JP7531079B2 - Battery charge/discharge control method - Google Patents

Description

The present invention relates to a method for measuring the concentration of reactive ions in an electrode active material of a storage battery, and also to a method and device for controlling charging and discharging using the measuring method.

The voltage changes as shown in Figure 2 until the battery returns to an equilibrium state after charging and discharging is stopped. This period is called the recovery transient voltage (TRV). This recovery transient voltage is a transient phenomenon from a non-uniform state to a uniform state in the battery. Experiments in which electrode design conditions such as the thickness of the electrode and the porosity of the electrode active material are changed have shown that this phenomenon occurs because the difference in the concentration of reactive species ions between the electrode surface and the depth of the electrode is eliminated. A current interruption method has been attempted in which this phenomenon is approximated to an exponential function, and the kinetic coefficient is calculated from the coefficient to predict the dilution of the concentration of reactive species ions occurring in the depth of the electrode. However, in order to measure accurately, it is necessary to measure the transient phenomenon for several minutes to several hours after cutting off the current, which is inconvenient for obtaining data to be sent to the control system of the device used.

Electrochemical Impedance Method, 2nd Edition, by Masayuki Itagaki, Maruzen Publishing Co., Ltd., pp. 4, 5, 77

Electrochemical Measurement Manual, Basics, Edited by the Electrochemical Society, Published by Maruzen Co., Ltd., P. 96

Latest Lithium-Ion Secondary Batteries, co-authored, published by Information Technology Co., Ltd., p. 40

Information Technology Corporation Home Page Lecturer's Column Shigeru Sano

The AC impedance measurement method is widely used to measure the internal resistance and electrode reaction resistance of storage batteries. It is a simple measurement method that only requires electrically connecting the positive and negative terminals of a battery to a measuring device, and can be measured in seconds unless extremely low frequencies are measured. A typical result is a semicircle as shown in Figure 3. According to Non-Patent Document 1, the equivalent circuit can be drawn as shown in Figure 4. This semicircle is called a Cole-Cole plot (Nyquist plot), and devices that can automatically draw it are commercially available from Toyo Corporation and others. However, since AC does not reflect the change in the concentration of reactive species ions that occurs when the movement of reactive species ions accompanying charging and discharging lags behind the electrode reaction, it was not possible to detect the extreme capacity decrease during high-current discharge or the generation of dendrites due to overcharging during high-current charging, which are important for battery charging and discharging control.

As described in Non-Patent Document 2, AC impedance measurement is to be started after the battery is left for several minutes to several hours after charging and discharging is stopped and the equilibrium potential is stabilized. If the measurement is performed before the equilibrium potential is stabilized, the measurement results are unstable and poorly reproducible, making it difficult to analyze the results.

In the AC measurement of power wiring, if the Cole-Cole plot is not a semicircle but a semiellipse as shown in Figure 1, the transmission line theory, which takes into account the difference in wiring resistance of each household, is applied, and the equivalent circuit is as shown in Figure 5. Even when the AC impedance measurement of a storage battery results in a semiellipse, the transmission line theory is applied as shown in Figure 6, with each part of the electrode active material of the storage battery being likened to each household of the power wiring. Since each part of the electrode active material has a difference in distance from the counter electrode, there is a difference in the electronic resistance and ionic migration resistance, and an equivalent circuit similar to the transmission line theory should be established. As shown in the explanatory diagram in Figure 7, multiple semicircles are translated in the real axis direction, and they are measured simultaneously to form a semiellipse, but when the electronic resistance and ionic migration resistance difference generated in each part are compared with the amount of movement of the real axis, they are significantly different. Therefore, the transmission line theory cannot be applied to the electronic resistance and ionic migration resistance alone to interpret the semiellipse of a real battery.

The semi-ellipse in the battery measurement results is sometimes explained by the diversity of electrode pore shapes, but since the electrodes of a battery have a porous structure with random pores, and most AC impedance measurement results are semicircular, there is a logical inconsistency in attributing the semi-ellipse to the diversity of pore shapes. Therefore, the conventional AC impedance measurement method does not have sufficient theoretical basis to obtain data to send to a charge/discharge control system.

When a semi-ellipse like that shown in FIG. 1 is obtained, curve fitting is performed using an equivalent circuit including a constant phase element (CPE) to obtain ion migration resistance and electrode reaction resistance. This is useful for comparing battery characteristics, but it is thought to be merely a mathematical process and is not backed by any physical or chemical phenomena, so it is inappropriate to use this data as data to send to a charge/discharge control system.

When a lithium-ion secondary battery is discharged, the lithium ions in the negative electrode active material dissolve into the electrolyte, pass through the separator, and electrophores to the positive electrode, where they are stored at the lithium ion insertion site in the positive electrode active material by reducing the valence of the transition metal in the positive electrode active material. If the discharge current is greater than the amount of lithium ions supplied by electrophoresis, the lithium ions in the positive electrode pores become insufficient and the concentration becomes low. If the concentration of lithium ions, which are reactive species on the electrode reaction surface, decreases, the positive electrode potential decreases based on the Nernst equation, which has long been known as an experimental rule and is represented by Chemical Formula 1 and shown in Figure 8. If the concentration in the positive electrode pores decreases, the battery voltage drops rapidly, falling below the reference discharge voltage, and the battery-operated device suddenly stops working.

When a lithium ion secondary battery is charged, the lithium ions stored in the positive electrode active material move to the negative electrode by electrophoresis, and are inserted between the graphene layers when the negative electrode active material is graphite, or into the micropores when the negative electrode active material is a porous carbon such as hard carbon. In the case of a large current charge where electrophoresis cannot keep up, the lithium ion concentration in the negative electrode pores decreases. When the concentration at the electrode reaction surface decreases, the insertion reaction potential decreases according to the Nernst equation shown in Chemical Formula 1 and illustrated in FIG. 8. When the concentration becomes one percent of the initial concentration (generally 1 mol/l), the electrode reaction potential decreases by about 120 mV. Since the difference between the deposition reaction potential of lithium metal and the insertion reaction potential of lithium ions is about 80 mV, the deposition reaction is higher than the insertion reaction into the negative electrode carbon, and the deposition reaction proceeds easily, and as shown in the schematic diagram of FIG. 9, metal deposition called dendrites grows and eventually reaches the positive electrode, causing the battery to explode and catch fire.

Thus, the difference in the concentration of reactive species ions in the electrolyte between the electrode surface and the inner part of the electrode causes problems, so when designing a battery, the electrode thickness, porosity, pore size, etc. are changed to design the battery so that the reactive species ion concentration is constant throughout the electrode. Under these design conditions, the limit value of the charge and discharge current is also determined, so that the reactive species ion concentration is never uneven in the battery at the beginning of use. However, as shown in the schematic diagram in Figure 10, SEI (solid electrolyte interface) is generated in the electrode pores, and the scum produced by the peeling and dissolution of the SEI floats and accumulates in the pores, narrowing the actual pore size, preventing the movement of reactive species ions, and causing a difference in the reactive species ion concentration between the electrode surface and the inner part of the electrode, which should not occur in the design. As a result, a sudden decrease in capacity and a serious malfunction such as explosion and fire occur.

If it were possible to detect the difference in concentration of reactive ions between the surface and deep inside of an electrode from the outside, it would be possible to determine the optimum discharge current and optimum charge current in a charge/discharge control system, and prevent a rapid decrease in capacity or explosion and fire due to dendrite shorts. The present invention provides the means for this.

When AC impedance is measured after the equilibrium potential has stabilized sufficiently after charging and discharging, two semicircles are formed in the lithium-ion secondary battery. Experiments in which the size of the electrode opposite to the electrode being measured is changed have revealed that the semicircle on the high frequency side reflects the negative electrode, and the semicircle on the low frequency side reflects the positive electrode. In a normal design, the semicircle reflecting the negative electrode is smaller than the semicircle reflecting the positive electrode. When the two circles overlap, the semicircles may appear almost as one. Conversely, if the negative electrode is made extremely large, only one semicircle reflecting the positive electrode will be formed. Immediately after charging and discharging, the uneven state inside the battery becomes uniform, which is a transient phenomenon. If the measurement is performed during this period, the semicircle shown in Figure 1 will be obtained, rather than the semicircle shown in Figure 3. The semicircle will extend over a wide range different from the position where the semicircle reflecting the negative electrode and the semicircle reflecting the positive electrode overlap. The height of the semicircle in the imaginary axis direction depends on the size of the original semicircle, and it is possible to distinguish whether the semicircle extended in the real axis direction is a semicircle reflecting the negative electrode or a semicircle reflecting the positive electrode. Since the positive semicircle is large and has a large effect, the positive semicircle often stretches. If you measure it separately at balanced voltage, you can calculate the elongation of the actual axis from the difference between the two.

The value obtained by converting the extension in the real axis direction into a voltage is the change in equilibrium potential at each part of the electrode. When the change in equilibrium potential of the reaction at each part is added to the equivalent circuit shown in Figure 6, which applies the transmission line theory to a storage battery, the equivalent circuit becomes as shown in Figure 11. The correction of the equilibrium voltage change in Figure 11 is not a voltage but is described as a correction resistance value that is consistent on the equivalent circuit. The gist of the present invention is this equivalent circuit, and a method, system, and device for providing data obtained by numerical processing based on this equivalent circuit to a charge/discharge control system to control charge/discharge.

In a lithium ion secondary battery, among the elements described as changing depending on each part in the equivalent circuit of FIG. 11, the change in the charge transfer reaction resistance and the Warburg impedance is very small during the measurement time, and the ion transfer resistance in each pore and the electronic resistance in each pore are much smaller than the amount of change from a semicircle to a semiellipse, and the difference between each part can be ignored. Therefore, they can be included in the ion transfer resistance in the electrolyte and the electronic resistance of the active material, the current collector, the terminals, etc., and the equivalent circuit can be simplified and written as shown in FIG. 12. Compared with FIG. 4, the difference between the semicircle and the semiellipse is only the difference in the equilibrium potential of each part. The difference in equilibrium potential is calculated from the difference between the semicircle and the semiellipse obtained by a previous experiment or curve fitting, and the concentration is calculated by the Nernst equation shown in Chemical Formula 1 and illustrated in FIG. 8, so that the data on dilution in the inner part of the electrode can be obtained. By providing the data on dilution to a charge/discharge control device, the charge/discharge current can be controlled to prevent a sudden decrease in discharge capacity or the generation of dendrites due to dilution. In addition, since slag formation occurs equally at the positive and negative electrodes, if dilution progresses at one electrode, it can be considered that dilution is also progressing at the other electrode.

When the charge/discharge state of reactive ions in the electrode, especially between the active materials in the positive electrode, becomes uneven, a difference occurs in the equilibrium potential at each part, and the equilibrium potential must be corrected. Accuracy can be improved by measuring the charge state at each part separately and correcting it. Generally, this difference is smaller than the difference in the concentration of reactive ions in the electrolyte, so it can be ignored. In the case of iron olivine-based positive electrodes, where the positive electrode potential hardly changes during charge/discharge, this error becomes very small and can be measured with greater accuracy.

In the present invention, since a semicircle or semiellipse can be drawn sufficiently by only measuring the high frequency side of the AC impedance measurement, a measurement time of 1 second or less is sufficient. Since it is a short-time measurement, it is not limited to after the charge/discharge is cut off, and it may be superimposed on a direct current. A high-speed analysis technique has also been developed that can obtain results similar to AC measurements by mathematically processing the results obtained from instantaneous pulse current, such as by Fourier transform, and it is possible to perform measurements even during high-speed pulse charging/discharging. If the pulse waveform generated by the switching regulator of the charging/discharging device is used for mathematical processing, such as Fourier transform, the device should be able to be further simplified.

According to the present invention, even if dilution of the electrolyte occurs under circumstances different from the original design, such as when slag, which is a product of the peeling and dissolution of the SEI due to charge/discharge or deterioration over time, precipitates in the pores in the electrode active material, by providing this data to the charge/discharge control system, it is possible to control the charge/discharge current and prevent a sudden decrease in discharge capacity or the occurrence of dendrites.

As deterioration progresses in lithium ion secondary batteries, a phenomenon called sudden death occurs in which the capacity suddenly disappears, which is a very difficult problem to solve in practical use, but since most of these sudden deaths are caused by dilution deep inside the electrodes, the present invention can measure dilution, which can be detected in advance and serious practical problems can be prevented. Furthermore, even in the case of reusing lithium ion secondary batteries, it becomes possible to predict sudden death or explosion and fire after deterioration, and the expected life of used batteries can be presented.

Results of AC impedance measurements during recovery voltage transients. Battery discharge curve and recovery transient voltage results Battery results using AC measurement method Battery equivalent circuit Equivalent circuit based on transmission line theory Equivalent circuit applying transmission line theory to storage batteries A semiellipse made up of multiple overlapping semicircles Nernst Equation Diagram of dendrite formation during rapid charging Illustration of dendrite formation caused by floating sediment of slag Equivalent circuit for correcting the equilibrium potential for changes in ion concentration in each pore Simplified equivalent circuit diagram for equilibrium potential correction Figure 7. A semi-ellipse with actual measurements plotted on the real axis

Lead wires are spot welded to the positive and negative terminals of a commercially available 1000 mAh 3.7 V square lithium ion secondary battery for mobile phones, and connected to an AC impedance measuring device (Solartron 1280 manufactured by Toyo Corporation). When the open circuit voltage is measured before charging and discharging, the potential is set, and AC impedance measurement is performed, two semicircles are obtained, the small semicircle on the high frequency side is due to the negative electrode, and the large semicircle on the low frequency side is due to the positive electrode. In this connection state, a discharge current of 2 A is applied to both terminals for 10 minutes, and AC impedance measurement is performed at a frequency range of 1 MHz to 10 Hz with a current amplitude of 0.1 mA 1 second after the current application is completed. A semi-ellipse as shown in FIG. 13 is drawn at the same height as the semicircle of the negative electrode measured at the open circuit voltage. The semicircle can be estimated from the left quarter circle of the semi-ellipse in FIG. 13 , and can be read from the real axis (horizontal axis), and its resistance value is 400Ω. Similarly, the resistance value of the semi-ellipse can be read from the real axis (horizontal axis), which is four times 1600Ω, and the difference is 1200Ω. Because this is a numerical calculation on the real axis, Ohm's law can be applied, and multiplying it by the applied AC amplitude of 0.1mA gives 120mV. Substituting this into the Nernst equation shown in Figure 8, which is expressed as Chemical Formula 1, the initial input reactive species ion concentration was measured to be 1/100 of 1 mole/liter, or 0.01 mole/liter. The result was provided to the charge/discharge control system by a well-known method, and the charge/discharge system was able to appropriately control the charge current, thereby preventing the occurrence of malfunctions.

The present invention can detect a rapid decrease in battery capacity or the occurrence of dendrites, which is a serious defect, in advance, so if incorporated into a battery control system for an electric vehicle, the driving reliability of the electric vehicle will be improved, and it is highly likely to be used in the electric vehicle field. In addition, since the reliability of the battery life in the used lithium-ion secondary battery market will be improved, it is highly likely to be used in the used battery market.

The present invention is not limited to lithium ion secondary batteries, but can be applied to all batteries in which the concentration of reactive ions changes in the electrode pores, and is particularly useful for lead-acid batteries in which the sulfuric acid concentration is diluted in the pores during high-rate discharge.

Claims (2)

When the Cole-Cole plot of the AC impedance measurement results is an extended semi-ellipse, the difference between the diameter of the original semi-circle and the major axis of the semi-ellipse is converted into voltage, and this voltage difference is converted into the reactive ion concentration according to the Nernst equation. A charge/discharge control method for controlling charge/discharge based on reactive ion concentrations at each electrode portion measured by the concentration measurement method at each electrode portion according to claim 1.

Images