Dissolution, migration, and deposition of transition metal ions in Li-ion batteries exemplified by Mn-based cathodes – a critical review

Chun Zhan a, Tianpin Wu b, Jun Lu *a and Khalil Amine *a
aChemical Sciences and Engineering Division, Argonne National Laboratory, Argonne, IL 60439, USA. E-mail: junlu@anl.gov; amine@anl.gov
bX-Ray Science Division, Advanced Photon Source, Argonne National Laboratory, Argonne, IL 60439, USA

Received 31st October 2017 , Accepted 8th December 2017

First published on 14th December 2017


Unlike the revolutionary advances in the anodes of lithium-ion batteries from Li intercalation materials to Li alloy and/or conversion reaction materials, the development of the cathode is still dominated by the Li intercalation compounds. Transition metal ions are essential in these cathodes as the rapid redox reaction centers, and one of the biggest challenges for the TM-based cathodes is the capacity and power fading especially at an elevated temperature, which is directly associated with the dissolution–migration–deposition (DMD) process of TMs from the cathode materials. This process not only alters the surface structure of the cathode materials, but more importantly, changes the SEI composition at the anode side. There is no doubt that the TM-DMD issue should be addressed thoroughly to unlock the potential of these compounds to enable a prolonged battery lifetime. This review article mainly focuses on research activities with regard to the DMD process in TM-based cathode materials. In the first four sections, we choose Mn-based cathodes as an example to discuss how Mn DMD relates to the capacity fade of the cell, and what possible approaches might suppress the DMD process by modification of the electrode or electrolyte. In the fifth section, we discuss the TM DMD process in Ni-, Co-, Fe- and V-containing cathode materials. This article reviews the frontier electrochemical research on TM-based cathodes and summarizes the progress and challenges, thereby helping to advance future R&D of LIBs.


image file: c7ee03122j-p1.tif

Chun Zhan

Dr Chun Zhan is a postdoctoral appointee in the Division of Chemical Sciences and Engineering at the Argonne National Laboratory. Her research interests focus on electrochemical energy storage technology, with the main focus on cathode materials of lithium ion batteries. Dr Zhan earned her PhD degree in Physical Chemistry in the Department of Chemistry of Tsinghua University in 2014.

image file: c7ee03122j-p2.tif

Tianpin Wu

Dr Tianpin Wu is the Principal Beamline Scientist working at beamline 9-BM of the Advanced Photon Source, Argonne National Laboratory. Her expertise are in energy storage and catalyst material characterization by X-ray Absorption Spectroscopy. She received her undergraduate degree in Chemistry from the University of Science and Technology of China, and her PhD degree in Physical and Analytical Chemistry from the University of Utah. Following two and half years of postdoctoral research in the Division of Chemical Sciences and Engineering at the Argonne National Laboratory, she joined the X-ray Science Division as a physicist in 2012.

image file: c7ee03122j-p3.tif

Jun Lu

Dr Jun Lu is a chemist at the Argonne National Laboratory. His research interests focus on the electrochemical energy storage and conversion technologies, with the main focus on beyond Li-ion battery technology. Dr Lu earned his bachelor degree in Chemistry Physics from the University of Science and Technology of China in 2000. He completed his PhD from the Department of Metallurgical Engineering at the University of Utah in 2009. He was the awardee of the first DOE-EERE postdoctoral fellow under the Vehicle Technologies Program from 2011 to 2013. Dr Lu has authored/co-authored more than 200 peer-reviewed research articles and has filed over a dozen patents and patent applications.

image file: c7ee03122j-p4.tif

Khalil Amine

Dr Khalil Amine is an Argonne Distinguished Fellow and the manager of the Advanced Battery Technology team at the Argonne National Laboratory, where he is responsible for directing research and development of advanced materials and battery systems. He is the deputy director of the US-China Clean Energy Research Center and a committee member of U.S. National Research Council and National Academies of Sciences. He is a recipient of the Scientific American's Top Worldwide 50 Researcher Award (2003), the U.S. Federal Laboratory Award for Excellence in Technology Transfer (2009), and the DOE Vehicle Technologies Office Award (2013), and is the five-time recipient of the R&D 100 Award.



Broader context

A new world of smart energy is emerging, with transformative developments in renewable energy techniques, smart grids and electric vehicles. As the heart of electric vehicles, Li-ion batteries are developing rapidly with higher energy density and longer lifetime. The cathode of a Li-ion battery is the source of lithium ions and thus the capacity-determining part, which is dominated by the Li intercalation compounds based on transition metal ions. The transition metal ions are supposed to remain in the cathode during repeated Li+ intercalation–deintercalation, but they have been detected to dissolve from the cathode, migrate in the electrolyte, and deposit on the anode. This dissolution–migration–deposition (DMD) process is believed to be directly associated with the capacity fading of the battery. In this paper, we review the research activities on the mechanistic understanding and solutions to overcome the TM DMD process, from the earliest discoveries to the latest progress. Considerable material engineering has been conducted to solve the DMD issue, but the underlying mechanisms of the TM DMD process are still under debate. A thorough understanding of the DMD process using state-of-the-art techniques with critical thinking is required to unlock the potential of the cathode materials to enable a prolonged battery lifetime.

1. Introduction

The rechargeable lithium-ion battery (LIB) is one of the most successfully commercialized electrochemical power sources owing to its high energy and power density and high energy efficiency compared to other secondary batteries, such as lead–acid, nickel–cadmium (Ni–Cd), and nickel–metal hydride (NiMH) systems. With their applications extending from portable devices to electric vehicles (EVs) since the late 1990s, state-of-the-art LIBs require additional improvements for better durability and lower cost to meet the need of the automotive industry. For instance, the United States Advanced Battery Consortium (USABC) set the goals for EVs at a calendar life of 15 years and at an operating temperature from −30 to 52 °C, which are beyond the electrochemical performance of today's LIBs. Clearly, the cost and performance limitations of the existing Li-ion battery technologies seriously hinder the rapid transition to EVs.

The concept of rechargeable LIBs evolved from the Li battery with a Li-ion insertion material as the cathode and Li metal as the anode,1 which attracted great interest due to the high energy density of Li metal. In 1976, Whittingham succeeded in building the first commercial rechargeable lithium battery with the Li–TiS2 couple.2 However, the safety issues of the Li battery induced by Li dendrite growth on the Li anode and the consequent short circuit (if the dendrite penetrates the separator and directly contacts the cathode) severely limited its practical applications. After that, the second generation of rechargeable Li batteries had two Li-ion insertion electrodes, avoiding the use of Li metal. This first LIB3 was successfully developed and demonstrated4 thanks to the discovery of the layered LiCoO2 cathode5 and graphite anode.6 The commercialization of rechargeable LIBs in the 1990s by Sony Corporation to power portable electronic devices, including cellular phones and laptop computers, sparked a revolution in battery technology. This revolution marked a massive swing away from relatively low-voltage, water-based energy storage systems, for instance, Ni–Cd and NiMH batteries.

A rechargeable LIB consists of two Li-ion intercalation electrodes with a non-aqueous electrolyte in between for ionic conduction. The electric and chemical energies in a Li-ion cell are interconverted via reversible de-intercalation/intercalation processes of Li ions between the cathode and anode along with electrons travelling via the external circuit simultaneously, as shown in Fig. 1. In principle, the delithiation/lithiation processes in the Li-ion cell are expected to be completely reversible, indicating that the cell capacity is fully recovered after a charge–discharge cycle. However, in reality, numerous irreversible chemical and/or electrochemical reactions occur in the electrodes and on the electrode surface, for example, the decomposition of the electrolyte and the structural disorder or dissolution of the electrode materials.7 Most of these irreversible reactions lead to the loss of the active material and increase of internal resistance, and consequently, the fade of the overall energy and power of the cell.


image file: c7ee03122j-f1.tif
Fig. 1 A schematic of a lithium-ion battery.

The cathode is the source of lithium ions and the capacity-determining part of Li-ion cells. It thus plays a vital role in battery performance considering that carbon in the form of graphite is the current choice as the anode material. Such electrodes need to have stable structures over a wide compositional range so that as much lithium as possible can be extracted and re-inserted during repeated charging and discharging to maximize the energy density of the cell. Since the early studies on layered LiCoO2, several alternative cathode materials have been exploited over the years, including spinel electrodes derived from LiMn2O4,8–10 olivine LiFePO4,11 layered mixed-metal oxides (LiCoxNiyAl1−xyO2 (NCA)),12 LiMnxNi1−xO2,13 LiNiyMnzCo1−yzO2,14,15 and high-energy Li-rich manganese-based oxides (xLi2MnO3·(1 − x)LiMO2).16,17 The Ni-containing materials provided a higher practical capacity (for example, 160–180 mA h g−1 for NCA)12 than LiCoO2; however, their thermal instability on delithiation due to the presence of the high-valence Ni compromises the safety of the Li-ion cells. Olivine LiFePO4 electrodes are significantly more stable to lithium extraction than the layered Co- and Ni-based electrodes, but they deliver relatively low practical capacities (100–150 mA h g−1) above 3 V.18 A family of high-energy manganese-based cathodes fabricated by structurally integrating a Li2MnO3 stabilizing component into an electrochemically active LiMO2 (M = Mn, Ni, and Co) electrode has been intensively tested as a potential cathode material for EVs because the excess lithium in these composites boosts the specific capacity of the cell up to 250 mA h g−1 between 4.6 and 2.5 V.17 Another advantage of these high-energy cathode materials is their low cost and low toxicity due to their relatively high Mn content. However, the cells containing these high-energy cathode materials face severe capacity and voltage fading. Nevertheless, it is clear that the transition metal (3d) elements in these cathode materials determine the cell capacity, power density, and safety.

Transition metal (TM) ions are supposed to remain in the cathode during repeated Li+ intercalation–deintercalation. However, a small amount of TM element has been detected in the electrolyte and on the anode after long-term cell cycling, especially at elevated temperatures.19,20 In other words, the TM ions dissolve from the cathode, migrate through the electrolyte and deposit on the anode. This dissolution–migration–deposition (DMD) process is believed to be directly associated with the capacity fading of the battery.21,22 This is believed to be the case for TM-based cathodes such as LiMn2O4, LiMO2 (M = Co, Ni, Mn, etc.), and Li2MnO3 (Fig. 2). These cathodes are attractive owing to their low cost, relatively low toxicity, and thermal stability.23,24 However, one of the biggest challenges to commercially realize lithium-ion batteries with TM-containing cathodes is the capacity and power fade of the cell at elevated temperatures. TM DMD is thought to not only alter the surface structure of these cathode materials, but more importantly change the solid–electrolyte interface (SEI) composition at the anode side.25–27 Clearly, the underlying mechanisms of the TM DMD process and the related capacity fading must be addressed in order to further prolong the lifetime of Li-ion batteries.


image file: c7ee03122j-f2.tif
Fig. 2 Crystal structures of Mn-based cathode materials.

This review article will mainly focus on research activities with regard to the DMD process in TM-based cathode materials, especially how it relates to the capacity fade of the cell, and what possible approaches might suppress the DMD process by modification of the electrode or electrolyte. We chose Mn DMD as an example in this article because the electrochemistry of LiMn2O4 as the cathode material in Li-ion cells is well established, and the Mn DMD process has been intensively investigated. We will also discuss the TM DMD process in Ni-, Co-, and Fe-containing cathode materials. This article reviews the frontier electrochemical research on the TM-based cathode and evaluates and summarizes the progress and challenges at hand, thereby helping to advance future R&D of LIBs.

2. Mn dissolution and its relation to the capacity fading of Mn-based cells

The phenomenon of Mn dissolution from spinel LiMn2O4 in acidic media was discovered in 1981,28 even much earlier than the adoption of LiMn2O4 as a cathode material for Li-ion batteries.10 Through the treatment of the spinel LiMn2O4 with aqueous acid, Hunter28 reported the conversion of LiMn2O4 to nearly pure λ-MnO2 via chemical delithiation. The resulting λ-MnO2 preserves the structural framework of LiMn2O4, but with most of the lithium removed from the tetrahedral sites. Such a conversion leads to some contraction of the lattice in the spinel structure, as evidenced by a reduction in the lattice constant (a) from 8.24 to 8.03 Å. A disproportionation-type mechanism was proposed to interpret the conversion reaction, as shown in eqn (1).
 
2LiMn2O4 + 4H+ → 2Li+ + Mn2+ + 3λ-MnO2 + 2H2O (1)

With Li+ and Mn2+ being removed from the surface, more Li ions from the bulk LiMn2O4 diffuse to the surface, accompanied by electron hopping from Mn3+ to Mn4+, as shown in Fig. 3.


image file: c7ee03122j-f3.tif
Fig. 3 Schematic representation of the conversion of LiMn2O4 to λ-MnO2 in an acidic aqueous solution. Reprinted from ref. 28. Copyright © 1981 Elsevier Inc.

The slow dissolution of manganese from LiMn2O4 in a lithium-ion battery has also been observed using differential pulse polarography29 and X-ray adsorption spectroscopy (XAS),30 and the oxidation state of Mn dissolved in the electrolyte was identified to be divalent (Mn2+), while a recent study claimed that Mn(III) is the main soluble Mn ion in the electrolyte based on electronic paramagnetic resonance (EPR) plus inductively coupled plasma (ICP) spectroscopy and X-ray absorption near edge structure (XANES) spectroscopy.31 The dissolution of Mn was noticeably accelerated with an elevated temperature,21,25 a larger specific surface area,29,32 a higher acidity of the electrolyte,33 and a higher potential of the cathode.29,32 Moreover, in situ X-ray diffraction (XRD) and X-ray reflectivity (XRR) and ex situ transition electron microscopy (TEM) showed that the (110) crystal plane of LiMn2O4 is less stable compared to the (111) crystal plane and thus more subjected to Mn-ion dissolution, which is likely due to the looser arrangement of the Mn ions on the (110) plane,34 as shown in Fig. 4.


image file: c7ee03122j-f4.tif
Fig. 4 (a) A schematic of atomic stacking and surface reactions for (b) (111) and (c) (110) crystal planes of LiMn2O4 characterized by in situ XRD and XRR measurements and ex situ TEM observations. Reprinted from ref. 34. Copyright © 2010 American Chemical Society.

The electrochemical performance of a LiMn2O4 cell could be significantly altered even with a small amount of manganese (typically at the ppm level in the electrolyte) dissolved from the cathode.35,36 Therefore, the key questions are: how does the Mn(II) dissolve from LiMn2O4 in an electrochemical cell? Does it share the same mechanism, i.e. the disproportionation-type process, as the chemical dissolution in the aqueous acidic media? What is the correlation between Mn dissolution and the capacity fading of the cell? To address these issues, extensive research has been devoted to both experiments and theoretical calculations.37–39 The disproportionation reaction (as shown in eqn (1)) was initially proposed by Gummow et al.40 to interpret the dissolution of Mn from the LiMn2O4 cathode material, based on the finding that the oxidation state of Mn dissolved in the electrolyte is 2+, while the concentration of Mn(IV) on the spinel surface increases along with the dissolution. In most cases, this mechanism has been proved to be valid and consistent with the experimental findings that the dissolution of Mn accelerates with the increase of the acid concentration in the electrolyte, the specific surface area of the active material particles, and the temperature. Therefore, this mechanism is still accepted and widely used to interpret the observed capacity fading in LiMn2O4 cells.

However, disproportionation of Mn(III) cannot explain the increased Mn dissolution at high cathode potentials at which the concentration of Mn(III) decreases. Therefore, other mechanisms have been proposed to explain the Mn dissolution process at high charge potentials, including phase transformation21 and chemical lithiation/protonation.41 The phase transformation mechanism (as shown in Fig. 5) can address the Mn dissolution from the spinel at the high-potential plateau where a two-phase structure coexists, as described below:21

 
LixMn2O4 → LixMn2−yO4−y + yMnO2(soluble) (2)


image file: c7ee03122j-f5.tif
Fig. 5 A semi-quantitative Li–Mn–O phase diagram. The lines with arrows show the two routes for phase transformation via the loss of MnO or Mn2O3·MnO. Reprinted from ref. 21. Copyright © 1997 The Electrochemical Society.

At high temperature, the above reaction will accelerate because the two-phase structure is sensitive to temperature. Even the direct dissolution of Mn(III) can occur in this situation, as shown in eqn (3).

 
4LiMn(III)Mn(IV)O4 → Li4Mn(IV)5O12 + Mn(II)O·Mn(III)2O3 (soluble) (3)

The formation of a Mn3O4 phase on the charged LiMn2O4 was confirmed by scanning transmission electron microscopy.42 The above reaction is consistent with the recent report from the Arubach group showing that Mn(III) rather than Mn(II) is the main soluble Mn ion species in the electrolyte.31

In a few cases, chemical lithiation and protonation could also contribute to Mn dissolution during the aging of LiMn2O4 at high temperature,41 as is evident by the formation of a protonated Mn-deficient λ-LixHyMnO2 phase. The generation of the protonated phase mediated by the formation of a Li-rich spinel was proposed as follows:

 
image file: c7ee03122j-t1.tif(4)
 
image file: c7ee03122j-t2.tif(5)
 
image file: c7ee03122j-t3.tif(6)

The protonated λ-MnO2 is partially electrochemically inactive and thus acts as a passivation layer on the surface of the aged spinel. At a potential of 4.2 V vs. Li, the delithiated spinel becomes more oxidative, and thus accelerates the generation of protons in the electrolyte solvents according to the following scheme:

 
image file: c7ee03122j-t4.tif(7)

As a result of the above redox reaction between the delithiated spinel and the electrolyte solvent, Mn dissolution demonstrates a dramatic increase as the state of charge (SOC) approaches the end of charging.43

Although there is considerable debate on the mechanism of Mn dissolution from the spinel during the electrochemical operation of the cell, it is generally accepted that the capacity fading of the cell is directly associated with Mn dissolution. When Mn dissolution is accelerated, i.e. by a higher temperature or a higher surface area, the capacity fading of the cell declines accordingly.21,29 The loss of active material (Mn3+) from the cathode is a straightforward reason for the capacity loss caused by Mn dissolution. However, the material loss accounts for only 20–33% of the overall capacity loss of the cell.29 An AC impedance test on cycled cathodes demonstrated that a considerable portion of the capacity loss can be attributed to the concomitant increase of resistance in the cathode due to the segregation of contact sites of a three-phase zone (spinel/carbon/electrolyte) where Li-ion intercalation/deintercalation takes place.29,38 This impedance rise of the LiMn2O4 cathode due to Mn dissolution during cell cycling was also confirmed by Arubach and co-workers,22 who observed a continuous increase in interfacial impedance when the LiMn2O4 cathode was polarized at 4.5 V.

Other Mn-based cathodes were also studied to understand the Mn dissolution process and its influence on the electrochemical performance. These data provide strong evidence that the existence of Mn(III) in the cathode is the main reason for Mn dissolution. Choi et al. systematically investigated Mn-doped layered oxides, orthorhombic LiMnO2, and 5 V spinel LiMn1.5Ni0.5O4.44 The results demonstrated that cathode materials containing Mn3+, such as layered LiMn0.8Cr0.2O2, LiMnO2, and spinel LiMn2O4, showed high Mn dissolution and faster capacity fading; whereas LiMn0.5−xNi0.5−yCo2yO2 and LiMn1.5Ni0.5O4, in which manganese exists as Mn4+, exhibited low or moderate Mn dissolution and, therefore, relatively good capacity retention upon cycling. Thackeray and coworkers45 studied Mn dissolution in a Li- and Mn-rich layered material (0.3Li2MnO3·0.7LiMn0.333Ni0.333Co0.333O2) under acidic treatment, where the oxidation state of manganese is also tetravalent. No detectable change in the Mn/Ni/Co ratio in the material was observed, suggesting that dissolution did not occur readily due to the relatively stable Mn(IV) in the acid. However, during the initial charge of this composite at a high potential, Li2MnO3 converted to MnO2. In the following discharge/charge cycles, a spinel-like component gradually formed when the Li ions entered the MnO2. In this case, it is reasonable to expect that Mn dissolution from the spinel will occur. The Mn dissolution from the Mn4+-only MnO2 is similar. The MnO2 system is very versatile due to its various crystal structures, such as α, β, γ, δ, λ, and ramsdellite phases.46,47 During the lithiation of MnO2, Mn4+ is reduced to Mn3+, and thus may dissolve in the electrolyte via the disproportionation reaction, similar to what occurs in the Zn–MnO2 battery.48

The Mn dissolution in Mn(II)-containing cathode materials was reported by Martha et al.49 They found that a very small amount of Mn (0.02%) was dissolved into the electrolyte even at an elevated temperature (60 °C), and the dissolution seemed to be saturated after 14 days. Considering that Mn(II) in LiMn0.2Fe0.8PO4 should be subjected to an acid attack from the electrolyte, a passivation film was likely formed on the surface due to the dissolution of Mn(II), and this film plays a critical role in preventing the Mn dissolution, although this hypothesis has not been investigated in detail yet.

3. Mn deposition on the anodes and its relation to the capacity fade of the cells

Once the Mn ions dissolve in the electrolyte, they will migrate to the anode driven by the concentration gradient and/or electric field force and thereby deposit on the anode. The deposition of Mn on the anode was first detected by Rutherford backscattering spectroscopy (RBS)50 in the LiMn2O4/Li cell and then further confirmed by energy-dispersive X-ray spectroscopy (EDX),51 secondary ion mass spectrometry (SIMS),52 ICP Auger electron spectroscopy (AES),53 X-ray photoelectron spectroscopy (XPS),54 and X-ray absorption fine structure (XAFS) spectroscopy.26,27,55 The data indicated that the capacity fade of the graphite anode was directly related to Mn deposition, and this process dominated the overall performance fade of the cell.51 In cell testing, when the aged LiMn2O4 cathode and the graphite anode were cycled against fresh Li-metal electrodes, the cathode retained about 70% of the capacity of the fresh one, but the graphite anode did not show any ability to sustain capacity upon cycling. However, after removing the Mn deposit with acid treatment, the anode recovered almost all of its initial capacity. Moreover, Amine et al.19 found that the impedance of the negative electrode (graphite anode) significantly increased when even a small amount of Mn was present in the electrolyte, and the impedance rise of the graphite anode dominated the overall impedance fade of the LiMn2O4/graphite system, supporting the hypothesis that the graphite anode is the main source of capacity fading. Since then, the deposition of Mn on the anode has been intensively investigated as the crucial factor giving rise to the capacity fade of Mn-based LIBs. The focus of these investigations is to understand the driving force of Mn deposition and to determine how Mn deposition causes capacity fading.

Because the operation potentials of commonly used anodes such as lithium metal and graphite are lower than the tabulated redox potential of the Mn2+/Mn couple, an electrochemical50 or chemical53 reduction mechanism was proposed to explain Mn deposition on the anode side. In the electrochemical reduction model, the soluble Mn in the electrolyte (written as Mn2+ for simplification) is electroplated on the anode, which can be expressed as:

 
Mn2+ + 2e → Mn (8)

The Mn2+/Mn redox potential of 1 mM MnCl2 in a 1 M LiPF6 + ethylene carbonate/dimethyl carbonate (EC/DMC) electrolyte was determined by cyclic voltammetry to be about 1.58 V vs. Li+/Li. This potential is lower than the tabulated redox potential of the Mn2+/Mn couple (−1.185 V vs. SHE, i.e. 1.86 V vs. Li+/Li) because of the low activity and concentration of the Mn2+ ions in the electrolyte.51 Similarly, the chemical reduction model attributes the deposition of Mn to the reductive property of the lithiated anode, and the reaction is expressed as:53

 
yMn2+ + LixC6yMn(on C) + Lix−2yC6 + 2yLi+ (9)

Based on the reduction mechanism, the capacity fade of the cell was attributed to the accelerated irreversible decomposition of the electrolyte caused by the deposited Mn(0) metal. As adding Mn2+ to the electrolyte could lead to an increase in irreversible capacity (and decrease in columbic efficiency) of the graphite anode, simulation suggested that the electrocatalysis properties56,57 or electrical conductivity58 of the Mn metal resulted in more electrolyte decomposition compared to a situation without Mn deposition. Consequently, the decomposition products on the Mn surface gradually cover the anode surface, thus interfering with the lithium intercalation into the graphite and the increase in the anode impedance.56,57 Meanwhile, the formation reaction of inactive materials consumed Li ions, and the loss of the active Li ions could also lead to capacity fade with cycling.53 This explains why the Li/LiMn2O4 cell normally shows better cycling performance than the graphite/LiMn2O4 cell at the same elevated temperature,19 as the Li metal anode can always provide extra active Li ions during the discharge of the cell.

In addition, Delacourt58 suggested that Mn(0) acts as the intermediate for Mn deposition as MnCO3:

 
Mn2+ + 2e → Mn0 (10)
 
Mn0 + EC → MnCO3 + C2H4 (11)

A conversion reaction could occur between MnCO3 and Li+ as follows:

 
MnCO3 + 2Li+ + 2e ↔ Li2CO3 + Mn0 (12)

This reaction occurs reversibly during cycling and changes the morphology of the SEI continuously. The resulting cracks and pores increase the electrochemically active surface area in the SEI layer and also provide pathways for solvated Li+ transportation, while the Mn metal in the SEI layer serves as an electron carrier, as shown in Fig. 6. As a result, the Mn-containing porous SEI accelerates the electrolyte deposition and, therefore, the capacity fading.


image file: c7ee03122j-f6.tif
Fig. 6 Schematics of the mechanism proposed for SEI growth on the copper electrode during (a) formation in a Mn-free electrolyte, (b) formation in a Mn-contaminated electrolyte, and (c) cycling of a Mn-contaminated SEI in a Mn-free electrolyte. “S” stands for a solvent molecule. Reprinted with permission;58 Copyright © 2013 The Electrochemical Society.

Although this reduction mechanism was not questioned for over 20 years, Mn metal deposited on the anode was not actually detected until 2012.59 In this work, 200 ppm Mn2+ was added into the electrolyte, and Mn 2p XPS showed that Mn(0) was deposited on the edge plane of highly oriented pyrolytic graphite (HOPG) after 48 h of potential-state discharge at 5 mV vs. Li+/Li. However, it is questionable whether these results can be extrapolated to practical LiMn2O4/graphite systems in which the Mn2+ is gradually dissolved from the cathode while cycling, instead of existing at a high concentration in the electrolyte in the very beginning.

The Mn deposition mechanism in practical Mn-based batteries was systematically analyzed in a previous work of our group.26 In this work, the Mn was dissolved from the spinel cathode gradually during cycling, so that the Mn deposits on the graphite anode, as measured by ICP-AES. Mn deposition was found to occur on electrodes with a potential much higher than 1.85 V vs. Li+/Li (e.g., 3.55 V vs. Li+/L for a Li1−xFePO4 anode), as shown in Fig. 7a, as well as on fully delithiated graphite (cycled in a Mn-free electrolyte and then polarized at the delithiated state) via storage in the Mn-containing electrolyte. More importantly, XANES spectra indicate that the oxidation state of Mn on the anode is 2+, independent of the cycle number or the Li insertion potential of the different anodes, as shown in Fig. 7a and b. These results show that Mn2+ is able to deposit on the anode without the driving force of low potential. Based on the XAS results together with an in situ impedance test of the graphite anode upon cycling, we proposed an ion-exchange model (Fig. 8) for the Mn deposition mechanism: the Mn2+ ions in the electrolyte can accumulate in the anode SEI layer via the ion-exchange reaction with the mobile Li+ ions in the SEI, and subsequently block the lithium diffusion path, leading to the increased impedance of the graphite anode and hence the capacity fade of the battery.


image file: c7ee03122j-f7.tif
Fig. 7 X-ray absorption near-edge spectra (XANES): (a) normalized XANES of the Mn K-edge for commercial MnO, Mn2O3, Mn foil, and different anode samples harvested after 100 cycles and (b) unnormalized XANES of the Mn K-edge for mesocarbon microbead (MCMB) anode samples harvested after different cycles. Reprinted with permission;26 Copyright © 2013 Macmillan Publishers Limited.

image file: c7ee03122j-f8.tif
Fig. 8 Ion-exchange model: the proposed ion-exchange model for Mn deposition on the graphite anode during the Mn DMD process in the graphite/LMO cell. Reprinted with permission;26 Copyright © 2013 Macmillan Publishers Limited.

Intrigued by our questioning of the conventional reduction mechanism, several groups have sought to understand the Mn deposition in batteries under practical operation. The results reveal that the oxidation state of Mn deposited on the graphite anode is likely to be dependent on the lithiation state of the anode. Only Mn2+ ions are present on a fully delithiated graphite anode that has been held for 48 hours at a constant voltage of 1.4 V vs. Li+/Li after 200 electrochemical cycles, and it appears that the Mn2+ ion center in the delithiated graphite electrode is closely associated with the lithium carbonate phase (Fig. 9).27 However, on a graphite anode after cycling between 3 V and 4.8 V vs. Li1.05Mn2O4, α-Mn metal nanoparticles were observed at the SEI/graphite interface while Mn2+ species such as MnF2 were detected at the topmost surface layer of the SEI.60 The authors did not clarify the lithiation state, but it is rational to suppose that the cells were disassembled at the end of cycling at 3 V, and therefore, the anode was likely to be highly but not fully delithiated. On the fully lithiated-graphite anodes, both Mn(0) and Mn(II) can be probed by XANES on a graphite anode cycled either in a Mn-enriched electrolyte or against Mn-containing cathodes.55 Shkrob et al.27 claimed that the Mn(0) deposits in an unidentified state different from the atomic scale, nanometer scale or mesoscale Mn(0) clusters. It is likely that (at least) some of the Mn(II) is reduced during lithiation of the anode, and the reduced Mn is oxidized to Mn(II) in the delithiation process. However, only detecting Mn(0) on the anode does not mean that the reduction reaction from Mn(II) (in the electrolyte) to Mn(0) (on the anode) is the only driving force of the deposition. Because Mn(II) can be detected on an anode with high potential as well as on the delithiated graphite stored in Mn2+-containing electrolyte, it is undeniable that the Mn can be directly deposited on the anode as Mn(II) via ion exchange26 or chemisorption,27 and then the Mn(II) may be reduced to Mn(0) at the delithiated state. To further understand this matter, in situ analysis of the oxidation state of Mn deposited on the anode is imperative, although it will require intricate experimental design to eliminate the effect of Mn from the cathode and/or the electrolyte.


image file: c7ee03122j-f9.tif
Fig. 9 Two models of transport and deposition of the Mn2+ ions in the Li ion batteries. (a) Mn2+ ions are solvated by carbonate molecules in the same fashion as Li+ ions and drift along the field lines penetrating through the outer (organic) SEI (ii) and become deposited into the inner (mineral) SEI layer at the graphite particle surface (i). (b) Electrolyte decomposition products with chelating groups reach the positive electrode and form neutral complexes of Mn(II) that diffusively migrate to the graphite surface by passing the outer SEI, and these complexes become chemisorbed on the surface of lithium carbonate crystallites in the inner SEI layer that serves as an ion exchanger. Reprinted with permission;27 Copyright © 2014 American Chemical Society.

In summary of the above two sections, Mn dissolution–migration–deposition in the Mn-based LIBs (the LiMn2O4/graphite system, for instance) is represented in Fig. 10. The DMD process is dramatically enhanced by an elevated temperature; therefore, it is the main cause for the poor high-temperature cycling performance of the Mn-based LIBs. Determination of the detailed mechanism of the DMD process and understanding of how it leads to the fade of the electrochemical performance are still under debate. Generally, on the cathode side, the dissolution of Mn leads to the loss of the active material and increased impedance of the cathode; on the anode side, Mn deposition severely degrades the SEI layer and results in a significant impedance rise for the anode. The incremental increase in the impedance of the electrodes reveals the hindered charge transfer and transport in the lithiation/delithiation process, thus leading to the capacity fade.


image file: c7ee03122j-f10.tif
Fig. 10 A schematic of Mn dissolution, migration, and deposition processes in a LIB. Reprinted with permission;107 Copyright © 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

In addition, as Mn DMD is able to increase the impedance of both the cathode and anode, it is conceivable that DMD can remarkably affect the long-term power performance,61 which is one of the key properties of EV batteries. In fact, the correlation between the power performance or conductivity of the Mn battery and the Mn DMD process is very complicated. For instance, a higher electronic conductivity can be realized by increasing the carbon content in the composite LiMn2O4 electrode. However, Mn dissolution is aggravated at the same time because the electrolyte solvents are oxidized on the carbon surface and more acidic species are generated, which in turn causes conductivity loss of the electrode.29 A similar contradictory situation also exists between the ionic conductivity and Mn dissolution. The lithium ions can easily diffuse through the (110) planes exposed on the particle surfaces as their orientations are aligned to Li diffusion channels.62 Unfortunately, they are also the planes most vulnerable to Mn dissolution due to the loss of atomic arrangement.34,63 Therefore, the optimized trade-off between the conductivity and Mn dissolution is critical to the design of LiMn2O4 with both high power performance and long cycling life.

Due to the negative impact of Mn dissolution, migration, and deposition on the electrochemical performance, especially at elevated temperatures, numerous studies have been carried out to minimize the temperature-enhanced Mn DMD process, aiming to develop a high-power, long-lifetime LIB for electric vehicles, as will be reviewed in the following two sections.

4. Suppressing Mn dissolution

As Mn dissolution takes place at the interface between the cathode and electrolyte, stabilizing the cathode structure against the attack of acid from the electrolyte is critical to suppressing the dissolution process. Modification of the cathode is one of the most commonly used approaches to improve the cycling performance of the Mn-based cathode materials.

The morphology and crystal stability of the pristine cathode material can be optimized by adjusting the parameters of the synthesis methods. The annealing temperature29 and a Li precursor64 were observed to affect the particle size and thus the surface area of spinel LiMn2O4 prepared by the conventional solid-state method, which would impact Mn dissolution. Meanwhile, new synthesis routes such as the resorcinol–formaldehyde method have been applied to prepare stoichiometric LiMn2O4 powders with lower Mn dissolution compared to conventional methods.61 However, the optimized spinel still suffers from Mn dissolution due to the impact of Mn(III) in the cathode materials. Modification of the Mn-based cathode material is required to further suppress Mn dissolution, which can be achieved mainly via three approaches: surface coating, bulk doping, and surface doping.

Since the first study of surface coating of LiMn2O4 by Li-bearing oxide glasses Li2O3:B2O3,65 hundreds of studies have been carried out on coating the cathode with many other oxides,66–74 lithiated metal oxides,20,67,75–77 fluorides,78–80 polymers,81 or phosphate.82,83 The surface-coated cathodes showed improved cycling performance at elevated temperatures because the coating materials are tougher against the attack of HF or can consume the HF and convert it to more stable species. The preparation methods and electrochemical performance of the surface-coated spinel LiMn2O4 have been reviewed by Yi et al.84 However, further efforts are required to solve the issues of surface coatings such as their non-uniform deposition, poor chemical/physical bonding, and difficulties in controlling the depth of the deposition.85

Cationic and anodic doping at the bulk level has improved the cycling performance of spinel LiMn2O4. The doping metallic ions usually have an oxidation state lower than 3+, such as Na+, Li+, Ni2+, Co2+, and Al3+,40,86–92 and therefore keep the average valence of Mn higher than 3.5. An increase in the Mn valence content results in a lower level of Mn dissolution and improvement in cyclability of the spinel, but it also means a lower Mn3+ content. Because the intercalation of Li+ ions is accompanied by conversion from Mn3+ to Mn4+, the loss of Mn3+ inevitably leads to a sacrifice in the initial capacity. On the other hand, both the capacity and cycling performance can be improved if electrochemically active cations (such as Ni2+ and Co2+) are substituted for the Mn. For instance, in Ni-substituted spinel LiMn1.5Ni0.5O4,90 in which Mn exists as Mn4+, the Ni2+/Ni3+ and Ni3+/Ni4+ couples show two plateaus at about 4.7 and 4.8 V, respectively, and contribute to the overall capacity. As a result, 5 V spinel LiMn1.5Ni0.5O4 has attracted much interest for its advantages in terms of both capacity and cycling performance.93–95 However, this cathode material requires an electrolyte with a larger electrochemical window to work at such a high potential and also leads to safety issues due to the high chemical activity of the Ni4+. In addition, anodic substitutions like F doping lead to an increase in the average valence of Mn and thus a higher Mn3+/Mn4+ ratio, but this doping is capable of stabilizing the crystal structure.92 Therefore, the combination of anionic and cationic doping enhances both the capacity and the cycling performance of the spinel.96

One previous work of our group reported a surface doping concept, taking advantage of both surface coating and bulk doping.97 In the surface doping approach, an electrochemically inactive cation, for instance Ti4+, is incorporated into the surface (a few nanometers thick only) of the spinel LiMn2O4 to form a cation-doped surface layer, as shown in Fig. 11. Similar to the bulk doping approach, the surface doping layer stabilizes the spinel crystal structure, and also protects the bulk LiMn2O4 from acid corrosion like a surface coating layer. However, unlike a surface coating layer, which imposes a physical barrier, the surface-doped layer maintains the electrochemically active spinel structure and therefore maintains the charge transportation channels on the surface. Moreover, the similarity of the crystal structure of the bulk and the surface doping layer minimized the possible phase segregation or separation upon cycling. As a consequence, the surface-doped LiMn2O4 demonstrates an enhanced electrochemical performance in terms of both cyclability and capacity at elevated temperatures.


image file: c7ee03122j-f11.tif
Fig. 11 High-resolution TEM images of the surface-treated LiMn2O4 particles. (a and b) High-resolution TEM images of the surface-doped LiMn2O4 particles showing uniform structures from the surface to the interior (scale bar, 2 nm). (c) Selected area diffraction pattern along [112] for the region shown in (b). (d) TEM image with regions indicated for the corresponding EELS data shown in (e) (scale bar, 5 nm). (f) High-resolution TEM image of the surface-coated LiMn2O4 particles (scale bar, 5 nm). (g) EELS spectra from regions A and B in (f). Reprinted with permission;97 Copyright © 2013 Macmillan Publishers Limited.

Besides cathode modification, a few electrolyte additives, such as butylamine,98 N,N′′-dicyclohexylcarbodiimide,99 N,N-diethylamino trimethylsilane,100 and hexamethyldisilazane,101 were also applied to suppress Mn dissolution by scavenging the water and/or HF impurities in the electrolyte. Another approach to protect the cathode using electrolyte additives is to form a protective film on the cathode. For instance, lithium bis(oxalato)borate (LiBOB) was found to effectively suppress Mn dissolution from Li2MnO4,19 and it is proposed that the BOB and Mn2+ coordinate with each other and form an insoluble and stable surface layer on the cathode.102 In addition, some very interesting work was also carried out by using separators containing HF-scavengers, such as CaCO3103 and pyridine,104 to reduce the Mn dissolution from LiMn2O4.

5. Suppressing Mn deposition

Manganese deposition can be effectively suppressed by preventing direct contact between the anode and the Mn2+ in the electrolyte, but there are only a few studies applying this approach. Xiao et al.105 carried out Al2O3 coating on both the LiMn1.5Ni0.5O4 cathode and the graphite anode to mitigate the Mn DMD process. In Mn 2p XPS of the graphite anode after cycling, the surface coating of the anode was found to be much more beneficial in preventing Mn deposition. As a result, the surface-coated anode exhibited better capacity retention upon cycling.

Using an additive to alter the morphology and composition of the SEI layer is one of the most economical ways to improve the cycling performance of Li-ion batteries. But understanding how these additives affect the Mn deposition reaction is still lacking. Komaba et al.54 found that 2-vinylpyridine (VP) formed a polymer layer at ca. 0.9 V before the typical SEI formation at ca. 0.8 V, and an SEI modified by poly(2-vinylpyridine) blocked Mn reduction, but a vinylene carbonate (VC) additive hardly eliminated the degradation caused by Mn(II). Cho et al.106 observed that VC is able to suppress the Mn deposition, the impedance increase, the open-circuit potential change of graphite, and the overall capacity loss of the spinel/graphite system at the fully charged state, while fluoroethylene carbonate (FEC) does not. They explain that the VC helps to form an SEI layer containing polymer-like compounds, which protects the anode against the attack of Mn ions in the electrolyte, while the FEC favorable for the SEI layer formation mostly consists of LiF, which is not so protective, as shown in Fig. 12. One of our recent studies showed that the amount of Mn deposited on the anode is increased due to the VC or FEC additives. The improvement in the cycling performance should be attributed to the enhanced robustness of the additive-modified SEI layer, according to in situ electrochemical impedance spectroscopy. During the SEI formation process, the VC or FEC increases the concentration of mobile Li ions in the SEI via polymerization reactions. As a result, the additive-modified SEI layer is able to maintain sufficient Li transportation paths even if more Mn ions deposit on the anode via Li–Mn ion exchange.107


image file: c7ee03122j-f12.tif
Fig. 12 Schematics for the functional roles of the FEC- and VC-derived SEI. Reprinted with permission;106 Copyright © 2005 The Electrochemical Society.

Some organic or inorganic ligands, such as cyclic ethers108 and NH4+,109 are able to lower the chemical activity of Mn2+ by forming Mn2+ complexes, so that the deposition potential of Mn from the complex can be decreased. Cyclic ethers such as 1 wt% 1,4,7,10,13,16-hexaoxacyclooctadecane (18C6) and 1.4 wt% 4,7,13,16,21,24-hexaoxa-1,10-diazabicyclo hexacosane (C222) were found to effectively suppress the Mn deposition (detected by XPS as shown in Fig. 13) and the consequent impedance rise of the HOPG electrode cycled in Mn2+-containing electrolyte.108 In addition, NH4+ was reported to remarkably improve the coulombic efficiency of a graphite anode in the electrolyte with Mn(ClO4)2.109 Although it seems reasonable that the Mn2+ additive would trap the Mn2+ in the electrolyte and suppress the Mn deposition on the anode, this effect has not been confirmed in the literature. These additives can eventually improve the cycling performance, especially for full cells, likely due to the electrochemical instability of the Mn2+ at the operation potentials of the electrodes.


image file: c7ee03122j-f13.tif
Fig. 13 Mn 2p spectra of the edge plane HOPG electrodes after the 5th potential cycle and AC impedance measurements in 1 M LiClO4/EC + DEC containing 2 wt% VC and 100 ppm Mn with and without 1 wt% 18C6 or 1.4 wt% C222. The HOPG electrode was removed from the electrochemical cells after they were left in an open circuit condition for 1 h. Reprinted with permission;108 Copyright © 2013 The Electrochemical Society.

Additive ions with specific adsorption on the surface of Mn metal were also suggested to overcome the capacity fading induced by Mn deposition. As the Mn metal showed less electrochemical activity in an electrolyte with I or Br, Komaba et al.109 believed that these two ions can suppress the irreversible reactions on the Mn(0)/electrolyte interphase by specific adsorption on the surface of Mn metal deposited on the anode. In this case, the Mn deposition is not inhibited, but the electrocatalytic effects are suppressed.

6. DMD in Ni-, Co-, Fe- and V-containing cathodes

The correlation between the Mn DMD mechanism and the capacity fade of the Mn-based cathodes provides a guideline to investigate the instability of Ni-, Co-, or Fe-containing cathodes, especially at elevated temperatures. Cobalt dissolution from LiCoO2,110 the first successfully commercialized LIB cathode material, was detected following the early studies of Mn dissolution. The Co loss from the LiCoO2 cathode was found to increase sharply when the potential was higher than 4.2 V. This loss was well correlated with the irreversible transition from the rhombohedral phase to the monoclinic phase when x < 0.5 in Li1−xCoO2. The concurrent migration of Co ions to the Li layer is a possible route for the observed dissolution. Besides high potential, Co dissolution is also enhanced in the electrolyte with HF contaminants, high temperatures, and a polyvinylidene fluoride binder in the composite electrode.111 As CoO2 and Co3O4 were detected on the surface of aged LiCoO2, the following reactions were proposed to elucidate the Co dissolution:
 
image file: c7ee03122j-t5.tif(13)
 
Li2O + 2HF → 2LiF + H2O (14)

Derived from LiCoO2, mixed nickel–manganese–cobalt dioxides (LiNi1−yzMnyCozO2) have been the fastest developing commercial cathodes in recent years due to their balanced capacity and stability, and they are model materials to understand the Mn/Co/Ni DMD process. TM dissolution from LiNi1/3Co1/3Mn1/3O2 (NCM) increased dramatically at a potential higher than 4.5–4.6 V, at which a considerable capacity loss occurred.112 The causal links between TM dissolution and the oxidative potential could be (i) increasingly exposed surfaces due to particle cracks; (ii) the widening of the TM–O band due to the oxygen loss;113 and (iii) the possible phase transition at deep delithiation, which is similar to the case of LiCoO2.114 The concentration of the three TM ions dissolved in the electrolyte showed the following decreasing order: Mn > Co > Ni. The low Ni dissolution could be related to the stable NiO-like rocksalt phase generated at high potential via loss of oxygen.108,115,116 TM deposition was detected on the anode by SIMS,117 and the TM oxidation states were identified to be 2+ by operando XAS.118 The content of Mn is higher than Co and Ni, consistent with the TM concentration in the electrolyte. TM deposition induced a dramatic impedance rise, which could be explained by the change in morphology, constitution, or thickness of the SEI layer on the anode.26,117

Iron dissolution and deposition have been studied to understand the poor cycling performance of the LiFePO4 electrode at elevated temperatures. With regard to LIBs, LiFePO4 is the first cathode material with low cost and plentiful, environmentally benign elements.119 It has excellent stability at room temperature in LiPF6 electrolyte, while at elevated temperatures, the stability of LiFePO4 in HF-containing electrolytes depends upon the synthesis route and conditions used for processing.111,120 Surface corrosion and iron dissolution were observed directly using field-emission scanning electron microscopy (SEM) and EDX on a flat surface of the LiFePO4 particles, and the carbon coating is confirmed to significantly improve the stability of LiFePO4.121 Iron deposition on the graphite anode surfaces was detected by EDX,122 which correlated well with the impedance rise of the anode upon cycling at elevated temperature.123

Vanadium-based oxides were among the earliest studied cathode materials for LIBs in the 1970s,124 and have been intensively investigated since then due to their high capacity, low cost and high abundance in the Earth's crust.125,126 The main focus of the research on V-based cathodes is to improve the cycling stability and rate capacity.127,128 V dissolution from V2O5 was confirmed by the detection of V on the Li metal anode by EDS129 and the color change of the electrolyte after cycling.130,131 This V dissolution is believed to be one of the main reasons for the poor stability of this cathode.132

7. Challenges and perspectives

Up to now, in spite of the numerous studies on TM DMD, the detailed mechanism of the dissolution and deposition process is still under debate. The main difficulty is detecting and identifying the TM dissolved in the electrolyte and deposited on the anode without altering their status in the electrochemical environment. With the development of advanced analytical methods such as in situ X-ray and neutron techniques, more reliable and conclusive results can be expected to give further insights into the DMD process and thus an improved understanding of how it causes the capacity and impedance fade.

Moreover, the correlation between the phase stability and TM dissolution from the cathode has not been paid enough attention. For instance, most of the work on suppression of Mn dissolution to prevent cathode/electrolyte contact or increase the average valence of the Mn ions is based on the disproportionation mechanism. In fact, the phase transition induced Mn(III) dissolution was reported as early as 199721 and has been supported by recent work,31 as mentioned above in Section 2. There has been little subsequent work on stabilizing the two-phase structure. A recent study on layered Li(NixMnyCoz)O2 confirmed that the phase stability at an elevated temperature and high potential can be improved by Li–TM disorder.133 Therefore, it can be expected that tuning the Li–Mn disorder in spinel LiMn2O4 may stabilize the two-phase structure and consequently suppress the Mn dissolution.

8. Conclusions

Overcoming the electrochemical performance degradation induced by TM dissolution–migration–deposition is crucial to transform LIBs into power sources for EVs, meeting the requirements of long lifetime and a wide working temperature range. In this paper, using spinel LiMn2O4 as the model electrode, we reviewed research activities on the mechanism and solutions to overcome the TM DMD process, from the earliest discoveries to the latest progress. The DMD process generally occurs in the following steps: (1) TM ions dissolve into the electrolyte from the cathode via a disproportionation reaction or phase transition; (2) the TM ions diffuse and/or migrate from the cathode to the anode through the electrolyte; and (3) the TM ions are deposited on the anode and accumulate. The impedance rise of the anode induced by TM deposition is the primary reason for the overall impedance increase and capacity fade with cycling. The main strategy to prolong the durability of the TM-based cathodes is to suppress the DMD process, including the structural/surficial modification of the cathode, electrolyte additives, and the surface coating of the anode. Bulk doping and surface coating of the cathode have been practically applied in the battery industry, while the solutions directly dealing with TM deposition are still in a preliminary stage. A deeper understanding of the TM deposition mechanism by the application of state-of-the-art techniques with critical thinking will help to open up new avenues to understand and eventually manipulate the TM DMD process.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors gratefully acknowledge support from the U. S. Department of Energy (DOE), Office of Energy Efficiency and Renewable Energy, Vehicle Technologies Office. Argonne National Laboratory is operated for DOE Office of Science by UChicago Argonne, LLC, under contract number DE-AC02-06CH11357.

Notes and references

  1. G. N. Lewis and F. G. Keyes, J. Am. Chem. Soc., 1913, 35, 340–344 CrossRef CAS.
  2. M. S. Whittinghan, Science, 1976, 192, 1126–1127 Search PubMed.
  3. M. Lazzari and B. Scrosati, J. Electrochem. Soc., 1980, 127, 773–774 CrossRef CAS.
  4. A. Yoshino, K. Sanechika and T. Nakajima, Jp. Pat., JP1989293, 1985 Search PubMed.
  5. K. Mizushima, P. C. Jones, P. J. Wiseman and J. B. Goodenough, Mater. Res. Bull., 1980, 15, 783–789 CrossRef CAS.
  6. R. Yazami and P. Touzain, J. Power Sources, 1983, 9, 365–371 CrossRef CAS.
  7. J. Vetter, P. Novák, M. R. Wagner, C. Veit, K. C. Möller, J. O. Besenhard, M. Winter, M. Wohlfahrt-Mehrens, C. Vogler and A. Hammouche, J. Power Sources, 2005, 147, 269–281 CrossRef CAS.
  8. M. M. Thackeray, W. I. F. David, P. G. Bruce and J. B. Goodenough, Mater. Res. Bull., 1983, 18, 461–472 CrossRef CAS.
  9. M. M. Thackeray, P. J. Johnson, L. A. de Picciotto, P. G. Bruce and J. B. Goodenough, Mater. Res. Bull., 1984, 19, 179–187 CrossRef CAS.
  10. J. M. Tarascon, E. Wang, F. K. Shokoohi, W. R. McKinnon and S. Colson, J. Electrochem. Soc., 1991, 138, 2859–2864 CrossRef CAS.
  11. A. K. Padhi, K. Nanjundaswamy and J. Goodenough, J. Electrochem. Soc., 1997, 144, 1188–1194 CrossRef CAS.
  12. K. Shaju, G. S. Rao and B. Chowdari, Electrochim. Acta, 2002, 48, 145–151 CrossRef CAS.
  13. M. E. Spahr, P. Novák, B. Schnyder, O. Haas and R. Nesper, J. Electrochem. Soc., 1998, 145, 1113–1121 CrossRef CAS.
  14. Z. Liu, A. Yu and J. Y. Lee, J. Power Sources, 1999, 81–82, 416–419 CrossRef CAS.
  15. M. Yoshio, H. Noguchi, J.-i. Itoh, M. Okada and T. Mouri, J. Power Sources, 2000, 90, 176–181 CrossRef CAS.
  16. Z. Lu, D. MacNeil and J. Dahn, Electrochem. Solid-State Lett., 2001, 4, A191–A194 CrossRef CAS.
  17. M. M. Thackeray, S.-H. Kang, C. S. Johnson, J. T. Vaughey, R. Benedek and S. A. Hackney, J. Mater. Chem., 2007, 17, 3112–3125 RSC.
  18. A. S. Andersson, J. O. Thomas, B. Kalska and L. Häggström, Electrochem. Solid-State Lett., 2000, 3, 66–68 CrossRef CAS.
  19. K. Amine, J. Liu, S. Kang, I. Belharouak, Y. Hyung, D. Vissers and G. Henriksen, J. Power Sources, 2004, 129, 14–19 CrossRef CAS.
  20. S.-C. Park, Y.-M. Kim, S.-C. Han, S. Ahn, C.-H. Ku and J.-Y. Lee, J. Power Sources, 2002, 107, 42–47 CrossRef CAS.
  21. Y. Xia, Y. Zhou and M. Yoshio, J. Electrochem. Soc., 1997, 144, 2593–2600 CrossRef CAS.
  22. D. Aurbach, M. D. Levi, K. Gamulski, B. Markovsky, G. Salitra, E. Levi, U. Heider, L. Heider and R. Oesten, J. Power Sources, 1999, 81–82, 472–479 CrossRef CAS.
  23. J. R. Dahn, E. W. Fuller, M. Obrovac and U. von Sacken, Solid State Ionics, 1994, 69, 265–270 CrossRef CAS.
  24. Z. Zhang, D. Fouchard and J. R. Rea, J. Power Sources, 1998, 70, 16–20 CrossRef CAS.
  25. A. Blyr, A. Du Pasquier, G. Amatucci and J. M. Tarascon, Ionics, 1997, 3, 321–331 CrossRef CAS.
  26. C. Zhan, J. Lu, A. Jeremy Kropf, T. Wu, A. N. Jansen, Y.-K. Sun, X. Qiu and K. Amine, Nat. Commun., 2013, 4, 2437 Search PubMed.
  27. I. A. Shkrob, A. J. Kropf, T. W. Marin, Y. Li, O. G. Poluektov, J. Niklas and D. P. Abraham, J. Phys. Chem. C, 2014, 118, 24335–24348 CAS.
  28. J. C. Hunter, J. Solid State Chem., 1981, 39, 142–147 CrossRef CAS.
  29. D. H. Jang, Y. J. Shin and S. M. Oh, J. Electrochem. Soc., 1996, 143, 2204–2211 CrossRef CAS.
  30. Y. Terada, Y. Nishiwaki, I. Nakai and F. Nishikawa, J. Power Sources, 2001, 97-8, 420–422 CrossRef.
  31. A. Banerjee, Y. Shilina, B. Ziv, J. M. Ziegelbauer, S. Luski, D. Aurbach and I. C. Halalay, J. Am. Chem. Soc., 2017, 139, 1738–1741 CrossRef CAS PubMed.
  32. L.-F. Wang, C.-C. Ou, K. A. Striebel and J.-S. Chen, J. Electrochem. Soc., 2003, 150, A905–A911 CrossRef CAS.
  33. D. H. Jang and S. M. Oh, J. Electrochem. Soc., 1997, 144, 3342–3348 CrossRef CAS.
  34. M. Hirayama, H. Ido, K. Kim, W. Cho, K. Tamura, J. i. Mizuki and R. Kanno, J. Am. Chem. Soc., 2010, 132, 15268–15276 CrossRef CAS PubMed.
  35. V. Manev, B. Banov, A. Momchilov and A. Nassalevska, J. Power Sources, 1995, 57, 99–103 CrossRef CAS.
  36. A. Momchilov, V. Manev, A. Nassalevska and A. Kozawa, J. Power Sources, 1993, 41, 305–314 CrossRef CAS.
  37. R. Benedek and M. Thackeray, Electrochem. Solid-State Lett., 2006, 9, A265–A267 CrossRef CAS.
  38. J. Park, J. H. Seo, G. Plett, W. Lu and A. M. Sastry, Electrochem. Solid-State Lett., 2011, 14, A14–A18 CrossRef CAS.
  39. K. Suzuki, Y. Oumi, S. Takami, M. Kubo, A. Miyamoto, M. Kikuchi, N. Yamazaki and M. Mita, Jpn. J. Appl. Phys., 2000, 39, 4318 CrossRef CAS.
  40. R. J. Gummow, A. Dekock and M. M. Thackeray, Solid State Ionics, 1994, 69, 59–67 CrossRef CAS.
  41. A. Du Pasquier, A. Blyr, P. Courjal, D. Larcher, G. Amatucci, B. Gerand and J. M. Tarascon, J. Electrochem. Soc., 1999, 146, 428–436 CrossRef CAS.
  42. D. Tang, Y. Sun, Z. Yang, L. Ben, L. Gu and X. Huang, Chem. Mater., 2014, 26, 3535–3543 CrossRef CAS.
  43. S. S. Zhang, Front. Energy Res., 2014, 2, 59 Search PubMed.
  44. W. Choi and A. Manthiram, J. Electrochem. Soc., 2006, 153, A1760–A1764 CrossRef CAS.
  45. C. S. Johnson, N. Li, C. Lefief, J. T. Vaughey and M. M. Thackeray, Chem. Mater., 2008, 20, 6095–6106 CrossRef CAS.
  46. C. Johnson, M. Mansuetto, M. Thackeray, Y. Shao-Horn and S. Hackney, J. Electrochem. Soc., 1997, 144, 2279–2283 CrossRef CAS.
  47. M. M. Thackeray, M. H. Rossouw, A. de Kock, A. P. de la Harpe, R. J. Gummow, K. Pearce and D. C. Liles, J. Power Sources, 1993, 43, 289–300 CrossRef CAS.
  48. B. Lee, C. S. Yoon, H. R. Lee, K. Y. Chung, B. W. Cho and S. H. Oh, Sci. Rep., 2014, 4, 6066 CrossRef CAS PubMed.
  49. Surendra K. Martha, J. Grinblat, O. Haik, E. Zinigrad, T. Drezen, James H. Miners, I. Exnar, A. Kay, B. Markovsky and D. Aurbach, Angew. Chem., Int. Ed., 2009, 48, 8559–8563 CrossRef CAS PubMed.
  50. J. Tarascon, W. McKinnon, F. Coowar, T. Bowmer, G. Amatucci and D. Guyomard, J. Electrochem. Soc., 1994, 141, 1421–1431 CrossRef CAS.
  51. A. Blyr, C. Sigala, G. Amatucci, D. Guyomard, Y. Chabre and J. M. Tarascon, J. Electrochem. Soc., 1998, 145, 194–209 CrossRef CAS.
  52. D. Abraham, T. Spila, M. Furczon and E. Sammann, Electrochem. Solid-State Lett., 2008, 11, A226–A228 CrossRef CAS.
  53. H. Tsunekawa, S. Tanimoto, R. Marubayashi, M. Fujita, K. Kifune and M. Sano, J. Electrochem. Soc., 2002, 149, A1326–A1331 CrossRef CAS.
  54. S. Komaba, T. Itabashi, T. Ohtsuka, H. Groult, N. Kumagai, B. Kaplan and H. Yashiro, J. Electrochem. Soc., 2005, 152, A937–A946 CrossRef CAS.
  55. S. R. Gowda, K. G. Gallagher, J. R. Croy, M. Bettge, M. M. Thackeray and M. Balasubramanian, Phys. Chem. Chem. Phys., 2014, 16, 6898–6902 RSC.
  56. N. Kumagai, S. Komaba, Y. Kataoka and M. Koyanagi, Chem. Lett., 2000, 1154–1155 CrossRef CAS.
  57. S. Komaba, N. Kumagai and Y. Kataoka, Electrochim. Acta, 2002, 47, 1229–1239 CrossRef CAS.
  58. C. Delacourt, A. Kwong, X. Liu, R. Qiao, W. Yang, P. Lu, S. Harris and V. Srinivasan, J. Electrochem. Soc., 2013, 160, A1099–A1107 CrossRef CAS.
  59. M. Ochida, Y. Domi, T. Doi, S. Tsubouchi, H. Nakagawa, T. Yamanaka, T. Abe and Z. Ogumi, J. Electrochem. Soc., 2012, 159, A961–A966 CrossRef CAS.
  60. X. Xiao, Z. Liu, L. Baggetto, G. M. Veith, K. L. More and R. R. Unocic, Phys. Chem. Chem. Phys., 2014, 16, 10398–10402 RSC.
  61. K. M. Shaju and P. G. Bruce, Chem. Mater., 2008, 20, 5557–5562 CrossRef CAS.
  62. E. Hosono, T. Kudo, I. Honma, H. Matsuda and H. Zhou, Nano Lett., 2009, 9, 1045–1051 CrossRef CAS PubMed.
  63. R. Benedek, M. M. Thackeray, J. Low and T. Bučko, J. Phys. Chem. C, 2012, 116, 4050–4059 CAS.
  64. G. G. Amatucci, C. N. Schmutz, A. Blyr, C. Sigala, A. S. Gozdz, D. Larcher and J. M. Tarascon, J. Power Sources, 1997, 69, 11–25 CrossRef CAS.
  65. G. G. Amatucci, A. Blyr, C. Sigala, P. Alfonse and J. M. Tarascon, Solid State Ionics, 1997, 104, 13–25 CrossRef CAS.
  66. Y.-K. Sun, K.-J. Hong and J. Prakash, J. Electrochem. Soc., 2003, 150, A970–A972 CrossRef CAS.
  67. M. M. Thackeray, C. S. Johnson, J. S. Kim, K. C. Lauzze, J. T. Vaughey, N. Dietz, D. Abraham, S. A. Hackney, W. Zeltner and M. A. Anderson, Electrochem. Commun., 2003, 5, 752–758 CrossRef CAS.
  68. J.-S. Kim, C. S. Johnson, J. T. Vaughey, S. A. Hackney, K. A. Walz, W. A. Zeltner, M. A. Anderson and M. M. Thackeray, J. Electrochem. Soc., 2004, 151, A1755–A1761 CrossRef CAS.
  69. Z. R. Zhang, Z. L. Gong and Y. Yang, J. Phys. Chem. B, 2004, 108, 17546–17552 CrossRef CAS.
  70. J.-M. Han, S.-T. Myung and Y.-K. Sun, J. Electrochem. Soc., 2006, 153, A1290–A1295 CrossRef CAS.
  71. S. Lim and J. Cho, Electrochem. Commun., 2008, 10, 1478–1481 CrossRef CAS.
  72. Z. Yang, W. Yang, D. G. Evans, Y. Zhao and X. Wei, J. Power Sources, 2009, 189, 1147–1153 CrossRef CAS.
  73. K. A. Walz, C. S. Johnson, J. Genthe, L. C. Stoiber, W. A. Zeltner, M. A. Anderson and M. M. Thackeray, J. Power Sources, 2010, 195, 4943–4951 CrossRef CAS.
  74. H. Sclar, O. Haik, T. Menachem, J. Grinblat, N. Leifer, A. Meitav, S. Luski and D. Aurbach, J. Electrochem. Soc., 2012, 159, A228–A237 CrossRef CAS.
  75. S.-C. Park, Y.-S. Han, Y.-S. Kang, P. S. Lee, S. Ahn, H.-M. Lee and J.-Y. Lee, J. Electrochem. Soc., 2001, 148, A680–A686 CrossRef CAS.
  76. Z. Liu, H. Wang, L. Fang, J. Y. Lee and L. M. Gan, J. Power Sources, 2002, 104, 101–107 CrossRef CAS.
  77. D. Q. Liu, X. Q. Liu and Z. Z. He, Mater. Chem. Phys., 2007, 105, 362–366 CrossRef CAS.
  78. J. Cao, G. Cao, H. Yu, J. Xie and X. Zhao, Rare Met., 2011, 30, 39–43 CrossRef CAS.
  79. J. Li, L. Wang, Q. Zhang and X. He, J. Power Sources, 2009, 190, 149–153 CrossRef CAS.
  80. S. Zhao, Y. Bai, Q. Chang, Y. Yang and W. Zhang, Electrochim. Acta, 2013, 108, 727–735 CrossRef CAS.
  81. R. Vidu and P. Stroeve, Ind. Eng. Chem. Res., 2004, 43, 3314–3324 CrossRef CAS.
  82. C. Qing, Y. Bai, J. Yang and W. Zhang, Electrochim. Acta, 2011, 56, 6612–6618 CrossRef CAS.
  83. D. Liu, Z. He and X. Liu, Mater. Lett., 2007, 61, 4703–4706 CrossRef CAS.
  84. T.-F. Yi, Y.-R. Zhu, X.-D. Zhu, J. Shu, C.-B. Yue and A.-N. Zhou, Ionics, 2009, 15, 779–784 CrossRef CAS.
  85. N. M. Hagh, F. Cosandey, S. Rangan, R. Bartynski and G. G. Amatucci, J. Electrochem. Soc., 2010, 157, A305–A319 CrossRef CAS.
  86. D. Song, H. Ikuta, T. Uchida and M. Wakihara, Solid State Ionics, 1999, 117, 151–156 CrossRef.
  87. Y.-S. Lee, N. Kumada and M. Yoshio, J. Power Sources, 2001, 96, 376–384 CrossRef CAS.
  88. I. Taniguchi, D. Song and M. Wakihara, J. Power Sources, 2002, 109, 333–339 CrossRef CAS.
  89. K. W. Kim, S.-W. Lee, K.-S. Han, H. J. Chung and S. I. Woo, Electrochim. Acta, 2003, 48, 4223–4231 CrossRef CAS.
  90. B. Markovsky, Y. Talyossef, G. Salitra, D. Aurbach, H.-J. Kim and S. Choi, Electrochem. Commun., 2004, 6, 821–826 CrossRef CAS.
  91. D. Aurbach, B. Markovsky, Y. Talyossef, G. Salitra, H.-J. Kim and S. Choi, J. Power Sources, 2006, 162, 780–789 CrossRef CAS.
  92. A. Gutierrez and A. Manthiram, J. Electrochem. Soc., 2013, 160, A901–A905 CrossRef CAS.
  93. J. C. Arrebola, A. Caballero, M. Cruz, L. Hernán, J. Morales and E. R. Castellón, Adv. Funct. Mater., 2006, 16, 1904–1912 CrossRef CAS.
  94. L. Zhou, D. Zhao and X. D. Lou, Angew. Chem., Int. Ed., 2012, 124, 243–245 CrossRef.
  95. J. Xiao, X. Chen, P. V. Sushko, M. L. Sushko, L. Kovarik, J. Feng, Z. Deng, J. Zheng, G. L. Graff and Z. Nie, Adv. Mater., 2012, 24, 2109–2116 CrossRef CAS PubMed.
  96. Q. Luo and A. Manthiram, J. Electrochem. Soc., 2009, 156, A84–A88 CrossRef CAS.
  97. J. Lu, C. Zhan, T. Wu, J. Wen, Y. Lei, A. J. Kropf, H. Wu, D. J. Miller, J. W. Elam, Y.-K. Sun, X. Qiu and K. Amine, Nat. Commun., 2014, 5, 5693 CrossRef CAS PubMed.
  98. M. Saidi, F. Gao, J. Barker, C. Scordilis-Kelley and M. Y. Saidi, US Pat., US5846673-A, 1998 Search PubMed.
  99. A. Koiwai, T. Shiga and K. Takechi, US Pat., US6077628-A, 2000 Search PubMed.
  100. K. Takechi and T. Shiga, US Pat., US6235431-B1, 2001 Search PubMed.
  101. H. Yamane, T. Inoue, M. Fujita and M. Sano, J. Power Sources, 2001, 99, 60–65 CrossRef CAS.
  102. S. S. Zhang, J. Power Sources, 2006, 162, 1379–1394 CrossRef CAS.
  103. S. S. Zhang, K. Xu and T. R. Jow, J. Solid State Electrochem., 2003, 7, 492–496 CrossRef CAS.
  104. A. Banerjee, B. Ziv, Y. Shilina, S. Luski, D. Aurbach and I. C. Halalay, ACS Energy Lett., 2017, 2, 2388–2393 CrossRef CAS.
  105. X. Xiao, D. Ahn, Z. Liu, J.-H. Kim and P. Lu, Electrochem. Commun., 2013, 32, 31–34 CrossRef CAS.
  106. I. H. Cho, S. S. Kim, S. C. Shin and N. S. Choi, Electrochem. Solid-State Lett., 2010, 13, A168–A172 CrossRef CAS.
  107. C. Zhan, X. Qiu, J. Lu and K. Amine, Adv. Mater. Interfaces, 2016, 3, 1500856 CrossRef.
  108. M. Ochida, T. Doi, Y. Domi, S. Tsubouchi, H. Nakagawa, T. Yamanaka, T. Abe and Z. Ogumi, J. Electrochem. Soc., 2013, 160, A410–A413 CrossRef CAS.
  109. S. Komaba, B. Kaplan, T. Ohtsuka, Y. Kataoka, N. Kumagai and H. Groult, J. Power Sources, 2003, 119-121, 378–382 CrossRef CAS.
  110. G. G. Amatucci, J. M. Tarascon and L. C. Klein, Solid State Ionics, 1996, 83, 167–173 CrossRef CAS.
  111. D. Aurbach, B. Markovsky, G. Salitra, E. Markevich, Y. Talyossef, M. Koltypin, L. Nazar, B. Ellis and D. Kovacheva, J. Power Sources, 2007, 165, 491–499 CrossRef CAS.
  112. H. Zheng, Q. Sun, G. Liu, X. Song and V. S. Battaglia, J. Power Sources, 2012, 207, 134–140 CrossRef CAS.
  113. J. A. Gilbert, I. A. Shkrob and D. P. Abraham, J. Electrochem. Soc., 2017, 164, A389–A399 CrossRef CAS.
  114. I. Buchberger, S. Seidlmayer, A. Pokharel, M. Piana, J. Hattendorff, P. Kudejova, R. Gilles and H. A. Gasteiger, J. Electrochem. Soc., 2015, 162, A2737–A2746 CrossRef CAS.
  115. S. Watanabe, M. Kinoshita, T. Hosokawa, K. Morigaki and K. Nakura, J. Power Sources, 2014, 258, 210–217 CrossRef CAS.
  116. D. P. Abraham, R. D. Twesten, M. Balasubramanian, J. Kropf, D. Fischer, J. McBreen, I. Petrov and K. Amine, J. Electrochem. Soc., 2003, 150, A1450–A1456 CrossRef CAS.
  117. D. P. Abraham, T. Spila, M. M. Furczon and E. Sammann, Electrochem. Solid-State Lett., 2008, 11, A226–A228 CrossRef CAS.
  118. J. Wandt, A. Freiberg, R. Thomas, Y. Gorlin, A. Siebel, R. Jung, H. A. Gasteiger and M. Tromp, J. Mater. Chem. A, 2016, 4, 18300–18305 CAS.
  119. M. S. Whittingham, Chem. Rev., 2004, 104, 4271–4302 CrossRef CAS PubMed.
  120. M. Koltypin, D. Aurbach, L. Nazar and B. Ellis, J. Power Sources, 2007, 174, 1241–1250 CrossRef CAS.
  121. J. Wang, J. Yang, Y. Tang, R. Li, G. Liang, T.-K. Sham and X. Sun, J. Mater. Chem. A, 2013, 1, 1579–1586 CAS.
  122. H. Song, Z. Cao, X. Chen, H. Lu, M. Jia, Z. Zhang, Y. Lai, J. Li and Y. Liu, J. Solid State Electrochem., 2013, 17, 599–605 CrossRef CAS.
  123. K. Amine, J. Liu and I. Belharouak, Electrochem. Commun., 2005, 7, 669–673 CrossRef CAS.
  124. M. S. Whittingham, J. Electrochem. Soc., 1976, 123, 315–320 CrossRef CAS.
  125. Y. Liu, E. Uchaker, N. Zhou, J. Li, Q. Zhang and G. Cao, J. Mater. Chem., 2012, 22, 24439–24445 RSC.
  126. M.-S. Balogun, Y. Luo, F. Lyu, F. Wang, H. Yang, H. Li, C. Liang, M. Huang, Y. Huang and Y. Tong, ACS Appl. Mater. Interfaces, 2016, 8, 9733–9744 CAS.
  127. X. Xu, Y.-Z. Luo, L.-Q. Mai, Y.-L. Zhao, Q.-Y. An, L. Xu, F. Hu, L. Zhang and Q.-J. Zhang, NPG Asia Mater., 2012, 4, e20 CrossRef.
  128. D. Chao, X. Xia, J. Liu, Z. Fan, C. F. Ng, J. Lin, H. Zhang, Z. X. Shen and H. J. Fan, Adv. Mater., 2014, 26, 5794–5800 CrossRef CAS PubMed.
  129. G. Sudant, E. Baudrin, B. Dunn and J.-M. Tarascon, J. Electrochem. Soc., 2004, 151, A666–A671 CrossRef CAS.
  130. S.-R. Li, S.-Y. Ge, Y. Qiao, Y.-M. Chen, X.-Y. Feng, J.-F. Zhu and C.-H. Chen, Electrochim. Acta, 2012, 64, 81–86 CrossRef CAS.
  131. W. Cheng, G. Zeng and M. Niederberger, J. Mater. Chem. A, 2015, 3, 2861–2868 CAS.
  132. N. A. Chernova, M. Roppolo, A. C. Dillon and M. S. Whittingham, J. Mater. Chem., 2009, 19, 2526–2552 RSC.
  133. J. Zheng, T. Liu, Z. Hu, Y. Wei, X. Song, Y. Ren, W. Wang, M. Rao, Y. Lin, Z. Chen, J. Lu, C. Wang, K. Amine and F. Pan, J. Am. Chem. Soc., 2016, 138, 13326–13334 CrossRef CAS PubMed.

This journal is © The Royal Society of Chemistry 2018