different types of multi-metal compounds is achieved, and the basic data for toxicity assessment and risk prediction are provided.
Patent Information
- Application Number
- CN202411708626.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-27
- Publication Date
- 2025-04-04
AI Technical Summary
The prior art is difficult to quickly and efficiently evaluate the toxicity of multi-metal compounds, especially when the proportion of different metal elements is present, resulting in difficulty in toxicity and risk assessment.
By establishing a biotoxicity prediction model based on element characteristics related to the composition of multi-metal compounds, the energy dispersion X-ray spectrometer of scanning electron microscope is used to characterize the type and relative proportion of metal elements, and combining the biotoxicity experimental data of a mixture of single metal elements and multi-metal elements, an integrated addition and interaction model (IAI) is established for toxicity prediction.
It achieves rapid, accurate and efficient prediction of the toxicity of multi-metal compounds, and is suitable for different types of multi-metal compounds, providing basic data for toxicity assessment, risk prediction and regulation.
Smart Images
Figure CN119763711A_ABST
Abstract
Description
Technical field:
[0001] The present invention relates to the technical field of toxicity prediction of toxic substances, and in particular to a biological toxicity prediction model and method of multi-metal compounds. Background technology:
[0002] As the cornerstone of the renewable energy industry, lithium batteries have experienced rapid development. Multi-metal compounds, such as lithium nickel cobalt manganese oxide metal oxide composed of different contents of nickel, cobalt and manganese, are key components of lithium batteries. Due to their excellent characteristics of high energy density, their application in lithium batteries and their market share are increasing. However, due to the lack of protective measures during the production and recycling process, the rapid development of the production and application of multi-metal compounds has increased their potential inhalation exposure risks, which has aroused people's concern about the human health risks of multi-metal compounds, and the potential risks of multi-metal compounds need to be thoroughly evaluated.
[0003] After inhalation exposure, multi-metal compounds will release the metal elements in them. The toxicity of the released metal element mixture is an important source of the toxicity of multi-metal compounds. However, multi-metal compounds, such as lithium nickel cobalt manganese oxide, have developed rapidly and are complex and diverse. Different types of multi-metal compounds are mainly characterized by different types and proportions of metal elements. This complexity has brought great difficulties to toxicity and risk assessment. Experimental determination of the toxicity of each type of multi-metal compound is time-consuming, labor-intensive, difficult to operate, and has poor feasibility. Therefore, it is urgent to construct a representative, fast and efficient toxicity prediction method suitable for different types of multi-metal compounds. However, few studies have reported toxicity prediction methods that can be used for metal ion mixtures with different proportions of metal elements to predict the toxicity of multi-metal compounds. Summary of the invention:
[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a simple, fast, accurate and efficient method for predicting the toxicity of multi-metal compounds, such as lithium nickel cobalt manganese oxide. This method can predict the toxicity of other multi-metal compounds containing the same metal elements but different proportions based on the type and proportion of metal elements in a multi-metal compound, thereby providing the necessary basic data for toxicity assessment, risk prediction and supervision of multi-metal compounds.
[0005] In order to achieve the above-mentioned object, the present invention provides a toxicity prediction method for a multi-metal compound. The prediction method performs biological toxicity evaluation based on the element characteristics related to the element composition of the multi-metal compound to obtain biological toxicity experimental data, and then performs prediction model building on the biological toxicity experimental data obtained above, and compares the results of the prediction model with the toxicity experimental data to screen the optimal model. The prediction method specifically includes the following steps:
[0006] (1) Characterization of the physicochemical properties of the metal element types and relative proportions of multinary metal compounds: For a certain type of multinary metal compound, one of the multinary metal compounds is selected as the basic compound for modeling. The total number of toxic metal element types n and the relative proportion of the amount or atomic percentage of each toxic metal element contained in the multinary metal compound are obtained from the known chemical equation or by using the energy dispersive X-ray spectrometer (EDS) of a scanning electron microscope;
[0007] (2) Testing the biological toxicity of the released elements: Based on the n types of toxic metal elements obtained in step (1) and the relative proportions of the toxic metal elements, two exposure conditions are used: exposure to a single toxic metal element and exposure to a toxic metal element mixture in which all toxic metal elements are mixed in proportion (the concentration of the toxic metal element mixture is expressed as the sum of the concentrations of each single toxic metal element therein), and the biological toxicity (such as inhibition rate percentage, IR) on alveolar macrophage viability at different exposure doses is measured; Origin software is used to fit the logistic method to establish a dose-response relationship equation with the exposure dose as the independent variable and the biological toxicity of alveolar macrophage viability as the dependent variable, and the slope pi and IC of the dose-response relationship equation of each toxic metal element are obtained based on the dose-response relationship equation. 50，i Dose, maximum inhibition rate IR max，i and the minimum inhibition rate IR min，i , IC of toxic metal element mixture 50，mixture Dose, maximum inhibition rate IR min，mixture and the minimum inhibition rate IR max，mixture ;
[0008] Among them, the single toxic metal dose-response relationship equation of logistic fitting is shown in formula ①
[0009]
[0010] In formula ①, IR(c i ) is the percentage of inhibition of the i-th single toxic metal element on the activity of alveolar macrophages, IR min，i is the minimum inhibitory rate of the ith single metal on the activity of alveolar macrophages, IR max，i is the maximum inhibition rate of the ith single metal on the activity of alveolar macrophages, c i is the concentration of the ith single metal element, IC 50，i is the concentration of the ith single metal element when the inhibition rate on the activity of alveolar macrophages is 50%, and pi is the slope of the dose-response equation of the ith metal element;
[0011] The dose-response relationship equation of the toxic metal element mixture fitted by logistic is shown in formula ②
[0012]
[0013] In formula ②, IR(c mixture ) is the percentage of inhibition of toxic metal element mixture on alveolar macrophage activity, IR min，mixture is the minimum inhibitory rate of the toxic metal element mixture on the activity of alveolar macrophages, IR max，mixture is the maximum inhibition rate of the toxic metal element mixture on the activity of alveolar macrophages, c mixture is the concentration of toxic metal element mixture, IC 50，mixture is the concentration of metal element mixture at which the inhibition rate of alveolar macrophage activity is 50%, p mixture is the slope of the dose-response relationship equation;
[0014] (3) Establish five different quantitative prediction models for biological toxicity
[0015] 1) Concentration Additive Model (CA):
[0016] The concentration addition model (CA) is shown in formula ③
[0017]
[0018] In formula ③, IC x，mixture is the concentration of the toxic metal element mixture when the inhibition rate of the toxic metal element mixture on the activity of alveolar macrophages is x, L i is the proportion of the ith toxic metal element in the total toxic metal elements, IC x，i is the concentration of the metal element when the inhibition rate of the i-th metal element on the activity of alveolar macrophages is x, which can be calculated based on the dose-response relationship equation of the i-th metal element in step (2);
[0019] 2) Independent Action Model (IA):
[0020] The independent action model (IA) is shown in formula ④
[0021]
[0022] In formula ④, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i ) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); ci is calculated based on the proportion of the ith toxic metal element in the total toxic metal elements when the concentration of the toxic metal element mixture is c mixture is the concentration of the ith metal, and n represents the number of toxic metal elements in the compound;
[0023] 3) Effect Additive Model (ES):
[0024] The effect addition model (ES) model equation is shown in formula ⑤
[0025]
[0026] In formula ⑤, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i ) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); c i is the ratio of the ith toxic metal element to the total toxic metal elements in the toxic metal element mixture with a concentration of c mixture is the concentration of the ith metal, and n represents the number of toxic metal elements in the compound;
[0027] 4) Integrated Additive Model (IAM):
[0028] The model equation of the integrated additive model (IAM) is shown in formula ⑥
[0029]
[0030] In formula ⑥, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of 50，i The IC of the ith metal is obtained based on the dose-response equation of the ith metal element in the metal mixture in step (2). 50 Dose; c i is the ratio of the ith toxic metal element to the total toxic metal elements in the toxic metal element mixture with a concentration of c mixture The concentration of the i-th metal at time ; where p i is the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2), where n represents the number of toxic metal elements in the compound;
[0031] 5) Integrated Additive Interaction Model (IAI):
[0032] The IAI model equation is shown in formula ⑦
[0033]
[0034] In formula ⑦, c mixture is the concentration of the metal element mixture, c i is the concentration of the ith metal element in the metal element mixture, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of K i is a function that describes the effect of the ith metal element on the total toxicity, IC 50，i is the concentration of the i-th metal element obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%, IC 50，mixture The concentration of the metal element mixture obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%; where p i is the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2), where n represents the number of toxic metal elements in the compound;
[0035] (4) Model accuracy evaluation: The prediction results of the above models were compared with the toxicity test data. On the one hand, different concentrations of toxic metal element mixtures were set, and the concentration of each single metal was calculated based on the proportion of each toxic metal element in the total toxic metal elements obtained in step (1). i ; Then, based on the dose-response equation of the ith metal element in the metal mixture in step (2), IR (c i ), the IC of the i-th metal element is known 50，i , slope p i IC of metal element mixture 50，mixture , substitute the above four prediction models, namely, independent action model (IA), effect addition model (ES), integrated additive model (IAM) and integrated additive interaction model (IAI), to calculate the toxicity prediction value of each model; compare the calculated biological toxicity prediction value of the metal element mixture with the biological toxicity value of the metal element mixture at the concentration calculated by the dose-response relationship equation based on the experimental data of the metal element mixture according to step (2); and calculate the value by R square (R 2 ) and root mean square error (RMSE) were used to evaluate the goodness of fit and prediction accuracy of each model;
[0036] On the other hand, different biological toxicity values of metal element mixtures are set, and the inhibition rate of the ith metal element on the activity of alveolar macrophages is calculated based on formula ① in step (2) at the metal element concentration under the set inhibition rate, and the concentration addition model (CA) is brought into the concentration addition model to obtain the predicted metal element mixture concentration under the set inhibition rate; the set inhibition rate is further compared with the biological toxicity value of the metal element mixture concentration calculated according to the dose-response relationship equation based on the experimental data of the metal element mixture in step (2); the R square (R 2 ) and root mean square error (RMSE) were used to evaluate the goodness of fit and prediction accuracy of each model;
[0037] R 2 The closer it is to 1, the smaller the RMSE value is and the better the prediction ability is, based on which the model with the best prediction accuracy is selected.
[0038] Among them, R 2 The calculation of is shown in formula ⑧
[0039]
[0040] The calculation of RMSE is shown in formula ⑨
[0041]
[0042] In formulas ⑧ and ⑨, for the above four prediction models, namely, the independent action model (IA), the effect addition model (ES), the integrated additive model (IAM), and the integrated additive interaction model (IAI), is the inhibition rate of the metal element mixture on alveolar macrophage activity predicted by each prediction model; for the concentration addition model (CA), is the set inhibition rate; IR is the experimental value of the inhibition rate of alveolar macrophage activity calculated based on the dose-effect relationship equation of the metal element mixture in step (2) at the same concentration of the metal element mixture, The mean of a series of experimental values of inhibition rate on alveolar macrophage activity calculated based on the metal element mixture dose-effect relationship equation in step (2) at a series of metal element mixture concentrations for comparison.
[0043] More specifically, the present invention provides a method for establishing and screening a risk assessment toxicity quantitative prediction model for a typical multi-metal compound, nickel cobalt manganese oxide, in the lithium battery industry, comprising the following steps:
[0044] (1) Characterization of the physical and chemical properties of metal element types and relative proportions of multi-metal compounds: For lithium nickel cobalt manganese oxide, LiNi 0.8 Co 0.1 Mn0.1 O2 is used as the basic compound for modeling. On the one hand, it can be based on LiNi 0.8 Co 0.1 Mn 0.1 O2 chemical formula, the toxic metal elements are Ni, Co and Mn (Li is not included in the following because it is a light metal, metabolized rapidly in the human body and is basically non-toxic), and the relative proportions of the amounts of Ni, Co and Mn are 8 / 1 / 1 respectively; on the other hand, the energy dispersive X-ray spectrometer (EDS) of a scanning electron microscope can also be used to characterize LiNi 0.8 Co 0.1 Mn 0.1 The relative proportion of the metal element type and the atomic percentage of the metal element in the O2 multi-metal compound;
[0045] (2) Testing the biological toxicity of the released elements: According to the metal element types and relative proportions of the metal elements obtained in step (1), two exposure conditions were used: exposure to Ni, Co and Mn metal elements and exposure to a metal element mixture in which all metal elements were mixed at a ratio of Ni / Co / Mn=8 / 1 / 1 (the concentration of the metal element mixture was expressed as the sum of the concentrations of Ni, Co and Mn metal elements), and the biological toxicity (such as inhibition rate percentage, IR) on alveolar macrophage viability at different exposure doses was determined; Origin software was used to fit the logistic method to establish a dose-response relationship equation with the exposure dose as the independent variable and the biological toxicity of alveolar macrophage viability as the dependent variable, and the slope pi and IC of the dose-response relationship equation for each metal element were obtained based on the dose-response relationship equation. 50，i Dose, maximum inhibition rate IR max，i and the minimum inhibition rate IR min，i , IC of metal element mixture 50，mixture Dose, maximum inhibition rate IR min，mixture and the minimum inhibition rate IR max，mixture ;
[0046] Among them, the single metal dose-response relationship equation of logistic fitting is shown in formula ①
[0047]
[0048] In formula ①, IR(c i ) is the percentage of inhibition of the i-th single metal element on the activity of alveolar macrophages, IR min，i is the minimum value, IR max，i is the maximum value, c i is the concentration of the ith single metal element, IC 50，iis the concentration of the ith single metal element when the inhibition rate on the activity of alveolar macrophages is 50%, pi is the slope of the dose-response equation of the ith metal element; i = 1, 2, 3, representing Ni, Co and Mn, respectively;
[0049] The dose-response relationship equation of the metal element mixture fitted by logistic fitting is shown in formula ②
[0050]
[0051] In formula ②, IR(c mixture ) is the percentage of inhibition of the metal element mixture on the activity of alveolar macrophages, IR min，mixture is the minimum value, IR max，mixture is the maximum value, c mixture is the concentration of metal element mixture, IC 50，mixture is the concentration of metal element mixture at which the inhibition rate of alveolar macrophage activity is 50%, p mixture is the slope of the dose-response relationship equation;
[0052] (3) Establish five different quantitative prediction models for biological toxicity
[0053] 1) Concentration Additive Model (CA):
[0054] The concentration addition model (CA) is shown in formula ③
[0055]
[0056] In formula ③, IC x，mixture is the concentration of the metal element mixture when the inhibition rate of the metal element mixture on the activity of alveolar macrophages is x, L i is the proportion of the ith toxic metal element in the total toxic metal elements; IC x，i is the concentration of the metal element when the inhibition rate of the i-th metal element on the activity of alveolar macrophages is x, which can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); n=3, i=1, 2, 3, representing Ni, Co and Mn respectively;
[0057] 2) Independent Action Model (IA):
[0058] The independent action model (IA) is shown in formula ④
[0059]
[0060] In formula ④, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); c i is calculated based on the proportion of the ith metal element in the total toxic metal elements when the concentration of the metal element mixture is c mixture is the concentration of the ith metal, and n represents the number of toxic metal elements in the compound;
[0061] 3) Effect Additive Model (ES):
[0062] The effect addition model (ES) model equation is shown in formula ⑤
[0063]
[0064] In formula ⑤, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i ) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); c i is the ratio of the ith metal element to the total toxic metal elements in the metal element mixture with a concentration of c mixture is the concentration of the ith metal, and n represents the number of toxic metal elements in the compound;
[0065] 4) Integrated Additive Model (IAM):
[0066] The model equation of the integrated additive model (IAM) is shown in formula ⑥
[0067]
[0068] In formula ⑥, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of 50，i The IC of the ith metal is obtained based on the dose-response equation of the ith metal element in the metal mixture in step (2). 50 Dose; c i is the ratio of the ith metal element to the total toxic metal elements in the metal element mixture with a concentration of c mixture The concentration of the i-th metal at time ; where p iis the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2), where n represents the number of toxic metal elements in the compound;
[0069] 5) Integrated Additive Interaction Model (IAI):
[0070] The IAI model equation is shown in formula ⑦
[0071]
[0072] In formula ⑦, c mixture is the concentration of the metal element mixture, c i is the concentration of the ith metal element in the metal element mixture, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of K i is a function that describes the effect of the ith metal element on the total toxicity, IC 50，i is the concentration of the i-th metal element obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%, IC 50，mixture The concentration of the metal element mixture obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%; where p i is the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2), where n represents the number of toxic metal elements in the compound;
[0073] (4) Model accuracy evaluation and optimal model screening: The prediction results of the above models are compared with the toxicity test data. On the one hand, different concentrations of metal element mixtures are set, and the proportion of each metal element in the total toxic metal elements is obtained according to step (1): L1 = L Ni =8 / 10, L2=L Co =1 / 10, L3=L Mn =1 / 10, calculate the concentration c of each single metal i ; Then, based on the dose-response equation of the ith metal element in the metal mixture in step (2), IR (c i ), the IC of the i-th metal element is known 50，i , slope p i IC of metal element mixture 50，mixture, substitute the above four prediction models, namely, independent action model (IA), effect addition model (ES), integrated additive model (IAM) and integrated additive interaction model (IAI), to calculate the toxicity prediction value of each model; compare the calculated biological toxicity prediction value of the metal element mixture with the biological toxicity value of the metal element mixture at the concentration calculated according to the dose-response relationship equation based on the experimental data of the metal element mixture in step (2), and calculate the value by R square (R 2 ) and root mean square error (RMSE) were used to evaluate the goodness of fit and prediction accuracy of each model;
[0074] On the other hand, different biological toxicity values of metal element mixtures are set, and the inhibition rate of the ith metal element on the activity of alveolar macrophages is calculated based on formula ① in step (2) at the metal element concentration under the set inhibition rate, and the concentration addition model (CA) is brought into the concentration addition model to obtain the predicted metal element mixture concentration under the set inhibition rate; the set inhibition rate is further compared with the biological toxicity value of the metal element mixture concentration calculated according to the dose-response relationship equation based on the experimental data of the metal element mixture in step (2); the R square (R 2 ) and root mean square error (RMSE) were used to evaluate the goodness of fit and prediction accuracy of each model;
[0075] R 2 The closer it is to 1, the smaller the RMSE value is and the better the prediction ability is. Based on this, the model with the best prediction accuracy is selected.
[0076] Among them, R 2 The calculation of is shown in formula ⑧
[0077]
[0078] The calculation of RMSE is shown in formula ⑨
[0079]
[0080] In formulas ⑧ and ⑨, for the above four prediction models, namely, the independent action model (IA), the effect addition model (ES), the integrated additive model (IAM), and the integrated additive interaction model (IAI), is the inhibition rate of the metal element mixture on alveolar macrophage viability predicted by each prediction model. For the concentration addition model (CA), is the set inhibition rate; IR is the experimental value of the inhibition rate of alveolar macrophage activity calculated based on the dose-effect relationship equation of the metal element mixture in step (2) at the same concentration of the metal element mixture, The mean of a series of experimental values of inhibition rate on alveolar macrophage activity calculated based on the metal element mixture dose-effect relationship equation in step (2) at a series of metal element mixture concentrations for comparison;
[0081] The present invention also provides a toxicity prediction model for lithium nickel cobalt manganese oxide metal oxide, which is an integrated additive interaction model IAI:
[0082] The IAI model equation is shown below:
[0083]
[0084] In the formula, n=3, i=1, 2, 3, representing Ni, Co and Mn respectively; c i is the concentration of the ith toxic metal element released by the lithium nickel cobalt manganese oxide to be tested; IR(c mixture ) is a mixture of toxic metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of K i is a function describing the effect of the ith metal element on the total toxicity, K1 = 0.3621, K2 = 0.2625, K3 = 1.3969; IC 50，1 =0.3714mM, IC 50，2 =0.2692mM, IC 50，3 =1.4325mM; p'=1.3996.
[0085] When the optimal model screened out by the present invention is actually applied, it is only necessary to detect the concentration of each toxic metal element released by the compound to be tested and substitute it into the model to obtain the toxicity of the compound.
[0086] Compared with the prior art, the present invention is based on the physicochemical properties of the metal element types and the relative proportions of the metal elements in the multi-metal compound, and further combines the biological toxicity experimental data of single metal elements and multi-metal element mixtures to establish a toxicity prediction IAI model; the method of the present invention is simple and has high prediction accuracy. The established IAI model is suitable for toxicity assessment of other types of multi-metal compounds with the same metal elements but different metal element proportions, and provides a quick and effective prediction method for evaluating the biological toxicity of multi-metal compounds. Description of the drawings:
[0087] Figure 1 The present invention provides a flow chart of a toxicity prediction method for risk assessment of multi-metal compounds.
[0088] Figure 2 The multi-metal compound LiNi 0.8 Co 0.1 Mn 0.1Dose-response curves of logistic fitting of the concentration and cell viability inhibition rate of Ni, Co, and Mn alone and the metal element mixture exposure in a relative ratio of 8:1:1 in O2, where A is single metal exposure and B is metal element mixture exposure.
[0089] Figure 3 The present invention relates to a dose-effect curve of a multi-metal compound toxicity prediction model and a model screening result diagram, wherein A is a multi-metal compound LiNi 0.8 Co 0.1 Mn 0.1 The dose-response curves of the exposure concentration of the metal element mixture of Ni, Co, and Mn in the relative ratio of 8:1:1 in O2 and the logistic fitting of the inhibition rate of cell viability, as well as the dose-response curves predicted by the CA, IA, ES, IAM, and IAI models; B is the results of screening the five models using RMSE; C is the results of screening the five models using R 2 Filter 5 model results.
[0090] Figure 4 The multi-metal compound toxicity prediction model of the present invention is for the multi-metal compound LiNi 1 / 3Co 1 / 3 Mn 1 / 3 O2 and LiNi 0.5 Co 0.2 Mn 0.3 Comparison of the toxicity prediction accuracy of O2-releasing element mixtures, where A is based on the multi-metal compound LiNi 1 / 3 Co 1 / 3 Mn 1 / 3 The dose-response curve of the exposure concentration of the metal element mixture of Ni, Co, and Mn in a relative ratio of 1:1:1 in O2 and the logistic fitting of the inhibition rate of cell viability and the dose-response curve predicted by the IAI model; B is based on the multi-metal compound LiNi 0.5 Co 0.2 Mn 0.3 The dose-response curve of the exposure concentration of the metal element mixture of Ni, Co, and Mn in the relative ratio of 5:2:3 in O2 and the logistic fitting of the inhibition rate of cell viability and the dose-response curve predicted by the IAI model; C is the comparison of the accuracy of IAI prediction of the toxicity of the element mixture released by three multi-metal compounds using RMSE; D is the comparison of the accuracy of IAI prediction of the toxicity of the element mixture released by three multi-metal compounds using R 2 The accuracy of IAI in predicting the toxicity of element mixtures released from three multi-metal compounds was compared. Specific implementation method:
[0091] The present invention is described in detail below through specific embodiments in conjunction with the accompanying drawings.
[0092] Embodiment 1:
[0093] This embodiment relates to a method for predicting the toxicity of a multi-metal compound. For lithium nickel cobalt manganese oxide, LiNi 0.8 Co 0.1 Mn 0.1 O2 is the basic compound for modeling. Based on the toxic metal element composition and relative proportion of the compound, combined with the toxicity data of single elements and element mixtures, five toxicity prediction models are established and screened, such as Figure 1 As shown; the specific steps are as follows:
[0094] (1) Characterization of the physical and chemical properties of metal element types and relative proportions of multi-metal compounds: On the one hand, based on LiNi 0.8 Co 0.1 Mn 0.1 O2 chemical formula, the toxic metal elements are Ni, Co and Mn (Li is not included in the subsequent analysis because Li is a light metal, which is rapidly metabolized in the human body and is basically non-toxic), and the relative proportions of the toxic metal elements Ni, Co and Mn are 8:1:1 respectively; on the other hand, the energy dispersive X-ray spectrometer (EDS) of the scanning electron microscope is used to characterize LiNi 0.8 Co 0.1 Mn 0.1 The toxic metal elements of O2 multi-metal compounds are also Ni, Co and Mn, and the relative ratio of the atomic percentage of the metal elements is 8.4:0.9:0.7, which is close to 8:1:1 in the chemical formula. Therefore, Ni:Co:Mn=8:1:1 is uniformly adopted, with Ni accounting for 8 / 10, Co accounting for 1 / 10, and Mn accounting for 1 / 10;
[0095] (2) Testing the biological toxicity of the released elements: Based on the three types of metal elements Ni, Co and Mn and the relative ratios of the metal elements Ni, Co and Mn obtained in step (1), single metal element Ni, Co and Mn were exposed, as well as a metal element mixture of Ni, Co and Mn in a ratio of 8:1:1 was exposed, and the inhibition rate of alveolar macrophage (MH-S cell line) activity at exposure concentrations of 0.1, 1, 10, 50, 100, 200, 400, 800, 1000, 2000 and 3000 μM for 24 hours was measured; based on the exposure concentrations and the measured inhibition rates, as shown in FIG. Figure 2As shown in the figure, the logistic method in Origin software was used to fit the dose-response relationship equations ① and ②, with the exposure dose (c) as the independent variable and the inhibition rate of alveolar macrophage activity (IR) as the dependent variable. The slope pi and IC of the dose-response relationship equation of each metal element were obtained according to the dose-response relationship equation. 50，i Dose, maximum inhibition rate IR max，i and the minimum inhibition rate IR min，i , IC of metal element mixture 50，mixture Dose, maximum inhibition rate IR min，mixture and the minimum inhibition rate IR max，mixture ;
[0096] Among them, the single metal dose-response relationship equation of logistic fitting is shown in formula ①
[0097]
[0098] In formula ①, IR(c i ) is the percentage of inhibition of the i-th single metal element on the activity of alveolar macrophages, IR min，i is the minimum inhibition rate, IR max，i is the maximum inhibition rate, c i is the concentration of the ith single metal element, IC 50，i is the concentration of the ith single metal element when the inhibition rate on the activity of alveolar macrophages is 50%, pi is the slope of the dose-response equation of the ith metal element; i = 1, 2, 3, representing Ni, Co and Mn, respectively;
[0099] The dose-response relationship equation of the metal element mixture fitted by logistic fitting is shown in formula ②
[0100]
[0101] In formula ②, IR(c mixture ) is the percentage of inhibition of the metal element mixture on the activity of alveolar macrophages, IR min，mixture is the minimum inhibition rate, IR max，mixture is the maximum inhibition rate, c mixture is the concentration of metal element mixture, IC 50，mixture is the concentration of metal element mixture at which the inhibition rate of alveolar macrophage activity is 50%, p mixture is the slope of the dose-response relationship equation;
[0102] For Ni, i = 1, IR min，1 =-0.1341,IR max，1 =85.5001, p1=1.0898, IC 50，1=0.3714mM;
[0103] For Co, i = 2, IR min，2 =-23.8700,IR max，2 =78.3894, p2=2.1870, IC 50，2 =0.2692mM;
[0104] For Mn, i=3, IR min，3 =-12.6202,IR max，3 =97.0164, p3=0.9221, IC 50，3 =1.4325mM;
[0105] For a mixture of Ni, Co and Mn in a ratio of 8:1:1, IR min，mixture =4.4518,IR max，mixture =122.9492, p mixture =1.3526,IC 50，mixture =1.0255mM;
[0106] (3) Establishment of five quantitative prediction models for biological toxicity
[0107] 1) Concentration addition model (CA) is shown in formula ③
[0108]
[0109] In formula ③, IC x，mixture is the concentration of the metal element mixture when the inhibition rate of the metal element mixture on the activity of alveolar macrophages is x, L i is the proportion of the ith toxic metal element in the total toxic metal elements; IC x，i is the concentration of the metal element when the inhibition rate of the i-th metal element on the activity of alveolar macrophages is x, which can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); n = 3, i = 1, 2, 3, representing Ni, Co and Mn respectively
[0110] For the concentration addition model (CA), firstly, based on step (1), the proportions of Ni, Co and Mn metal elements in the total toxic metal elements are obtained as L1=8 / 10, L2=1 / 10 and L3=1 / 10; then, according to the dose-response relationship equation of each metal element in step (2), the concentrations of each metal element when the inhibition rate of alveolar macrophage activity is 5, 10, 15, 20, 15, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95 are calculated; finally, according to formula ③, the predicted concentration values of the metal element mixture when the inhibition rate of the metal element mixture on alveolar macrophage activity is 5, 10, 15, 20, 15, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95 are calculated, and the model prediction value fitting curve is as follows: Figure 3 As shown in A;
[0111] 2) Independent Action Model (IA):
[0112] The independent action model (IA) is shown in formula ④
[0113]
[0114] In formula ④, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i ) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); c i is calculated based on the proportion of the ith metal element in the total toxic metal elements when the concentration of the metal element mixture is c mixture The concentration of the i-th metal at the time; n represents the number of toxic metal elements in the compound; n = 3, i = 1, 2, 3, representing Ni, Co and Mn respectively;
[0115] For the independent action model (IA), firstly, based on step (1), the proportions of Ni, Co and Mn metal elements in the total toxic metal elements were obtained: L1 = 8 / 10, L2 = 1 / 10, and L3 = 1 / 10. For the convenience of subsequent comparison, a series of concentrations of toxic metal element mixtures (131.0333, 225.8604, 315.1819, 403.7896, 494.0667, 587.6882, 686.1103, 790.7878, 903.3132, 1025.5386, 1159.7133, 1308.6 635, 1476.0462, 1666.7315, 1887.4040, 2147.5551, 2461.2079, 2850.1123, 3350.1416, 4025.4764 μM), calculate the concentration c1, c2, c3 of each toxic metal element; then based on the dose-response relationship equation of each metal element in step (2), calculate the inhibition rate IR(c1), IR(c2), IR(c3) of each toxic metal element on the activity of alveolar macrophages; finally, according to formula ④, the predicted value of the independent action model (IA) is obtained, and the fitting curve is as follows: Figure 3 As shown in A;
[0116] 3) Effect Additive Model (ES):
[0117] The effect addition model (ES) model equation is shown in formula ⑤
[0118]
[0119] In formula ⑤, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i ) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); c i is the ratio of the ith metal element to the total toxic metal elements in the metal element mixture with a concentration of c mixture The concentration of the i-th metal at the time; n represents the number of toxic metal elements in the compound; n = 3, i = 1, 2, 3, representing Ni, Co and Mn respectively;
[0120] For the effect addition model (ES), firstly, based on step (1), the proportions of Ni, Co and Mn metal elements in the total toxic metal elements are obtained: L1 = 8 / 10, L2 = 1 / 10, L3 = 1 / 10. According to a series of concentrations of the set metal element mixture (131.0333, 225.8604, 315.1819, 403.7896, 494.0667, 587.6882, 686.1103, 790.7878, 903.3132, 1025.5386, 1159.7133, 1308.6635, 1476.0462, 1666.7315, 1887.4040, 2147.5551, 2461.2079, 2850.1123, 3350.1416, 4025.4764 μM), calculate the concentration c1, c2, c3 of each metal element; then based on the dose-response relationship equation of each metal element in step (2), calculate the inhibition rate IR(c1), IR(c2), IR(c3) of each metal element on the activity of alveolar macrophages; finally, according to formula ⑤, the predicted value of the effect addition model (ES) is obtained, and the fitting curve is as follows: Figure 3 As shown in A;
[0121] 4) Integrated Additive Model (IAM):
[0122] The model equation of the integrated additive model (IAM) is shown in formula ⑥
[0123]
[0124] In formula ⑥, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of 50，i The IC of the ith metal is obtained based on the dose-response equation of the ith metal element in the metal mixture in step (2). 50 Dose; c i is the ratio of the ith metal element to the total toxic metal elements in the metal element mixture with a concentration of c mixture The concentration of the i-th metal at time ; where p i is the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2); n represents the n toxic metal elements in the compound; n=3, i=1, 2, 3, representing Ni, Co and Mn respectively;
[0125] For the integrated additive model (IAM), firstly, based on the slopes p of the dose-response equations of Ni, Co and Mn metal elements in step (3), Ni=p1,p Co =p2, p Mn =p3, according to the formula The metal element mixture p′=1.3996 is calculated, and then based on step (1), the proportions of Ni, Co and Mn metal elements in the total toxic metal elements are obtained as L1=8 / 10, L2=1 / 10, L3=1 / 10, and the concentrations of each metal element c1, c2, c3 are calculated based on a set series of metal element mixture concentrations; finally, based on the IC values of Ni, Co and Mn metal elements obtained in step (2), 50 Dosage IC 50，1 、IC 50，2 、IC 50，3 According to formula ⑥, the prediction value of the integrated additive model (IAM) is obtained, and the fitting curve is as follows Figure 3 As shown in A;
[0126] 5) Integrated Additive Interaction Model (IAI):
[0127] The IAI model equation is shown in formula ⑦
[0128]
[0129] In formula ⑦, c mixture is the concentration of the metal element mixture, c i is the concentration of the ith metal element in the metal element mixture, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of K i is a function that describes the effect of the ith metal element on the total toxicity, IC 50，i is the concentration of the i-th metal element obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%, IC 50，mixture The concentration of the metal element mixture obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%; where p i is the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2); n represents the n toxic metal elements in the compound; n=3, i=1, 2, 3, representing Ni, Co and Mn respectively;
[0130] For the integrated additive interaction model (IAI), the ICs of Ni, Co and Mn metal elements obtained in step (2) are first calculated. 50 Dosage IC 50，1 、IC 50，2 、IC50，3 IC with metal element mixture 50 Dosage IC 50，mixture , calculate the K of each metal element i The values K1 = 0.3621, K2 = 0.2625, K3 = 1.3969; further based on the slopes p1, p2, p3 of the dose-response equations of the Ni, Co and Mn metal elements, according to the formula The calculated p′ of the metal element mixture is 1.3996. Finally, based on the proportions of Ni, Co and Mn metal elements in the total toxic metal elements obtained in step (1), L1=8 / 10, L2=1 / 10, L3=1 / 10, the concentrations of each metal element c1, c2, c3 are calculated according to a set series of metal element mixture concentrations. The predicted value of the integrated additive interaction model (IAI) is obtained according to formula ⑩. The predicted value fitting curve is as follows: Figure 3 As shown in A;
[0131] (4) Evaluation of model accuracy and screening of the optimal model: On the one hand, the predicted values calculated by the four models IA, ES, IAM and IAI in step (3) are compared with the experimental values calculated by the dose-response relationship equation based on the experimental data of the metal element mixture in step (2); on the other hand, a series of inhibition rates (5, 10, 15, 20, 15, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95) are set, and the metal element concentrations of Ni, Co and Mn metal elements on the activity of alveolar macrophages under the set inhibition rate are calculated based on formula ① in step (2), and the metal element concentrations under the set inhibition rate are brought into the concentration addition model (CA) to obtain the predicted metal element mixture concentration under the set inhibition rate; based on the metal element mixture concentration calculated by the CA model in step (3), the set series of inhibition rates are compared with the experimental values calculated by the dose-response relationship equation based on the experimental data of the metal element mixture in step (2) under the predicted concentration; the R square (R 2 ) and root mean square error (RMSE) were used to evaluate the goodness of fit and prediction accuracy of each model. 2 The closer it is to 1, the smaller the RMSE value is and the better the prediction ability is. The results are as follows Figure 4 B and 4C, the R of model IAI 2 =0.9682 is closest to 1, and RMSE = 5.1389 is the smallest value, so the model with the best prediction accuracy is the integrated additive interaction model (IAI);
[0132] The optimal prediction model IAI selected is as follows:
[0133]
[0134] where c mixture is the concentration of the metal element mixture, c i is the concentration of the ith metal element in the metal element mixture, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, n = 3, i = 1, 2, 3, representing Ni, Co and Mn respectively; K1 = 0.3621, K2 = 0.2625, K3 = 1.3969; IC 50，1 =IC 50，Ni =0.3714mM, IC 50，2 =IC 50，Co =0.2692mM, IC 50，3 =IC 50，Mn =1.4325mM, p'=1.3996.
[0135] Embodiment 2:
[0136] This embodiment 2 is related to the accuracy evaluation of the optimal model IAI established in embodiment 1, and the application of the optimal model established in embodiment 1 to other multi-metal compounds with similar compositions LiNi 1 / 3 Co 1 / 3 Mn 1 / 3 The toxicity prediction accuracy of O2; the specific steps are as follows:
[0137] (1) Characterization of the physical and chemical properties of metal element types and relative proportions of multi-metal compounds: Based on LiNi 1 / 3 Co 1 / 3Mn 1 / 3 O2 chemical formula, the types of toxic metal elements are Ni, Co and Mn, and the relative proportions of the amounts of Ni, Co and Mn metal elements are 1:1:1, respectively, with Ni accounting for 1 / 3, Co accounting for 1 / 3, and Mn accounting for 1 / 3;
[0138] (2) Toxicity prediction based on the IAI model: In a series of toxic metal element mixture concentrations (204.3038, 298.0248, 376.5055, 448.9535, 519.1889, 589.4769, 661.5253, 736.8874, 817.1957, 904.3618, 1000.8137, 1109.8448, 1236.1979, 1387. 1436, 1574.6737, 1820.5344, 2169.8407, 2738.6406, 3983.4849, 39198.6435 μM), according to the proportion of each metal element, the concentration c1, c2, c3 of each metal element is calculated, where n=3, i=1, 2, 3, representing Ni, Co and Mn respectively; K1=0.3621, K2=0.2625, K3=1.3969; IC 50，1 =0.3714mM, IC 50，2 =0.2692mM, IC 50，3 =1.4325mM, p'=1.3996. Substituting into the IAI model of Example 1, the LiNi 1 / 3 Co 1 / 3 Mn 1 / 3 The predicted value of the inhibition rate of O2 on alveolar macrophage activity and the dose-response curve predicted by the model are as follows: Figure 4 As shown in A;
[0139] (3) Experimental test of biological toxicity of released elements: Based on the three types of metal elements Ni, Co and Mn obtained in step (1) and the relative ratio of metal elements Ni, Co and Mn, a metal element mixture prepared by mixing Ni, Co and Mn in a ratio of 1:1:1 was used for exposure, and the inhibition rate of alveolar macrophage (MH-S cell line) activity at exposure concentrations of 0.1, 1, 10, 50, 100, 200, 400, 800, 1000, 2000 and 3000 μM was measured; based on the exposure concentration and the measured inhibition rate, the logistic method in Origin software was used to fit and establish a dose-response relationship equation with the exposure dose (c) as the independent variable and the inhibition rate of alveolar macrophage activity (IR) as the dependent variable, and the IC of the metal element mixture was obtained according to the dose-response relationship equation. 50 Dose, dose response curve such as Figure 4 As shown in A;
[0140] Among them, the dose-response relationship equation of logistic fitting is shown in formula ②
[0141]
[0142] For a mixture of Ni, Co and Mn in a ratio of 1:1:1, IR min =1.1490,IR max =100.0577, p=1.9789, IC 50，mixture =0.9044mM;
[0143] Substitute the concentrations of the toxic metal element mixture used in the prediction model (204.3038, 298.0248, 376.5055, 448.9535, 519.1889, 589.4769, 661.5253, 736.8874, 817.1957, 904.3618, 1000.8137, 1109.8448, 1236.1979, 1387.1436, 1574.6737, 1820.5344, 2169.8407, 2738.6406, 3983.4849, 39198.6435 μM) into the above formula ② to obtain the experimental values of the inhibition rate of the corresponding concentrations;
[0144] (4) Model verification: The predicted value obtained in step (2) is compared with the experimental value obtained in step (3) and the R square (R 2 ) and root mean square error (RMSE) were used to evaluate the goodness of fit and prediction accuracy of each model. Figure 4 C, 4D shows that IAI for LiNi 1 / 3 Co 1 / 3 Mn 1 / 3 R of the comparison between the predicted and experimental values of O2 2 =0.9464, RMSE = 6.6783, compared with LiNi 0.8 Co 0.1 Mn 0.1 O2 is similar, indicating that according to Example (1) the multi-metal compound LiNi 0.8 Co 0.1 Mn 0.1 The IAI model established by O2 is applicable to other multi-metal compounds with the same element type, such as LiNi 1 / 3 Co 1 / 3 Mn 1 / 3 The biological toxicity prediction of O2 has good accuracy, and the present invention has good applicability to the screening and establishment process of a toxicity prediction method for multi-metal compounds.
[0145] Embodiment 3:
[0146] This embodiment 3 is related to the accuracy evaluation of the optimal model IAI established in embodiment 1, and the application of the optimal model established in embodiment 1 to other multi-metal compounds with similar compositions LiNi 0.5 Co 0.2Mn 0.3 The toxicity prediction accuracy of O2; the specific steps are as follows:
[0147] (1) Characterization of the physical and chemical properties of metal element types and relative proportions of multi-metal compounds: Based on LiNi 0.5 Co 0.2 Mn 0.3 O2 chemical formula, the types of toxic metal elements are Ni, Co and Mn, and the relative proportions of the amounts of Ni, Co and Mn metal elements are 5:2:3, respectively, with Ni accounting for 5 / 10, Co accounting for 2 / 10, and Mn accounting for 3 / 10;
[0148] (2) Toxicity prediction based on the IAI model: In a series of toxic metal element mixture concentrations (213.8637, 279.0153, 342.7312, 406.8976, 472.8374, 541.6838, 614.5557, 692.6739, 777.4673, 870.6965, 974.6221, 1092.2553, 1227. 7566, 1387.1059, 1579.3035, 1818.6804, 2129.7764, 2558.9810 μM), according to the proportion of each metal element, the concentration c1, c2, c3 of each metal element is calculated. It is known that n = 3, i = 1, 2, 3, representing Ni, Co and Mn respectively; K1 = 0.3621, K2 = 0.2625, K3 = 1.3969; IC 50，1 =0.3714mM, IC 50，2 =0.2692mM, IC 50，3 =1.4325mM, p'=1.3996. Substituting into the IAI model of Example 1, the LiNi 0.5 Co 0.2 Mn 0.3 The predicted value of the inhibition rate of O2 on alveolar macrophage activity and the dose-response curve predicted by the model are as follows: Figure 4 B
[0149] (3) Experimental test of biological toxicity of released elements: Based on the three types of metal elements Ni, Co and Mn obtained in step (1) and the relative ratio of metal elements Ni, Co and Mn, a metal element mixture of Ni, Co and Mn in a ratio of 5:2:3 was used for exposure, and the inhibition rate of alveolar macrophage (MH-S cell line) activity at exposure concentrations of 0.1, 1, 10, 50, 100, 200, 400, 800, 1000, 2000 and 3000 μM was measured; based on the exposure concentration and the measured inhibition rate, the logistic method in Origin software was used to fit and establish a dose-response relationship equation with the exposure dose (c) as the independent variable and the inhibition rate of alveolar macrophage activity (IR) as the dependent variable, and the IC of the metal element mixture was obtained according to the dose-response relationship equation. 50 Dose, dose response curve such as Figure 4 As shown in B;
[0150] Among them, the dose-response relationship equation of logistic fitting is shown in formula ②
[0151]
[0152] For a mixture of Ni, Co and Mn in a ratio of 5:2:3, IR min =-6.6471,IR max =109.5932, p=1.5273, IC 50，mixture =0.9001mM;
[0153] Substitute the concentrations of the metal element mixture used in the prediction model (213.8637, 279.0153, 342.7312, 406.8976, 472.8374, 541.6838, 614.5557, 692.6739, 777.4673, 870.6965, 974.6221, 1092.2553, 1227.7566, 1387.1059, 1579.3035, 1818.6804, 2129.7764, 2558.9810 μM) into the above formula ② to obtain the experimental values of the inhibition rate of the corresponding concentrations;
[0154] (4) Model verification: The predicted value obtained in step (2) is compared with the experimental value obtained in step (3) and the R square (R 2 ) and root mean square error (RMSE) were used to evaluate the goodness of fit and prediction accuracy of each model. Figure 4 C, 4D shows that IAI for LiNi 0.5 Co 0.2 Mn 0.3 R of the comparison between the predicted and experimental values of O2 2=0.9270, RMSE = 7.0416, and LiNi 0.8 Co 0.1 Mn 0.1 O2 and LiNi 1 / 3 Co 1 / 3 Mn 1 / 3 O2 is similar, indicating that according to Example (1) the multi-metal compound LiNi 0.8 Co 0.1 Mn 0.1 The IAI model established by O2 is applicable to other multi-metal compounds with the same element type, such as LiNi 0.5 Co 0.2 Mn 0.3 The biological toxicity prediction of O2 has good accuracy, and the present invention has good applicability to the screening and establishment process of a toxicity prediction method for multi-metal compounds.
Claims
1. A method for predicting the toxicity of a polynary metal compound, characterized in that: The specific steps include: (1) Characterization of the physicochemical properties of metal element types and relative proportions of multinary metal compounds: For a certain type of multinary metal compound, one of the multinary metal compounds is selected as the basic compound for modeling. From the known chemical equation or by using the energy dispersive X-ray spectrometer of a scanning electron microscope, the total number of toxic metal element types n and the relative proportion of the amount or atomic percentage of each toxic metal element substance contained in the multinary metal compound are obtained; (2) Testing the biological toxicity of the released elements: Based on the n types of toxic metal elements obtained in step (1) and the relative proportions of the toxic metal elements, two exposure conditions are used: exposure to a single metal element and exposure to a metal element mixture in which all metal elements are mixed in proportion, to determine the biological toxicity to alveolar macrophage viability at different exposure doses, such as the inhibition rate percentage IR; Origin software is used to fit the exposure dose as the independent variable and the biological toxicity of alveolar macrophage viability as the dependent variable to establish a dose-response relationship equation, and the slope pi and IC of the dose-response relationship equation for each metal element are obtained based on the dose-response relationship equation. 50，i Dosage and IC of Metal Mixtures 50，mixture dose; Among them, the single metal dose-response relationship equation of logistic fitting is shown in formula ①: In formula ①, IR(c i ) is the percentage of inhibition of the i-th single metal element on the activity of alveolar macrophages, IR min，i is the minimum inhibitory rate of the ith single metal on the activity of alveolar macrophages, IR max，i is the maximum inhibition rate of the ith single metal on the activity of alveolar macrophages, c i is the concentration of the ith single metal element, IC 50，i is the concentration of the ith single metal element when the inhibition rate on the activity of alveolar macrophages is 50%, and pi is the slope of the dose-response equation of the ith metal element; The dose-response relationship equation of the metal element mixture fitted by logistic fitting is shown in formula ②. In formula ②, IR(c mixture ) is the percentage of inhibition of the metal element mixture on the activity of alveolar macrophages, IR min，mixture is the minimum inhibitory rate of metal element mixture on alveolar macrophage activity, IR max，mixture is the maximum inhibition rate of metal element mixture on alveolar macrophage activity, c mixture is the concentration of metal element mixture, IC 50，mixture is the concentration of metal element mixture at which the inhibition rate of alveolar macrophage activity is 50%, p mixture is the slope of the dose-response relationship equation; (3) Establish five different quantitative prediction models for biological toxicity 1) Concentration Additive Model CA: The concentration addition model CA is shown in formula ③ In formula ③, IC x，mixture is the concentration of the metal element mixture when the inhibition rate of the metal element mixture on the activity of alveolar macrophages is x, L i is the proportion of the ith toxic metal element in the total toxic metal elements, IC x，i is the concentration of the metal element when the inhibition rate of the i-th metal element on the activity of alveolar macrophages is x, which can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); 2) Independent Action Model IA: The IA model is shown in formula ④ In formula ④, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i ) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); c i is calculated based on the proportion of the ith metal element in the total toxic metal elements when the concentration of the metal element mixture is c mixture is the concentration of the ith metal, and n represents the number of toxic metal elements in the compound; 3) Effect addition model ES: The ES model equation is shown in formula ⑤ In formula ⑤, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i ) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); c i is the ratio of the ith metal element to the total toxic metal elements in the metal element mixture with a concentration of c mixture is the concentration of the ith metal, and n represents the number of toxic metal elements in the compound; 4) Integrated Additive Model IAM: The IAM model equation is shown in formula ⑥ In formula ⑥, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of 50，i The IC of the ith metal is obtained based on the dose-response equation of the ith metal element in the metal mixture in step (2). 50 Dose; c i is the ratio of the ith metal element to the total toxic metal elements in the metal element mixture with a concentration of c mixture The concentration of the i-th metal at time ; where p i is the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2), where n represents the number of toxic metal elements in the compound; 5) Integrated additive interaction model IAI: The IAI model equation is shown in formula ⑦ In formula ⑦, c mixture is the concentration of the metal element mixture, c i is the concentration of the ith metal element in the metal element mixture, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of K i is a function that describes the effect of the ith metal element on the total toxicity, IC 50，i is the concentration of the i-th metal element obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%, IC 50，mixture The concentration of the metal element mixture obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%; where p i is the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2), where n represents the number of toxic metal elements in the compound; (4) Model accuracy evaluation: The prediction results of the above models were compared with the toxicity experimental data, and the fit and prediction accuracy of each model were evaluated by R square and root mean square error; R 2 The closer it is to 1, the smaller the RMSE value is and the better the prediction ability is, based on which the model with the best prediction accuracy is selected.
2. A method for predicting the toxicity of lithium nickel cobalt manganese oxide metal oxide, characterized in that: The following steps are involved: (1) Characterization of the physical and chemical properties of metal element types and relative proportions of multi-metal compounds: For lithium nickel cobalt manganese oxide, LiNi 0.8 Co 0.1 Mn 0.1 O2 is used as the basic compound for modeling, based on LiNi 0.8 Co 0.1 Mn 0.1 The chemical formula of O2 shows that the toxic metal elements are Ni, Co and Mn, and the proportions of the substances of Ni, Co and Mn in the total toxic metal elements are L1=8 / 10, L2=1 / 10, and L3=1 / 10; (2) Testing the biological toxicity of the released elements: According to the metal element types and relative proportions of the metal elements obtained in step (1), two exposure conditions were used: exposure to Ni, Co and Mn metal elements, and exposure to a metal element mixture in which all metal elements were mixed at a ratio of Ni / Co / Mn=8 / 1 / 1, and the biological toxicity (such as inhibition rate percentage, IR) on alveolar macrophage viability at different exposure doses was measured; Origin software was used to fit the exposure dose as the independent variable and the biological toxicity of alveolar macrophage viability as the dependent variable to establish a dose-response relationship equation, and the slope pi and IC of the dose-response relationship equation for each metal element were obtained based on the dose-response relationship equation. 50，i Dose, maximum inhibition rate IR max，i and the minimum inhibition rate IR min，i , IC of metal element mixture 50，mixture Dose, maximum inhibition rate IR min，mixture and the minimum inhibition rate IR max，mixture ; Among them, the single metal dose-response relationship equation of logistic fitting is shown in formula ① In formula ①, IR(c i ) is the percentage of inhibition of the i-th single metal element on the activity of alveolar macrophages, IR min，i is the minimum value, IR max，i is the maximum value, c i is the concentration of the ith single metal element, IC 50，i is the concentration of the ith single metal element when the inhibition rate on the activity of alveolar macrophages is 50%, pi is the slope of the dose-response equation of the ith metal element; i = 1, 2, 3, representing Ni, Co and Mn, respectively; The dose-response relationship equation of the metal element mixture fitted by logistic fitting is shown in formula ② In formula ②, IR(c mixture ) is the percentage of inhibition of the metal element mixture on the activity of alveolar macrophages, IR min，mixture is the minimum value, IR max，mixture is the maximum value, c mixture is the concentration of metal element mixture, IC 50，mixture is the concentration of metal element mixture at which the inhibition rate of alveolar macrophage activity is 50%, p mixture is the slope of the dose-response relationship equation; (3) Establish five different quantitative prediction models for biological toxicity 1) Concentration Additive Model CA: The CA model is shown in formula ③ In formula ③, IC x，mixture is the concentration of the metal element mixture when the inhibition rate of the metal element mixture on the activity of alveolar macrophages is x, L i is the proportion of the ith toxic metal element in the total toxic metal elements; IC x，i is the concentration of the metal element when the inhibition rate of the i-th metal element on the activity of alveolar macrophages is x, which can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); n=3, i=1, 2, 3, representing Ni, Co and Mn respectively; 2) Independent Action Model IA: The IA model is shown in formula ④ In formula ④, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i ) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); c i is calculated based on the proportion of the ith metal element in the total toxic metal elements when the concentration of the metal element mixture is c mixture is the concentration of the ith metal, and n represents the number of toxic metal elements in the compound; 3) Effect addition model ES: The ES model equation is shown in formula ⑤ In formula ⑤, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at different concentrations, IR(c i ) is the i-th metal element at its concentration c i The inhibition rate of alveolar macrophage activity when the concentration of iodine is 0.04 can be calculated based on the dose-response relationship equation of the i-th metal element in step (2); c i is the ratio of the ith metal element to the total toxic metal elements in the metal element mixture with a concentration of c mixture is the concentration of the ith metal, and n represents the number of toxic metal elements in the compound; 4) Integrated Additive Model IAM: The IAM model equation is shown in formula ⑥ In formula ⑥, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of 50，i The IC of the ith metal is obtained based on the dose-response equation of the ith metal element in the metal mixture in step (2). 50 Dose; c i is the ratio of the ith metal element to the total toxic metal elements in the metal element mixture with a concentration of c mixture The concentration of the i-th metal at time ; where p i is the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2), where n represents the number of toxic metal elements in the compound; 5) Integrated additive interaction model IAI: The IAI model equation is shown in formula ⑦ In formula ⑦, c mixture is the concentration of the metal element mixture, c i is the concentration of the ith metal element in the metal element mixture, IR(c mixture ) is a mixture of metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of K i is a function that describes the effect of the ith metal element on the total toxicity, IC 50，i is the concentration of the i-th metal element obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%, IC 50，mixture The concentration of the metal element mixture obtained in step (2) at which the activity of alveolar macrophages is inhibited by 50%; where p i is the slope of the dose-response equation of the i-th metal element in the metal element mixture, obtained according to the dose-response equation of each metal element in step (2), where n represents the number of toxic metal elements in the compound; (4) Model accuracy evaluation: The prediction results of the above models were compared with the toxicity test data, and the fit and prediction accuracy of each model were evaluated by R square and root mean square error; R 2 The closer it is to 1, the smaller the RMSE value is and the better the prediction ability is, based on which the model with the best prediction accuracy is selected.
3. A toxicity prediction model for lithium nickel cobalt manganese oxide metal oxide, characterized in that: The model is an integrated additive interaction model IAI: The IAI model equation is shown below In the formula, n=3, i=1, 2, 3, representing Ni, Co and Mn respectively; c i is the concentration of the ith toxic metal element released by the lithium nickel cobalt manganese oxide to be tested; IR(c mixture ) is a mixture of toxic metal elements in c mixture The inhibition rate of alveolar macrophage activity at the concentration of K i is a function describing the effect of the ith metal element on the total toxicity, K1 = 0.3621, K2 = 0.2625, K3 = 1.3969; IC 50，1 =0.3714mM, IC 50，2 =0.2692mM, IC 50，3 =1.4325mM; p'=1.3996.