Secret: Application of Electrochemical Simulation Technology in Lithium Battery Research

With the introduction of the ban on fuel trucks in various countries, the status of new energy vehicles has become more stable. As the core power source of electric vehicles , lithium-ion batteries are increasingly sought after by the market. In the production process, lithium-ion batteries involve the selection and matching of materials such as positive electrode, electrolyte, negative electrode and separator, and the selection of design parameters of the pole piece. The battery process involves chemical reaction, mass transfer, conduction, heat generation and other processes. It can be seen that the lithium ion battery is a very complicated system.

Exploring lithium-ion batteries by means of experiments is an effective means. Especially with the continuous improvement of characterization methods, we can get more and more information about the influence of design parameters and working conditions on battery performance. It is undeniable that in the development process of lithium-ion batteries, there are too many design parameters and heavy experimental tasks; the influence of each parameter on battery performance is not clear, the experimental design has certain blindness, and sometimes there will be time-consuming and laborious funds. But it is a thankless phenomenon. The opportunity to improve this situation is to apply battery simulation technology to the battery.

Lithium-ion battery simulation technology can use equivalent circuit model, semi-empirical model, electrochemical model and so on. Simulation techniques based on electrochemical models are a good solution to the problems mentioned above. As a supplement to the experiment, the electrochemical simulation can simulate various schemes before the experiment, and can also simulate the charging and discharging process of the battery under different working conditions, which helps the researchers to understand the internal process of the battery. At the same time, the experimental results can also point out the shortcomings of the simulation and promote the continuous development of the simulation model. It can be said that the simulation makes the experiment even more powerful, and the experiment makes the simulation icing on the cake.

Simply talk about the electrochemical model. The electrochemical model is mainly composed of three processes of mass transfer, conduction and electrochemical reaction. The governing equations are shown in the following table. From the complexity degree, the electrochemical model has a single particle model, a quasi-two-dimensional model, a two-dimensional model, and a three-dimensional model. Commonly used is the quasi-two-dimensional model, based on this model, can achieve a variety of purposes including battery design, charge and discharge performance, battery internal resistance (polarization) analysis. In order to reduce the amount of calculation when predicting battery life, a single particle model is often used.

1. Application of Simulation Technology in Battery Design

In the battery design process, in addition to the inherent properties of the positive and negative materials, electrolyte and diaphragm, there are many design parameters to consider, such as positive and negative particle size (r), pole piece thickness (L), pole piece porosity. (ε) and so on. Marc Doyle and others used simulation technology to simulate the magnification of Sony's LiCoO2/EC, PC, and LiPF6/graphite batteries, and the battery rate performance was very similar to the test results. The figure below compares the test results of the charge and discharge curves at different magnifications with the simulation results.

Venkat Srinivasan et al. used simulation technology to study the effect of particle size on the power density of LiFePO4 half-cells. It was found that the use of small-diameter cathode materials is beneficial to increase the power density of batteries, which provides a direction for the development of high-power batteries. The author also used the discharge platform of LiFePO4 to mark the ohmic overpotential, reaction overpotential and diffusion overpotential during the constant current discharge of the battery, and found the reason why the platform became a slope when the large rate discharge occurred, and provided the battery internal resistance. Ideas.

In the battery development process, the model can be used to analyze the relationship between each design parameter and battery performance, determine the main influencing factors, and then conduct experiments on this factor, which can greatly reduce the amount of experiment.

2. Simulation of side reactions and lithium deposition in batteries

When the side reaction cannot be ignored in the model, a Bulter-Volmer formula describing the side reaction needs to be added as shown below. Of course, if the side reaction causes other changes, such as an increase in the thickness of the surface layer of the particle, an increase in resistance, etc., additional consideration is required.

In the LiMn2O4 half-cell, the self-discharge caused by the co-embedding side reaction (irreversible) of the electrolyte solvent (PC) and lithium ions was studied. The low-speed CV curve was used as the model calibration standard, and the transfer coefficient of the side reaction was used as the variable parameter. For batteries with different active loadings, the resulting side reaction transfer coefficients are different. It is difficult to control and monitor the side reactions in the battery. It is sometimes an effective means to obtain the physical and chemical parameters related to the side reactions by using the model and parameter identification.

Lithium is one of the main culprits in battery safety and capacity reduction. Theoretically, when the lithium potential is lower than 0V, lithium is precipitated. In fact, since the reaction requires driving force, there is a certain overpotential, and the lithium potential of the negative electrode deviates from 0V. In the lithium deposition model, in addition to the need to add a formula for describing the lithium reaction BV, the effect of lithium deposition on the capacity and the effect of the deposited layer on the surface layer of the particle are also considered. The research on LiMn2O4/graphite full battery shows that N/P is an effective method to inhibit lithium deposition. The larger the particle size, the easier it is to precipitate lithium. The thicker the pole piece, the easier it is to decompose lithium. The lithium precipitation mainly occurs at the end of constant current charging. At the constant pressure stage, the phenomenon of lithium is rapidly weakened and disappeared. The figure below shows the effect of the thickness of the pole piece, the particle size, and the charge cut-off voltage on the amount of lithium deposited.

In addition, other side reactions, such as decomposition of the electrolyte, formation of an SEI film on the negative electrode, formation of irreversible products in the electrode, etc., can be explored using simulation techniques.

3. Battery internal resistance

The DCR and EIS, which are often used to describe the internal resistance of the battery, can be described by models for both internal resistances.

Small disturbance signals are required during the EIS test to ensure that the system maintains (quasi-) steady state and the input signal is linear with the output signal. Therefore, during the modeling process, the interior of the battery is considered to be in a steady state process and is linearly responsive. Based on these assumptions and the difference between the real part and the imaginary part of the impedance, the governing equations of the electrochemical model are corrected to obtain the EIS model. With the aid of EIS simulation, the effects of diffusion, electrochemistry and other processes on EIS can be studied. The effects of electrochemical activity and conductivity of electrode materials on EIS can also be studied. The two electrodes of the whole battery can also be investigated separately. It is a very convenient means.

The simulation of DCR, in simple terms, is to change the charge and discharge mode in the electrochemical model, and to change the constant current charge or discharge to pulse charge or discharge. Andreas Nyman gave a calculation of the polarization when analyzing the various polarizations of the battery in his article, and based on this, calculated LiNi0.8Co0.15Al0.05O2/LiPF6, EC:EMC 3:7/MAG-10 system The proportion of different polarizations. Very helpful for us to understand the internal polarization of the battery. The figure below shows the decomposition of the polarization in the paper.

Internal resistance is a very important performance indicator of the battery, which has an important impact on the fast charging, heat generation and aging of the battery. If the ohmic internal resistance, reaction internal resistance, and diffusion internal resistance of the positive and negative electrodes and the electrolyte in the battery can be clearly analyzed by the model, it is very advantageous for improving the battery performance.

4. Battery life prediction

There are many reasons for the capacity decay of lithium-ion batteries, such as collapse of material structure, consumption of lithium by side reactions, consumption of lithium by SEI, and increase in internal resistance and lithium deposition. For the convenience of calculation, only one or two reasons for attenuation are considered in the general model.

The single-particle model is a simplification of the alignment of the two-dimensional model: all the active particles in the pole piece are considered to be the same, that is, the internal lithium ion concentration distribution is the same as the external environment.

The lifetime decay is attributed to the fact that the electrolyte solvent is reduced to consume lithium and the anode film resistance is increased. Gang Ning et al. quantitatively studied the effect of the discharge depth (DOD) on the discharge cut-off voltage, the charge cut-off voltage versus Li loss and The effect of internal resistance and the effect of the number of cycles on capacity and internal resistance. The simulation results are in line with our basic understanding of the battery, which has the advantage of quantifying these effects.

Some researchers believe that lithium is present in most of the charge and discharge process, and the inflection point of the battery capacity decay rate (excessive from the linear attenuation zone to the nonlinear attenuation zone) is caused by lithium deposition. The basic idea is: in the first few cycles of the cycle, the formation of the SEI film causes a decrease in the local porosity of the negative electrode near the separator, which increases the potential gradient of the local electrolyte, which creates conditions for lithium deposition, and further increases the porosity by lithium deposition. , forming a positive feedback that eventually leads to an exponential decay of capacity. Based on this consideration, a life model of capacity decay caused by SEI growth and lithium deposition is established. When predicting the lifetime of NCM622/EC/EMC (3:7 by wt.)+2% wt% VC/graphite system, although there are some small errors in the prediction of charge and discharge curves during the cycle (see the left figure below), ; The accuracy of the capacity prediction during the cycle is high (as shown in the right figure below). The model results show that the potential of the negative electrode electrolyte gradually increases from the separator/negative electrode end to the negative electrode/copper foil end; the electrode potential distribution also conforms to this trend, and there is no lithium deposition in the fresh battery. When the cycle reaches 1000 laps, lithium has been formed. The lithium is first formed in the negative electrode region close to the separator, and the negative electrode potential is the lowest at the end of the constant current charging, and the lithium is most easily precipitated.

In addition, models can be used to estimate capacity loss. For example, it is assumed that the capacity loss mainly comes from the offset of the positive and negative SOC intervals during the charging and discharging process and the loss of active material, and the activity is identified by the life model and the measured discharge curve with the positive and negative SOC and the positive and negative active materials at the beginning of the discharge as variables. The amount of material loss and the positive and negative SOC at the start of discharge. The contribution of the positive and negative electrodes to the capacity decay can be quantitatively analyzed.

The life prediction of the battery is carried out by electrochemical model. Although the model is more complicated, the model is based on the actual process inside the battery, so the accuracy is high. Using the model to explore the main cause of capacity loss, it is quick and convenient to disassemble and test the battery after the cycle.

The above briefly introduces the main functions of the electrochemical model in lithium-ion battery simulation, but the electrochemical model can do much more than this. Others such as power, temperature rise, safety, etc. can be explored using models. Although letting us build our own electrochemical model will have many difficulties in understanding the internal processes of the battery, solving partial differential equations and nonlinear equations, and coupling the physics field, but now the commercial software Comsol can help us quickly establish electrochemical Models reduce the effort required for the modeling process.