METHOD FOR TESTING ONE OR MORE MICROBATTERIES AND SYSTEM FOR PERFORMING THIS TEST METHOD

Description

[0001] The field of the invention is that of the electrical testing of electronic or technological devices produced using microelectronic technology: integrated circuits, sensors, elementary devices (resistance, diode, capacitance, etc.).

[0002] More specifically, it is an electrical testing method dedicated to a certain category of micro-energy storage components, commonly called microbatteries.

[0003] A microbattery is made by successive deposition on a substrate of (i) first current collector, (ii) first electrode, (iii) electrolyte, (iv) second electrode, and (v) second current collector. Encapsulation by deposition of additional layers, or by cap transfer is necessary to protect the device from chemical reactivity with oxygen and water vapor.

[0004] Generally speaking, microbatteries can be classified into three categories based on the type of negative electrode (or anode), often deposited above the electrolyte: anode made of Li +< ion storage material: these are often transition metal oxides (Vox, TiOx, NiOx, etc.) or crystallogenic materials (Si, Ge, C or mixtures); metallic lithium anode: a layer of lithium is deposited in metallic form above the collector; anode made of a metal inert to lithium: this configuration is often called "Li-free" with reference to the article J. Electrochem. Soc.-2000-Neudecker-517-23 . In this case a metallic lithium anode is formed during the first charge of the battery by electrodeposition of Li+ ions (from the cathode) between the electrolyte and the second current collector.

[0005] The latter configuration is the most advantageous in the case of integrating microbatteries on a silicon substrate and using micro-fabrication techniques: it offers the presence of lithium metal (the most efficient anode in terms of energy density, potential and cyclability) without the disadvantages associated with its presence during the process (sensitivity to air, incompatibility with etching and photolithography solutions).

[0006] Generally, the most widely used and reliable electrical sorting and testing protocol consists of (i) testing the microbattery over several complete charge and discharge cycles (going from fully charged to fully discharged) and (ii) comparing the capacity values, voltage profiles and internal resistance to reference values.

[0007] This approach nevertheless has several limitations: in the context of microbatteries, and more particularly microbatteries "Li-free", Cycling generates structural variations at the level of the active stack, in this case the formation of metallic lithium making it very sensitive to air and complex to continue in a technological manufacturing process.

[0008] In the case of larger batteries, the application of this method remains very rare as it requires long testing periods, which represents a significant cost.

[0009] Sorting can be based on one or more microbattery parameters without having to cycle them. Among the most commonly used electrical parameters are: open circuit voltage or OCV (for " O pen C circuit Voltage"). This voltage corresponds to that measured at the microbattery level upon leaving the factory, and before any electrical stress. Generally, it gives an idea of ​​the state of charge of the microbattery. This is checked for example in the case of microbatteries with a metallic lithium anode or a storage anode. In the case of microbatteries "Li-free", It was experimentally observed that OCV exhibited significant variability within a plate and from plate to plate, and that no correlation was possible between OCV and state of charge as shown in the figure 1 which illustrates the variations in OCV for a microbattery-type component made on a batch of 7 wafers. The variability is significant both within a wafer and from wafer to wafer. The variability is linked to the fact that the lithium anode is not yet formed at this stage, and the OCV measurement provides information on the state of the electrolyte / electrode interfaces only, without any link to the state of charge or future electrical behavior. Consequently, this parameter is not relevant for establishing a sorting of all microbattery configurations. internal resistance: this corresponds to all the resistive contributions taking place within the structure of the microbattery; it is the sum of the resistance of the electrolyte, the charge transfers at the electrodes and the interfaces. This parameter is not relevant for sorting in the case of microbatteries "Li-free" for the same reason as mentioned above, the metallic lithium anode not being formed.

[0010] The challenge of sorting in this specific case is to be able to have electrical parameters that can be used to classify microbatteries, including those in "Li-free" without having to carry out cycling, at the risk of only being able to sort after the end of the manufacturing process. Indeed, sorting on the basis of cycling requires that the microbattery be encapsulated, either at the plate level, or after cutting and placing in the case, in order to avoid structural changes and sensitivity to air linked to the presence of metallic lithium. Intermediate sorting to reduce manufacturing costs and identify failure modes is not possible, and this is what we are trying to achieve in this specific case.

[0011] US9209496 proposes a sorting method consisting of measuring the open circuit voltage or OCV (for Open Circuit Voltage) of a battery, discharging it for a certain period and comparing the end of discharge potential with that of the OCV. It is clear that in the case of microbatteries "Li-free", This sorting method is not suitable since (i) the OCV does not provide any information regarding the condition of the battery, and (ii) the discharge of the batteries "Li-free" does not allow their potential to be modified in a stable manner since the microbattery is completely discharged at the time of the test.

[0012] Document US6526361 proposes a sorting method consisting of applying several successive pulses (either an application of current pulses or voltage pulses) with a relaxation period between each pulse to batteries, measuring the associated voltage response and sorting based on this response. This approach is not applicable to microbatteries of the type "Li-free" since it requires a stable initial voltage state.

[0013] The paper “A comparative study of sorting methods for Lithium-ion batteries” by Xiaoyu Li et al, XP032671721 and the paper “Fundamentals, impedance, and performance of solid-state Ll-metal microbatteries” by Collins John et al, XP012247065 disclose methods for sorting Lithium-ion batteries.

[0014] In this context and to overcome the aforementioned problems, the present invention relates to a test method particularly well suited to microbatteries of the type "Li-Free", for which the methods of the known art are not compatible. This method allows non-degrading and rapid tests to be carried out, in addition this method can be applied to one or a plurality of devices tested in parallel.

[0015] More specifically, the subject of the present invention is a method for testing at least one micro-energy storage device comprising a metallic lithium anode formed by electro-deposition of Li +< ions on a metal inert with respect to lithium ions, an electrolyte and a cathode, said method comprising a sequence of test steps during the manufacture of said anode, said sequence comprising: a measurement step Ph-a of the initial voltage OCV of said open-circuit device making it possible to obtain at an instantt0 , the value of the initial voltage of said micro-energy storage device; if said initial voltage is non-zero, a first charging step Ph-b including: o the application of a current Ib for a while tb allowing a first part of said device to be charged during a minimum charging time to form a first thickness of lithium layer as an anode; o the measurement of the voltage Vb at the terminals of said device; the determination of the internal resistance Rb being relative to the considered state of charge of said device; if the voltage Vb is located in an interval Vb min -Vb max defined according to the architecture of the component and if the resistance Rb is located in an interval Rb min - Rb max defined according to the component architecture, a stabilization step Ph-c comprising: o the continuation of the application of at least one current lcfor a while tc allowing a second part of said device to be charged during a second charging time tc to form a second thickness of lithium layer as an anode; o the measurement of the voltage Vc at the terminals of said device; if the voltage Vc is located in an interval Vc min -Vc max defined according to the component architecture, a retention step Ph-d including: o the application of a zero current for a time td ; or the measurement of voltage V aux terminals of the device over time; o the determination of the parameter Vd with Vd = ΔV / Δt ; o yes Vd is located in an interval Vd min -Vd max defined according to the architecture of the component, said device is tested valid.

[0016] According to variants of the invention: the first charge step involves a series of voltage measurements Vbi for periods of time Δtbi spaced out in time Tbi allowing the curve to be determined Vbi as a function of time and allowing the voltage value to be determined Vb which can be defined as the maximum value of the tensions Vbi or the average of the tensions Vbi ; resistance Rb can be determined after each time tbi by the report Vbi / lc of the curve Vbi depending on the time t.

[0017] According to variants of the invention, the durations Δtbi are of the order of 10 seconds, the times Tbi being of the order of 0.1 seconds.

[0018] According to variants of the invention, the first thickness is of the order of a few nanometers and preferably 1 to 2 nanometers.

[0019] According to variants of the invention: The stabilization step involves a series of voltage measurements Vci for periods of time Δtci spaced out in time Tci to determine the voltage value Vc defined as the average of the tensions Vci.

[0020] According to variants of the invention, the durations Δtci are of the order of 50 seconds, the times Tci being of the order of 1 second.

[0021] According to variants of the invention, the second thickness is less than 10 nanometers and preferably equal to 5 nanometers.

[0022] According to variants of the invention: The retention step involves a series of voltage measurements Vdi for periods of time Δtdi spaced out in time Tdi allowing the curve to be determined Vdi as a function of time; the parameter Vd is determined by the slope of the curve Vdi as a function of time t.

[0023] According to variants of the invention, the durations Δtdi are of the order of 10 seconds, the times Tdi being of the order of 1 second.

[0024] The invention also relates to a system for testing at least one energy micro-storage device, implementing the testing method of the invention, comprising: a current generator connected to said device; a voltmeter connected to said device; a controller connected to said current generator and to said voltmeter.

[0025] According to variants of the invention, the current generator is programmable and the voltmeter is programmable.

[0026] According to variations of the invention, the controller comprises a stopwatch, a memory and a processor.

[0027] The invention also relates to a set of a plurality of energy micro-storage device testing systems, implementing the method according to the invention, each system comprising: a programmable current generator connected to said device; a programmable voltmeter connected to said device, each current generator and each voltmeter being connected to a controller.

[0028] The invention also relates to a set of a plurality of energy micro-storage device testing systems, implementing the testing method according to the invention, each system comprising: a programmable current generator connected to said device; a programmable voltmeter connected to said device, each current generator and each voltmeter being connected to a single central controller.

[0029] According to variants of the invention, the test system is such that each controller or the central controller comprises a timer, a memory and a processor.

[0030] The test system of the invention may comprise a central controller comprising a memory, each controller managing the test phases for each test system, recording the voltage measurement results in its memory, the central controller retrieving said measurement results to record them in its memory.

[0031] The invention will be better understood and other advantages will appear on reading the description which follows, given without limitation and thanks to the appended figures among which:

[0032] [ Fig 1 ] illustrates the variations of the OCV for a microbattery type component produced on a batch of 7 wafers, and more precisely a Henry diagram, in which the ordinate is the Normit function of x, described in detail in the document accessible via the link: http: / / www.modulad.fr / numero-35 / Excelense-grenier-35 / DiagrammedeHenry.pdf;

[0033] [ Fig 2 ] illustrates the evolution of the voltage at the terminals of a microbattery device as a function of time during all the steps of the test method according to the invention;

[0034] [ Fig 3 ] illustrates a flowchart comprising all of the successive test phases in the test method according to the invention;

[0035] [ Fig 4 ] illustrates an example of a system for testing a plurality of micro-energy storage devices according to the invention, each device being connected to a controller;

[0036] [ Fig 5 ] illustrates an example of a system for testing a plurality of micro-energy storage devices according to the invention, each device being connected to the same central controller.

[0037] In general, the test method proposed in the present invention aims to overcome the various aforementioned drawbacks by using a non-degrading and rapid test technique. Advantageously, this technique can be applied to one or a plurality of devices tested in parallel.

[0038] The test method of the invention uses the test sequence described below, linking the different steps after validation of a criterion in the previous step, thus making it possible to conclude as soon as possible on a faulty device.

[0039] The micro-energy storage device is hereinafter referred to as a microbattery.

[0040] It is described in the context of an example of a lithium-based microbattery comprising a metal anode which may be made of metal, for example titanium, and comprises a solid electrolyte which may be made of LiPON (Lithium Phosphorus Oxynitride) and a cathode which may be made of LiCoO 2 , for example.

[0041] This technique is advantageous because it applies four phases for each device tested which are triggered one after the other when a validation criterion is ensured. These four phases detailed below are called, initial voltage measurement phase OCV, charging phase, stabilization phase and retention phase. The figure 2 illustrates the sequence of these phases by showing the evolution of the voltage at the terminals of the device tested as a function of time in an example of a method described in more detail below.

[0042] This sequence of test phases is described in detail in the remainder of the description and is also illustrated by the flowchart shown in figure 3 .

[0043] The test system is carried out on a device or the plurality of test systems can be carried out in parallel on a set of devices. The means used to implement the test method of the invention can be, according to a first variant of the invention, the following means and this for the test of each device Dn , only one of these devices Dn being represented in figure 4 These means include a programmable current generator An , a programmable voltmeter Vn and a controller Cn . The device An has an anode connection Dn-Anode and a cathode connection Dn-cathode , the generator An : a positive connection An-H and a negative connection An-L, the voltmeter: a positive connection Vn-H and a negative connection Vn-L.

[0044] The test method is to perform the microbattery device test by connecting the connections An-H of the current generator, Vn-H of the voltmeter and Dn-Anode of the microbattery device together, the connections An-L of the current generator, Vn-L of the voltmeter and Dn-Cathode of the same microbattery device together. This basic system can be repeated a plurality of times to test a plurality of devices.

[0045] Each basic system can be controlled by a controller Cn comprising a processor Pn , a stopwatch Crn and a memory Mn , this controller is connected to each voltmeter Vn and each current generator An and becomes an integral part of this system as also shown by the figure 4 ; in this case all controllers Cn of the plurality of test systems are themselves connected to a central controller Cc to control all test systems. This configuration is advantageous in that the algorithm used offers greater flexibility and simpler implementation.

[0046] According to another variant of the invention, the plurality of systems can be controlled by a central controller Cc including a stopwatch Crc and a memory Mc connected to all voltmeters Vn and to all current generators An as shown in the figure 5 .

[0047] Thus the means of control can be ensured either by a central controller or by a set of individual controllers. In all cases, the controller used has in its memory Mn Or Mc , the characteristic times at which voltage measurements are made by voltmeters Vn and time measurements by chronometers Crn or the stopwatch Crc. The controller used records these time and voltage measurements in its memory Mn Or Mc.

[0048] In the case where there is only the central controller Cc, This manages the test sequences for the plurality of test systems and records the measurement results in its memory Mc ; in case there is a controller Cn associated with each test system, each controller Cn manages the test sequence for each test system, it records the measurement results in its memory Mn , then the central controller retrieves all the measurement results and saves them in its memory Mc.

[0049] The test method according to the invention begins with a first phase Ph-a including the measurement of the initial tension OCV of each device using the voltmeter Vn connected to each device, each of these measurements is recorded in the memory Mn of each controller Cn of each system, or in memory Mc of Central Control Cc. This first phase allows for example to obtain the parameter OCV (Open Circuit Voltage) of each device, from the measurement of this voltage at the instant t0. Only measuring a value equal to zero is useful in the case of "Li-free" microbatteries, because it signifies a failure in short-circuit mode and allows the tests for the affected components to be stopped. The Phase Ph-a resumed on the organization chart of the figure 3 illustrates this first phase which can be conclusive and stop the test on a faulty device.

[0050] Then, during a second phase called the charging phase Ph-b each current generator An connected to each device Dn applies a current Ib to each device from a time t0. This current allows a minimal part of the device to be charged, which has the effect of not damaging the device. More specifically, this operation consists of initiating the migration of Li+ ions, and forming a continuous nanometric thickness (for example 1 to 2 nm) of lithium in the anode, and consequently the transition from a "Li-free" configuration to metallic Li, without structural impact on the component, which ensures both the relevance of the sorting a posteriori as well as the possibility of doing this sorting at any time during the manufacturing process.

[0051] This second phase allows for example to obtain the parameter Vb by extracting the maximum voltage obtained from the measurements recorded during this phase. The components are considered compliant if the voltage Vb is located in an interval V bmin -V bmax defined according to the component architecture. The interval V bmin -V bmax is essentially defined by the redox couple involved, more specifically by the potential of the cathode (or positive electrode) with respect to a metallic lithium electrode. For example, in the case of a LiCoO2 cathode this interval will correspond to 3.85-3.89V, while in the case of a Li4Ti5O12 cathode, this interval will correspond to 1.4-1.6V. The phase Ph-b resumed on the organization chart of the figure 3 illustrates this second phase which can be conclusive and allow the test to be continued or stopped if the device does not meet the expected values ​​and is considered non-compliant.

[0052] Then during a third phase called the stabilization phase Ph-c, the charging phase process Ph-b continues with different predetermined times which can be for example times spaced 1 second apart for 50 seconds, each current generator An connected to each device Dn continues to apply a current lc (which can be equal to Ib ) to each device Dn. The purpose of this step is to adjust the thickness of the nanoscale Li layer to the needs of the electrical tests. Several thicknesses can be of interest. The maximum sufficient thickness can typically be less than 10 nm, and preferably less than 5 nm. It should be noted that the total thickness of metallic lithium that will be used in the case of a complete cycling corresponds to 5000 nm, the proportion used for the test is approximately 1 / 1000th.

[0053] This is a great advantage of this method, which allows conclusive tests to be carried out from the very beginning of the formation of the lithium metal electrode and does not require the completion of the manufacturing process in order to test the microbattery device.

[0054] This third phase allows, for example, to obtain the Vc parameter by extracting the average of the voltage values ​​obtained from the measurements recorded during this phase. The components are considered compliant if the Vc parameter is located in an interval V cmin -V cmax defined based on the component architecture. The determination of V cmin -V cmax is essentially linked to the theoretical variation of the potential of the microbattery after the injection of a quantity of charge corresponding to Ic*temps. This amount is very low (less than 0.1% of the total charge quantity that the microbattery can store) and in the case of a functional microbattery, the values V cmin -V cmax are substantially similar (+ / -10%) to the values V bmin -V bmax . In the case of a non-functional microbattery, the gap between the two potential ranges will be greater. The phase Ph-c resumed on the organization chart of the figure 3 illustrates this third phase which can be conclusive and allows the test to be continued or stopped if the device does not meet the expected values ​​and is considered non-compliant.

[0055] Other parameters can be considered during this phase: for example slope, or delta V.... The value of the slope (V cmin -V cmax ) / Ic allows to give an estimate of the internal resistance of the microbattery at this precise state of charge, in the same way the difference between V cmin And V cmax or the ratio V cmin / V cmax can provide information on the charging kinetics of the microbattery.

[0056] Then during a fourth phase called the retention phase Ph-d each current generator applies zero current, and the phase process Ph-b and the phase Phc continues with predetermined times spaced for example by 1 second for 10 seconds. This fourth phase allows for example to obtain the parameter Vd by extracting the slope (derivative) of the voltage versus time curve obtained from the measurements recorded during this phase. The components are considered compliant, if the parameter Vd is located in an interval V dmin -V dmax defined according to the component architecture. The phase Ph-d appearing on the organization chart of the figure 3 illustrates this fourth phase which allows us to conclude that the device tested is compliant. This phase allows us to evaluate the kinetics and the amplitude of relaxation of the microbattery after the previous charging phase. Ideally, the relaxation is limited to an equivalent drop in potential à lc*Rint (internal resistance), this value being very low given the low current used.

[0057] Ideally: Vb = Vc = Vd

[0058] The set of parameters previously obtained can also be processed according to statistical laws to characterize the plurality of elements.

[0059] The test method according to the invention is advantageous because it makes it possible to obtain on devices during manufacture (directly on wafer), a set of parameters characterizing a plurality of devices in one time, if we take the examples of predetermined values ​​given in the phases Ph-a has Ph-d, of 1 minute and 10 seconds, and this without being destructive for the devices, allowing the manufacturing stages to then be completed.

Claims

1. A method for testing at least one energy micro-storage device comprising a anode made of metallic lithium formed by electrodeposition of Li+ ions on a metal inert to lithium ions, an electrolyte and a cathode, characterized in that said method comprises a sequence of testing steps during the manufacture of said anode, said sequence containing: - a step of measuring (Ph-a) the initial voltage (OCV) of said device in open-circuit mode in order to obtain, at a time t0, the value of the initial voltage of said energy micro-storage device; - if said initial voltage is non-zero, a first charging step (Ph-b) comprising: o applying a current Ib for a time tb in order to charge a first portion of said device for a minimum charging time so as to form a first lithium layer thickness at the anode; o measuring the voltage Vb across the terminals of said device; determining the internal resistance Rb; - if the voltage Vb is located within an interval Vbmin-Vbmax defined on the basis of the architecture of the component and if the resistance Rb is located within an interval Rbmin-Rbmax defined on the basis of the architecture of the component, a stabilization step (Ph-c) comprising: o continuing to apply at least one current Ic for a time tc in order to charge a second portion of said device for a second charging time tc so as to form a second lithium layer thickness at the anode; o measuring the voltage Vc across the terminals of said device; - if the voltage Vc is located within an interval Vcmin-Vcmax defined on the basis of the architecture of the component, a retention step (Ph-d) comprising: o applying a zero current for a time td; o measuring the voltage V across the terminals of the device over time; o determining the parameter Vd, where Vd = ΔV / Δt; o if Vd is located within an interval Vdmin-Vdmax defined on the basis of the architecture of the component, said device is tested as being valid.

2. The testing method according to claim 1, wherein: - the first charging step contains a series of measurements of voltages Vbi during periods Δtbi spaced by times Tbi in order to determine the curve Vbi as a function of time and in order to determine the value of the voltage Vb which can be defined as the maximum value of the voltages Vbi or the average of the voltages Vbi; - the resistance Rb may be determined after each time tbi by the ratio Vbi / Ic of the curve Vbi as a function of time t.

3. The testing method according to claim 2, wherein the periods Δtbi are of the order of 10 seconds, the times Tbi being of the order of 0.1 second.

4. The testing method according to one of claims 1 or 2, wherein the first thickness is of the order of a few nanometres and preferably 1 to 2 nanometres.

5. The testing method according to one of claims 2 to 4, wherein: - the stabilization step contains a series of measurements of voltages Vci during periods Δtci spaced by times Tci in order to determine the value of the voltage Vc defined as the average of the voltages Vci.

6. The testing method according to claim 5, wherein the periods Δtci are of the order of 50 seconds, the times Tci being of the order of 1 second.

7. The testing method according to one of claims 1 to 6, wherein the second thickness is less than 10 nanometres and preferably equal to 5 nanometres.

8. The testing method according to one of claims 1 to 7, wherein: - the retention step contains a series of measurements of voltages Vdi during periods Δtdi spaced by times Tdi in order to determine the curve Vdi as a function of time; - the parameter Vd is determined by the gradient of the curve Vdi as a function of time t.

9. The testing method according to claim 8, wherein the periods Δtdi are of the order of 10 seconds, the times Tdi being of the order of 1 second.

10. A system for testing at least one energy micro-storage device, implementing a method for testing at least one energy micro-storage device comprising an anode made of metallic lithium formed by electrodeposition of Li+ ions on a metal inert to lithium ions, an electrolyte and a cathode, according to one of claims 1 to 9, said testing method comprising a sequence of testing steps during the manufacture of said anode, said sequence containing: - a step of measuring (Ph-a) the initial voltage (OCV) of said device in open-circuit mode in order to obtain, at a time t0, the value of the initial voltage of said energy micro-storage device; - if said initial voltage is non-zero, a first charging step (Ph-b) comprising: o applying a current Ib for a time tb in order to charge a first portion of said device for a minimum charging time so as to form a first lithium layer thickness at the anode; o measuring the voltage Vb across the terminals of said device; determining the internal resistance Rb; - if the voltage Vb is located within an interval Vbmin-Vbmax defined on the basis of the architecture of the component and if the resistance Rb is located within an interval Rbmin-Rbmax defined on the basis of the architecture of the component, a stabilization step (Ph-c) comprising: o continuing to apply at least one current Ic for a time tc in order to charge a second portion of said device for a second charging time tc so as to form a second lithium layer thickness at the anode; o measuring the voltage Vc across the terminals of said device; - if the voltage Vc is located within an interval Vcmin-Vcmax defined on the basis of the architecture of the component, a retention step (Ph-d) comprising: o applying a zero current for a time td; o measuring the voltage V across the terminals of the device over time; o determining the parameter Vd, where Vd = ΔV / Δt; o if Vd is located within an interval Vdmin-Vdmax defined on the basis of the architecture of the component, said device is tested as being valid; said testing system comprising: - a current generator configured to be connected to said device; - a voltmeter configured to be connected to said device; - a controller connected to said current generator and said voltmeter.

11. The testing system according to claim 10, wherein the current generator is programmable and the voltmeter is programmable.

12. The testing system according to one of claims 10 or 11, wherein the controller comprises a timer, a memory and a processor.

13. An assembly of a plurality of systems for testing an energy micro-storage device (Dn) according according to one of claims 10 to 12, implementing the method according to one of claims 1 to 9, each system comprising: - a current generator (An), possibly programmable, and configured to be connected to said device (Dn); - a voltmeter (Vn), possibly programmable, and configured to be connected to said device (Dn); - each current generator (An) and each voltmeter (Vn) being connected to a controller (Cn).

14. The testing system according to one of claims 10 to 12, wherein each controller or a central controller comprises a timer, a memory and a processor.

15. The testing system according to claim 12 and claim 14, comprising a central controller (Cc) having a memory (Mc), each controller (Cn) managing the test phases for each testing system, recording the voltage measurement results in its memory (Mn), the central controller recovering said measurement results so as to record them in its memory (Mc).