Realtime sensing of batteries

Abstract

A thin-film battery sensor includes a thin-film substrate; a common electrode, a second electrode, and a third electrode on the thin-film substrate; a first resistor on the thin-film substrate and coupled to the common electrode and the second electrode; and a second resistor on the thin-film substrate and coupled to the common electrode and the third electrode. A third resistor can be included that is coupled to the common electrode at one end and a corresponding electrode at another end. Each resistor is configured to function as a heater and / or thermistor to deposit heat onto the battery and detect local temperature variations. A readout integrated circuit (ROIC) is coupled to the sensor to apply a modulated current source and readout temperature modulation amplitude for determining thermal conductivity and thermal diffusivity tensors, which are used for estimation of state of health, state of charge, and state of safety of the battery.

Description

REALTIME SENSING OF BATTERIESCROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of US Provisional Application No. 63 / 643,028, filed on May 6, 2024.BACKGROUND

[0002] The use of battery technology, including the use of lithium-based, sodium- ion-based, and solid-state-based batteries, in a variety of applications (e.g., vehicles, electrical energy' storage systems, large appliances, etc.) is on the rise. Currently, there is no real-time non-invasive technique to assess the state-of-health and state-of-safety of batteries and reliable methods to estimate the state-of-charge of batteries during operation are limited. Current estimation methods are prone to error accumulation over a large number of cycles due to dependence of measured parameters on instantaneous battery pack temperature. Accordingly, the increase in the use of battery technology has exacerbated the need for accurate determination of the state-of-charge, state-of-health, and state-of- safety of these batteries to effectively recycle, reuse, and / or repurpose the batteries.BRIEF SUMMARY

[0003] Systems and techniques for real-time determination of the state-of-charge, state-of-health, and state-of-safety of batteries are disclosed herein. Advantageously, using thin-film battery' sensors that can be integrated either on the exterior or interior of the battery cell, metrics including thermal conductivity along different directions, thermal diffusivity along different directions and temperature on individual battery cells in a battery can be determined instantaneously and over a period of time. These metrics can be compared to charge-discharge cycling datasets to determine the state-of-charge, state-of- health, and state-of-safety' of the battery. Furthermore, even if the individual battery cells in the battery are no longer fit for their current use, because the state-of-charge, state-of- health. and state-of-safety of the battery are able to be determined, those individual battery cells may be fit to be recycled, reused, and / or repurposed in a more effective manner than previous systems allowed.

[0004] A thin-film battery sensor includes a thin-film substrate, a common electrode, a first resistor on the thin-film substrate and coupled to the common electrode, a second electrode on the thin-film substrate and coupled to the first resistor, a secondresistor on the thin-film substrate and coupled to the common electrode, and a third electrode on the thin-film substrate and coupled to the second resistor. A third resistor can be included on the thin-film substrate and coupled to the common electrode at one end and a corresponding electrode at another end.

[0005] In some cases, the device further includes a readout integrated circuit (ROIC) that includes an oscillator, a current source, and a number of lock-in amplifiers corresponding to each resistor on the thin-film substrate. The ROIC is coupled to the common electrode and each of the other electrodes coupled to the resistors on the thin- film substrate to provide a readout of voltage from the thin-film batten sensor.

[0006] Each resistor is configured to function as a localized heater and a thermistor to detect local temperature variation. In some cases, the ROIC can include circuitry to further provide a readout of the temperature in each resistor (e.g., as temperature modulation amplitude). Temperature modulation amplitude can be obtained from measured voltage.

[0007] A method of determining a state of health, state of charge, and state of safety of each cell in a lithium-based battery can include receiving a readout of the current, the voltage, and the temperature of a thin-film battery sensor physically coupled to a cell of the lithium-based battery' over a period of time, comparing the readout to one or more charge-discharge cycling datasets, and determining a state of health, state of charge, and state of safety of each cell in a lithium-based battery based on the comparison between the readout for that cell across the period of time and the one or more charge-discharge cycling datasets

[0008] This Summary' is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figures 1A and IB illustrate a two-probe thin-film battery sensor.

[0010] Figure 2 illustrates a four-probe thin-film battery sensor.

[0011] Figure 3 illustrates an example of internal layers of a battery' cell.

[0012] Figure 4 illustrates a thin-film battery' sensor coupled to a readout integrated circuit.

[0013] Figure 5 illustrates an example circuit for a readout integrated circuit.

[0014] Figure 6 illustrates a plurality of thin-film battery sensors coupled to a readout integrated circuit.

[0015] Figure 7 illustrates an example implementation of a multiplexer of a readout integrated circuit.

[0016] Figure 8 illustrates a flow diagram for determining a state of health, state of charge, and state of safety of each cell in a battery.

[0017] Figure 9 illustrates an example system for determining a state of health, state of charge, and state of safety of each cell in a battery.

[0018] Figure 10 illustrates a thin-film battery' sensor on a lithium-based pouch battery cell.

[0019] Figures 11A and 11B illustrate the preliminary results during two standard consecutive charge-discharge cycles at a rate of 1C on a fresh cell.

[0020] Figures 12A and 12B illustrate the preliminary results of accelerated degradation cycling.

[0021] Figure 13 illustrates an example calibration curve containing the thermal conductivity anisotropy ratios of Kxx / Kzz and Kyy / Kzz as a function of state-of-health (SOH).DETAILED DESCRIPTION

[0022] Systems and techniques for real-time determination of the state-of-charge. state-of-health, and state-of-safety of batteries are disclosed herein. Advantageously, using thin-film battery' sensors, metrics including thermal conductivity' along different directions, thermal diffusivity along different directions and temperature on individual battery cells in a battery can be determined instantaneously and over a period of time. These metrics can be compared to charge-discharge cycling datasets to determine the state- of-charge, state-of-health, and state-of-safety of the battery. Furthermore, even if the individual battery cells in the battery are no longer fit for their current use, because the state-of-charge, state-of-health, and state-of-safety of the battery are able to be determined, those individual battery cells may be fit to be recycled, reused, and / or repurposed in a more effective manner than previous systems allowed.

[0023] Figures 1A and IB illustrate two-probe thin-film battery' sensors. In Figure 1A, a two-resistor design is shown and in Figure IB, a three-resistor design is shown.

[0024] Referring to Figure 1A. a battery sensor 100 includes a thin-film substrate 102, a common electrode 104, a first resistor 106 on the thin-film substrate 102 andcoupled to the common electrode 104, a second independent electrode 108 on the thin- film substrate 102 and coupled to the first resistor 106, a second resistor 110 on the thin- film substrate 102 and coupled to the common electrode 104, and a third independent electrode 112 on the thin-film substrate 102 and coupled to the second resistor 110. The common electrode 104 can be disposed around a region of the thin-film substrate 102. In some cases, current is sourced to the first resistor 106 to generate localized heating, and resistance measurements are done on both the first resistor 106 and the second resistor 110 to measure temperature variations at their respective locations in response to the deposited heat.

[0025] Referring to Figure IB, a battery sensor 150 includes a thin-film substrate 152, a common electrode 154, a first resistor 156 on the thin-film substrate 152 and coupled to the common electrode 154, a second independent electrode 158 on the thin- film substrate 152 and coupled to the first resistor 156, a second resistor 160 on the thin- film substrate 152 and coupled to the common electrode 154, a third independent electrode 162 on the thin-film substrate 152 and coupled to the second resistor 160, a third resistor 164 on the thin-film substrate 152 and coupled to the common electrode 154. and a fourth independent electrode 166 on the thin-film substrate 152 and coupled to the third resistor 164. The common electrode 154 can be disposed around a region of the thin-film substrate 152. In some cases, current is sourced to the first resistor 156 to generate localized heating, and resistance measurements are done on all three resistors (156. 160, and 164) to measure temperature variations at their respective locations in response to the deposited heat.

[0026] Referring to Figures 1A and IB, the two-probe thin-film battery' sensors 100, 150 are considered a “two-probe"’ battery' sensors because each resistor includes two electrodes (e.g., a common electrode 204 and a corresponding, independent electrode for each resistor).

[0027] Figure 2 illustrates a four-probe thin-film battery' sensor. Referring to Figure 2, a four-probe battery' sensor 200 includes a thin-film substrate 202, a common electrode 204, a first resistor 206 on the thin-film substrate 202 and coupled to the common electrode 204, a first independent electrode 208 on the thin-film substrate 202 and coupled to the first resistor 206, a second independent electrode 210 on the thin-film substrate 202 and coupled to the first resistor 206, and a third independent electrode 212 on the thin-film substrate 202 and coupled to the first resistor 206. The battery sensor 200 further includes a second resistor 214 on the thin-film substrate 202 and coupled to the common electrode 204, a fourth independent electrode 216 on the thin-film substrate 202 and coupled to thesecond resistor 214. a fifth independent electrode 218 on the thin-film substrate 202 and coupled to the second resistor 214. and a sixth independent electrode 220 on the thin-film substrate 202 and coupled to the second resistor 214. The battery sensor 200 further includes a third resistor 222 on the thin-film substrate 202 and coupled to the common electrode 204, a seventh independent electrode 224 on the thin-film substrate 202 and coupled to the third resistor 222, an eighth independent electrode 226 on the thin-film substrate 202 and coupled to the third resistor 222, and a ninth independent electrode 228 on the thin-film substrate 202 and coupled to the third resistor 222. The common electrode 204 can be disposed around a region of the thin-film substrate 202.

[0028] Here, the four-probe thin-film battery' sensor 200 is considered a “four- probe” battery’ sensor because each resistor includes four electrodes (e.g., a common electrode 204 and three corresponding independent electrodes). The additional two electrodes for each resistor enables parallel access to the two ends of the resistor. For example, the common electrode 204 and the third independent electrode 212 can be coupled to a same end of the first resistor 206. In addition, the first independent electrode 208 and the second independent electrode 210 can be coupled to the same end of the first resistor 206 opposite the end to which the common electrode 204 and the third independent electrode 212 are coupled.

[0029] For the battery sensors described herein, including those as shown in Figures 1A. IB, and 2, the thin-film substrate of the batter sensor can be a flexible polyimide substrate. The top surface of the sensor can also be coated with a thin film insulating layer for packaging purposes. In some cases, the components (e.g., resistors and electrodes) are deposited on a side of the of the flexible polyimide substrate and an opposite side of the flexible polyimide substrate is configured to attach to a cell of a battery. In some cases, the components of the battery sensor are made of chromium, titanium, platinum, gold, or vanadium oxide. In some cases, a total thickness of the battery sensor is about 50 pm and a length and w idth of the battery sensor can be millimeter to centimeter in scale. In some cases, the length and width can be any measurement within a range from 1 mm to 25 mm. In some cases, the described battery sensor can be microfabricated to below 1 mm in lateral dimensions.

[0030] The two-probe thin-film battery sensors (e.g., battery sensor 100, 150 of Figures 1A and IB) and the four-probe thin-film battery sensor (e.g., batten' sensor 200 of Figure 2) are each examples of battery sensors that provide an advantage over previous battery sensors. In any case as disclosed herein, a battery sensor can include two or moreresistors, with each resistor coupled to two or more electrodes (e.g., a common electrode and one or more independent electrodes). Indeed, the disclosed battery sensors can be used to measure the anisotropic thermal conductivity tensor (K) elements, and / or anisotropic thermal diffusivity (a) elements of the batteries as stimulated and detected by the resistors on the thin film. In some cases, the K and / or a tensor elements depend on the composition and microstructure of active materials and interfaces within the cell, which are affected by both SOC and SOH. The disclosed battery sensors can be used to source localized heat to the battery cell and measure the resulting temperature variations (e.g., including temperature modulation amplitude) at different locations to measure the K and / or a tensor elements.

[0031] For example, molecular dynamics simulations of lithium-based battery cells illustrate that the degree of lithiation of graphite in the negative electrode, which is a key indicator of remaining cell capacity7, alters the through-plane and in-plane thermal conductivities unevenly. In addition, key mechanisms of lithium-based batterydegradation, including lithium plating, dendrite formation, solid electrolyte interphase (SEI) growth, structural decomposition, and transition metal dissolution, are all expected to alter one or more elements of the K and / or a tensors. Therefore, a ratiometric measurement of K and / or a elements with an appropriate calibration can provide a method for determining SOH and SOC.

[0032] In some cases, each resistor functions as a heater (e.g., to provide heat to the corresponding battery cell on which the battery sensor is attached to) and / or a thermistor to detect local temperature, which can be monitored over time to detect local temperature variation on the corresponding battery- cell. The ability- of the sensors to detect the established temperature gradients in response to the deposited heat (e.g., while acting as a thermistor) can be referred to as a ? tensor sensing technique. The K tensor sensing technique relies on sourcing a small sinusoidally-modulated current at frequency f to resistor 1, which generates a heat flow (Joule heating) at frequency 2 / (and in some cases If can range from 0. 1 to 1 Hz) while simultaneously measuring temperature oscillation amplitude and phase at 2 / ' using all three resistors via lock-in amplifiers on a readout integrated circuit (e g., as described below). The three temperature oscillation amplitudes (“temperature modulation amplitude”) can determine the diagonal elements of K tensorKxx, Kyy. Kzz) in real time. In some cases, a sampling rate can be every one to ten seconds.

[0033] Figure 3 illustrates an example of internal layers of a battery cell. Referring to Figure 3, a battery cell 300 includes a plurality of layers 302. Each layer 302 includesa specific composition and structural properties. The inventors have discovered that through-plane thermal conductivity of lithium-based batteries is one order magnitude smaller than the in-plane thermal conductivity of the lithium-based batteries. Note that in Figure 3, in-plane thermal conductivity includes the X and Y direction, whereas through- plane thermal conductivity includes the Z direction. Advantageously, through the use of the battery sensors described above, the diagonal elements of A tensor (Xxv, Kyy. Kzz) can be determined in real time. These values change unevenly as a function of state-of-heath (SOH) and state-of-charge (SOC). This physical principle described with respect to lithium-based batteries is applicable to other batteries, such as sodium-ion-based batteries and solid-state-based batteries, as well. Accordingly, a ratiometric measurement of these diagonal elements can be used as an identifier for SOH and SOC in many types of batteries.

[0034] Figure 4 illustrates a thin-film battery sensor coupled to a readout integrated circuit. Referring to Figure 4, a battery sensor 400 (e.g., battery sensor 150 of Figure IB) is electrically coupled to a readout integrated circuit (ROIC) 420. The ROIC 420 includes an oscillator 422 and a current source 424 for generating a sine wave and three lock-in amplifiers 426, 428, 430 corresponding to each of the three resistors (e.g.. resistors 410. 412, 414) of the battery sensor 400. The ROIC 420 is electrically coupled to the common electrode 402, the second independent electrode 404, the third independent electrode 406, and the fourth independent electrode 408 of the battery sensor 400 to provide a readout of voltage (e.g., for obtaining temperature modulation amplitude) through the first resistor 410, the second resistor 412, and the third resistor 414. Furthermore, based on the sine wave generated by the ROIC 420, one or more of the resistors 410, 412, 414 acts as a heater and / or a thermistor to enable the ROIC to detect local temperature variations.

[0035] Figure 5 illustrates an example circuit for a readout integrated circuit. Referring to Figure 5, circuit 500 illustrates an example implementation of readout integrated circuit 420 and includes a modulated (e.g., sine wave) current source 502 and three lock-in amplifiers 506, 508, 510 (e.g., one for each resistor of a battery sensor). In addition, instrumentation amplifiers (e.g., instrumentation amplifiers 522, 524, 526. 534, 536) are included for capturing the signals across the resistors. The first resistor 512 of the battery sensor (e.g., first resistor 410 of Figure 4) is electrically coupled to the first lock- in amplifier 506 via a common electrode (e.g., common electrode 402 of Figure 4) at connection point CPI and a second independent electrode (e.g., the second independent electrode 404 of Figure 4) at connection point CP2. The second resistor 514 of the battery sensor is electrically coupled to the second lock-in amplifier 508 via the common electrodeat connection point CP3 and a third independent electrode (e.g., the third independent electrode 406 of Figure 4) at connection point CP4. The third resistor 516 of the battery sensor is electrically coupled to the third lock-in amplifier 510 via the common electrode at connection point CP5 and a fourth independent electrode (e.g., the fourth independent electrode 408 of Figure 4) at connection point CP6.

[0036] Connection points CPI. CP2. CP3, CP4, CP5, and CP6 can be coupled to the appropriate electrodes of a battery sensor via any suitable means. For example, conductive wires can be bonded to the electrodes. Of course, other packaging methods can be used.

[0037] In operation, the sinusoidally-modulated current source 502 at frequency f is electrically coupled to the first resistor 512. which generates a heat flow (Joule heating) at frequency 2 / (e.g., ranging from 0.1 to 1 Hz) in both the forward and backward current direction while simultaneously measuring temperature oscillation amplitude and phase at a 2 / component using all three resistors via lock-in amplifiers on the readout integrated circuit.

[0038] Furthermore, the multiplication of / and 2 / ’ results in a 3 / ’ signal, which is proportional to the temperature oscillation amplitude (“temperature modulation amplitude”) of the corresponding resistor (e.g., first resistor 512, second resistor 514, and / or third resistor 516). At the ROIC 500, a matching resistor 518 is included in series with the first resistor 512 to remove the / component from the modulated current source 502. This can be performed by connecting instrumentation amplifier 522 to the nodes at the ends of the matching resistor 518 (and optional calibration resistor 520, which may be implemented by a potentiometer or other variable / modifiable resistor) using an opposite polarity to the connection of the instrumentation amplifier 524 that is coupled to the first resistor 512 at connection points CPI and CP2. The output of the two instrumentation amplifiers 522 and 524 are coupled to the inputs of a third instrumentation amplifier 526 to ultimately remove the / component before being input to the first lock-in amplifier 506. Specifically, by taking the Fourier transform across the first resistor 512, the / component is removed by the matching resistor 518 (in combination with a calibration resistor 520, when used). Accordingly, the first lock-in amplifier 506 is not saturated by the / component, leaving only the 3 / component for measurement. The first lock-in amplifier 506 also gets the reference signal from the current source 504 to “lock-in” the phase and frequency. The third harmonic 528 can be chosen for extraction for the first lock-inamplifier 506 by appropriate selection in the lock-in amplifier settings 506. The amplitude 530 of the oscillation at 2 / is also chosen.

[0039] The instrumentation amplifier 534 for the second resistor 514 and instrumentation amplifier 536 for the third resistor 516 can sense the difference in voltage between a reference voltage across a matching resistor and the second resistor 514 or third resistor 516. respectively. Similar to the matching resistor 518 for the first resistor 514, the matching resistors for the second resistor 514 and third resistor 516 can further include a calibration resistance (e.g., by a fixed or variable resistor). The second lock-in amplifier 508 and the third lock-in amplifier 510 each have the second harmonic 532 chosen for extraction. The amplitude 530 of the oscillation at 2 / is also chosen.

[0040] The signals detected by the ROIC enable the detection of changes in the signal passing across the battery from the heated first resistor 512 to the other resistors 514, 516

[0041] In some cases, the ROIC is integrated into a battery7management system for one or more batteries having state-of-charge, state-of-health, and state-of-safety being monitored. In some cases, the ROIC is a separate component that is, for example, printed on a printed circuit board and coupled to the battery7sensor via shielded wires. In some cases, the ROIC is printed on the thin-film substrate of the battery7sensor. In some cases, the signal of the amplitude 530 values are recorded (e.g., before or after the calibration for the corresponding resistor at the chosen frequency) by the battery management system (or microcontroller or computer) and converted to values of temperature oscillation amplitude (“temperature modulation amplitude”). Subsequently, the temperature oscillation amplitudes are input to a calibration to obtain thermal conductivity tensor elements. Similarly, the phase output from lock-in amplifier can be input to a calibration to obtain the thermal diffusivity tensor elements. The thermal conductivity and thermal diffusivity values are used for estimation of state of health, state of charge, and state of safety7of the battery.

[0042] In some cases, a single ROIC can be coupled to multiple battery7sensors (and their corresponding batteries). Figure 6 illustrates a plurality of thin-film battery sensors coupled to a readout integrated circuit. Referring to Figure 6, a plurality7of thin- film battery sensors 602 can be coupled to a single ROIC 610. The ROIC 610 includes an oscillator 612, a current source 614, and three lock-in amplifiers 616, 618, 620 similar to that described with respect to Figures 4 and 5. However, ROIC 610 further includes a multiplexer 622 to enable the single ROIC 610 to perform read out functionality for theplurality of thin-film battery sensors 602. As described in more detail with respect to the example implementation shown in Figure 7, the ROIC can be electrically coupled to a common electrode, a second independent electrode, a third independent electrode, and a fourth independent electrode of each battery sensor 602 via the multiplexer 622 to provide a readout of voltage through the first resistor Tl, the second resistor T2, and the third resistor T3 of each battery sensor 602. Furthermore, each resistor Tl, T2, and T3 acts as a heater and a thermistor to detect local temperature variation (e.g., providing a measurement of anisotropic thermal conductivity or diffusivity tensor elements of a battery' cell).

[0043] Figure 7 illustrates an example implementation of a multiplexer of a readout integrated circuit. Referring to Figure 7, a multiplexer 700 of a readout integrated circuit includes connections to a plurality of battery sensors. In the example implementation of Figure 7, the multiplexer 700 is formed of three 8: 1 or 16: 1 multiplexers (MUXs). One MUX selectively couples one or both of the electrodes of a first resistor 702 of each batterysensor (e.g.. common electrode and corresponding independent electrode or just the corresponding independent electrode) to the ROIC (e.g.. at CPI and CP2 connections of Figure 5), one MUX selectively couples the one or both of the electrodes of a second resistor 704 of each battery' sensor (e.g., common electrode and corresponding independent electrode or just the corresponding independent electrode) to the ROIC (e.g., at CP3 and CP4 connections of Figure 5). and one MUX selectively couples the one or both of the electrodes of a third resistor 706 of each battery sensor (e.g., common electrode and corresponding independent electrode or just the corresponding independent electrode) to the ROIC (e.g., at CP5 and CP6 connections of Figure 5). Control signals 708 are received by the multiplexer 700 to control input selection of the MUXs in order to sample each resistor (e.g., in a looping fashion). The control signals 708 may come from a battery management system or a stand-alone controller associated with the ROIC. The ground connection 710 of each resistor 702, 704, 706 is to the common electrode of each battery sensor.

[0044] Figure 8 illustrates a flow diagram for determining a state of health, state of charge, and state of safety of each cell in a battery'. Referring to Figure 8, a method 800 of determining a state of health, state of charge, and state of safety of each cell in a battery includes receiving (802) a readout of a temperature modulation amplitude from each thin- film battery sensor physically coupled to a cell of the battery over a period of time, comparing (804) the readout to one or more charge-discharge cycling datasets (805), anddetermining (806) a state of health, state of charge, and state of safety of each cell in a battery based on the comparison between the readout for that cell across the period of time and the one or more charge-discharge cycling datasets. The thin-film battery sensor can be implemented as described with respect to any of Figures 1A, IB, and 2. The readout of the temperature modulation amplitude can be implemented using a ROIC such as described with respect to any of Figures 4, 5. 6, and 7.

[0045] In some cases, neural network state estimators are trained using the one or more charge-discharge cycling datasets on the state of health, state of charge, and state of safety of each cell in a battery'. Accordingly, the method 800 may be carried out by a neural network that is trained on the one or more charge-discharge cycling datasets. In some cases, the charge-discharge cycling dataset is specific to a type (e.g., size, shape, weight, charge capacity, and / or electrode materials) of each cell in the battery.

[0046] Figure 9 illustrates an example system for determining a state of health, state of charge, and state of safety of each cell in a battery. Referring to Figure 9, a system 900 for determining a state of health, state of charge, and state of safety of each cell in a battery includes an interface 902 that is connected to one or more ROICs 920 for wired or wireless communications such as receiving voltage, current, and / or temperature readings (e.g., temperature modulation amplitude) from one or more battery' sensors. The interface 902 can also include wired or wireless interfaces for communicating with a battery management system, as well as interfaces for communicating with the “outside world” (e.g., external networks). In some cases, the system 900 further includes other interfaces, such as a user interface for providing user-understandable output (e.g., audio, visual) and receiving user input. In some cases, a user interface can include a display on which the results of the determination of SOH, SOC, and SOS can be displayed as well as suitable input device interfaces for receiving user input (e.g., mouse, keyboard, microphone).

[0047] The system 900 further includes one or more processors 904 and a corresponding storage system 906 storing instructions 908 for determining a state of health, state of charge, and state of safety’ of each cell in a battery (e.g., instructions for performing method 800 of Figure 8) and storing data such as charge-discharge cycling datasets 910. The one or more processors 904 can include one or more of any’ suitable processing devices (“processors”), such as a microprocessor, central processing unit (CPU), graphics processing unit (GPU), field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), logic circuits, state machines, application-specific standard products (ASSPs), system-on-a-chip systems (SOCs),complex programmable logic devices (CPLDs), etc. Storage system 906 can include any suitable storage media that can store the instructions 908 for determining a state of health, state of charge, and state of safety of each cell in a battery. Suitable storage media for storage system 906 includes random access memory, read only memory, magnetic disks, optical disks, CDs, DVDs, flash memory, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other suitable storage media. As used herein, “storage,” “memory storage,” and “storage media” do not consist of transitory, propagating waves. Instead, these terms refer to non-transitory media.

[0048] Laboratory Testing

[0049] The inventors conducted experiments to measure the diagonal elements of K tensor on a pouch cell using the techniques described above (and circuitry such as described with respect to Figures 4-7). The cell under test was an NMC with 25 Ah capacity (Enertech Inc.) and dimensions of 10 inches by 6 inches by 0.05 inches, as illustrated in Figure 10. The inventors used a heating amplitude of 100 mW at a heating frequency of 2 / = 0.1 Hz, generating a peak temperature change of <1°C. The measured temperature oscillation gradients in x. y. and z directions in combination with the thermal boundary conditions imposed by connecting the perimeter of the cell to cooled aluminum blocks (e.g., simulating cooling plates in EV packs) allowed the inventors to extract T.v, Kyy, Kzz using a simple heat construction model.

[0050] Figures 11A and 1 IB illustrate the preliminary results during two standard consecutive charge-discharge cycles at a rate of 1 C on a fresh cell. Specifically, Figure 11 A illustrates the cell voltage and SOC curves and Figure 1 IB illustrates the Kxx, Kyy, K::values. As can be seen, the in-plane (Kxx. Kyy) and through-plane (Kzz) thermal conductivity values vary unevenly as a function of SOC. This observation allows the inventors to identify SOC using a sufficient calibration.

[0051] Figures 12A and 12B illustrate the preliminary results of accelerated degradation cycling. In this experiment, a fresh cell was subject to an overvoltage charge of 4.7V and discharge of 2.35V at a rate of 1.6C to cause accelerated degradation primarily caused by Li-plating and electrolyte decomposition. Specifically, Figure 12A illustrates the charge-discharge capacity curves as a function of cycle number measured directly using a battery cycler and Figure 12B illustrates the median ratios of Kxx / Kz: and KyyIKzz as a function of cycle number. The shaded areas represent the 95% confidence interval of these ratiometric signals. As can be seen, both signals in Figure 12B have a strong correlation with the measured discharge capacity in Figure 12A. Furthermore, theconfidence intervals obtained from the standard deviation of the signals seem to increase in a predictive manner a few cycles before event 2 occurs, during which a sharp drop in charge capacity was observed. The ratiometric thermal conductivity signals continue to show large variation after this event, which may indicate an abrupt change in the internal structure / composition of the cell.

[0052] Figure 13 illustrates an example calibration curve containing the thermal conductivity anisotropy ratios of Kxx / Kzzand Kyy / Kzzas a function of state-of-health (SOH). SOH is defined as the ratio of discharge capacity at each cycle to the discharge capacity level at cycle 1. For the specific cell chemistry and degradation conditions tested here, a simple linear fit correlates the thermal conductivity anisotropy ratio to the SOH (e.g.. fitted lines in Figure 13). This shows the feasibility of using ratiometric thermal conductivity values as identifiers for SOH and remaining useful life estimation.

[0053] Although the subject matter has been described in language specific to structural features and / or acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as examples of implementing the claims and other equivalent features and acts are intended to be within the scope of the claims.

Claims

CLAIMSWhat is claimed is:

1. A thin-film battery sensor, comprising: a thin-film substrate; a common electrode on the thin-film substrate; a first resistor on the thin-film substrate and coupled to the common electrode; a second electrode on the thin-film substrate and coupled to the first resistor; a second resistor on the thin-film substrate and coupled to the common electrode; and a third electrode on the thin-film substrate and coupled to the second resistor.

2. The thin-film battery sensor of claim 1, further comprising: a fourth and a fifth electrode on the thin-film substrate and coupled to the first resistor; and a sixth and a seventh electrode on the thin-film substrate and coupled to the second resistor.

3. The thin-film battery sensor of claim 2, further comprising: a third resistor on the thin-film substrate and coupled to the common electrode; and an eighth electrode on the thin-film substrate and coupled to the third resistor.

4. The thin-film battery sensor of claim 3, further comprising a ninth electrode on the thin-film substrate and coupled to the third resistor.

5. The thin-film battery sensor of any of claims 1-4, further comprising: a readout integrated circuit (ROIC) comprising a modulated current source and a number of lock-in amplifiers corresponding to each resistor on the thin-film substrate, wherein the ROIC is coupled to electrodes including the common electrode, the second electrode, and the third electrode to provide a readout of voltage from the thin-film battery sensor.

6. The thin-film battery sensor of claim 5, wherein at least one resistor of the thin- film battery sensor is configured by the ROIC to function as a heater and at least oneresistor of the thin-film battery' sensor is configured by the ROIC to function as a thermistor.

7. The thin-film battery sensor of claim 1, wherein the thin-film substrate is a flexible substrate, wherein components of the thin-film battery sensor are deposited on a side of the flexible substrate and an opposite side of the flexible substrate is configured to attach to a cell of a battery.

8. The thin-film battery' sensor of claim 7, wherein components of the thin-film battery sensor are made of chromium, titanium, platinum, gold, or vanadium oxide or other temperature sensitive materials.

9. A system for determining state of health, state of charge, and state of safety of a battery comprising: a plurality of thin-film battery sensors, each thin-film battery’ sensor configured to couple to a cell in a battery, each thin-film battery sensor comprising: a thin-film substrate: a common electrode on the thin-film substrate; a first resistor on the thin-film substrate and coupled to the common electrode; a second electrode on the thin-film substrate and coupled to the first resistor; a second resistor on the thin-film substrate and coupled to the common electrode; a third electrode on the thin-film substrate and coupled to the second resistor; a third resistor on the thin-film substrate and coupled to the common electrode; and an eighth electrode on the thin-film substrate and coupled to the third resistor, and a readout integrated circuit (ROIC) comprising a modulated current source and a number of lock-in amplifiers corresponding to each resistor on the thin-film substrate, wherein the ROIC is coupled to the common electrode, the second electrode, the third electrode, and the eighth electrode of each thin-film battery’ sensor of the plurality of thin-film battery sensors to provide a readout of voltage of each of the plurality of thin-film battery sensors.

10. The system of claim 9, wherein each resistor of the plurality of thin-film battery sensors is configured to function as a heater and / or a thermistor to detect local temperature variation, wherein the readout of the ROIC indicates a temperature modulation amplitude sensed by each resistor of the plurality of thin-film battery sensors.

11. The system of claim 10, further comprising: a processor; and a storage system storing a charge-discharge cycling dataset and instructions for determining a state of health, state of charge, and state of safety, wherein the instructions for a state of health, state of charge, and state of safety stored on the storage system, when executed by the processor, direct the system to: receive the readout of the current, the voltage, and the temperature of each thin- film battery sensor from the ROIC over a period of time; compare the readout to the charge-discharge cycling dataset; and determine a state of health, state of charge, and state of safety of each cell in the battery based on the comparison between the readout of the thin-film battery sensor of that cell across the period of time and the charge-discharge cycling dataset.

12. The system of claim 11, wherein the charge-discharge cycling dataset is specific to a type of each cell in the battery.

13. The system of claim 9, wherein each of the plurality of thin-film battery sensors further comprises: a fourth and a fifth electrode on the thin-film substrate and coupled to the first resistor; a sixth and a seventh electrode on the thin-film substrate and coupled to the second resistor; and an eighth and a ninth electrode on the thin-film substrate and coupled to the third resistor.

14. The system of claim 13, wherein the ROIC is further coupled to the fourth electrode, the fifth electrode, the sixth electrode, the seventh electrode, the eighth electrode, and the ninth electrode.

15. The system of any claims 9-14, wherein the thin-film substrate of each of the plurality of thin-film battery sensors is a flexible substrate, wherein components of each of the plurality of thin-film battery sensors are deposited on a side of the flexible substrate and an opposite side of the flexible substrate is configured to attach to a cell of a battery.

16. A method of determining a state of health, state of charge, and state of safety of each cell in a battery, comprising: receiving a readout of a temperature modulation amplitude from each thin-film battery sensor physically coupled to a cell of the battery over a period of time, wherein each thin-film battery sensor comprises: a thin-film substrate; a common electrode on the thin-film substrate; a first resistor on the thin-film substrate and coupled to the common electrode; a second electrode on the thin-film substrate and coupled to the first resistor; a second resistor on the thin-film substrate and coupled to the common electrode; a third electrode on the thin-film substrate and coupled to the second resistor; a third resistor on the thin-film substrate and coupled to the common electrode; and an eighth electrode on the thin-film substrate and coupled to the third resistor; comparing the readout to one or more charge-discharge cycling datasets; and determining a state of health, state of charge, and state of safety of each cell in a battery based on a comparison between the readout for that cell across the period of time and the one or more charge-discharge cycling datasets.

17. The method of claim 16. wherein the readout of the temperature modulation amplitude is received from a readout integrated circuit (ROIC) that applies a modulatedcurrent source to at least one resistor of the thin-film battery sensor and reads out signals across each resistor of the thin-film battery sensor.

18. A thin-film battery sensor for a lithium-based battery cell structured to provide measurements of anisotropic thermal conductivity or diffusivity tensor elements of the lithium-based battery cell, wherein the thin-film battery sensor is structured according to any of claims 1-8.

19. A system coupled to the thin-film battery sensor for identifying a state of charge, state of health, and / or state of safety of individual battery cells based on measurements received from corresponding thin-film battery sensors, each thin-film battery sensor structured according to claim 18.

Images