At 23:00 beijing time on August 22, 2022, the team of Researcher Zhang Wei and Professor Chen Jun of Nankai University, in collaboration with the team of Professor Ma Zifeng and Researcher Li Linsen of Shanghai Jiao Tong University, published an article in the journal Chem entitled “Stabilizing lattice oxygen in slightly Li-enriched nickel oxide cathodes towards high-energy.” batteries” new study.
In this study, the irreversible phase transition and oxygen evolution problem of lithium nickel oxide layered cathode material were solved by introducing a micro-rich lithium strategy, so that it has both high energy density, long cycle stability and thermal stability. The first authors of this paper are Zhou Zheng and Wang Han, and the corresponding authors are researchers Li Linsen and Zhang Wei.
Taking this opportunity, Xiao Ke Chemistry interviewed Zhang Wei, a researcher at Nankai University. He will not only introduce the content of the work, but also talk about the story behind the work.
Q: We know that lithium-ion batteries have been widely used in mobile phones, electric vehicles, daily home appliances and other equipment for daily use. The main step in the development of high-performance lithium-ion batteries is to improve the cathode material. So among so many cathode materials, why would you choose lithium nickel oxide (LiNiO2) cathode material?
Researcher Wei Zhang:
The main development goal of commercial batteries is to increase the energy density of batteries. In simple terms, the battery should be able to store enough energy after a single charge to ensure that appliances such as mobile phones work for a long time. In addition, the safety and low cost of the battery is also an important indicator of the commercialization of the battery, because no one is willing to encounter accidents such as fire or explosion when using the battery, and also hopes to spend less money to enjoy better products. Compared with the current commercial cathode materials, only the layered cathode materials (LiCoO2, LiNi1-x-yCoxMnyO2, LiNi1-x-yCoxAlyO2, etc.) have high energy density. But these materials all contain cobalt, and due to geopolitics and prominent environmental and ethical issues in the co mining process, there is a growing concern about the price of Co. If the development of cobalt-free ultra-high nickel layered cathode materials, not only can reduce costs, but also the high nickel content can further improve the energy density of the battery. Therefore, under the wave of “de-Co-liter Ni”, cobalt-free ultra-high nickel materials (Ni content greater than 90%), such as LiNi1-xMnxO2, LiNi1-xAlxO2 and LiNiO2 (LNO), will become the mainstream choice of the future positive electrode. Unfortunately, the cyclic and thermal stability of cobalt-free ultra-high nickel materials is very poor, which seriously hinders its process towards commercialization.
The reasons for the poor performance of cobalt-free ultra-high nickel materials are mainly related to the structural instability on the surface of the material. First of all, the surface structure will undergo an irreversible phase transition from the layered phase to the rock-salt phase during the charging process. Compared with the layered structure, the lithium storage capacity of the rock salt phase is poor, so its continuous increase is bound to lead to the decay of the cathode capacity. At the same time, the phase transition process is accompanied by oxidation of lattice oxygen, which releases O2 on the surface. The released O2 reacts with the electrolyte oxidation, which will release a large amount of heat, resulting in the battery thermal runaway. The formation of O2 cannot be avoided, because the O2-/Ni3+/4+ energy band overlaps, so when the surface is deeply delithium, O2- must lose electrons. Surface irreversible phase transitions and surface lattice oxygen instability are common to all layered cathode materials. In particular, the problem of O2 precipitation on the surface of layered cathode materials has been an unsolved problem in the battery field since the commercialization of LiCoO2 cathode in 1991.
In view of the above problems, our Nankai University team selected LNO as a research model in 2019 and planned to propose a common solution strategy. LNO was chosen because for cobalt-free ultra-high nickel layered cathodes, the higher the Ni content, the higher the energy density, but the worse the cycle stability and thermal stability, and the LNO has the highest energy density and the worst stability. Therefore, if the surface problem of LNO can be solved, then presumably this strategy can also be applied to other layered cathode materials. On the other hand, the elemental composition of LNO is simple and contains only Ni, so the interference of other elements (Mn, Al, etc.) on the evolution of surface structure can be excluded when analyzing the material structure-activity relationship.
Q: Since you want to make a breakthrough in a problem that has existed for more than 30 years, the process must be very difficult. So how did you come up with a new strategy? What effect can this strategy achieve?
Researcher Wei Zhang:
Methods such as lattice doping and surface coating have been proposed to solve the surface problem of LNO (Figures 1B and 1C), but these methods have sacrificed capacity in exchange for improved cycle performance, and they cannot completely hinder surface O2 release. We note that in a lithium-rich manganese-rich base-like cathode material (about 20% lithium-rich), the vacancy cluster formed in the body phase during charging can lock in oxygen without escaping. However, this oxygen-locking phenomenon only occurs in the bulk phase, and the surface of lithium-rich manganese-based materials still cannot avoid the problem of lattice oxygen instability. So is there a way for us to make the LNO surface also rich in lithium, thereby inhibiting oxygen escape? But we must control the amount of lithium-rich, otherwise the amount of lithium will be too large, so that the surface of the LNO will fall into the same situation as the lithium-rich manganese group. Based on the above thinking, the team of Shanghai Jiao Tong University used Ni(OH)2 as a precursor to synthesize a micro-lithium-rich lithium nickel oxide (Li1.04Ni0.96O2, LR-LNO, lithium-rich 4%) in the lipo-rich Li2SO4-LiOH molten salt reaction environment. This is a new material that has not yet been synthesized, because traditional high-temperature solid-phase sintering methods either obtain lithium-poor lithium nickel oxide (Li1-xNiO2) or obtain lithium nickel oxide (e.g., Li).[Li1/3Ni2/3]O2, spatial group structure C2/m). Neutron powder diffraction results show that LR-LNO has the same R ̄3m layered structure as lean lithium LNO (Figure 1D). Micro excess Li ions in LR-LNO are randomly distributed in the Ni layer (Figure 1E). The high-angle circular darkfield image (HAADF) also confirms that LR-LNO maintains a layered structure consistent with LNO (Figure 1F).
Figure 1: Traditional lithium-poor LNO structure and micro-rich lithium LR-LNO structure
Although the lithium content has increased by only 4%, this small component change gives the LR-LNO a significantly enhanced surface lattice oxygen stability. In situ differential electrochemical mass spectrometry found that lithium-poor LNOs produced large amounts of O2 and CO2 during charging (Figure 2A), while LR-LNO produced only a small amount of CO2 without O2 release (Figure 2B). By comparing the O2 emissions from other high-energy density layered oxide cathode materials (Figure 2C), this micro-rich lithium strategy is clearly more effective than other modification strategies such as doping, surface coating, and core-shell design. In addition, the thermal stability of LR-LNO has also been significantly improved. We performed thermal runaway tests on soft pack batteries in a fully charged state using an accelerated calorimetry (Figure 2D) and found that the thermal runaway temperature of LR-LNO, T2 (253.2 °C), was nearly 80 °C higher than that of the depleted lithium LNO (179.0 °C), comparable to the commercial NCM811 (259.8 °C). This will greatly facilitate the thermal management of lithium nickel oxide materials when applied to battery packs, and also improve the overall battery safety.
Figure 2: O2 release in LR-LNO is sufficiently suppressed and thermal stability is improved
In addition to enhanced thermal stability, the LR-LNO also has excellent cycle stability. Figure 3A shows that at a 0.1 C magnification, the first turn specific capacity of the LR-LNO is as high as 233.7 ± 1.7 mAh g-1, and the coulomb efficiency is as high as 96.2 ± 0.4%, which is comparable to the lean lithium LNO capacity (236.7 ± 1.2 mAh g-1). Compared to other NCM cathodes, LR-LNO provides a higher specific energy (over 900 Wh kg-1), an increase of 20% compared to the NCM811 (inset in Figure 3A). At the same time, LR-LNO also exhibits better high-magnification performance than LNO (Figure 3B). In addition, the LR-LNO also exhibited quite excellent cycle stability, with the half-cell capacity rate maintaining a capacity ratio of up to 92.3% after 100 cycles of charge and discharge at 1 C (Figure 3C), surpassing other doped or coated high nickel cathode materials (Figure 3D). To further evaluate the long-cycle stability of the material under actual operating conditions, we assembled a single-piece soft-pack battery (negative pole graphite) and performed a long-cycle test at 0.5 C after three cycles of activation at 0.1 C (Figure 3E). LNO soft pack batteries show rapid capacity decay (more than 20% capacity loss after 35 laps), which is consistent with the performance of previous LNO full batteries. However, the LR-LNO soft pack battery still has a capacity retention rate of 80% after more than 400 cycles. In addition, we have improved the cyclic stability of Li1.05 (Ni0.95Co0.025Mn0.025) 0.95O2 cathode materials through the lithium-rich strategy (LR-Ni95 in Figure 3C and Figure 3D), demonstrating that this strategy can be applied to other ultra-high nickel layered oxides.
Figure 3: Excellent electrochemical properties of LR-LNO
Q: From the performance test results, the emergence of LR-LNO breaks the traditional view of ultra-high nickel cathode materials – higher Ni content leads to higher energy density, but at the same time, cycling performance and thermal stability performance will also be worse. Why do small changes in chemical composition produce such a huge difference in performance?
Researcher Wei Zhang:
As explained earlier, macroscopic properties are closely related to the structural evolution of the positive surface. Therefore, we want to study the micro-mechanism behind the performance, the most direct way is to directly observe the structural changes on the surface of the material under the working conditions of the battery. But the phase transition region on the surface is only about a few tens of nanometers. It’s not just us, it’s estimated that most scientists who study the positive electrode are eager to know how changes have occurred in such a tiny area of the positive electrode surface when the battery is working. Our Nankai University team published an in situ microcell device in 2020 that can perform constant current charge and discharge within the electron microscope, while using the electron energy loss spectroscopy technology of the electron energy loss spectroscopy technology of the electron microscope to track the dynamic process of lithium ion transport in the electrode in real time. Therefore, on this basis, with the help of the microscopic structure characterization ability of electron microscopy, the surface microregistration analysis method is used to select surface microregions for structural determination. The technical means are available, but a fatal problem was encountered in the process of in situ electron diffraction experiments. The original in situ microcell device was only allowed to rotate around one axis, and the resulting polycrystalline electron diffraction pattern could not distinguish between a structurally similar layered structure and a rock-salt phase (Figure 4). Single-crystal electron diffraction patterns from the stratiform and rock-salt phases can precisely distinguish the phase structure (Figure 5), but this requires the addition of an additional axis of rotation to the in situ device, allowing us to select any one of the positive particles and tilt it to a specific crystal orientation. This requires us to redesign the new in-situ microcell device, which is very difficult. Fortunately, after 18 months of repeated debugging, we successfully prepared an in situ electrochemical reaction device that can be double-tilted in the electron microscope, not only allowing the battery to be usedIt works normally inside the transmission electron microscope and also has the function of simultaneous tilting in the x and y directions (Figures 6A, B). In addition, due to the delicate and complex assembly process of in situ microcells, the team deliberately built a drying room so that we could quickly and efficiently assemble the batteries in the air.
Figure 4: Polycrystalline electron diffraction pattern based on a single-tilt in situ microcell.
Figure 5: Single crystal electron diffraction patterns along a specific crystal orientation can be zoned in layered phases and rock-salt facies
Using the low-voltage transmission electron microscope of the new material structure analysis platform of Nankai University, we performed in situ selective electron diffraction experiments on the 10 nm region of the surface of LNO and LR-LNO particles when they were charged and discharged at a constant current (Figures 6C, F). The results show that the LNO-depleted lithium-poor undergoes a phase transition process from the layered to the rock-salt phase when charging, and the rock-salt phase cannot be converted back to the layered phase during the discharge process. Unlike LNO, LR-LNO maintains a layered structure during charge-discharge, and no rock-salt facies are formed on the surface. Subsequently, we used spherical aberration correction electron microscopy to acquire the atomic images of the surface of the LNO and LR-LNO after the first turn of charging. After the first turn of charging, the rock-salt phase width of the LNO surface increases from 4 nm to 8 nm (to the right of the white dotted line in Figure 7A). In contrast, the width of the rock-salt facies on the surface of the LR-LNO particles is about 2 nm (to the right of the white dotted line in Figure 7C), which is essentially indistinguishable from the surface part of the original sample (1.6 nm, Figure 1F). This indicates that the growth of the rock salt facies on the surface of LR-LNO has been inhibited. The above results explain why the cyclic performance of LR-LNO is much better than that of LNO.
Figure 6: In situ SAED observation of phase transitions on the surface of individual LR-LNO and LNO particles in a constant current charge-discharge test
In addition, we were surprised to find that the bright spot intensity (representing the Ni content in each atomic column) in the atomic image of the charged LR-LNO was unevenly distributed (Figure 7C), while the Ni layer signal intensity in the LNO was uniform (Figure 7B). This indicates that the LR-LNO has undergone intralayer migration of Ni while charging, thus causing Ni to converge. For the surface phase transition process of the positive electrode of the layer, the general consensus is that the transition from the layered phase to the rock salt phase must require interlayer migration between Ni. The ultra-high resolution atom image proves that LR-LNO does not undergo Inter-layer migration, but rather intra-layer migration, which is not difficult to understand why the surface of LR-LNO is not easy to form a rock-salt facies.
Figure 7: Atomic structure of charging state LNO and LR-LNO particles in the first week
In addition to the reversible surface phase transition, we are also very curious about the valence state of LR-LNO surface lattice oxygen. We extracted a series of O electron energy loss spectra (EELS) from the surface of the two charged particles (LNO in Figure 8A and LR-LNO in Figure 8B). The results showed that oxygen vacancies (VO) and rock-salt facies were present on the surface of the charged LNO particles (Figures 8D and 8E). The EELS of LR-LNO shows that in addition to vo and lithoglin phases in the surface area, there are also molecules O2 (Figures 8F and 8G), which indicates that the LR-LNO surface forms O2 during charging, but O2 is not released, but is bound in the near-surface lattice. In addition, we also collected EELS of charging state LR-LNO particles after 30 laps of operation, and the molecule O2 can still be detected near the surface (Figures 8C, 8H, and 8I).
Figure 8: Near-surface lattice-bound O2 of a charging state LR-LNO particle
Based on the above results, we summarized the surface structure evolution of LNO and LR-LNO respectively, and more intuitively told everyone how the slight lithium enrichment in the original lattice of LR-LNO is how to stabilize the surface lattice oxygen during charging and improve its electrochemical properties.
Conventional lithium-poor LNOs delithinate during charging, and the O in the near-surface lattice undergoes oxidation and then loses in the form of O2 (Figure 9A-B). After O2 loss, the remaining VO facilitates interlayer migration of Ni, eventually forming a rock-salt facies on the surface of the material (Figure 9B-C). This phase transition is irreversible because the lost O2 can no longer be returned to the rock-salt facies (Figure 9C-D). Then as the cycle progresses, O2 is continuously released, and the thickness of the rock salt phase continues to increase, eventually leading to a rapid decline in battery performance.
For LR-LNO, although there are only trace amounts of Li, they play an important role. First, both the Li ions in the Li layer and the extra Li ions in the Ni layer can be shed on charge (Figure 9E), leaving a vacancy in the Ni layer after the explusion. These vacancies facilitate intra-Ni migration rather than inter-layer migration (Figure 9F). Intralayer migration replaces interlayer migration, and the rock salt phase cannot be formed. What’s more, intralayer migration causes vacancies to accumulate, forming clusters of vacancies that bind O2 to the near-surface lattice (Figure 9G). During discharge, as the Li ions return to the lattice vacancies, the bound O2 is reduced, so the layered phase structure is maintained (Figure 9H).
Therefore, the highly reversible lattice oxygen redox reaction and lattice oxygen binding mechanism explain why the cyclic and thermal stability of LR-LNO have been significantly improved. In addition, during the electrochemical cycle, very few Ni ions migrate to the Li layer due to intralayer migration of Ni. This is critical for the rapid conduction of Li ions in the Li layer, which explains why LR-LNO has excellent magnification performance.
Figure 9: Surface phase transition dynamics of LNO and LR-LNO
Q: How do you comment on the impact of this work?
Researcher Wei Zhang:
By constructing a lithium-rich molten salt reaction environment, this study introduces a small amount of Li into the Ni layer, thereby obtaining a small excess of Li-LNO cathode material, which significantly improves the cycle stability and thermal stability of LNO material while maintaining a high specific capacity. The intralayer migration of Ni in delithium LR-LNO promotes the formation of vacancy clusters, which not only stabilizes the oxidative formation of O2 in the near-surface lattice, but also effectively inhibits the stratification to rock-salt phase transition that destroys the electrochemical properties. This study not only “rekindles hope” at the lithium nickel oxide cathode, but also reveals the importance of understanding the complex interaction between cation migration and O redox (especially on the surface of the particles), which is an important inspiration for the study of various alkali metal oxide cathode materials. Since the surface lattice of O-containing cathode materials is usually unstable during cycling, the above method stimulates technological innovation in cathode materials by fine-tuning the composition, allowing researchers to more fully grasp the relationship between battery performance and microscopic mechanisms. In addition, in addition to the field of batteries, the bi-in-situ TEM technology will also help to study the complex phase transition mechanism in a wide range of fields such as catalysis and metallurgy.
Finally, we would like to thank Nankai University’s new substance structure analysis platform for its strong technical support for our work. The platform is equipped with two spherical difference correction electron microscopes as the core, equipped with 11 sets of electron microscopes, atomic force microscopes, plasma element Raman spectroscopy and other instruments, the platform also develops cutting-edge in situ characterization methods, creates advanced instruments and equipment and scientific devices, so that we can use high space-time energy resolution characterization methods to accurately analyze the microstructure and new reaction mechanisms of new substances. The research was supported by the key research and development programs of the Ministry of Science and Technology (2021YFB2500300, 2021YFB3800300) and related projects of the National Natural Science Foundation of China (52072185, 22008154). (Source: Science Network)
Related paper information:https://doi.org/10.1016/j.chempr.2022.07.023