Method for estimating the self-discharge of a lithium battery

Abstract

A method for determining the self-discharge current of a lithium-ion battery provided with a positive electrode, a negative electrode, and an electrolyte arranged between the positive and negative electrodes includes charging the battery until a metal lithium layer is formed between the electrolyte and the negative electrode, measuring the open-circuit voltage of the battery at two moments, and determining the self-discharge current from the variation of the voltage measured between the two moments.

Claims

1. A method for determining a self-discharge current comprising the steps of: providing a lithium-ion battery provided with a positive electrode, a negative electrode comprising a lithium ion insertion material, and a solid type electrolyte arranged between the positive and negative electrodes, the lithium ion insertion material being in direct contact with the solid type electrolyte, charging the lithium-ion battery until a metal lithium layer is formed between the solid type electrolyte and the negative electrode, measuring the open-circuit voltage of the lithium-ion battery at two moments without connecting the battery between the two moments, and determining the self-discharge current from the variation of the measured voltage between the two moments.

2. The method according to claim 1, wherein the positive electrode comprises a quantity of lithium greater than the quantity of lithium that the negative electrode is capable of inserting.

3. The method according to claim 1, wherein the charge parameters are selected to cause a flow of lithium ions greater than the maximum flow of lithium ions capable of diffusing within the negative electrode.

4. The method according to claim 1, wherein the negative electrode comprises a material having a lithium chemical diffusion coefficient smaller than 10.sup.9 cm.sup.2/s.

5. The method according to claim 1, wherein the negative electrode comprises a lithium ion insertion material selected from the group consisting of Si, Al, Ge, SnO, LiNiO.sub.2, and indium lead oxide.

6. The method according to claim 1, comprising a full first charging step of the lithium-ion battery configured to saturate the lithium ion insertion material followed by an additional charging step configured to form the metal lithium layer.

7. The method according to claim 1, wherein voltage measurements between two moments are made while the potential of the negative electrode is set to 0.

8. The method according to claim 1, wherein voltage measurements between two moments are achieved when relaxation of Li+ concentration in the positive electrode is reached.

9. A method for determining the self-discharge current of a lithium-ion battery provided with a positive electrode, a negative electrode comprising a lithium ion insertion material, and a solid type electrolyte arranged between the positive and negative electrodes, comprising the steps of: providing the lithium-ion battery provided with a positive electrode, a negative electrode comprising a lithium ion insertion material, and a solid type electrolyte arranged between the positive and negative electrodes, the lithium ion insertion material being in direct contact with the solid type electrolyte, charging the lithium-ion battery so that a significant quantity of Li+ ions is transferred to the negative electrode during a short time period, until a metal lithium layer is formed between the solid type electrolyte and the negative electrode, measuring the open-circuit voltage of the lithium-ion battery at two moments without connecting the battery between the two moments, and determining the self-discharge current from the variation of the measured voltage between the two moments.

10. The method according to claim 9, wherein the positive electrode comprises a quantity of lithium greater than the quantity of lithium that the negative electrode is capable of inserting.

11. The method according to claim 9, wherein the charge parameters are selected to cause a flow of lithium ions greater than the maximum flow of lithium ions capable of diffusing within the negative electrode.

12. The method according to claim 9, wherein the negative electrode comprises a material having a lithium chemical diffusion coefficient smaller than 10.sup.9 cm.sup.2/s.

13. The method according to claim 9, wherein the negative electrode comprises a lithium ion insertion material selected from the group consisting of Si, Al, Ge, SnO, LiNiO.sub.2 and indium lead oxide.

14. The method according to claim 9, comprising a full first charging step of the lithium-ion battery configured to saturate the lithium ion insertion material followed by an additional charging step configured to form the metal lithium layer.

15. The method according to claim 9, wherein voltage measurements between two moments are made while the potential of the negative electrode is set to 0.

16. The method according to claim 9, wherein voltage measurements between two moments are achieved when relaxation of Li+ concentration in the positive electrode is reached.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Other advantages and features will become more clearly apparent from the following description of particular embodiments of the invention given for non-restrictive example purposes only and represented in the appended drawings, in which:

(2) FIG. 1, previously described, represents a conventional lithium battery architecture;

(3) FIG. 2, previously described, schematically represents the operation in charge mode of a Li.sub.xTiOS/LiPON/Si battery and the lithium ion concentration gradient X.sub.Li in the electrodes, before and after relaxation of the battery;

(4) FIG. 3, previously described, represents the variation of open-circuit voltage V.sub.CO of the Li.sub.xTiOS/LiPON/Si battery along time after a nominal charge at 2.6 V;

(5) FIG. 4 represents the charge of a lithium-ion battery until a metal electrode forms, enabling to rapidly determine the battery leakage current;

(6) FIG. 5 represents potential V.sub.+ of the positive electrode and potential V.sub. of the negative electrode according to the battery state of charge SOC, in a test charge and a nominal operation cycle; and

(7) FIG. 6 represents the variation of open-circuit voltage V.sub.CO of the battery of FIG. 4 along time.

DESCRIPTION OF A PREFERRED EMBODIMENT OF THE INVENTION

(8) It is here provided to shorten the self-discharge estimation time by decreasing the time required for the battery relaxation. To achieve this, the battery is charged until a metal lithium layer forms on the negative electrode. This metal layer, at the negative electrode/electrolyte interface, enables to set the potential of the negative electrode. Thus, the relaxation time only depends on the lithium diffusion in the positive electrode. The diffusion being faster in the positive electrode than in the negative electrode, the relaxation is more rapidly reached. The self-discharge current or leakage current of the battery is then determined by measuring the open-circuit voltage.

(9) A lithium-ion battery comprising a positive electrode 2, a negative electrode 6, and an electrolyte 4 arranged between electrodes 2 and 6 is first provided, as illustrated in FIG. 1. Electrodes 2 and 6 comprise a lithium ion insertion material, for example, TiOS for positive electrode 2 and Si for negative electrode 6. Further, at least one of these two materials is lithiated, that is, contains lithium. Lithium is for example incorporated into the positive electrode (Li.sub.xTiOS) during the manufacturing of the battery. The electrolyte preferably is of solid type, for example, LiPON.

(10) In FIG. 4, the lithium-ion battery is charged until a metal lithium layer 8 forms between electrolyte 4 and negative electrode 6.

(11) In a preferred embodiment, metal layer 8 is formed by a full charge of the battery, the battery being imbalanced in terms of lithium ion insertion capacity. This means that positive electrode 2 initially contains more Li.sup.+ ions than negative electrode 6 can store.

(12) During the charge, the Li.sup.+ ions originating from electrode 2 insert into electrode 6 until saturation thereof. Then, the Li.sup.+ ions of electrode 2, in excess with respect to the capacity of electrode 6, are electrodeposited on the surface of electrode 6, thus forming layer 8.

(13) FIG. 5 represents the variation of potential V.sub.+ of the Li.sub.xTiOS electrode and of potential V.sub. of the Si electrode according to the battery state of charge SOC, in a test charge and a nominal operation cycle.

(14) Points A and A show the initial state of the electrodes, for example, just after manufacturing of the battery. The battery state of charge SOC is zero, it is thus discharged. The TiOS electrode contains a quantity of lithium corresponding to a potential V.sub.+ of 1.7 V with respect to the Li.sup.+/Li reference potential. Conversely, the Si electrode comprises no lithium, its potential is maximum (V.sub.=1 V with respect to the Li.sup.+/Li reference potential).

(15) During the charge, potential V.sub.+ increases given that the Li.sub.xTiOS electrode releases Li.sup.+ ions. Conversely, potential V.sub. decreases because the Si electrode charges with Li.sup.+ ions.

(16) At B and B, the battery is fully charged (SOC=100%). Voltage V.sub.+ is 2.6 V (vs. Li.sup.+/Li) and potential V.sub. reaches 0 V (vs. Li.sup.+/Li), which means that the negative electrode is saturated with lithium. Portions AB and AB actually correspond to a nominal charge of the battery, that is, a charge in normal operation.

(17) The charge continues beyond points B and B, for example, at C and C as shown in dotted lines in FIG. 5. Thus, other Li.sup.+ ions are extracted from electrode 2. However, they cannot insert into electrode 6 since said electrode is saturated. They are thus electrodeposited at the silicon surface and form metal layer 8. Potential V.sub. is maintained at 0 V (vs. Li.sup.+/Li) while potential V.sub.+ keeps on increasing to reach 2.9 V (vs. Li.sup.+/Li). Layer 8 then constitutes a reference potential. In the present example, this potential is 0 V (vs. Li.sup.+/Li).

(18) The battery is then placed in open circuit and potential V.sub.CO at the terminals of the battery is measured. Potential V.sub.CO here corresponds to potential V.sub.+ of the positive electrode since potential V.sub. of the negative electrode is zero. When the slope of curve V.sub.CO (t) becomes constant, the positive electrode is considered to have reached its balance. The slope, corresponding to a voltage decrease per unit time, then accounts for the battery leakage current. Voltage V.sub.CO is measured at two moments, after which the self-discharge current is calculated from the voltage variation between these two moments.

(19) The self-discharge can also be expressed in terms of capacity, by using charts of voltage V.sub.CO versus the state of charge, SOC.

(20) After having determined the self-discharge, the battery can be normally used, according to nominal charge and discharge cycles. An operating range defined by a low cut-off voltage DD (in discharge mode) and a high cut-off voltage BB (in charge mode) is generally set. The hysteresis observed on the curves of FIG. 5 (voltage jumps at A and B) is due to the internal resistances and to the overvoltages of the reactions.

(21) The high cut-off voltage preferably corresponds to the saturation threshold of the negative electrode (or insertion limit of lithium in the negative electrode), at points B and B in FIG. 5. The high cut-off voltage for example is V.sub.BAT=V.sub.+(B)V.sub.(B)=2.6 V.

(22) The low cut-off voltage of the battery varies according to the electrode materials and to the device that it powers. Generally, the low cut-off voltage corresponds to a partial discharge of the battery. This means that part of the Li.sup.+ ions is immobilized in the silicon electrode. It is for example located at points D and D of FIG. 5 and is V.sub.BAT=V.sub.+(D)(D)=1.7 V. This corresponds to a percentage of immobilized Li.sup.+ ions of approximately 23%.

(23) FIG. 6 represents in solid line open-circuit voltage V.sub.CO of the battery of FIG. 4, according to the time elapsed after a charge at 2.9 V. The curve of FIG. 3 has been copied on the diagram in dotted lines, for comparative purposes.

(24) It can be observed that the voltage stabilization occurs much faster in the case of the battery provided with the metal layer. Indeed, the slope of the curve becomes constant after one hour only, while at least three hours are necessary in the case of a nominal charge at 2.6 V. The time for estimating the self-discharge is thus considerably decreased with respect to conventional techniques.

(25) The above-described method for determining the self-discharge current is preferably carried out in the context of quality controls performed immediately after the battery manufacturing. Indeed, the self-discharge current is an important datum to assess the reliability of a battery. The first battery charge is thus used to form the metal lithium layer to rapidly determine its leakage current, after which the battery is normally used.

(26) In an alternative embodiment, the charge parameters are selected so that a significant quantity of Li.sup.+ ions is transferred to electrode 6 during a short time period. The flow of Li.sup.+ ions from the positive electrode to the negative electrode is then greater than the maximum flow of Li.sup.+ ions capable of diffusing within the negative electrode. Since electrode 6 cannot insert such a quantity within such a short time, a metal lithium layer forms at its surface. Potential V.sub. reaches 0 V (vs. Li.sup.+/Li) without for the battery to be totally charged (SOC<100%). In this case, the battery may be balanced in terms of insertion capacity.

(27) In other words, the battery is charged at a sustained rate to saturate the negative electrode at its surface only, by a sufficient quantity to persist during the relaxation time of the positive electrode. The battery is preferably charged at a high constant voltage, for example, at 2.9 V for thirty minutes.

(28) The method for determining the self-discharge current is particularly advantageous for batteries having a negative electrode with a chemical lithium diffusion coefficient smaller than 10.sup.9 cm.sup.2/s.

(29) Many variations and modifications of the method for determining the self-discharge current will occur to those skilled in the art. The method has been described in relation with a LiTiOS/LiPON/Si battery. However, other insertion materials may be used, especially LiCo.sub.2, LiMn.sub.2O.sub.4, LiNi.sub.0.5Mn.sub.1.5O.sub.4 for the positive electrode and Si, Al, Ge, SnO, LiNiO.sub.2, indium lead oxide for the negative electrode. The voltage values may further vary according to the nature of the electrode materials and to the envisaged application.