Towards an understanding of the role of hyper-branched oligomers coated on cathodes, in the safety mechanism of lithium-ion batteries

Abstract

Self-terminated hyper-branched oligomers (STOBA) were coated and then melted on a Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode to form a dense polymer film at high temperatures. The physical and structural changes of the polymer layer at different temperatures and charge conditions were investigated by nitrogen adsorption–desorption, X-ray photoelectron spectroscopy, resistance measurements, scanning electron microscopy, and solid-state ⁷Li-NMR and ¹³C-NMR spectroscopy in order to improve the understanding of the role of the STOBA layer in the enhancement of the safety mechanism of lithium ion batteries. The morphological change of the STOBA layer from the porous to nonporous state at the temperature of a thermal runaway of a battery was demonstrated. The change in the resistance values at high temperatures revealed that the STOBA coating is helpful for the prevention of internal short-circuiting and thermal runaway. Most importantly, the ⁷Li-NMR results acquired at a very high spinning speed (50 kHz) allow the monitoring of the subtle changes in the local environments of the Li⁺ ions and their interaction and mobility in the STOBA–cathode interface as functions of temperature and charge states. The combined characterization results improve the understanding of how the STOBA layer can contribute to the safety features of lithium ion batteries.

---

Introduction

The use of fossil fuels raises numerous environmental and economic concerns, and increases the demand for clean and renewable energy systems, such as solar cells, fuel cells, batteries, and wind power generators.^1–4 Automobiles are one of the major sources of greenhouse gases because of the fossil fuels they consume. Therefore, the transition to an electrified transportation system should be a societal goal to curb emissions, and rechargeable battery systems may provide a feasible route to achieve this objective. Various battery systems have been developed and commercialized over the past few decades.^5–7 Among them, the most successful rechargeable battery technology is the lithium-ion battery (LIB), which is widely used in most of today's electronic equipment and devices, and currently in electric vehicles (EVs) or hybrid electric vehicles.^8–10 However, the increasing use of these batteries raises safety concerns, as several cases of fire-related accidents caused by LIBs (or packs of them) have been reported that occurred in personal electronic devices and EVs. Therefore, increasing the safety of LIBs is of paramount importance.

The safety of LIBs may be compromised under different abuse conditions, such as mechanical abuse behaviors (nail penetration, dropping, crushing, vibration, and so on), electrochemical abuse behaviors (short circuits, overcharge, over-discharge, gas generation, and so on), and thermal abuse (external heating, flame attack, hot combustion gases, and so on).^11,12 Mechanical exploitation may cause immediate failure of the battery leading to thermal runaway. Electrochemical abuse conditions such as overcharge/overdischarge or short circuits will result in the formation of dendritic lithium on the anode, dissolution of the current collector, decomposition of electrolytes with gas and heat generation, and finally, thermal runaway of the battery. Furthermore, batteries that are subject to external heating may undergo thermal runaway more rapidly.

To alleviate the safety problems of LIBs, many protection techniques and methods have been used externally or internally. The external protection mechanisms that are employed include current interruptive devices (CID), positive temperature coefficient (PTC) devices, current limiting fuses, and diodes (blocking/bypass) to minimize potential hazards from batteries under normal operating conditions.^13–15 CIDs break the internal electrical connection when the internal pressure reaches a set value. PTC disks are normally put in the cell header to limit high currents. PTC elements display a large increase in resistance upon a rapid temperature rise and block the flow of current at the battery terminal. Current limiting fuses can be used when a continuous discharge is not favored. However, it is still not possible to make the LIB 100% safe because of the occasional failures of the external hardware systems that occur mostly in abnormal conditions. Furthermore, external devices result in a system that is complex, and they also add weight, volume, and extra cost to the battery pack, which are not favorable for batteries with high energy or power densities. In addition, the external devices cannot quickly respond to complex internal chemistry changes such as increase in resistance, interfacial phenomena, and corrosion of current collectors. Thus, the development of a reliable internal protection mechanism is highly desirable to improve the safety of LIBs. Presently, improvements to the internal protection mechanisms are mainly focused on the individual components, including cathode and anode materials, separators, current collectors, and electrolytes to make the battery system hazard proof.^16–20 The redox shuttle is one of the efficient internal protection methods that can be used to prevent overcharge abuse in a battery.^21–23 In some cases, shutdown separators,^24–26 flame retardant electrolyte additives,^27,28 thermally stable electrode materials,^29,30 and thermoresponsive microspheres deposited onto anodes or separators³¹ are used as internal protection mechanisms to improve battery safety.

Among the different abuse conditions, thermal runaway has received much attention because of its harmful effect on LIB applications.^32,33 Thermal overheating, internal short-circuits and over-charging lead to thermal runaway and are common safety concerns, especially among heavy-duty and high power applications. The thermal runaway process is initiated by the breakdown of the solid electrolyte interphase layer, which leads to a large electrolyte reduction at the surface of the lithiated graphite anode. The cathode materials release oxygen and trigger the interfacial oxidation of the electrolytes above 180 °C, leading to a high rate thermal runaway with a steep temperature rise, which could reach several hundred degrees within a few seconds.

In an attempt to solve the thermal runaway problem, the Industrial Technology Research Institute (ITRI) in Taiwan has successfully developed a self-terminated oligomer with hyper-branched architecture (STOBA) that can be coated on cathode materials to suppress the risk of thermal runaway.^34–36 As a new material, however, the properties of STOBA have not been fully studied. Lin et al. have shown that the STOBA coating dramatically reduced thermal runaway without much effect on the electrochemical performance of LIBs.³⁵ The nail penetration test clearly showed that the STOBA coating suppressed the temperature from rising sharply. The temperature of a cell assembled with a bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode without the STOBA coating was raised over 700 °C within 2 seconds.³⁵ Furthermore, rapid heat generation in a cell may be caused by accidental internal or external short-circuits. However, a detailed analysis has not been carried out to understand how the STOBA layer prevents short-circuits, and thus, the thermal runaway in a battery. In this study, the STOBA–cathode interface at the fully charged state and the structural changes of the STOBA layer at different temperatures were analyzed in order to improve the understanding of its safety mechanism.

Experimental section

Synthesis of the STOBA-coated cathode and anode materials

The synthesis of STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ materials was carried out according to the proprietary processes developed by ITRI³⁴ and also that described by Lin et al.³⁵ Briefly, STOBA was dispersed in N-methyl-2-pyrrolidone (NMP, Aldrich), keeping a STOBA weight fraction of 2.0% in the mixture. Then, Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles (Shenzhen Tianjiao Technology Co., Ltd) were added to the STOBA solution and stirred at 70 °C for 60 min. Afterwards, a NMP solution containing polyvinylidene fluoride (PVdF, T1300, Kureha America Inc.) was added to the STOBA/Li(Ni_0.4Co_0.2Mn_0.4)O₂ solution, which was stirred for an additional 30 min. Finally, conductive agents including TIMREX® KS6 graphite (Timcal) and carbon black (Super P®, Timcal) were added to the mixture, which was stirred for another 30 min. The final composition of the cathode material was 89 wt% Li(Ni_0.4Co_0.2Mn_0.4)O₂ (together with 2 wt% STOBA for the STOBA-coated sample), 4 wt% PVdF, and 7 wt% KS6 and Super P. The resulting slurry was blade-coated on aluminum foil and dried overnight in an oven to obtain the active cathode material. To avoid any aggregation of STOBA during the coating process, the temperature of the solution was maintained at 70 °C and a minimum amount of STOBA was used. Although the process results in a satisfactory layer of STOBA that is coated over the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles, the layer thickness varies slightly in different regions of the particles. The anode material MesoCarbon MicroBeads (MCMB) was purchased from China Steel Chemical Corp., Taiwan. The anode active layer consisted of 95 wt% MCMB, 1 wt% Super P carbon, and 4 wt% PVdF binder on a Cu current collector. LiPF₆ (1 M) in ethylene carbonate (EC)/ethyl methyl carbonate (EMC) (1 : 2 vol%) (Zhangjiagang Guotai-Huarong) with 2 wt% vinylene carbonate (VC) was used as the electrolyte.

Assembling and disassembling of the battery

A pouch-type LIB was fabricated using a STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode, MCMB anode, and 1 M LiPF₆ in EC/EMC with 2 wt% VC as the electrolyte. A porous polyethylene film was used as a separator. The battery was then charged to 100% state of charge (SOC) at 4.2 V under a constant current of 0.1 C. The design capacity of the pouch cell was 8 mA h. Then, the battery was disassembled in a glove box under an Ar atmosphere. The recovered STOBA coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ electrode with the electrolyte was sealed in two high pressure Ti tubes and thermally treated separately at 150 °C and 180 °C for 30 min. The thermally treated electrode samples were stored in the glove box for further characterizations.

Characterizations

N₂ adsorption–desorption isotherms were measured at 77 K on a Micromeritics ASAP 2020 analyzer. The samples were degassed for 24 h before the measurements. The pore size was determined from an analysis of the adsorption branch of the isotherm using the Barrett–Joyner–Halenda method. Pore volumes were obtained from the volumes of N₂ adsorbed at or near P/P₀ = 0.95.

Thermogravimetric analysis (TGA) was carried out under a nitrogen environment at a heating rate of 10 °C min⁻¹ from room temperature to 850 °C using a PerkinElmer TGA 7 thermogravimetric analyzer.

X-ray photoelectron spectroscopy (XPS) analysis was conducted on PHI 5000 VersaProbe (ULVAC-PHI), which employed a focused X-ray source of 30 W.

Scanning electron micrographs (SEM) of the STOBA sample at different temperatures were obtained using a FEI Nova NanoSEM 230 field emission electron microscope at an accelerating voltage of 10 kV. A secondary electron image of the cross-sectional micrograph of the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ sample was taken using a FEI Nova 200 Nanolab focused ion beam scanning electron microscope (FIB-SEM).

The resistance measurements of the bare and STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ samples were carried out in a four-probe setup using a programmable low-ohm meter DU-5010 (Delta United Instrument Co. Ltd). The recovered sample after disassembling the cell was washed with dimethyl carbonate and dried in a vacuum chamber overnight using a dry pump inside a glove box. The resistance of the dried sample was measured in a dry room with a dewpoint of −40 °C.

Solid-state ⁷Li-NMR and ¹³C-NMR experiments were performed on a Varian Infinity Plus 500 NMR spectrometer equipped with Chemagnetics 1.2 mm (for ⁷Li) and 5 mm (for ¹³C) magic angle spinning (MAS) T3 probes. The Larmor frequencies for ⁷Li and ¹³C nuclei are 193.7 and 125.4 MHz, respectively. The ⁷Li MAS spectra were acquired with a high spinning speed of 50 kHz in order to achieve better spectral resolution. The ⁷Li chemical shift was externally referenced to 1 M LiCl solution at 0 ppm. The Hartmann-Hahn matching condition for ¹H → ¹³C cross-polarization (CP) MAS experiments was determined using admantane. The ¹³C chemical shift was externally referenced to tetramethylsilane at 0 ppm.

Results and discussion

Pore size distribution and stability of STOBA

Nitrogen adsorption–desorption measurements were carried out to investigate the changes in the pore structure of the STOBA materials under thermal treatment at different temperatures. The pore size distribution of the STOBA samples at different temperatures is shown in Fig. 1, which illustrates that the STOBA material without thermal treatment possesses a bimodal pore structure with pore sizes of 1.9 and 2.4 nm. The STOBA sample after thermal treatment at 150 °C also shows a similar bimodal pore structure. Interestingly, when the sample was thermally treated at 180 °C, no significant pores in the sample were observed. The thermal runaway of a LIB takes place above 180 °C because of the exothermic reaction of cathode materials and release of O₂, which triggers interfacial oxidation of the electrolytes leading to an accelerated increase in temperature.^11,37 It was found that the porosity of STOBA nearly disappeared at approximately 180 °C as revealed by the nitrogen adsorption–desorption measurements, which may indicate that the STOBA chains were crosslinked to each other and then melted at high temperatures. As a result, STOBA became nonporous in nature after thermal treatment at 180 °C. The nonporous state of STOBA may restrict the direct contact of the electrolyte with oxygen and prevent the cathode–electrolyte reaction, and thus is helpful for the suppression of the thermal runaway.

Fig. 1 Pore size distribution curves of (a) pure STOBA, and STOBA thermally treated at (b) 150 °C and (c) 180 °C, obtained from nitrogen adsorption–desorption measurements.

TGA was carried out to measure the thermal stability of the STOBA material. The TGA curve of STOBA is shown in Fig. S1 (ESI†). Initially, approximately 3% weight loss was observed up to 300 °C because of the small amount of physisorbed water in the STOBA sample that was gained while transferring it to the instrument. After 350 °C, the STOBA sample gradually lost weight because of the decomposition of the polymeric unit. At 850 °C, the STOBA sample retained approximately 48 wt% even though some polymeric units decomposed. The TGA result reveals that the STOBA material is thermally stable up to 300 °C, which is beneficial for preventing any heat-related activity in batteries.

XPS analysis of the STOBA-coated cathode

The XPS study was carried out with fresh samples that did not undergo any charge–discharge testing. Fig. 2 displays the XPS spectra of the bare and STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode samples. The spectrum of the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ sample shows a strong N 1s peak at the binding energy of 400.2 eV which is because of the nitrogen attached with the fundamental building-block of STOBA.³⁵ The O 1s spectra are composed of two components at 529.6 and 531.6 eV. The main peak at 529.6 eV in the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ sample is from the characteristic O²⁻ ions of the crystalline lattice. The second peak at 531.6 eV is attributed to the oxygen-containing species of the surface layers that possess coordination deficiencies, i.e., the defects in the surface associated with sites where the coordination number of oxygen ions is smaller than in the regular sites.³⁸ The peak intensity at 531.6 eV increased and that at 529.6 eV decreased when STOBA was coated on the Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode. This may be because of the increase in the surface activity, as the O 1s peaks of STOBA are also expected in this binding energy range. The Co 2p spectrum shows a well defined profile with the 2p_3/2 and 2p_1/2 components at binding energies of 780.4 and 795.4 eV, respectively. The main binding energy components and small satellite peak indicate that cobalt ions are in the 3+ oxidation state in an oxygen environment.^37,39

Fig. 2 XPS spectra of the (a) STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode and (b) bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode with core peaks.

The Ni spectrum is characterized by an intense and complicated satellite structure. The Ni spectrum exhibits two peaks from 2p_3/2 and 2p_1/2 at 854.4 and 872.6 eV, respectively. The low binding energy peak is in agreement with the formation of NiO, which confirms the presence of Ni²⁺ in the compound.³⁹ The satellite peaks have energies that are approximately 6 eV higher than the main 2p_3/2–1/2 doublet. The wide peak centered at 854.4 eV may be because of the superposition of Ni²⁺ and Ni³⁺ signals. The Mn 2p spectrum depicts two distinct peaks related to Mn 2p_1/2 and Mn 2p_3/2 at binding energies of 653.6 and 642.4 eV, respectively, for the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ sample.

The two peaks of the Mn 2p_3/2–1/2 doublet are assigned to the Mn⁴⁺ ions. The intensities of the Co, Ni, and Mn peaks are reduced effectively when STOBA was coated on the Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode, suggesting the deposition of the STOBA layer on the surface of the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles (Fig. 2a). A strong F peak at 687.6 eV was observed because of the presence of PVdF in the cathode material, which acts as a binder.³⁸ The intensity of the F peak was also reduced after STOBA was coated on the surface of the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles.

To obtain information regarding the uniformity of the STOBA layer, the XPS study was also carried out to measure the intensity at three different points in the STOBA-coated cathode sample, with each point approximately 1 cm apart from the other. The XPS spectra of the core peaks of the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode sample at three different points are shown in Fig. S2 (ESI†). As shown in Fig. S2,† the peak intensities of the core elements at three different points are nearly the same, which suggests that the STOBA layer is coated in a relatively uniform manner on the surface of the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles because of the ability of XPS to explore the surface layer of the sample by a few nanometers in depth.

Resistance measurements

To understand how the electrochemical behavior of the STOBA layer changes after thermal treatment at high temperatures, the resistances of the bare and STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathodes were measured from room temperature to 180 °C. As seen in Fig. 3, both samples exhibited nearly the same resistances at room temperature. The very low resistance value (approximately 0.5 mΩ) suggests good conduction of the samples. The resistances of both samples slowly increase with an increase in temperature, exhibiting a small difference in resistance values for the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode (152 mΩ) and the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode (185 mΩ) at 150 °C. Above 150 °C, however, the resistance of the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode sample increased sharply to 3.5 Ω in comparison to 1.2 Ω for the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode at 180 °C. These resistances of the STOBA-coated and the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode samples were 7000 and 2400 times higher, respectively, than the resistances at room temperature. The thermal runaway of a battery generally occurs in the temperature range of 150–180 °C, and therefore, only two temperatures of 150 °C and 180 °C were chosen, in addition to room temperature, for evaluating the resistances. Furthermore, the difference in the resistance values was not high enough between room temperature and 150 °C to significantly impact the current flow. It suggests that the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode material generates oxygen and triggers interfacial oxidation at approximately 180 °C, leading to an increase in the resistance. In the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode sample, however, the STOBA material melts at approximately 180 °C and forms a dense nonporous film on the cathode surface to break the conductive network, leading to a steep rise in resistance that is almost three times higher than that of the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode. This causes a drop in the current, which limits the heat generation in the cell and prevents thermal runaway. Furthermore, the STOBA coating prevents an internal short-circuit of the battery, which is also a crucial safety issue. As the resistance of the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode is not very high at 180 °C, the cell with the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode may be more prone to cause a short-circuit.

Fig. 3 Resistance measurements at different temperatures for the (a) STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode and (b) the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode.

Morphological study

The morphologies of the STOBA samples at different temperatures and the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ sample are shown in Fig. 4. The STOBA material without thermal treatment exhibited a particle morphology of fibrous porous structures (Fig. 4a). After thermal treatment at 150 °C, the fibrous STOBA particles started to melt, as is clearly observed in Fig. 4b. When the STOBA sample was thermally treated at 180 °C, the particles were completely melted (Fig. 4c). Therefore, it can be inferred that the STOBA material without thermal treatment is porous enough to transport Li-ions freely, and thus, does not significantly affect the capacity.³⁵ When the STOBA sample was thermally treated at 150 °C, however, the fibrous particle structure started to melt. As a result, the pore volume of the STOBA layer was slowly reduced, which is consistent with the observations revealed by the nitrogen adsorption–desorption measurements. In the case of the STOBA sample after thermal treatment at 180 °C, the fibrous STOBA particles have melted completely and the porous structure disappeared, suggesting the complete blockage of movement of lithium ions and electrons while it is used as a coating layer on the cathode of Li-ion batteries. Fig. 4d depicts the FIB-SEM image of the cross-section of a STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ particle, and the enlarged version of its edge is shown in Fig. 4e. The STOBA layer on the surface of the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particle is clearly visible, with an average layer thickness of 30 nm.

Fig. 4 SEM images of STOBA material (a) without thermal treatment and with thermal treatment at (b) 150 °C and (c) 180 °C, and (d) cross-section of a STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ particle. The marked region of (d) is shown as an enlarged image in (e) illustrating the STOBA coating thickness on the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particle.

Based on the STOBA layer thickness and Li(Ni_0.4Co_0.2Mn_0.4)O₂ particle size, the minimum amount of STOBA sample required to coat the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles has been estimated. A detailed estimation is given in the ESI.† It was found that approximately 0.8 wt% of STOBA sample is sufficient to properly coat the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles. A histogram of the particle numbers versus the particle size is given in Fig. S3 (ESI†) with an inset SEM image of the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles. It was observed that the majority of the particles are 5 to 10 nm in size with a nearly spherical shape. The minimum amount of STOBA was estimated using an ideal spherical cathode particle, and therefore, there may be possible errors because of the different size and shape (near spherical) of the particles as well as the varying thickness of the STOBA coating layer (Fig. 4e). However, the amount of STOBA used (2 wt%) in the present study was more than twice the calculated amount (0.8 wt%), which should be sufficient to coat the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles efficiently after compensating for possible errors. This established that the Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles were completely covered with the STOBA layer.

Generally, thermal runaway of standard commercial LIBs begins at approximately 180 °C. However, the STOBA particles completely melt at that temperature and form a dense conformal polymer film on the surface of the Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode.³¹ The molten STOBA particles are expected to wet the interface and provide an ion-insulating barrier by restricting the charge transfer between the cathode and the electrolyte. Therefore, there cannot be direct electrode–electrolyte reaction, and O₂ will be barred from direct contact with the electrolyte, thus preventing the cell from explosion at and above 180 °C. Unlike some commercial shutdown separators that shrink and risk the shorting of the electrode, the melting and formation of conformal film on the cathode surface by the STOBA material prevents any such types of battery-related accidents caused by the thermal runaway. The SEM results are also supported by the observation found in nitrogen adsorption–desorption analysis.

⁷Li MAS NMR analysis of the STOBA-coated cathode

⁷Li MAS NMR measurements were carried out to study the Li⁺ interaction with the STOBA-coated cathode. Fig. 5 shows the ⁷Li MAS NMR spectra of the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode and the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathodes without and with thermal treatment at 150 and 180 °C. All of the samples were in the 100% SOC state before the measurements. The peak observed at approximately −0.1 ppm is because of the interaction of Li⁺ with the oxygen atoms of the Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode. A close examination of these ⁷Li-NMR spectra revealed that there was an additional small peak observed at 136 ppm for the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode. The inset of Fig. 5 shows the enlarged version of the peak at 136 ppm. The interaction of Li⁺ ions with the N and O of the STOBA material gives the weak peak at 136 ppm. As the amount of STOBA is only 2.0% in comparison to the amount of Li(Ni_0.4Co_0.2Mn_0.4)O₂, the intensity of this peak is very weak. The intensity of the peak becomes weaker after the STOBA-coated sample was thermally treated at 150 °C for 30 min, and it completely disappeared after thermal treatment at 180 °C.

Fig. 5 ⁷Li MAS NMR spectra at 100% SOC condition of (a) bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode, (b) STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode, and thermally treated (c) STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode at 150 °C/30 min and (d) STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode at 180 °C/30 min. The spinning sidebands are marked with asterisks. The inset shows the enlarged version of the 136 ppm peak.

This phenomenon is illustrated in Scheme 1. At room temperature, the STOBA layer was porous enough to allow the Li⁺ ions to pass freely at the time of the charge–discharge cycles. When the sample was thermally treated at 150 °C, the STOBA layer started to melt and slowly blocked the pores and reduced the pore volumes as confirmed from Fig. 1 and 4. Therefore, the interaction of Li⁺ with N and O atoms also decreases because of the presence of less lithium ions in the STOBA layer, and thus a weaker peak is presented at 136 ppm. Finally, when the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ sample was thermally treated at 180 °C for 30 min, the STOBA layer completely melted with no significant pores in the material, and thus blocked the charge transfer through the layer. The crosslinking and melting of the STOBA layer that resulted from the thermal treatment not only stops the Li⁺ ion transfer but also suppresses the moving of electrons. In such a situation, Li⁺ ions and electrons can possibly reside inside the Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode, as illustrated in the right part of Scheme 1.

Scheme 1 Schematic diagram of the change in morphology of the STOBA layer in the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode after thermal treatment.

Another noticeable change in the ⁷Li MAS NMR spectra is the change in the linewidths of the main peak at −0.1 ppm. The ⁷Li MAS NMR linewidth of the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode at 100% SOC possesses 0.62 kHz. The linewidth increases to 2.93 kHz in the 2 wt% STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode at 100% SOC. This suggests that the Li⁺ ions are more mobile in the bare Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode than the STOBA-coated cathode. It has been reported that the discharge capacity decreases slightly with the STOBA-coated cathode as compared to the bare cathode.³⁵ These observations imply that the STOBA coating slightly influences the lithium ion mobility and may affect the overall capacity of the battery. The linewidths again narrow down to 2.5 and 0.5 kHz for the STOBA-coated cathode samples treated at 150 °C and 180 °C, respectively. When the STOBA material melts at these temperatures, the lithium ions would be extruded from this layer and confined within the cathode. As a result, the lithium ions may move more freely inside this layered cathode structure, which resulted in narrow linewidths.

Solid-state ¹³C CPMAS NMR analysis of STOBA

Solid-state ¹³C-NMR measurements were employed to investigate the backbone structure of the STOBA material. Fig. 6 shows the ¹³C CP MAS NMR spectra of pure STOBA material and the STOBA samples thermally treated at 150 °C and 180 °C. The peak at 179 ppm was assigned to the carbon adjacent to the oxygen of the modified maleimide unit, while the peak at 143 ppm was assigned to the carbon from the benzene ring of STOBA (Scheme 1). The carbon of the benzene ring that was attached to the nitrogen of the modified maleimide unit was observed at 131 ppm. The peak at 42 ppm was because of the methylene carbon between the two benzene rings. The peak at 31 ppm was ascribed to the quaternary carbon attached to the methyl group of the modified maleimide unit, and the peak at 19 ppm was because of the methyl carbon attached to the modified maleimide unit. In the ¹³C CP MAS NMR spectra, a characteristic change in the peak intensity was observed for the peaks at 31 and 19 ppm. In the pure STOBA sample without thermal treatment, the peak at 31 ppm appeared as a shoulder only, while the intensity of the peak at 19 ppm was very low. When the sample was thermally treated at 150 °C for 30 min, however, the corresponding peaks became more prominent. The peak intensity was increased further, and finally, sharp peaks were observed for the sample thermally treated at 180 °C for 30 min. These changes suggested that the STOBA chains were crosslinked with each other when the temperature was increased, leading to the increase in the degree of crosslinking. As the STOBA material melted and the degree of crosslinking was increased at a higher temperature, the chain mobility was reduced and became more rigid in favor of ¹H to ¹³C dipolar transfer via CP to increase the intensity of the peaks at 31 and 19 ppm. The ¹³C CP MAS NMR results are also consistent with the phenomenon described in Scheme 1.

Fig. 6 ¹³C CP MAS NMR spectra of (a) pure STOBA and the STOBA material thermally treated at (b) 150 °C and (c) 180 °C acquired at a spinning speed of 9 kHz. The spinning sidebands are marked with asterisks.

Conclusions

The self-terminated oligomers with hyper-branched architecture (STOBA) coated on the Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode material were studied at different temperatures to gain a deeper insight into the behavioral changes of the coating layer and to explore their possible role in suppressing the thermal runaway problem in LIBs. The nitrogen adsorption–desorption analysis and SEM confirmed the morphology change of the STOBA layer from a porous to an almost nonporous layer with the rise in temperature. The XPS analysis indicated mainly 2+, 3+, and 4+ oxidation states for nickel, cobalt, and manganese, respectively, in the Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode material, and confirmed the deposition of STOBA material on the Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode surface, as the intensities of the corresponding peaks were drastically reduced after STOBA coating. The change in resistances at high temperatures gives a better understanding of the role of STOBA in preventing thermal runaway and short-circuits. The very high spinning ⁷Li MAS NMR spectra not only demonstrated how the local environments of the lithium ions vary at different temperatures, but also revealed the possible phase change in the STOBA–cathode interface layer that can help to suppress the thermal runaway process of Li-ion batteries. The ¹³C CP MAS NMR results confirmed the increased rigidity of the STOBA chains at a higher temperature because of melting and enhanced crosslinking of the carbon chains. The present work allowed a deeper insight into the role of the STOBA-coated cathode materials, which can be used to improve the safety of LIBs.

Acknowledgements

The authors gratefully acknowledge the financial support from the Ministry of Science and Technology of Taiwan under the grant number NSC 102-2113-M-008-006-MY3. One of the authors, H.-M. L., acknowledges the Ph.D. scholarship (2012-C1-2885) from the Industrial Technology Research Institute of Taiwan.

Notes and references

1. S. Zhang, X. Yang, Y. Numata and L. Han, Energy Environ. Sci., 2013, 6, 1443.
2. D. Hart, J. Power Sources, 2000, 86, 23.
3. B. Dunn, H. Kamath and J. M. Tarascon, Science, 2011, 334, 928.
4. M. Z. Jacobson and C. L. Archer, Proc. Natl. Acad. Sci. U. S. A., 2012, 109, 15679.
5. F. Y. Cheng, J. Liang, Z. L. Tao and J. Chen, Adv. Mater., 2011, 23, 1695.
6. X. L. Ji, S. Evers, R. Black and L. F. Nazar, Nat. Commun., 2011, 2, 325.
7. A. S. Arico, P. Bruce, B. Scrosati, J. M. Tarascon and W. Van Schalkwijk, Nat. Mater., 2005, 4, 366.
8. J. M. Tarascon and M. Armand, Nature, 2001, 414, 359.
9. M. S. Whittingham, Chem. Rev., 2004, 104, 4271.
10. V. Etacheri, R. Marom, R. Elazari, G. Salitra and D. Aurbach, Energy Environ. Sci., 2011, 4, 3243.
11. J. Wen, Y. Yu and C. Chen, Mater. Express, 2012, 2, 197.
12. P. G. Balakrishnan, R. Ramesh and T. Prem Kumar, J. Power Sources, 2006, 155, 401.
13. Y. Yamanaka and K. Miki, US Pat. No. 6642694, 2003.
14. S. Nakamura, US Pat. No. 6697259, 2004.
15. D. Doughty and E. P. Roth, Electrochem. Soc. Interface, 2012, 21(2), 37.
16. T. Wei, R. Zeng, Y. Sun, Y. Huang and K. Huang, Chem. Commun., 2014, 50, 1962.
17. H. Wu, G. Yu, L. Pan, N. Liu, M. T. McDowell, Z. Bao and Y. Cui, Nat. Commun., 2013, 4, 1943.
18. G. Venugopal, J. Power Sources, 2001, 101, 231.
19. S.-T. Myung, Y. Hitoshia and Y.-K. Sun, J. Mater. Chem., 2011, 21, 9891.
20. S.-H. Kim, K.-H. Choi, S.-J. Cho, E.-H. Kil and S.-Y. Lee, J. Mater. Chem. A, 2013, 1, 4949.
21. L. Zhang, Z. Zhang, P. C. Redfern, L. A. Curtiss and K. Amine, Energy Environ. Sci., 2012, 5, 8204.
22. L. Zhang, Z. Zhang, H. Wu and K. Amine, Energy Environ. Sci., 2011, 4, 2858.
23. Q. Wang, S. M. Zakeeruddin, I. Exnar and M. Grätzel, Electrochem. Commun., 2008, 10, 651.
24. M. Xiong, H. Tang, Y. Wang and M. Pan, Carbohydr. Polym., 2014, 101, 1140.
25. S. S. Zhang, J. Power Sources, 2007, 164, 351.
26. H. Lee, M. Yanilmaz, O. Toprakci, K. Fu and X. Zhang, Energy Environ. Sci., 2014 10.1039/c4ee01432d.
27. Q. Wang, J. Sun, X. Yao and C. Chen, Electrochem. Solid-State Lett., 2005, 8, A467.
28. H. F. Xiang, Q. Y. Jin, C. H. Chen, X. W. Ge, S. Guo and J. H. Sun, J. Power Sources, 2007, 174, 335.
29. H. Lee, M. G. Kim and J. Cho, Electrochem. Commun., 2007, 9, 149.
30. L. J. Fu, H. Liu, C. Li, Y. P. Wu, E. Rahm, R. Holze and H. Q. Wu, Solid State Sci., 2006, 8, 113.
31. M. Baginska, B. J. Blaiszik, R. J. Merriman, N. R. Sottos, J. S. Moore and S. R. White, Adv. Energy Mater., 2012, 2, 583.
32. Q. Wang, P. Ping, X. Zhao, G. Chu, J. Sun and C. Chen, J. Power Sources, 2012, 208, 210.
33. Y. Shigematsu, S. Kinoshita and M. Ue, J. Electrochem. Soc., 2006, 153, A2166.
34. C.-R. Yang, J.-P. Pan, C.-A. Chen and J.-M. Hsu, US Pat. No. 8137838, 2012.
35. C.-C. Lin, H.-C. Wu, J.-P. Pan, C.-Y. Su, T.-H. Wang, H.-S. Sheu and N.-L. Wu, Electrochim. Acta, 2013, 101, 11.
36. http://www.rdmag.com/RD100-Awards-Self-Terminating-Technology-Holds-Li-Ion-Energy-In-Check/.
37. D. P. Abraham, E. P. Roth, R. Kostecki, K. McCarthy, S. MacLaren and D. H. Doughty, J. Power Sources, 2006, 161, 648.
38. S. Verdier, L. E. Ouatani, R. Dedryvère, F. Bonhomme, P. Biensan and D. Gonbeau, J. Electrochem. Soc., 2007, 154, A1088.
39. X. Bie, L. Liu, H. Ehrenberg, Y. Wei, K. Nikolowski, C. Wang, Y. Ueda, H. Chen, G. Chen and F. Du, RSC Adv., 2012, 2, 9986.

---

Footnote

† Electronic supplementary information (ESI) available: TGA curve of STOBA, XPS spectra of the STOBA-coated Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode with the core peaks at three different points of the sample, histogram of the particle size of the Li(Ni_0.4Co_0.2Mn_0.4)O₂ cathode, estimation of minimum amount of STOBA needed to coat Li(Ni_0.4Co_0.2Mn_0.4)O₂ particles. See DOI: 10.1039/c4ra09220a

---