Direct observation of internal state of thermal runaway in lithium ion battery during nail-penetration test

Direct observation of internal state of thermal runaway in lithium ion battery during nail-penetration test

Journal of Power Sources 393 (2018) 67–74 Contents lists available at ScienceDirect Journal of Power Sources journal homepage: www.elsevier.com/loca...

3MB Sizes 0 Downloads 1 Views

Journal of Power Sources 393 (2018) 67–74

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Direct observation of internal state of thermal runaway in lithium ion battery during nail-penetration test

T

Tokihiko Yokoshimaa,∗∗, Daikichi Mukoyamaa, Fujio Maedaa, Tetsuya Osakaa,∗, Koji Takazawab, Shun Egusab, Satomi Naoic, Satoru Ishikurac, Koichi Yamamotoc a

Research Organization for Nano & Life Innovation, Waseda University, 513 Waseda-tsurumaki-cho, Shinjuku-ku, Tokyo, 162-0041, Japan Toshiba Infrastructure Systems & Solutions Corporation, 72-34, Horikawa-cho, Saiwai-ku, Kawasaki-shi, Kanagawa, 212-8585, Japan c National Institute of Technology and Evaluation, 1-22-16, Nankokita, Suminoe-ku, Osaka-shi, Osaka, 559-0034, Japan b

H I GH L IG H T S

G R A P H I C A L A B S T R A C T

X-ray scanner system for testing • An internal short circuits in LIBs was developed.

transmission moving images of • X-ray the layered LIB structure were obtained.

changing structures of LIB during • The nail-penetration test were directly observed.

thermal runaway in LIB was di• The rectly observed.

A B S T R A C T

A system using a high-speed and high-precision X-ray inspection system for testing internal short circuits in lithium-ion batteries (LIBs) was firstly developed for the safety test on LIBs. X-ray transmission moving images of the anode and cathode were obtained in the layered LIB structure. The nail-penetration test was chosen to test for the presence of an internal short circuit. This system would allow direct observation of smoke generation inside outside the battery, ballooning of the pouch, as well as changing layered structures of electrodes in real time. Since the results of a conventional nail-penetration test are indicated only by smoke generation, fire, or explosion, this new system allows electrode changes in the pouch to be observed for the first time. This system is expected to lead to great developments in the safety of LIBs.

1. Introduction There have been many recent reports of accidents and troubles associated with fires caused by lithium-ion batteries (LIBs) [1–4]. Since LIBs are characterized by high operation voltage, high energy density, and the use of flammable organic electrolytes, they are energy devices with the potential to catch fire [5–8]. Thus, LIBs designed for commercial use must pass safety standards, and safety tests, for example ∗

electrical test, environmental tests, and mechanical tests, are carried out to maintain these standards [9–11]. The nail-penetration test is a widely used LIB safety test for the evaluation of internal short circuits, which are one of the major causes of battery fires, because of simple and easy [12–23]. Since the main results of this test are smoke generation, fire, explosion, or the absence of all of the above. This type of safety test only checks whether an LIB is safe, but an analysis of the test results is not carried out from an academic viewpoint. Therefore, the

Corresponding author. Corresponding author. E-mail addresses: [email protected] (T. Yokoshima), [email protected] (T. Osaka).

∗∗

https://doi.org/10.1016/j.jpowsour.2018.04.092 Received 27 February 2018; Received in revised form 11 April 2018; Accepted 26 April 2018 0378-7753/ © 2018 Published by Elsevier B.V.

Journal of Power Sources 393 (2018) 67–74

T. Yokoshima et al.

independent measurement instrument located the laboratory or the factory is not easy if using the synchrotron radiation as X-ray source. Since X-rays can be transmitted through the viewing window in the LIB cell, details of the internal state of the layered structure can be observed directly. The X-ray transmission image in Fig. 1 clearly shows the layered structure and materials of the LIB cell to allow the layer of active materials, current collector, and separator to be determined. The nail-penetration test for LIB cell was conducted using a simple nail as the indenter, and the thermal runaway of the LIB during the test was observed in real time using the iSC Scanner. The analysis of thermal runaway was designed as the first step of the study when the battery bursts and flames. This analysis was also designed as the first step in the visualization of the internal state of the LIB during safety conditions.

direct observation of the internal state of the electrodes, electrolyte, and separator during the test is not easy in real time. Only the voltage and thermal and external conditions of the LIB are able to be observed in real time [24–27]. Visualization or observation of the internal state of a battery cell can be considered the most important safety measurement that is lacking in current approaches. Moreover, it is necessary to systematically and methodically analyse the cause of smoke, fire, or explosion as a result of thermal runaway in order to understand the behaviour of the battery's internal state during the test. Moreover, the behaviour of an LIB when it bursts into flames depends on the operating conditions, and details of the test results are not always reproducible [28,29]. Thus, a method for direct observation of the internal state when a LIB into flames during an internal short-circuit test has been keenly anticipated. The direct or indirect observation of thermal runaway using computer simulations [30,31], electrochemical impedance measurement [32], and synchrotron radiation [33] was reported. To address this need, we developed a high-speed and high-precision internal-state direct observation system for internal short-circuit tests of battery cells. This system has an X-ray scanner to allow for high-speed and high-precision direct observation of a battery's internal state, and the system can record X-ray transmission moving images of the electrode behaviour during the internal short-circuit test. This system has a changeable indenter for the internal short-circuit test, and the traditional nail-penetration test, blunt nail-penetration test, and forced internal short-circuit test with contaminants can be carried out with the selection of an appropriate indenter. This internal short-circuit observation system with X-ray scanner for battery cells, hereafter abbreviated as iSC Scanner, is the first system developed for detailed analysis of the electrode behaviour and its changes in real time. Fig. 1 shows a schematic illustration of the iSC Scanner. Micro X-ray point source was used in this system, transmission image of commercial LIB cell cannot be observed without using computed tomography technique (CT). X-ray from the point source was not transmitted though layered structure of electrodes clearly since X-ray spreads radially from the point source. The synchrotron radiation can be transmitted though layered structure of commercial battery cell [33]. However, development of the

2. Experimental 2.1. Preparation of LIB samples for X-ray transmission observation CELLSEED C‐5H (lithium cobalt oxide, LiCoO2, Nippon Chemical Industrial Co. Ltd.) and graphite CGB-10 (Nippon Graphite Industries, Co. Ltd) were used as the active materials for the cathode and anode, respectively. The cathode was prepared on both sides of an aluminium foil using LiCoO2, acetylene black (DENKA BLACK, DENKA Company Ltd.), Ketjen black (Lion Specialty Chemicals Co. Ltd.), and polyvinylidene difluoride, PVDF (KF Polymer, Kureha Corp.) in the ratio of 87:6:2:5 by weight. The thickness of the aluminium foil was 20 μm. The anode was prepared on both sides of a copper foil using graphite, acetylene black and PVDF in the ratio of 90:5:5 by weight. The thickness of the copper foil was 20 μm. The areal discharge capacities of the cathode and cathode were 1.0 and 1.1 mAh cm−2, respectively. The electrolyte used was 1 M lithium hexafluorophosphate (LiPF6) in a 1:1 (v:v) mixed solvent of ethylene carbonate (EC) and diethylene carbonate (DEC) (Kishida Chemical Co. Ltd.). UPORE UP3085 (Ube Industries Ltd), was used as a separator. A cell consisting of a cathode, a separator, and an anode was then assembled in a dry room and placed in an aluminum-laminated bag. Detail of preparation technique in ref

Fig. 1. Schematic illustration of nail-penetration system with the iSC scanner, an X-ray scanner for laminated lithium-ion battery. The image on the right is atypical X-ray transmission image of a laminated LIB cell recorded through the viewing window. 68

Journal of Power Sources 393 (2018) 67–74

T. Yokoshima et al.

acrylic plates to suppressing ignition of the cell and to keep any generated smoke from spreading to the rest of the system. After the test, the smoke was removed with an active carbon filter. Additional countermeasures were made to prevent scattering X-rays to the area outside the system.

[34] was referred. The cathode and anode for a standard cell had dimensions of 70 mm × 70 mm and 74 mm × 74 mm, respectively. The cell consisted of 17 electrode layers (cathode: 9 layers, anode: 8 layers). The unit cell consisted of a single-faced anode and a single-faced cathode facing each other. Each battery cell consisted of 16 unit cells (16 pairs of anodes and cathodes) since these were eight double-sided anodes and nine double-sided cathodes. All of the unit cells consisting of one anode–cathode pair were connected in parallel. The capacity of the standard cell was 800 mAh. To record the X-ray transmission images, a viewing window was prepared. The width of a part or the entire positive electrode was shortened from 70 to 6 mm, and the width of a part or the entire negative electrode was shortened from 74 to 10 mm. The depths of the positive and negative electrodes were not changed from the original values of 70 and 74 mm, respectively. To maintain the flatness and layered structure of each electrode, which had a narrow width, a polypropylene (PP) plate with thickness of 2 mm was used as the retainer plate, and one plate was set at the top and one at the bottom of the stack of electrodes. To prevent warping of the PP plate during the nail-penetration test, a through-hole with diameter of 2 mm was drilled at the centre of the PP plate for nail injection. X-ray observation during the nail-penetration test was carried out using this viewing window, which consisted of an electrode with width of 8–10 mm and a PP retainer plate. An LIB cell with capacity of 60 mAh was constructed from 9 sheets of positive electrodes (width: 8 mm) and 8 sheets of negative electrodes (width: 10 mm). An LIB cell with capacity of 420 mAh was constructed from 5 sheets of positive electrodes (width: 8 mm), 4 sheets of positive electrodes (width: 70 mm), 3 sheets of negative electrodes (width: 10 mm), and 5 sheets of negative electrodes (width: 74 mm). To form a 860 mAh LIB cell, a 60 mAh cell and a 800 mAh cell were connected in parallel, this stacked cell was used as the 860 mAh cell. The X-ray transmission measurement of the 860 mAh stacked cell during the nail-penetration test was carried out through the viewing window of the 60 mAh cell. Before nail-penetration test, two cycles of charging and discharging were initiated at a C-rate of 0.1 C.

2.3. Direct observation of internal state of LIB cell The X-ray transmission image of entire cell was measured by a largearea X-ray scanner (HITEX-150HI with XRD-0820AN, HITEX). The iSC Scanner was used in all nail-penetration tests. The moving images of the X-ray transmission were recorded at 250 flame s−1, and the nail-delivery speed was 3 mm s−1. In this test, the nail was sent through the sample at 1 shot s−1, and the sending pitch was 0.2 mm. This naildelivery condition was determined to obtain repeatability of the test results. Using this condition, the optical events of the nail penetration test in the same conditions were almost same, and the inner events were not same, but similar. The focus-to-image distance (FID) and the focusto-object distance (FOD) were ∼40 and ∼400 mm, respectively; the magnification of the image was about 10 × . The image resolution was 512 pixels × 512 pixels. The X-ray tube current and voltage were determined to obtain clear images; a typical current of 300 μA and a typical voltage of 150 kV were used. The cell voltage during the test was measured by an oscilloscope (WaveSurfer 10, Teledyne Lecroy). Sampling rate was 5,000 samples s−1. The temperature of cell surface, nail and inner case during the test was measured by a data logger (Wireless Logging Station LR8410, Hioki E.E. Corporation) with a type T thermocouple. The sampling rate and the measurement window were 0.5 s and 0–100 °C, respectively. The measurement point of the cell surface was 1 cm from centre of the cell in which was piercing point by nail. Because the battery cell had a pp plate acted as heat insulator, this value was not inner cell temperature. The measurement point of the nail was 1.5 cm from tip top of the nail. This measurement temperature was not the value at top tip of the nail, because the nail was very thick diameter compared with that of the tip and very far from the tip top. Thus, these values were used for qualitative change in the temperature. These measurements were carried out as a collaborative test of the National Laboratory for Advanced Energy Storage Technologies (NLAB), Global Centre for Evaluation Technology (GCET), and National institute of Technology and Evaluation (NITE).

2.2. Internal short-circuit observation system with X-ray scanner for battery cell: iSC scanner The nail-penetration testing device consisted of a nail, a servomotor for moving the nail, and a sample holder. The nail was prepared by a lathe from a stainless round bar; the diameter of nail was 5 mm, and the apex angle of the cone was 30°. The nail was moved by the servomotor, and accuracy of the position was ± 30 μm. The minimum pitch and the moving speed were 100 μm and 3 mm s−1, respectively. A five-axis stage was used for the sample holder, and the accuracy of the position was ± 10 μm. The internal short-circuit test observation system with Xray scanner for battery cell, i.e., iSC Scanner, consisted of a nail-penetration testing device, a high-speed and high-precision X-ray camera, and its safety restraints. An industrial micro X-ray CT scanner (TOSCANER-32252 μhd-HS (Type C), Toshiba IT & Control Systems Corporation) was used to record X-ray transmission moving images of the electrode behaviour during the nail-penetration test. This system consisted of a micro-focus X-ray source, a high-speed X-ray transmission camera, a high-precision CT camera, and a control system. An Xray image-intensifier camera unit, which comprised X-ray I.I. E5764REP1 (Toshiba Electron Tubes & Devices Co. Ltd.) and Memrecam GX-1 Plus (Nac Image Technology Inc.), was used as the ultra-high-speed Xray transmission camera for the moving images. The maximum recording speed of a moving image was 10,000 flame s−1. An X-ray image-intensifier camera unit, which comprised X-ray I.I. E5764SD-P1K (Toshiba Electron Tubes & Devices Co. Ltd.) and CT-305C (Toshiba Corporation), was used as the high-precision CT camera. The CT image could not be observed during the nail-penetration test; however, details of the image could be obtained without moving the sample. The nailpenetration testing device was set in this X-ray system. The LIB sample cell for the nail-penetration test was covered with a box made of thick

2.4. LIB cell and its electrode characterization X-ray CT images were obtained before and after the nail-penetration test with an industrial micro-CT scanner (TOSCANER-32300μFD-Z, Toshiba IT & Control Systems Corporation). The nail tip was observed with a digital microscope (RH-2000, HIROX Co. Ltd.). 3. Results and discussion 3.1. X-ray CT images Fig. 2 shows X-ray computed tomography (CT) images of the 60 mAh cell, 420 mAh cell, and 60 mAh cell in the 860 mAh stacked cell were obtained before and after the nail-penetration test. The 60 mAh cell and 860 mAh cell served as the small- and large-capacity reference cells, respectively, for the 420 mAh cell. The centre of the cell was pierced by a nail during the test. All but the first layer of the 60 mAh cell remained unchanged before and after the test (Fig. 2a–d). The first layer of the cathode was separated by the impact into two layers, but a clear signature of gas generation was not observed, and external observations showed that the cell was not changed. Fig. 3 shows the schematic circuit diagram where the each electrode layer is regarded as the closed loop circuit with the charged unit cell, resistance and switch triggering electrical local short circuit. The number of short circuits is defined by the number of layers 69

Journal of Power Sources 393 (2018) 67–74

T. Yokoshima et al.

Fig. 2. X-ray CT images of the battery cells before and after the nail-penetration test. (a–d) 60 mAh cell; (e–j): 420 mAh cell; (k–n) 60 mAh cell in stacked 860 mAh cell. The 3D images were selected from the moving image prepared for supplementary data (420 mAh cell: video 1, 60 mAh cell in stacked 860 mAh cell: video 2).

were separated into two layers by the impact (Fig. 2k–n, video 2), and the first cathode layer was broken. Moreover, particles of the active material spread to the interior of the cell, which means that the electrode far from the impinging nail was changed, and the heat reached all areas of the cell. The Iis and Ies in Fig. 3 were large compared to that of the 60 mAh cell, and the intense heat boiled the electrode in the entire cell because the large Ies was supplied to the 60 mAh cell from the 800 mAh cell (Fig. 3). A large change was observed in the layered structure. Supplementary video related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2018.04.092.

penetrated by the nail. When a short circuit was formed by a nail, this short circuit formed another unit cell, and Ils flowed through the inner short circuit that consisted of the two electrodes and the nail. The Ils flowed through the short circuit in the unit cell in which the circuit was shorted. Moreover, the Iis also flowed through the short circuit from the unit cell whose circuit was not shorted. The heat was generated by the current through the resistance, mainly Rn in the short circuit. This effect only led to small thermal damages, however. In the case of the 420 mAh cell, the layered structure was changed after the test, and the gap distance between the electrodes slightly increased. The cathodes of the 1st and 3rd layers were separated by the impact into two layers (Fig. 2e–h, video 1). The gap distance between the 7 cm × 7 cm electrodes also increased (Fig. 2i and j). These results are believed to be clear indicators of gas generation. It is thought that this change represents damaged caused by current-generated heat in the short circuit formed by the nail–electrode connection. The shortcircuit currents Ils and Iis in Fig. 3 were large compared to that of the 60 mAh cell, and the intense heating boiled the electrode. However, the generated gas moved to the interlayer space between the 7 cm × 7 cm electrodes. Based on the external observations, gas was released for only a short moment, and the pouch slightly ballooned as a result (video 3). Thus, a small gas generation occurred, and the well-known thermal runaway did not occur. In the case of 60 mAh cell in the 860 mAh stacked cell, gas release to the exterior occurred continuously and the pouch ballooned, indicating that the well-known thermal runaway occurred (video 4). As shown in Fig. 2, the layered structure was dramatically changed, and the gap distance between the electrodes increased. All of the cathodes layers

3.2. Direct observation of internal state The results of the nail tip piercing the first and second layers of the 420 mAh cell are shown in Fig. 4b–f and video 5. The image shown in Fig. 2c just became the new internal short circuit formed by the sending the nail a distance of 0.2 mm into the electrode layers, and the gap distance between the 1st and 2nd layers increased, which means that we observed micro-swelling behaviour in the electrode. After 200 ms (Fig. 4d), the gap distance decreased and the layered structure and the interlayer distance remained the same as before the short circuit occurred. In Fig. 4c, the tip of the nail can be seen over the 2nd layer; however, the tip surprisingly disappeared over the 2nd layer in the image in Fig. 4d. The radius of curvature of the tip of the nail changed from 20 μm before the test to 100 μm after the test (Fig. S3). The nail tip was very sharp before melting, and the current of the short circuit was converged to the area measuring several thousand μm−2 at the tip. In the case of 60 mAh cell, the radius of curvature of the tip of the nail was 70

Journal of Power Sources 393 (2018) 67–74

T. Yokoshima et al.

Fig. 3. Circuit model of internal short circuit of cell during nail-penetration test. Rn: the contact resistance between the nail surface and one pair of electrodes that formed the unit cell, Rin the internal resistance, Riw: the resistance of inner wiring, Rew: the resistance of external wiring, Ilc: the current from local internal short circuit, Iis: the current from internal short circuit, Ies: the current from external short circuit. The unit cell consisted of a single-faced anode and a single-faced cathode facing each other. Each battery cell consisted of 16 unit cells (16 pairs of anodes and cathodes). The number of short circuits is defined by the number of layers penetrated by the nail.

Fig. 5. Changes in temperature and voltage of cell during nail-penetration test. Triangles, square, and circles indicate temperature of cell surface, nail, and inner case, respectively. The nail delivery and nail-penetration results (video3) are shown in the figure. The numbers are number of times on the nail delivery. Filled circles are presented nail reach the layer from cross section. Filled triangle is presented the penetration of the layer by nail from cross section. The results of external observation and the rebel of high-speed X-ray transmission images are also shown. The transmission image was selected from the moving image in Fig. 4.

Fig. 4. High-speed X-ray transmission images of 420 mA h battery cell during nail-penetration test. The images were selected from the moving image prepared for supplementary data (video 5, 6). 71

Journal of Power Sources 393 (2018) 67–74

T. Yokoshima et al.

Fig. 6. High-speed X-ray transmission images of the 60mAh Cell in the 860 mA h stacked battery cell during nail-penetration test. The images were selected from image prepared for supplementary data (video 6–9).

often observed. Generally speaking, it is not easy to obtain reproducible result in the battery safety tests. The direct and microscopic observation of internal phenomena occurring inside of battery cell is very important to analysis of what really happened during the short circuit. Supplementary video related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2018.04.092. The nail pushed the electrode as it was sent through another 0.2 mm; however, the nail tip did not impinge the electrode and the 2nd electrode was under tension (Fig. 4g). The nail tip changed or turned blunt when compared with a new one, and it is thought that the nail could not impale the electrode, thus no inner short circuit became. It indicated that the result under nail test strongly depends upon the microscopic behaviour around local short circuit which cannot be seen from outside. The behaviour of second short-circuit formation is shown in Fig. 4h–l and video 6. these images showed the nail penetrating four layers of the electrode (Fig. 4i), and each layer of the four electrode layers crumpled. Moreover, external observation showed that the cell pouch slightly swelled. After further 200 ms (Fig. 4j), the interlayer distance decreased; however, the crumpled layer structure remained. After this short-circuit formation, the pouch quickly further swelled. However, gas generation to the cell exterior was not observed, while the cell voltage became unstable and randomly decreased. After 18 s, the gap distance between the electrode layers increased again. As shown in Fig. 4k, the whole electrode was punctured by the nail because of spontaneously swelled electrodes. The cell temperature increased rapidly and the voltage decreased moderately. After 20 s (Fig. 5), the large current of the internal short circuit and the swelling

not changed. These results suggest that the nail tip was melted by the converged flow of large electric discharged current or the heat generated by the contact between the nail and the electrode. After 800 ms (Fig. 4e), the gap distance between the 1st and 2nd layers increased again without moving the nail, and the 2nd layer of the negative electrode vibrated a few times. It suggests that boiling behaviour of electrolyte was observed due to heating. After 250 ms (Fig. 4f), the gap distance decreased again and the layered structure and the interlayer distances were unchanged from that before the short circuit. As shown in Fig. 5, the voltage decreased from 4.2 to 3.6 V when the nail tip was stuck to the electrode, then increased moderately to 3.8 V, and 2 s later, the value remained constant at 3.8 V. This means that the pass of the short circuit was once connected and immediately disconnected, and the short circuit was not formed again later. External observation of the cell during this period (Fig. 4b–f) indicates that white gas were generated and jetted out outside the cell. In Fig. 3, large current of Ils and Iis flowed through Rn at the moment the inner short circuit became, and this concentrated current destroyed the tip of the nail. When the tip of the nail melted and the short circuit was cut off, the electrolyte boiling halted to let electrode layers return to its original position. It is suggested that the nail surface and two electrode edges were in contact again at that moment, the electrolyte between the electrodes boiled as a result of the generated heat, and gas was released outside of the cell. Using this iSC Scanner, we were able to directly observe details of electrode behaviour during the nail-penetration test, and we believe that this is the first observation. The transient phenomena occurring in the order of a few milliseconds were clearly monitored. The contact between the two layers and the simultaneously boiling electrolyte were 72

Journal of Power Sources 393 (2018) 67–74

T. Yokoshima et al.

the pouch occurred at 18–23 s, however, gas released was not observed. With increase in the number of the layer in which the nail stuck, Ils increased and Iis decreased. The current density at per one short circuit decreased because of decreasing Iis. The heating depended on the current density passing through Rn. It is suggested that the generated gas to outside cell was observed due to generate large Iis though the one inner short circuit. It is suggested that no gas release was observed because current density per circuit of unit cell was low in the case of the second short circuit. Supplementary video related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2018.04.092. As shown in the above result, the formation of the internal short circuit and transient behaviour of electrode layers were clearly and microscopically observed through the nailing test. The image and moving picture were monitored and recorded using this iSC Scanner, representing the first detailed report of internal short circuit in a battery cell during the nail-penetration test. 3.3. Direct observation of thermal runaway The behaviour of first short-circuit formation of 60 mAh cell in 860 mA stacked cell caused by the nail tip piercing the first and second layers is shown in Fig. 6b–f and video 7. The gap distance between the 1st and 5th layers increased, which means that we observed microswelling behaviour in the electrode (Fig. 6c). Even though this indenter impaled only two layers, five layers were affected, and the electrolyte between the 1st and 5th layers boiled. The white gas was generated outside the cell during this period (Fig. 7c–f). The electrolyte in the cell boiled as a result of heating caused by the inner short circuit at the nail tip. The top of the nail tip was melted and the melted material moved to side of tip (Fig. 6e and f). Its radius of curvature was found to have increased from 20 to 200 μm, which was larger than that of the 420 mAh cell (Fig. S3). Compared with the 420 mAh cell, a large current was present, and it is believed that a large amount of heat was generated. Fig. 7 shows that the voltage decreased quickly and, the value returned, indicating that the circuit was once connected and then immediately disconnected. Thus, the gas generation only occurred for a very short time, and no thermal runaway occurred. Supplementary video related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2018.04.092. The behaviour of second short-circuit formation by nail penetrating 7 layers is shown in Fig. 6h–j and video 8. Before the nail was sent through the structure 7 times, the nail tip did not pierce the electrode and only tension was generated in the electrode layer (Fig. 6h). The distance between all observed layers increased (Fig. 6i) and all of the layers vibrated a few times, as the tension caused by the penetration nail was released. The electrolyte in the cell boiled as a result of heating caused by the inner short circuit at the nail tip owing to the increase in the distance between the electrode layers. External observation of the cell during this time (Fig. 6i and j) showed that white gas was generated outside the cell only once. Fig. 7 shows that during this time, the voltage decreased quickly and the value returned. This means circuit was once connected and immediately disconnected, and the short circuit was not formed again later. So the voltage and layered structure did not changed for the period between 23 and 32 s, however, the temperatures of the nail and cell slightly increased. After 32 s, electrode started to move toward the nail (video 9), and the electrode was punctured by the nail (Fig. 6l). External observation showed that gas was continuous generated outside the cell. After 32 s, the electrode moved quickly and the nail penetrated deeper and deeper through the electrode (Fig. 6m). The gas release also became strong. Moreover, pouch ballooned until the time reached 48 s. The temperature of all components increased and exceeded 100 °C. After 38 s, the state of the well-known thermal runaway continued, the temperature increased rapidly, and the voltage decreased moderately. It is thought that the internal short circuit was not completely cut off after 23 s, and that some current flowed in the

Fig. 7. Changes in temperature and voltage of cell during nail-penetration test: (a) wide time range, (b) initial time range. Triangles, square, and circlesindicate temperature of cell surface, nail, and inner case, respectively. The nail delivery and nail-penetration results (video 4) are shown in the figure. The numbers are number of times on the nail delivery. Filled circles are presented nail reach the layer from cross section. Filled triangle is presented the penetration of the layer by nail from cross section. The results of external observation and the rebel of high-speed X-ray transmission images are also shown. The transmission image was selected from the moving image in Fig. 6.

short circuit. The electrolyte was moderately heated at first, but after 32 s, it boiled and electrodes were moved by gas generation. The electrode was punctured by the nail as a result of its own movement. As a result, the contact resistance between the electrode and nail surface decreased, and the current rapidly increased. During the 5 s when this occurred, of the state of the electrode changed dramatically and the cell state exhibited thermal runaway. Supplementary video related to this article can be found at http:// dx.doi.org/10.1016/j.jpowsour.2018.04.092. These results represent the first direct observation of the internal state of thermal runaway in an LIB during nail-penetration test. The behaviour of the electrode during the initial stage was complexly changed. As a result, detailed measurement of the internal state of the cell is very important to analysis of the internal short-circuit test results as well as the thermal runaway phenomenon.

73

Journal of Power Sources 393 (2018) 67–74

T. Yokoshima et al.

4. Conclusion

applications for Li secondary batteries, Energy Environ. Sci. 4 (2011) 1986–2002. [8] M. Jacoby, Assessing the safety of lithium-ion batteries, Chem. Eng. News 91 (2013) 33–37. [9] Secondary Cells and Batteries Containing Alkaline or Other Non-acid Electrolytes Safety Requirements for Portable Sealed Secondary Lithium Cells, and for Batteries Made from Them, for Use in Portable Applications - Part 2: Lithium Systems, IEC 62133–2:2017, (2017-02-07) 2017-02-07. [10] Safety Tests for Portable Lithium Ion Secondary Cells and Batteries for Use in Portable Electronic Applications, JIS C 8714:2007, (2007-11-12) 2007-11-12. [11] Secondary Lithium Cells and Batteries for Use in Industrial Applications - Part 2: Tests and Requirements of Safety, JIS C 8715–2:2012, (2012-07-20) 2012-07-20. [12] K. Chiu, C. Lin, S. Yeh, Y. Lin, K. Chen, An electrochemical modeling of lithium-ion battery nail penetration, J. Power Sources 251 (2014) 254–263. [13] R. Zhao, J. Liu, J. Gu, Simulation and experimental study on lithium ion battery short circuit, Appl. Energy 173 (2016) 29–39. [14] B. Liu, S. Yin, J. Xu, Integrated computation model of lithium-ion battery subject to nail penetration, Appl. Energy 183 (2016) 278–289. [15] R. Zhao, J. Liu, J. Gu, A comprehensive study on Li-ion battery nail penetrations and the possible solutions, Energy 123 (2017) 392–401. [16] H. Maleki, J.N. Howard, Internal short circuit in Li-ion cells, J. Power Sources 191 (2009) 568–574. [17] T.D. Hatchard, S. Trussler, J.R. Dahn, Building a “smart nail” for penetration tests on Li-ion cells, J. Power Sources 247 (2014) 821–823. [18] S. Wilke, B. Schweitzer, S. Khateeb, S. Al-Hallaj, Preventing thermal runaway propagation in lithium ion battery packs using a phase change composite material: an experimental study, J. Power Sources 340 (2017) 51–59. [19] M. Ichimura, The safety characteristics of lithium-ion batteries for mobile phones and the nail penetration test, 29th Telecommunications Energy Conference, 2007, pp. 687–692. [20] Y. Shi, D.J. Noelle, M. Wang, A.V. Le, H. Yoon, M. Zhang, Y.S. Meng, J. Fan, D. Wu, Y. Qiao, Mitigating thermal runaway of lithium-ion battery through electrolyte displacement, Appl. Phys. Lett. 110 (2017) 063902. [21] W. Zhao, G. Luo, C.-Y. Wang, Modeling nail penetration process in large-format LiIon cells, J. Electrochem. Soc. 162 (2015) A207–A217. [22] W. Zhao, G. Luo, C.-Y. Wang, Modeling internal shorting process in large-format LiIon cells, J. Electrochem. Soc. 162 (2015) A1352–A1364. [23] P. Vyroubal, T. Kazda, R. Bayer, FEM model of the lithium ion battery nail test, ECS Trans. 74 (2016) 71–75. [24] X. Feng, J. Sun, M. Ouyang, F. Wang, X. He, L. Lu, H. Peng, Characterization of penetration induced thermal runaway propagation process within a large format lithium ion battery module, J. Power Sources 275 (2015) 261–273. [25] B. Lei, W. Zhao, C. Ziebert, N. Uhlmann, M. Rohde, H.J. Seifert, Experimental analysis of thermal runaway in 18650 cylindrical Li-Ion cells using an accelerating rate calorimeter, Batteries 3 (2017) 14. [26] C.F. Lopez, J.A. Jeevarajan, P.P. Mukherjeea, Experimental analysis of thermal runaway and propagation in lithium-ion battery modules, J. Electrochem. Soc. 162 (2015) A1905–A1915. [27] C. Lee, S. Lee, M. Tang, P. Chen, In situ monitoring of temperature inside lithiumion batteries by flexible micro temperature sensors, Sensors 11 (2011) 9942–9950. [28] Q. Wang, P. Ping, X. Zhao, G. Chu, J. Sun, C. Chen, Thermal runaway caused fire and explosion of lithium ion battery, J. Power Sources 208 (2012) 210–224. [29] T.M. Bandhauer, S. Garimell, T.F. Fuller, A critical review of thermal issues in lithium-ion batteries, J. Electrochem. Soc. 158 (2011) R1–R25. [30] P.T. Coman, E.C. Darcy, C.T. Veje, R.E. White, Numerical analysis of heat propagation in a battery pack using a novel technology for triggering thermal runaway, Appl. Energy 203 (2017) 189–200. [31] X. Feng, L. Lu, M. Ouyang, J. Li, X. He, A 3D thermal runaway propagation model for a large format lithium ion battery module, Energy 115 (2016) 194–208. [32] D.P. Finegan, M. Scheme, J.B. Robinson, B. Tjaden, I. Hunt, T.J. Mason, J. Millichamp, M.D. Michiel, G.J. Offer, G. Hinds, D.J.L. Brett, P.R. Shearing, Inoperando high-speed tomography of lithium-ion batteries during thermal runaway, Nature Comm. 6 (2015) 6924. [33] I. Uchida, H. Ishikawa, M. Mohamedi, M. Umeda, AC-impedance measurements during thermal runaway process in several lithium/polymer batteries, J. Power Sources 119–121 (2003) 821–825. [34] M. Tochihara, H. Nara, D. Mukoyama, T. Yokoshima, T. Momma, T. Osaka, Liquid chromatography-quadruple time of flight mass spectrometry analysis of products in degraded lithium-ion batteries, J. Electrochem. Soc. 162 (2015) A2008–A2015.

We developed an internal short-circuit test system of LIBs that uses high-speed and high-precision X-ray inspection. Using this system, we were able to observe, for the first time, the internal state of the LIB during the safety test. This system allowed us to capture transmission moving images of the anode and cathode in the layered LIB structure. During the nail-penetration test, we were able to directly and clearly observe changes in the nail tip, boiling of the electrolyte, gas generation, changes in the electrode, changes in the interlayer distance, formation of a hole, and other details. Moreover, the beginning of the thermal runaway was observed for the first time. By comparing the behaviour of the internal state with conventional real-time data during the test, we investigated the mechanism of the changes. These analysis results of thermal runaway represent an important step towards the prevention of battery fires. The system also offers an important step toward the visualization of safety tests and analysis of the thermal runway phenomenon. This system is expected to lead to great developments in the safety of LIBs. Conflicts of interest There are no conflicts of interest to declare. Acknowledgements We wish to acknowledge Mr. Shinya Suzuki, Mr. Norihiro Togashi, Dr. Masami Tomizawa, and Mr. Toshinori Uchida of Toshiba IT & Control Systems Corporation for the development of the iSC Scanner and its technical operation. We also wish to acknowledge Ms. Miwako Mizuno of Toshiba IT & Control Systems Corporation for measurements of the CT images. Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx. doi.org/10.1016/j.jpowsour.2018.04.092. References [1] JTSB, Aircraft Serious Incident Investigation Report, All Nippon Airways Co., LTD JA804A, 2014 Report No. AI2014–4Japan Transport Safety Board. [2] NTSB, Aircraft Incident Report, Auxiliary Power Unit Battery Fire. Japan Airlines Boeing 787-8, JA829J, Report No. PB2014-108867 National Transport Safety Board, 2014. [3] Press Release: Prevention of Accidents Involving Batteries, National Institute of Technology and Evaluation, Product Safety Technology Center, July 19, 2012, http://www.nite.go.jp/en/jiko/chuikanki/press/2012fy/12071901.html. [4] Lithium Battery Incidents Since 1991 Reported to the FAA, Federal Aviation Administration, (5/22/2017) https://www.faa.gov/about/office_org/headquarters_ offices/ash/ash_programs/hazmat/aircarrier_info/media/Battery_incident_chart. pdf. [5] P.G. Balakrishnan, R. Ramesh, T.P. Kumar, Safety mechanisms in lithium-ion batteries, J. Power Sources 155 (2006) 401–414. [6] D. Doughty, P. Roth, A general discussion of Li ion battery safety, Electrochem. Soc. Interface 21 (2012) 37–44. [7] G. Jeong, Y.-U. Kim, H. Kim, Y.-J. Kim, H.-J. Sohn, Prospective materials and

74