Saturation magnetic induction prediction and component design of iron-based amorphous alloy and ultrahigh saturation magnetic induction iron-based amorphous / nanocrystalline alloy

Abstract

The invention discloses saturation magnetic induction prediction and component design of an iron-based amorphous alloy and an ultrahigh saturation magnetic induction iron-based amorphous / nanocrystalline alloy. The core effects of ferromagnetic element content, alloy mixing enthalpy and poor electronegativity are disclosed by combining a machine learning model XGBoost with SHAP analysis, the amorphous forming ability is not deteriorated while the high ferromagnetic element content is kept, extremely low coercive force can be obtained after annealing, and collaborative optimization of high saturation magnetic induction, the amorphous forming ability and the coercive force is realized. Parameters used by the component design criterion provided by the invention do not need experimental data, and can be directly calculated from alloy components. High-performance components are rapidly screened in combination with a machine learning model, the research and development period and cost of a traditional trial and error method are remarkably reduced, systematic theoretical guidance is provided for development of high-performance soft magnetic materials, and industrial feasibility is improved.

Description

Technical Field

[0001] The present invention relates to the field of iron-based amorphous / nanocrystalline alloys, and specifically to the prediction of saturation magnetic induction of iron-based amorphous alloys, composition design, and ultra-high saturation magnetic induction iron-based amorphous / nanocrystalline alloys. Background Art

[0002] Soft magnetic materials play a crucial role in energy transmission and conversion in electrical and electronic systems. Facing the increasingly severe global energy challenges, modern electronic devices are shifting towards miniaturization, higher operating frequencies, and higher energy efficiency. In this context, developing materials with excellent soft magnetic properties is crucial for promoting these technological advancements. In particular, the market has an urgent need for soft magnetic materials with both high saturation magnetic induction intensity and low coercivity, which is crucial for meeting the dual requirements of high power density, low loss, and device miniaturization.

[0003] For decades, iron-based amorphous / nanocrystalline alloys have become the most promising candidate materials to address this challenge due to their excellent soft magnetic properties. However, the saturation magnetic induction of iron-based amorphous / nanocrystalline alloys is usually between 1.2 - 1.7 T, lower than that of silicon steel (1.8 - 2.0 T), which limits their application in high power density devices. In addition, the existing high saturation magnetic induction alloys (such as Nanomet, Hitperm) have relatively high coercivity (7 - 100 A / m) or have strict requirements for the quality of amorphous precursors and heat treatment processes, resulting in high industrial production costs and low yields. Moreover, the traditional composition regulation method for developing amorphous alloys lacks systematic theoretical guidance and is inefficient, leading to a long R & D cycle, high costs, and it is difficult to balance the contradictions between high saturation magnetic induction and poor amorphous formation ability and high coercivity.

[0004] The development of iron-based amorphous / nanocrystalline soft magnetic alloys began in the 1960s, and its core goal is to achieve a comprehensive performance balance of high saturation magnetic induction intensity, low coercivity, and amorphous formation ability through material design. Historically, various systems of iron-based amorphous / nanocrystalline soft magnetic alloys have been developed successively at home and abroad. The Fe-Si-B system (such as METGLAS 2605SA1, saturation magnetic induction ≈ 1.56 T) has become the mainstream material for power transformers, but its saturation magnetic induction is still lower than that of silicon steel (1.8 - 2.0 T). The Fe-Si-B-Cu-Nb system (Finemet, saturation magnetic induction ≈ 1.24 T) significantly improves the magnetic permeability and reduces the coercivity to 0.5 A / m. Subsequently, systems such as NANOPERM (Fe-Zr-B) and NANOMET (Fe-Si-B-P-Cu) have emerged successively, and the saturation magnetic induction has been continuously improved, but they face the trade-off problem between amorphous formation ability and magnetic properties.

[0005] Annealing is a key step in regulating the nanocrystalline structure. The traditional process uses two-stage annealing: the first stage eliminates internal stress, and the second stage induces the precipitation of α-Fe(Si) nanocrystals. Annealing under an external magnetic field can optimize the magnetic domain orientation and reduce the coercivity.

[0006] Due to the lack of theoretical guidance on the correlation between the composition and magnetic properties of amorphous alloys, the traditional alloy design idea based on composition regulation is to balance the glass-forming ability and magnetic properties through the synergistic effect of elements. For example, adding B / Si can improve the glass-forming ability, but excessive addition will reduce the saturation magnetic induction; adding Cu / Nb as a nucleating agent can inhibit grain coarsening; adding Co / Ni can enhance the magnetic moment, but increase the cost. Such methods rely extremely on experience and require a large number of trial-and-error experiments. In recent years, the application of machine learning in the field of materials science has risen rapidly, driving the transformation of this field. Machine learning has great potential in data-driven aspects, providing a more efficient way to accelerate the discovery of new materials, especially having great advantages in the field of complex disordered materials such as amorphous alloys.

[0007] The Chinese patent application with the publication number CN116364205A discloses a prediction method for Fe-based nanocrystalline alloys based on machine learning, which realizes the collaborative optimization of high magnetic induction intensity and low coercivity through a data-driven model. The technical solution first constructs a database containing 348 sets of alloy composition, process parameters, and performance data, screens the as-quenched amorphous samples and cleans the redundant data, and retains 20 key features such as annealing temperature, valence electron concentration (VEC), and non-magnetic valence electron concentration (VEC1). Using the random forest model and the forward selection method, it is determined that the core features of Bs are VEC1, Fe content, and electronegativity, and the core features of coercivity are VEC, annealing temperature, and electronegativity, and a prediction framework is established through the XGBoost model. Further combined with the genetic algorithm NSGA-II for multi-objective optimization, it reveals the design criteria that the saturation magnetic induction increases as VEC1 decreases when VEC1 ≤ 0.78, and the coercivity decreases as VEC decreases when VEC ≤ 7.12, and screens out the optimal composition Fe 81.4 Si 6.8 B 8.6 Cu 0.7 P 1.9 Nb 0.6 , and after the alloy is subjected to melt spinning (roll speed 50 m / s) and annealing, the saturation magnetic induction reaches 1.72 T and the coercivity is as low as 3.8 A / m.

[0008] Chinese patent application with publication number CN117457117A discloses a method and system for synthesizing high-frequency nanocrystalline alloys. The technical solution first constructs a dataset containing 393 alloy compositions, process parameters, and performance indicators, covering 15 key elements (Fe, Co, Ni, Si, B, etc.) and 22 characteristic parameters (such as valence electron concentration VEC, non-magnetic valence electron concentration VEC1, crystallization temperature range ΔT, etc.). The XGBoost algorithm is used to train the amorphous formation ability prediction model. Through interpretability analysis, VEC1, VEC, B content, and ΔT are selected as the core influencing factors, and design criteria are established: VEC1>0.71, VEC<7.29, ΔT>121°C, and B content is 6.85 - 11.54 at%. Based on this criterion, a new alloy (such as Fe 72.8 Si 12.7 B 9.9 Cu 1.5 Nb 1 Al 0.52 Mn 0.8 Ga 0.8 ) is used to prepare 14 - 16 μm ultra-thin amorphous ribbons by melting and single-roll quenching. After nanocrystallization at 560°C and magnetic field annealing, a composite structure of α-Fe(Si) nanocrystals and amorphous matrix is formed. The coercivity of this alloy is as low as 0.5 A / m, the loss at 0.5 T / 10 kHz is 2.3 W / kg, and the surface roughness Ra of the ribbon is <0.8 μm.

[0009] The following are the main disadvantages of preparing high saturation magnetic induction iron-based amorphous alloys by traditional techniques:

[0010] 1. It is difficult to balance high saturation magnetic induction and low coercivity. Traditional methods improve the saturation magnetic induction by increasing the Fe content (>80 at%), but a high Fe content will cause grain coarsening (grain size >30 nm), increasing the pinning effect of magnetic domain walls, thus significantly increasing the coercivity. For example, Fe 85 Si 1 B 9 P 3 Cu 1 alloy can reach a saturation magnetic induction of 1.8 T, but the coercivity is as high as 7 - 100 A / m. There is a strong negative correlation between high saturation magnetic induction and coercivity, making it difficult to optimize them simultaneously.

[0011] 2. The amorphous formation ability is limited. A high Fe content will reduce the amorphous formation ability of the alloy, resulting in the easy formation of crystallization phases during rapid solidification, especially the surface crystallization layer will deteriorate the soft magnetic properties. This limits the flexibility of alloy composition design and makes it difficult to further improve the saturation magnetic induction by adjusting the Fe content.

[0012] 3. The component design relies on experience, resulting in low R & D efficiency. Traditional methods rely on the trial-and-error method to optimize components, lacking quantitative analysis of key physical parameters, leading to a long R & D cycle, high R & D costs, and difficulty in quickly screening out high-performance alloy components.

[0013] Although the above-mentioned publicly disclosed patent applications develop new soft magnetic alloys based on machine learning methods, there are still deficiencies. The disadvantage of the technical solution provided in CN116364205A is that it only considers the characteristics affecting the saturation magnetic induction in the alloy components and does not take into account the amorphous formation ability; in the design criteria proposed in the technical solution of CN117457117A, the crystallization temperature range ΔT cannot be obtained from the alloy components and needs to be measured experimentally in advance, which is not conducive to more efficient development of new alloys; the saturation magnetic induction of the alloys designed in these two patent applications has not exceeded 1.8T, limiting the wide application of the alloys. Summary of the Invention

[0014] The present invention provides a method for predicting the saturation magnetic induction of iron-based amorphous alloys, component design, and ultra-high saturation magnetic induction iron-based amorphous / nanocrystalline alloys.

[0015] The present invention provides a method for predicting the saturation magnetic induction and a component design criterion for realizing iron-based amorphous soft magnetic alloys with high saturation magnetic induction through a data mining method based on machine learning, and designs a series of new iron-based amorphous alloys according to this criterion. These alloys have ultra-high saturation magnetic induction intensity (1.85 - 1.92T) and extremely low coercivity (1.2 - 6A / m) after annealing heat treatment under an external magnetic field, and at the same time consider the amorphous formation ability, significantly improving the practicality and industrial feasibility of the materials.

[0016] The machine learning method adopted by the present invention establishes a quantitative relationship between the content of ferromagnetic elements, the mixing enthalpy of the alloy, the electronegativity difference, and the saturation magnetic induction. The mixing enthalpy of the alloy considers the amorphous formation ability of the alloy, while the content of ferromagnetic elements and the electronegativity difference limit the types and proportions of ferromagnetic elements and doping elements. The parameters in the prediction method and component design criterion proposed by the present invention can all be directly obtained from the alloy components without any data obtained through prior experiments. The saturation magnetic induction of the iron-based amorphous alloys developed by the present invention can reach 1.85 - 1.92T, and has extremely low coercivity of 1.2 - 6A / m. Its excellent performance enables it to be more widely applied to various electronic devices requiring miniaturization and high energy density application scenarios.

[0017] The specific technical solutions are introduced as follows:

[0018] [1] A method for predicting the saturation magnetic induction of an iron-based amorphous alloy, comprising:

[0019] Establish a dataset of the composition of iron-based amorphous alloys and their corresponding saturation magnetic induction. Calculate the following six key physical characteristics corresponding to the composition of iron-based amorphous alloys: the content of ferromagnetic elements, the mixing enthalpy of the alloy, the electronegativity difference, the average atomic radius, the mixing entropy, and the average electronegativity;

[0020] Establish an XGboost machine learning model and train it. The input of the XGboost machine learning model includes the proportion of each element in the composition of the iron-based amorphous alloy and the corresponding six key physical characteristics, and the output is the saturation magnetic induction;

[0021] Obtain the composition of the iron-based amorphous alloy to be predicted, calculate its six key physical characteristics, and use the trained XGboost machine learning model to predict the saturation magnetic induction.

[0022] In the method for predicting the saturation magnetic induction of the iron-based amorphous alloy, the dataset may contain any one or more of the following elements used to form the iron-based amorphous alloy: Fe, B, Si, C, P, Co, Nb, Mo, Ni, Ga, Cu, Zr, Al, Dy, Cr, Y, Hf, Nd, Gd, W.

[0023] In the method for predicting the saturation magnetic induction of the iron-based amorphous alloy, the ferromagnetic element may be Fe, or Fe and Co.

[0024] The calculation expression of the content of the ferromagnetic element may be: C Fe =N Fe / N total where C Fe represents the content of the ferromagnetic element, N Fe represents the number of atoms of the ferromagnetic element in the composition of the iron-based amorphous alloy, and N total represents the total number of atoms in the composition of the iron-based amorphous alloy.

[0025] The calculation expression of the mixing enthalpy of the alloy may be: where ΔH mix represents the mixing enthalpy of the alloy, i and j are the element labels in the composition of the iron-based amorphous alloy, C i =N i / N total 、C j =N j / N total respectively represent the content of element i and element j in the iron-based amorphous alloy, N i 、N j respectively represent the number of atoms of element i and element j in the iron-based amorphous alloy, represents the mixing enthalpy between element i and element j, which can be obtained by looking up a table or calculated according to the well-known methods in the art.

[0026] The calculation expression of the electronegativity difference may be: where δ χ represents the electronegativity difference, χ i represents the electronegativity of element i, represents the average electronegativity.

[0027] The calculation expression of the average atomic radius can be: where represents the average atomic radius, r i represents the atomic radius of element i.

[0028] The calculation expression of the mixing entropy can be: ΔS mix = ∑-RC i ln(C i ), where ΔS mix represents the mixing entropy, R represents the gas constant, and its value is 8.314 J / (mol·K).

[0029] The calculation expression of the average electronegativity can be:

[0030] [2] A method for designing the composition of an iron-based amorphous alloy with high saturation magnetic induction, designing the composition of the iron-based amorphous alloy according to the following criteria:

[0031] C Fe > 75%, C Fe represents the content of ferromagnetic elements in the iron-based amorphous alloy, C Fe = N Fe / N total , N Fe represents the number of atoms of ferromagnetic elements in the composition of the iron-based amorphous alloy, N total represents the total number of atoms in the composition of the iron-based amorphous alloy;

[0032] -18.7 kJ / mol < ΔH mix < -14 kJ / mol, ΔH mix represents the alloy mixing enthalpy, i, j are element labels in the composition of the iron-based amorphous alloy, C i = N i / N total , C j = N j / N total respectively represent the content of element i and element j in the iron-based amorphous alloy, N i , N j respectively represent the number of atoms of element i and element j in the iron-based amorphous alloy, represents the mixing enthalpy between element i and element j, which can be obtained by looking up a table or calculated according to the well-known methods in the art;

[0033] δ χ< 0.07, δ χ represents the electronegativity difference, χ i represents the electronegativity of element i, represents the average electronegativity,

[0034] In the above criteria, C Fe > 75% significantly improves the saturation magnetic induction, -18.7 kJ / mol < ΔH mix < -14 kJ / mol balances the high saturation magnetic induction and the ability to form amorphous, δ χ < 0.07 can increase the number of unpaired electrons of iron atoms in the alloy, enhance the magnetic moment, and thus obtain the composition of the iron-based amorphous alloy with high saturation magnetic induction.

[0035] The above composition design criteria can be obtained by performing SHAP analysis on the machine learning model described in [1].

[0036] The composition design method of the iron-based amorphous alloy with high saturation magnetic induction, the elements constituting the iron-based amorphous alloy can be selected from: Fe, B, Si, C, P, Co, Nb, Mo, Ni, Ga, Cu, Zr, Al, Dy, Cr, Y, Hf, Nd, Gd, W.

[0037] The ferromagnetic element can be Fe, or Fe and Co.

[0038] [3] An iron-based amorphous / nanocrystalline alloy, in atomic ratio, the chemical formula is:

[0039] Fe 70-x Co 16 Ni x Si 3 B 11 , where the range of x is 0 - 1; or,

[0040] (Fe 100-x Co x ) 85.5 Ni 1.5 B 9-y P 3 C 1 Mo y , where the range of x is 18 - 25 (such as 20, etc.), and the range of y is 0 - 1 (such as 0.5, etc.); or,

[0041] (Fe 100-x Co x ) 85.5 Ni 1.5 B 9-y P 3 C 1 V y, where the range of x is 18 to 25 (e.g., 20, etc.), and the range of y is 0 to 1 (e.g., 0.5, etc.).

[0042] Furthermore, in terms of atomic ratio, the chemical formula of the iron-based amorphous / nanocrystalline alloy can be any of the following:

[0043] Fe 69 Co 16 Ni 1 Si 3 B 11 ,

[0044] (Fe 80 Co 20 ) 85.5 Ni 1.5 B 8.5 P 3 C 1 Mo 0.5 ,

[0045] (Fe 75 Co 25 ) 85.5 Ni 1.5 B 8.5 P 3 C 1 V 0.5 ,

[0046] (Fe 75 Co 25 ) 85.5 Ni 1.5 B 9 P 3 C 1 ,

[0047] (Fe 80 Co 20 ) 85.5 Ni 1.5 B 9 P 3 C 1 ,

[0048] (Fe 82 Co 18 ) 85.5 Ni 1.5 B 9 P 3 C 1 .

[0049] The coercivity of the iron-based amorphous / nanocrystalline alloy is less than 6 A / m, such as 1.2 A / m, 1.3 A / m, 1.7 A / m, 1.8 A / m, 2.3 A / m, 2.4 A / m, 3.2 A / m, 3.3 A / m, 3.4 A / m, 3.5 A / m, 3.6 A / m, 3.9 A / m, 4.0 A / m, 5.5 A / m, 5.6 A / m, etc., and further does not exceed 5.6 A / m.

[0050] The saturation magnetic induction of the iron-based amorphous / nanocrystalline alloy is not less than 1.85 T, such as 1.87 T, 1.89 T, 1.90 T, 1.92 T, 1.95 T, etc.

[0051] [4] The preparation method of the iron-based amorphous / nanocrystalline alloy according to [3] includes:

[0052] Prepare an amorphous precursor according to the chemical formula;

[0053] Anneal the amorphous precursor in an external magnetic field at 340 - 420 °C, preferably 360 - 380 °C. The coercivity can be further reduced at the preferred annealing temperature.

[0054] The amorphous precursor can be a strip and can be prepared by the melt spinning method.

[0055] The direction of the external magnetic field is preferably consistent with the length direction of the amorphous precursor, which is beneficial to reducing the coercivity.

[0056] The magnitude of the external magnetic field can be 100 - 10000 A / m, such as 400 A / m, etc.

[0057] The annealing time can be 10 - 20 minutes, such as 15 minutes, etc.

[0058] Compared with the prior art, the beneficial effects of the present invention are:

[0059] The method for predicting the saturation magnetic induction of the iron-based amorphous alloy of the present invention can accurately predict the saturation magnetic induction of the iron-based amorphous alloy only through the alloy composition.

[0060] The criteria characteristics and innovation points of the composition design of the present invention are: the three selected parameters can be simply calculated through the alloy composition without any experimental measurement; the selected range of the alloy mixing enthalpy enables the consideration of the amorphous formation ability of the composition when designing the alloy, which is beneficial to meeting the requirements of production and preparation.

[0061] The ultra-high saturation magnetic induction iron-based amorphous / nanocrystalline alloy of the present invention can be composed of 5 - 7 elements among Fe, Co, Ni, Si, B, P, C, Mo, V, etc., meeting the parameter range of the composition design criteria of the present invention.

[0062] Through the combination of the machine learning model XGBoost and SHAP analysis, the present invention reveals the core roles of the ferromagnetic element content, alloy mixing enthalpy, and electronegativity difference. By regulating these parameters, while maintaining a high ferromagnetic element content, the glass-forming ability is not deteriorated, grain coarsening is inhibited, and the magnetic domain pinning effect is reduced, thereby achieving a balance between an ultra-high saturation magnetic induction (1.85 - 1.92 T) and an extremely low coercivity (1.2 - 6 A / m) after annealing, breaking through the bottleneck that it is difficult to balance the two in traditional alloys, and realizing the synergistic optimization of high saturation magnetic induction, glass-forming ability, and coercivity.

[0063] The parameters (ferromagnetic element content, alloy mixing enthalpy, electronegativity difference) used in the composition design criterion proposed by the present invention do not require experimental data and can be directly calculated from the alloy composition. Combining with the machine learning model to quickly screen high-performance compositions significantly reduces the R & D cycle and cost of the traditional trial-and-error method, provides systematic theoretical guidance for the development of high-performance soft magnetic materials, and improves the industrial feasibility. Brief Description of the Drawings

[0064] Figure 1 It is a range diagram for determining the iron content by SHAP analysis in the specific implementation manner.

[0065] Figure 2 It is a range diagram for determining the alloy mixing enthalpy by SHAP analysis in the specific implementation manner.

[0066] Figure 3 It is a range diagram for determining the electronegativity difference by SHAP analysis in the specific implementation manner.

[0067] Figure 4 It is the X-ray diffraction (XRD) pattern of six amorphous precursor ribbons in the specific implementation manner.

[0068] Figure 5 It is the differential scanning calorimetry (DSC) curve of six amorphous precursor ribbons in the specific implementation manner.

[0069] Figure 6 It is the curve of the change in coercivity of six amorphous precursor ribbons after being treated at different annealing temperatures in the specific implementation manner.

[0070] Figure 7 It is the magnetization curve of six alloy examples at the optimal annealing temperature in the specific implementation manner. Specific Implementation Manner

[0071] The present invention will be further described below in conjunction with the attached drawings and specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and not to limit the scope of the present invention. The operating methods without specific conditions noted in the following embodiments are usually carried out under conventional conditions or according to the conditions recommended by the manufacturer.

[0072] A method for predicting the saturation magnetic induction of an iron-based amorphous alloy, comprising:

[0073] Establish a data set of the compositions of iron-based amorphous alloys and their corresponding saturation magnetic inductions. In this example, literature data of 536 different iron-based amorphous alloys are collected, including 20 elements that make up iron-based amorphous alloys: Fe, B, Si, C, P, Co, Nb, Mo, Ni, Ga, Cu, Zr, Al, Dy, Cr, Y, Hf, Nd, Gd, W. The saturation magnetization intensity ranges from 0.33 to 1.92 T. Calculate the following 6 key physical characteristics corresponding to the composition of the iron-based amorphous alloy: ferromagnetic element content, alloy mixing enthalpy, electronegativity difference, average atomic radius, mixing entropy, and average electronegativity.

[0074] The ferromagnetic element is Fe, or Fe and Co.

[0075] The calculation expression for the ferromagnetic element content is: C Fe = N Fe / N total where C Fe represents the ferromagnetic element content, N Fe represents the number of ferromagnetic element atoms in the composition of the iron-based amorphous alloy, and N total represents the total number of atoms in the composition of the iron-based amorphous alloy.

[0076] The calculation expression for the alloy mixing enthalpy is: where ΔH mix represents the alloy mixing enthalpy, i and j are the element labels in the composition of the iron-based amorphous alloy, C i = N i / N total and C j = N j / N total represent the contents of element i and element j in the iron-based amorphous alloy respectively, and N i and N j represent the numbers of atoms of element i and element j in the iron-based amorphous alloy respectively, represents the mixing enthalpy between element i and element j, which can be obtained by looking up a table or calculated according to the well-known methods in the art.

[0077] The calculation expression for the electronegativity difference is: where δ χ represents the electronegativity difference, χ i represents the electronegativity of element i, represents the average electronegativity.

[0078] The calculation expression for the average atomic radius is: where represents the average atomic radius, and r i represents the atomic radius of element i.

[0079] The calculation expression of the mixing entropy is: ΔS mix = ∑-RC i ln(C i ), where ΔS mix represents the mixing entropy, R represents the gas constant, and its value is 8.314 J / (mol·K).

[0080] The calculation expression of the average electronegativity is:

[0081] An XGboost machine learning model is established and trained. The input of the XGboost machine learning model includes the proportion of each element in the iron-based amorphous alloy composition (in this example, the proportion of 20 elements, and the elements not present are 0) and the corresponding 6 key physical characteristics, and the output is the saturation magnetic induction.

[0082] Obtain the composition of the iron-based amorphous alloy to be predicted, calculate its 6 key physical characteristics, and use the trained XGboost machine learning model to predict the saturation magnetic induction to achieve accurate prediction of the saturation magnetic induction.

[0083] Based on the above XGboost machine learning model, SHAP analysis shows that C Fe , ΔH mix , δ χ are the core factors affecting the saturation magnetic induction. When C Fe > 75%, the saturation magnetic induction is significantly improved ( Figure 1 ). When -18.7 kJ / mol < ΔH mix < -14 kJ / mol, the high saturation magnetic induction and the amorphous formation ability are balanced ( Figure 2 ). When δ χ < 0.07, the number of unpaired electrons of iron atoms in the alloy can be increased, and the magnetic moment can be improved ( Figure 3 ). Thus, the composition design criterion for the iron-based amorphous alloy with high saturation magnetic induction is:

[0084] C Fe > 75%, where C Fe represents the content of ferromagnetic elements in the iron-based amorphous alloy, C Fe = N Fe / N total , N Fe represents the number of ferromagnetic element atoms in the iron-based amorphous alloy composition, and N total represents the total number of atoms in the iron-based amorphous alloy composition;

[0085] -18.7 kJ / mol < ΔH mix < -14 kJ / mol, where ΔH mix represents the alloy mixing enthalpy, i and j are element labels in the composition of the iron-based amorphous alloy, C i = N i / N total 、C j = N j / N total respectively represent the content of element i and element j in the iron-based amorphous alloy, N i 、N j respectively represent the number of atoms of element i and element j in the iron-based amorphous alloy, represents the mixing enthalpy between element i and element j, which can be obtained by looking up tables or calculated according to well-known methods in the art;

[0086] δ χ < 0.07, δ χ represents the electronegativity difference, χ i represents the electronegativity of element i, represents the average electronegativity,

[0087] Using the above criteria, the composition design of iron-based amorphous alloys with high saturation magnetic induction can be carried out.

[0088] Based on the above composition design method, Co is further introduced to enhance the Fe-Co ferromagnetic exchange interaction, and the following systems (by atomic ratio) are designed:

[0089] Fe 70-x Co 16 Ni x Si 3 B 11 , where the range of x is 0 - 1; or,

[0090] (Fe 100-x Co x ) 85.5 Ni 1.5 B 9-y P 3 C 1 Mo y , where the range of x is 18 - 25 (such as 20, etc.), and the range of y is 0 - 1 (such as 0.5, etc.); or,

[0091] (Fe 100-x Co x ) 85.5 Ni 1.5 B 9-y P 3 C 1 V y , where the range of x is 18 - 25 (such as 20, etc.), and the range of y is 0 - 1 (such as 0.5, etc.).

[0092] Specific alloy example samples include (by atomic ratio): Fe 69 Co 16 Ni 1 Si 3 B 11 , (Fe 80 Co 20 ) 85.5 Ni 1.5 B 8.5 P 3 C 1 Mo 0.5 , (Fe 75 Co 25 ) 85.5 Ni 1.5 B 8.5 P 3 C 1 V 0.5 , (Fe 75 Co 25 ) 85.5 Ni 1.5 B 9 P 3 C 1 , (Fe 80 Co 20 ) 85.5 Ni 1.5 B 9 P 3 C 1 , (Fe 82 Co 18 ) 85.5 Ni 1.5 B 9 P 3 C 1 .

[0093] The raw materials used to prepare the above six alloy example samples are all commercially available high-purity raw materials, including pure iron, pure cobalt, pure nickel, pure silicon, ferro-phosphorus alloy, pure carbon, pure vanadium, and pure molybdenum. The raw materials meeting the above alloy compositions are vacuum melted to prepare a master alloy ingot, and the obtained master alloy ingot is prepared into an amorphous precursor strip by the melt spinning method. The amorphous precursor strip is annealed by the magnetic field heat treatment method, with a longitudinal magnetic field of 400 A / m applied (i.e., the direction of the applied magnetic field is consistent with the length direction of the amorphous precursor strip) to reduce the magnetic anisotropy and coercivity. The annealing temperature range is 340 - 420 °C, and the annealing time is 15 min. After the heat preservation is completed, it is cooled to room temperature to obtain an amorphous / nanocrystalline soft magnetic alloy.

[0094] The XRD patterns of the amorphous precursor strips of these alloys are as Figure 4 shown. All XRD patterns show a broad peak at 2θ = 44 - 45°, indicating mainly an amorphous structure. Figure 5The DSC curves of six amorphous precursor ribbons are shown, and their crystallization temperature ranges from approximately 360 °C to 380 °C. Therefore, the subsequent annealing temperature was set between 340 °C and 420 °C, and the annealing time was 15 minutes. The prepared amorphous precursor ribbons were annealed in an external magnetic field. Annealing can relieve the magnetic anisotropy energy fluctuations caused by internal stress, thereby making the locally induced anisotropy tend to be consistent. As the annealing temperature increases, the coercivity of the alloy shows a trend of first decreasing and then increasing, indicating that there is an optimal annealing temperature at which the coercivity decreases significantly from 13 - 27 A / m to 1.2 - 5.6 A / m, as Figure 6 shown. The coercivity is much lower than that of crystalline silicon steel. The magnetization curve of the annealed sample with the lowest coercivity is shown in Figure 7 . Table 1 shows the coercivity performance indexes of alloy examples at different annealing temperatures, in units of A / m.

[0095] Table 1

[0096]

[0097] The performance indexes of the saturation magnetic induction of each alloy example at the optimal annealing temperature:

[0098] Fe 69 Co 16 Ni 1 Si 3 B 11 1.92 T, the optimal annealing temperature is 380 °C;

[0099] (Fe 80 Co 20 ) 85.5 Ni 1.5 B 8.5 P 3 C 1 Mo 0.5 1.87 T, the optimal annealing temperature is 360 °C;

[0100] (Fe 75 Co 25 ) 85.5 Ni 1.5 B 8.5 P 3 C 1 V 0.5 1.90 T, the optimal annealing temperature is 360 °C;

[0101] (Fe 75 Co 25 ) 85.5 Ni 1.5 B 9 P 3 C 11.85T, the optimal annealing temperature is 380 °C;

[0102] (Fe 80 Co 20 ) 85.5 Ni 1.5 B 9 P 3 C 1 1.89T, the optimal annealing temperature is 380 °C;

[0103] (Fe 82 Co 18 ) 85.5 Ni 1.5 B 9 P 3 C 1 1.92T, the optimal annealing temperature is 380 °C.

[0104] In addition, it should be understood that after reading the above description of the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of this application.

Claims

1. A method for predicting the saturation magnetic induction of an iron-based amorphous alloy, characterized in that: include: Establish a data set of the composition and corresponding saturation magnetic induction of iron-based amorphous alloys, and calculate the following six key physical characteristics according to the composition of iron-based amorphous alloys: ferromagnetic element content, alloy mixing enthalpy, electronegativity difference, average atomic radius, mixing entropy and average electronegativity; Establish an XGboost machine learning model and train it. The input of the XGboost machine learning model includes the proportion of each element in the iron-based amorphous alloy composition and the corresponding six key physical characteristics, and the output is the saturation magnetic induction; The composition of the iron-based amorphous alloy to be predicted is obtained, the six key physical characteristics are calculated, and the saturation magnetic induction is predicted using the trained XGboost machine learning model.

2. The method for predicting the saturation magnetic induction of an iron-based amorphous alloy according to claim 1, characterized in that: The data set contains any one or more of the following elements used to constitute the iron-based amorphous alloy: Fe, B, Si, C, P, Co, Nb, Mo, Ni, Ga, Cu, Zr, Al, Dy, Cr, Y, Hf, Nd, Gd, and W.

3. The method for predicting the saturation magnetic induction of an Fe-based amorphous alloy according to claim 1, characterized in that: The ferromagnetic element is Fe, or Fe and Co; The calculation expression of the ferromagnetic element content is: Fe =N Fe / N total , where C Fe Indicates the content of ferromagnetic elements, N Fe Indicates the number of ferromagnetic element atoms in the iron-based amorphous alloy composition, N total It represents the total number of atoms in the iron-based amorphous alloy composition; The calculation expression of the alloy mixing enthalpy is: Where ΔH mix represents the alloy mixing enthalpy, i and j are element labels in the iron-based amorphous alloy composition, C i =N i / N total , C j =N j / N total Respectively represent the content of element i and element j in the Fe-based amorphous alloy, N i 、N j They represent the atomic numbers of element i and element j in the Fe-based amorphous alloy, represents the mixing enthalpy between element i and element j; The calculation expression of the electronegativity difference is: where δ χ Indicates poor electronegativity, χ i represents the electronegativity of element i, represents the average electronegativity; The calculation expression of the average atomic radius is: in represents the average atomic radius, r i represents the atomic radius of element i; The calculation expression of the mixing entropy is: ΔS mix =∑-RC i ln(C i ), where ΔS mix represents the mixing entropy, and R represents the gas constant; The calculation expression of the average electronegativity is:

4. A composition design method for an iron-based amorphous alloy with high saturation magnetic induction, characterized in that: The composition of Fe-based amorphous alloys is designed according to the following criteria: C Fe >75%, C Fe Indicates the content of ferromagnetic elements in Fe-based amorphous alloys, C Fe =N Fe / N total , N Fe Indicates the number of ferromagnetic element atoms in the iron-based amorphous alloy composition, N total It represents the total number of atoms in the iron-based amorphous alloy composition; -18.7kJ / mol<ΔH mix <-14 kJ / mol, ΔH mix represents the alloy mixing enthalpy, i and j are element labels in the iron-based amorphous alloy composition, C i =N i / N total , C j =N j / N total Respectively represent the content of element i and element j in the Fe-based amorphous alloy, N i 、N j They represent the atomic numbers of element i and element j in the Fe-based amorphous alloy, represents the mixing enthalpy between element i and element j; δ χ <0.07, δ χ It indicates poor electronegativity. χ i represents the electronegativity of element i, represents the average electronegativity, 5. The composition design method of the high saturation magnetic induction iron-based amorphous alloy according to claim 4, characterized in that: The elements constituting the iron-based amorphous alloy are selected from the group consisting of Fe, B, Si, C, P, Co, Nb, Mo, Ni, Ga, Cu, Zr, Al, Dy, Cr, Y, Hf, Nd, Gd, and W; The ferromagnetic element is Fe, or Fe and Co.

6. An iron-based amorphous / nanocrystalline alloy, characterized in that: In terms of atomic ratio, the chemical expression is: Fe 70-x Co 16 Ni x Si3B 11 , where x ranges from 0 to 1; or, (Fe 100-x Co x ) 85.5 Ni 1.5 B 9-y P3C1Mo y , where x ranges from 18 to 25 and y ranges from 0 to 1; or, (Fe 100-x Co x ) 85.5 Ni 1.5 B 9-y P3C1V y , where x ranges from 18 to 25 and y ranges from 0 to 1.

7. The iron-based amorphous / nanocrystalline alloy according to claim 6, characterized in that: In terms of atomic ratio, the chemical expression of the iron-based amorphous / nanocrystalline alloy is any one of the following: <h2 style=";text-align:left;direction:ltr">Fe<h2 style=";text-align:left;direction:ltr"> 69 <h2 style=";text-align:left;direction:ltr"> Co<h2 style=";text-align:left;direction:ltr"> 16 <h2 style=";text-align:left;direction:ltr"> Ni1Si3B<h2 style=";text-align:left;direction:ltr"> 11 <h2 style=";text-align:left;direction:ltr"> , (Faith 80 Co 20 ) 85.5 Nose 1.5 B 8.5 P3C1Mo 0.5 , (Fe 75 What 25 ) 85.5 There is no 1.5 B 8.5 P3C1V 0.5 , (Want 75 Co 25 ) 85.5 In 1.5 B9P3C1, (Want 80 Co 20 ) 85.5 In 1.5 B9P3C1, (Want 82 Co 18 ) 85.5 In 1.5 B9P3C1.

8. The iron-based amorphous / nanocrystalline alloy according to claim 6 or 7, characterized in that: The coercivity of the iron-based amorphous / nanocrystalline alloy is less than 6 A / m, and further does not exceed 5.6 A / m; The saturation magnetic induction of the iron-based amorphous / nanocrystalline alloy is not less than 1.85T.

9. The method for preparing an iron-based amorphous / nanocrystalline alloy according to any one of claims 6 to 8, characterized in that: include: preparing an amorphous precursor according to the chemical expression; The amorphous precursor is annealed in an external magnetic field at 340-420° C., preferably 360-380° C.

10. The preparation method according to claim 9, characterized in that: The amorphous precursor is a strip material, which is prepared by a melt rapid quenching method; The direction of the external magnetic field is consistent with the length direction of the amorphous precursor; The magnitude of the external magnetic field is 100 to 10000 A / m; The annealing treatment time is 10 to 20 minutes.

Images