Finite element simulation method for optimizing electrode slurry mixing process

Abstract

The invention discloses a finite element simulation method for optimizing an electrode slurry mixing process in the technical field of material mixing process optimization, and the method comprises the following steps: firstly, carrying out flow characteristic and motion mode analysis on particles in a charging basket, and preliminarily predicting the distribution condition and the mixing trend of the particles in the charging basket; then analyzing three aspects of movement of a single particle, collision of binary particles and movement of a particle group by combining a fluid mechanics effect and a particle mixing mechanism, and evaluating the mixing efficiency of the powder in the charging basket; and finally, determining the most appropriate grid size of the geometric model in combination with grid division modes of different dimensions in an analogue simulation process. By constructing a multi-scale simulation method capable of accurately simulating particle dynamics, a concentration field and a fluid force field in the electrode slurry mixing process, an accurate electrode slurry fluid algorithm model and an accurate regulation and control strategy are established, and controllable electrode preparation under different application conditions is achieved.

Description

Technical Field

[0001] The invention belongs to the technical field of material mixing process optimization, and in particular relates to a finite element simulation method for optimizing an electrode slurry mixing process. Background Art

[0002] Controllable preparation of electrode slurry has always been a technical bottleneck that needs to be overcome in the field of new energy batteries. The difficulty lies mainly in the fact that the production of electrode slurry is a complex system engineering. The slurry performance is affected by many factors such as raw materials, design, manufacturing equipment and process, and environment. Any defect may lead to the collapse of product performance. Among them, the stirring process is the basis for the high-quality completion of subsequent battery coating, rolling and other processes. It has the most obvious impact on battery performance, and the development of a mature stirring process does consume a lot of manpower and material resources.

[0003] The composition of the electrode slurry includes nano-lithium cobalt oxide (LCO), activated carbon (AC), conductive agent (mainly a mixture of SP conductive agent and CNT conductive agent), polyvinylidene fluoride (PVDF) and N-methyl-2-pyrrolidone (NMP) solvent. Since the preparation of electrode slurry involves multiple interrelated variables, such as stirring speed, time, temperature, solid content, etc., the preparation process is complicated and affected by many factors, so the preparation parameters are often unstable. At present, the mainstream controllable electrode slurry preparation parameters are mostly obtained through experiments. This method is not only time-consuming and labor-intensive, but also difficult to fully quantify a large number of interrelated variables. Summary of the invention

[0004] The present invention constructs a multi-scale simulation method that can accurately simulate the particle dynamics, concentration field, and fluid force field in the electrode slurry mixing process, establishes an accurate electrode slurry fluid algorithm model and regulation strategy, and realizes controllable electrode preparation under different application conditions.

[0005] A finite element simulation method for a preferred electrode slurry mixing process in this scheme includes the following steps:

[0006] S1. Powder particle size and interaction analysis: Analyze the flow characteristics and movement patterns of particles in the barrel, and preliminarily predict the distribution and mixing trend of particles in the barrel;

[0007] S2. Fluid mechanics effect and particle mixing mechanism: Combining the fluid mechanics effect and the particle mixing mechanism, the movement of single particles, the collision of binary particles and the movement of particle groups are analyzed to evaluate the mixing efficiency of the powder in the barrel;

[0008] S3. Meshing and simulation calculation: During the simulation process, the most appropriate mesh size for the geometric model is determined by combining meshing methods of different dimensions.

[0009] With the rapid development of computer technology, computer simulation has facilitated the development of hybrid capacitor manufacturing processes. The use of computer simulation technology can effectively study the manufacturing process and greatly shorten the research and development cycle. The present invention constructs a multi-scale simulation method that can accurately simulate the particle dynamics, concentration field, and fluid force field in the electrode slurry mixing process, establishes an accurate electrode slurry fluid algorithm model and regulation strategy, and realizes controllable electrode preparation under different application conditions.

[0010] The present invention conducts research on lithium cobalt oxide slurry based on precise grid division technology, and achieves matching of the model with the test results through continuous optimization.

[0011] Furthermore, in the analysis of S1, a continuous mobile phase is used instead of a solid powder phase to calculate the mixing and stirring process, and the viscosity of the mixed phase is controlled by a custom equation. A lower viscosity is set when the NMP liquid and powder phases are not mixed. After the mixing ratio is reached, the fluid viscosity corresponds to the viscosity at the shear rate.

[0012] Furthermore, STAR-CCM+ is used in S1 to calculate the mixing effect, and the standard k-ε turbulence model is used to simulate the powder-liquid mixing process in the mixing tank.

[0013] Furthermore, the unsteady Navier-Stokes equations with turbulence terms are used to describe the fluid flow in the stirred tank.

[0014] Furthermore, the semi-empirical formula is used to solve the k equation and the ε equation, as well as the relationship between the two and the turbulent eddy viscosity coefficient μ t relationship standards.

[0015] Further, the k equation is:

[0016]

[0017] The ε equation is:

[0018]

[0019] In the formula: The empirical constants are G 1ε =1.44, C 2ε =1.92; the degree to which the ε equation is affected by buoyancy depends on G 3ε , G 3ε =0;σ k =1.0;σ ε =1.3; ρ is density; t is time; u is velocity; μ is turbulent viscosity; x is the x-direction of the coordinate axis; subscripts i and j are free coordinates; G b is the generation term of turbulent kinetic energy k caused by buoyancy; G k is the stress source term caused by velocity gradient; YM is the pulsation expansion term in compressible turbulence; S k , S ε Define source items for users and select different values ​​according to actual conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is the geometric model of the stirring area of ​​the mixer of the present invention.

[0021] Figure 2 This is the process flow of the dry mixing, kneading and stirring strategy in the present invention.

[0022] Figure 3 This is the process flow of the wet mixing and stirring strategy for making glue in the present invention.

[0023] Figure 4 This is the process flow of the dry mixing glue kneading and stirring strategy in the present invention.

[0024] Figure 5 The corresponding shear rate-viscosity curves of the three strategies for simulating the mixing and dispersion process.

[0025] Figure 6 For strategy A 100s -1 Corresponding comprehensive powder distribution uniformity related data.

[0026] Figure 7 (a) is strategy A 100s -1、 10-step particle tracking distribution; (b) strategy A 100s -1 , 500step particle tracer distribution.

[0027] Figure 8 100s for strategy B -1 Corresponding comprehensive powder distribution uniformity related data.

[0028] Fig. 9 (a) is strategy B 100s -1、 10-step particle tracking distribution; (b) strategy B 100s -1 , 500step particle tracer distribution.

[0029] Fig.10 For C strategy 100s -1 Corresponding comprehensive powder distribution uniformity related data.

[0030] Fig.11 (a) is strategy C 100s -1、 10-step particle tracking distribution; (b) C strategy 100s -1 , 500step particle tracer distribution.

[0031] Fig.12Backscattering spectra of the slurry prepared by three strategies A, B, and C. DETAILED DESCRIPTION

[0032] The following is further described in detail through specific implementation methods:

[0033] 1. A finite element simulation method for optimizing electrode slurry mixing process

[0034] The present invention uses a continuous mobile phase as a solid powder phase to calculate the mixing and stirring process, and simulates three processes of dry mixing kneading and stirring (A strategy), wet mixing and stirring (B strategy) and dry mixing and kneading and stirring (C strategy) to optimize a mixing and dispersion method that is more suitable for this type of powder, thereby providing theoretical guidance for the production process. Figure 1 It is the geometric model adopted for the slurry mixing and dispersion simulation of the present invention.

[0035] In the simulation, the continuous mobile phase is used to replace the high-order powder for simulation. After comparing the calculation results, it is found that the traditional Euler particle suspension model is only applicable to the calculation of rheological properties of mixed liquids, which is not applicable to this model. Therefore, the continuous mobile phase is used as the solid powder phase to calculate the mixing process. The viscosity of the mixed phase is controlled by a custom equation. When the NMP liquid and powder phases are not mixed, a lower viscosity is set. After the mixing ratio is reached, the fluid viscosity corresponds to the viscosity at the shear rate.

[0036] The present invention uses STAR-CCM+ to calculate the mixing effect and adopts the standard k-ε turbulence model to simulate the powder-liquid mixing process in the mixing tank. The unsteady Navier-Stokes equation with turbulence terms is used to describe the fluid flow in the mixing tank. The semi-empirical formula is used to solve the k equation and the ε equation, as well as the relationship standard between the two and the turbulent eddy viscosity coefficient.

[0037] The turbulent kinetic energy equation (k equation) is:

[0038]

[0039] The ε equation is:

[0040]

[0041] In the formula: The empirical constants are G 1ε =1.44, C 2ε =1.92; the degree to which the ε equation is affected by buoyancy depends on G 3ε , G 3ε =0;σ k =1.0;σ ε=1.3; ρ is density; t is time; u is velocity; μ is turbulent viscosity; x is the x direction of the coordinate axis; subscripts i and j are free coordinates; G b is the generation term of turbulent kinetic energy k caused by buoyancy; G k is the stress source term caused by velocity gradient; Y M is the pulsation expansion term in compressible turbulence; S k , S ε Define source items for users and select different values ​​according to actual conditions.

[0042] 2. Simulation process based on the above method

[0043] In the present invention, LCO, AC, SP, CNT and PVDF are mixed / dispersed in a mass ratio of 85:9:2:1:3 using strategies A, B and C respectively to prepare electrode slurry. The above materials and NMP solvent are mixed in a vacuum planetary mixer, and the effects of the three slurrying processes on the properties of the slurry suspension are compared.

[0044] Strategy A is to premix all the powders and gradually add NMP solvent to them, so that the process has high shear stress. The specific process is as follows: Figure 2 Strategy B is to gradually add all the powders to the PVDF glue to apply low shear stress. The specific process is as follows: Figure 3 Strategy C is to pre-mix all powders (excluding PVDF) and gradually add PVDF glue to them. The difference between it and A is that PVDF is added to the slurry in batches. The specific process is as follows Figure 4 shown.

[0045] In the simulation mixing and dispersion process of the present invention, the stirrer speed is set to 1800 rpm, and the three stirring paddles simultaneously perform planetary motion at 70 rpm. The shear rate is set to 100 s according to the three strategies A, B, and C. -1 The mixing effect within 500 steps is calculated, and the mixing effect of the three strategies is calculated. The mixing viscosity of the powder and the mixed liquid is taken as the shear rate of 100s -1 The corresponding viscosity and shear rate viscosity curve are as follows: Figure 5 The viscosity of the slurry at different shear rates is shown in Table 1.

[0046] Table 1 Viscosity of slurry at different shear rates

[0047] <![CDATA[Shearing rate s -1 > A strategy corresponds to viscosity Pa·s B strategy corresponds to viscosity Pa·s C strategy corresponds to viscosity Pa·s 100 6850.693 1084.668 1325.664

[0048] 3. Simulation Results

[0049] Finally, the simulation obtained the mixing and dispersion state of the slurry under three stirring strategies A, B and C. Figure 6 and Figure 7 This is the simulated mixing and dispersion effect of strategy A. It can be seen from the figure that the tracer particles corresponding to various types of powders in the upper area are mixed evenly according to strategy A. After stirring for 500 steps, the powders pre-mixed evenly according to strategy A are mixed relatively fully.

[0050] Figure 8 and Fig. 9 This is the simulation mixing and dispersion effect of strategy B. It can be seen from the figure that according to strategy B, PVDF powder is pre-mixed in NMP solution, and LCO AC SP CNTs powder is added layer by layer. After 500 steps, there are some unevenly mixed powders and NMP liquids near the corners.

[0051] Fig.10 and Fig.11 This is the simulation mixing and dispersion effect of strategy C. It can be seen from the figure that according to strategy C, PVDF powder is pre-mixed in NMP solution, and LCO AC SP CNTs powder is added after being mixed evenly. After 500 steps, there are some unevenly mixed powders and NMP liquids near the corners.

[0052] In summary, through simulation research, it was found that the slurry prepared by strategy A is more evenly dispersed, and it is a slurry mixing and dispersion process that is more suitable for this system.

[0053] IV. Verification of the conformity of the industrial production process of the present invention with the simulation results

[0054] In order to further verify the correctness of the simulation results, the present invention conducted industrial production verification on the simulation results, and the test process and results are as follows.

[0055] (1) Dispersion stability analysis

[0056] The calculation formula of the slurry stability kinetic curve is as follows:

[0057]

[0058] As can be seen from formula (1), the stability index is a pure mathematical difference calculation formula, which reflects the comprehensive properties of the concentration and particle size change range of the sample during the entire static time. The larger the stability kinetic index and the larger the change range, the more unstable the system. The stability index is the result calculated based on the test data of the multiple light scattering stability analyzer. The slurry stability index (TSI) prepared by the three processes is shown in Table 2 below.

[0059] Table 2 Slurry stability index (TSI)

[0060]

[0061] It can be seen from Table 2 that the stability kinetic index of the slurry prepared by strategy A measured at different positions is relatively the smallest and the most stable. Therefore, the slurry prepared by this process strategy shows the most stable effect.

[0062] (2) Dispersion uniformity analysis

[0063] The main factors affecting the dispersion uniformity of the sample include the compatibility between the sample components, the dispersion process, etc. The dispersion uniformity of the suspension has a great influence on the overall stability of the sample, and it is one of the important factors leading to the later stratification of the sample. According to the principle of multiple light scattering, as time changes, the light intensity will change due to the instability of the sample, indicating that the sample has a problem of particle agglomeration, that is, the particle size and / or concentration of the particles have changed. In addition, this study uses the dispersion uniformity index and agglomeration index to quantify the dispersion uniformity of the slurry. The smaller the dispersion uniformity index and agglomeration index, the better the dispersion uniformity of the sample. Fig.12 The backscattering spectra of the slurries prepared by the three strategies are given. The horizontal axis of the spectrum is the height of the sample pool, the left side is the bottom of the sample pool, and the right side is the top of the sample pool. The vertical axis is the backscattered light intensity of the sample. It can be seen from the figure that the scattering spectra of the three strategies have a downward peak at the top, and an upward peak at the bottom, indicating that the top and bottom of the slurry have concentration changes, and the relationship between the backscattered light intensity and the concentration is positively correlated, that is, the backscattered light intensity increases with the increase of concentration and decreases with the decrease of concentration. Therefore, it shows that the concentration of the sample top decreases and the concentration of the sample bottom increases. This is a typical particle sedimentation phenomenon, and the samples in the middle do not overlap, and the particle size changes.

[0064] In order to quantitatively compare the dispersion uniformity of the slurries prepared by the three strategies, the test results of the slurry dispersion uniformity index are given in Table 3. As can be seen from Table 3, the dispersion uniformity index and agglomeration index of strategy A are the smallest, showing relatively good dispersion uniformity, indicating that the slurry prepared by this process is more conducive to the preparation of high uniformity electrodes.

[0065] Table 3 Slurry dispersion uniformity index

[0066]

[0067] The present invention also studied the coating uniformity of the slurries prepared by the above three processes, mainly analyzing and testing the uniformity of the weight of the prepared pole pieces, and analyzing the standard deviation and coefficient of variation of the pole pieces. As shown in Table 4 below, the pole piece weight obtained by the A strategy preparation process showed a relatively small standard deviation and coefficient of variation. The smaller the standard deviation and coefficient of variation, the closer the data is to the average value, that is, the better the consistency. Therefore, the pole piece obtained by the A strategy preparation process showed good uniformity, which is consistent with the slurry analysis results.

[0068] Table 4 Statistical analysis of pole piece consistency

[0069]

Claims

1. A finite element simulation method for optimizing electrode slurry mixing process, characterized in that The following steps are involved: S1. Powder particle size and interaction analysis: Analyze the flow characteristics and movement patterns of particles in the barrel, and preliminarily predict the distribution and mixing trend of particles in the barrel; S2. Fluid mechanics effect and particle mixing mechanism: Combining the fluid mechanics effect and the particle mixing mechanism, the movement of single particles, the collision of binary particles and the movement of particle groups are analyzed to evaluate the mixing efficiency of the powder in the barrel; S3. Meshing and simulation calculation: During the simulation process, the most appropriate mesh size for the geometric model is determined by combining meshing methods of different dimensions.

2. The finite element simulation method for optimizing the electrode slurry mixing process according to claim 1, characterized in that: In the analysis of S1, a continuous mobile phase is used instead of a solid powder phase to calculate the mixing and stirring process, and the viscosity of the mixed phase is controlled by a custom equation. A lower viscosity is set when the NMP liquid and powder phases are not mixed. After the mixing ratio is reached, the fluid viscosity corresponds to the viscosity at the shear rate.

3. The finite element simulation method for optimizing the electrode slurry mixing process according to claim 2, characterized in that: In S1, STAR-CCM+ is used to calculate the mixing effect, and the standard k-ε turbulence model is used to simulate the powder-liquid mixing process in the mixing tank.

4. The finite element simulation method for optimizing the electrode slurry mixing process according to claim 3, characterized in that: The unsteady Navier-Stokes equations with turbulence terms are used to describe the fluid flow in the stirred tank.

5. The finite element simulation method for optimizing the electrode slurry mixing process according to claim 4, characterized in that: The semi-empirical formula is used to solve the k equation and the ε equation, as well as the relationship between the two and the turbulent eddy viscosity coefficient μ. t relationship standards.

6. The finite element simulation method for optimizing the electrode slurry mixing process according to claim 5, characterized in that: The k equation is: The ε equation is: In the formula: The empirical constants are G 1ε =1.44, C 2ε =1.92; the degree to which the ε equation is affected by buoyancy depends on G 3ε , G 3ε =0;σ k =1.0;σ ε =1.3; ρ is density; t is time; u is velocity; μ is turbulent viscosity; x is the x-direction of the coordinate axis; subscripts i and j are free coordinates; G b is the generation term of turbulent kinetic energy k caused by buoyancy; G k is the stress source term caused by velocity gradient; Y M is the pulsation expansion term in compressible turbulence; S k , S ε Define source items for users and select different values ​​according to actual conditions.

Images