[PDF] High-voltage positive electrode materials for lithium-ion batteries

View - Download High-voltage positive electrode materials for lithium-ion batteries

Previous PDF

Next PDF

1

High-voltage positive electrode materials for

lithium-ion batteries

Wangda Li, Bohang Song, and Arumugam Manthiram*

Materials Science and Engineering Program and Texas Materials Institute

University of Texas at Austin

Austin, Texas 78712, USA

The ever-growing demand for advanced rechargeable lithium-ion batteries in portable electronics and electric vehicles has spurred intensive research efforts over the past decade. The key to sustain the progress in Li-ion batteries lies in the quest for safe, low-cost positive electrode (cathode) materials with desirable energy and power capabilities. One approach to boost the energy and power densities of the battery is to increase the output voltage while maintaining the high capacity, fast charge-discharge rate, and long service life. This review gives an account of the various emerging high-voltage positive electrode materials that have the potential to satisfy these requirements either in the short or long term, including nickel-rich layered oxides, lithium-rich layered oxides, high-voltage spinel oxides, and high- voltage polyanionic compounds. The key barriers and the corresponding strategies for the practical viability of these cathode materials are discussed along with the optimization of electrolytes and other cell components, with a particular emphasis on recent advances in the literature. A concise perspective with respect to plausible strategies for future developments in the field is also provided. ___________________________________ *Corresponding author: manth@austin.utexas.edu 2

Table of Contents

1. Introduction ...................................................................................................................................................... 4

2. The general implications of high-voltage electrochemical operation ...................................................... 7

2.1. Parasitic electrolyte oxidation at charged positive electrodes ............................................................ 9

2.2. Dissolution of active positive electrode materials .............................................................................13

2.3. Surface structural reconstruction and mechanical fracture of positive electrode materials .......15

2.4. Chemical crossover between the positive and negative electrodes ................................................17

2.5. Spontaneous reactions involving conductive carbon additives and their anodic instability at

high voltages ...................................................................................................................................................20

2.6. Instability of other cell components at extreme potentials .............................................................23

2.7. Exothermic breakdown of interphases on electrodes ......................................................................27

3. Nickel-rich layered LiNi1-xMxO2 ..................................................................................................................27

3.1. Structural and compositional aspects of LiNi1-xMxO2 ......................................................................29

3.2. Strategies to improve the cyclability of high-energy-density LiNi1-xMxO2 ....................................30

3.2.1. Surface coating and bulk compositional heterogeneity ............................................................31

3.2.2. Electrolyte additives and new bulk electrolyte solvents ...........................................................32

4. High-voltage spinel LiNi0.5Mn1.5O4 .............................................................................................................35

4.1. Structural and morphological aspects of LiNi0.5Mn1.5O4 .................................................................36

4.1.1. Cation ordering ...............................................................................................................................36

4.1.2. Surface chemistry and morphology .............................................................................................38

4.2. Strategies to address the rapid performance degradation of LiNi0.5Mn1.5O4 at high voltages ...40

4.2.1. Surface coating and doping ...........................................................................................................41

4.2.2. Electrolyte additives and alternative electrolyte combinations................................................42

4.2.3. Ion-selective membrane separators .............................................................................................46

5. Lithium-rich layered Li1+xM1-xO2 .................................................................................................................47

5.1. Structural, compositional, and morphological aspects of Li1+xM1-xO2 ..........................................48

5.1.1. Structural complexity .....................................................................................................................48

5.1.2. Peculiar oxygen redox reactivity ...................................................................................................50

5.1.3. General composition-morphology-property relationships ......................................................52

5.2. Challenges and corresponding strategies of Li1+xM1-xO2 for high-voltage Li-ion batteries .......53

5.2.1. Voltage decay ...................................................................................................................................54

5.2.2. Strategies for manganese-based compositions ...........................................................................56

5.2.3. Novel Li1+xM1-xO2 compounds ....................................................................................................59

3

6. High-voltage polyanionic compounds ........................................................................................................61

6.1. Redox energy variations in polyanionic compounds ........................................................................62

6.2. High-voltage olivine LiMPO4 and monolithic Li3M2(PO4)3 ............................................................64

6.3. High-voltage tavorite phosphate- and sulfate-based compounds and derivatives ......................68

6.3.1. LixMXO4Z with two electrons per metal redox processes ......................................................69

6.3.2. Fluorophosphates Li2MPO4F .......................................................................................................70

6.3.3. Fluorosulfates LiMSO4F ................................................................................................................71

6.4. Other high-voltage polyanionic compounds .....................................................................................72

7. Conclusions and perspectives ......................................................................................................................74

Acknowledgements ............................................................................................................................................77

References ............................................................................................................................................................78

Tables ...................................................................................................................................................................97

Figure captions ....................................................................................................................................................99

Figures ............................................................................................................................................................... 107

4

1. Introduction

Since the introduction of rechargeable lithium-ion batteries to the consumer market in the early 1990s, they have transformed global communication with the revolution in portable electronics. Currently, the Li-ion batteries are a well-established, efficient energy-storage technology in terms of energy and power densities, service life, and design flexibility.1-4 As the worldwide demand for energy and the environmental concerns and societal dependence on fossil fuels grow, efforts to electrifying the transportation sector are rapidly advancing, opening up another large market for Li-ion technology. Battery-powered vehicles, which involve a set of stringent requirements including driving ranges, cost, safety, and environmental compatibility, are expected to drastically reduce the air pollution imprint from conventional petroleum-driven internal combustion engines. Indeed, the energy density of a battery is one of the most essential parameters, and it determines the vehicle autonomy range. The first rechargeable Li-ion cell from almost twenty-five years ago delivered an energy density of ~ 150 Wh kg-1; today, the state-of-the-art Li-ion systems offer gravimetric and volumetric energy densities up to around, respectively, 260 Wh kg-1 and 780 Wh L-1, approaching the 300 driving mile range target for plug-in hybrid electric vehicles (PHEVs) and electric vehicles (EVs). Practical Li-ion batteries are based on insertion compounds for both positive and negative electrodes that allow reversible insertion and extraction of lithium ions free of major structural changes.2, 5-8 The output voltage of a Li-ion cell is determined by the difference in electrochemical potentials between the two electrode materials. Thus, design of proper electrode systems is of prime importance to realize a high energy density (capacity × voltage) to outperform other available battery systems. The first positive electrode material 5 chalcogenides were developed shortly after, which delivered average voltages of < 2.5 V vs. had to be dismissed for non-uniform lithium electroplating that poses unacceptable safety risks. Following the discovery of an oxide cathode material LiCoO2 operating at almost 4 V by Goodenough and co-workers in the 1980s,11 commercialization of the first rechargeable Li-ion cell was realized by Sony in the early 1990s with a carbonaceous anode material (petroleum coke) as an effective low-potential Li-ion host. To the present day, state-of-the- art Li-ion cells essentially maintain the original configuration, employing a successive generation of LiCoO2, i.e., nickel-based LiNi1-xMxO2 (M = Co and Mn/Al) as the cathode, and graphite (sometimes blended with small amounts of silicon) as the anode. These compounds are mixed with a small amount of conductive agents (e.g., carbon black) and a polymeric binder (e.g., polyvinylidene fluoride, PVDF) and tape casted onto a metal current collector (aluminum for the cathode and copper for the anode). The two electrodes are separated by a microporous membrane film (e.g., polypropylene, PP) and the whole battery is impregnated with an alkyl carbonate-based electrolyte solution with typically LiPF6 as the salt. Ideally, the electrolyte should not experience any major degradation during the course of operation (see section 2 below for details). In spite of increasing prospects for the practical deployment of large-capacity negative electrodes (e.g., silicon-based materials) over the years, their sole contribution to the energy increase of a Li-ion cell is limited by the mandatory presence of inert cell components after a certain threshold value (~ 800 ² 1000 mA h g-1).12 Unfortunately, development of their positive counterparts has largely lagged behind. To date, substantial research efforts from both the academic and industrial communities have focused on the design and optimization 6 of novel positive electrode materials with a large capacity (e.g. • 200 P$ O J-1) and/or high average voltage (e.g. • 4 9 vs. Li/Li+),13-19 the key determinant in further enhancing cell energy densities. Meanwhile, major attention has been directed to designing electrolyte systems that can accommodate a large voltage window for these electrodes.20, 21 Notable potential cathode candidates include nickel-rich layered oxides (LiNi1-xMxO2, M = Co, Mn and Al), lithium-rich layered oxides (Li1+xM1-xO2, M = Mn, Ni, Co, etc.), high-voltage spinel oxides (LiNi0.5Mn1.5O4), and high-voltage polyanionic compounds (phosphates, sulfates, silicates, etc.). Figure 1 illustrates the prospects and limitations of these cathode materials. LiNi1-xMxO2 with lower nickel content has already succeeded in commercialization (e.g., LiNi1/3Co1/3Mn1/3O2), and the Ni-rich compositions ((1-x) > 0.6) are being further optimized and on track to reach the 300 Wh kg-1 gravimetric energy density milestone in the near future. LiNi0.5Mn1.5O4 offers only a moderate energy density, but holds great promise for high- power applications if stable electrolyte systems to withstand its high operating voltage (4.7 V vs. Li/Li+) could be realized. The use of lithium-rich manganese-based Li1+xM1-xO2 with high capacity (~ 250 mA h g-1) has prompted massive interests in recent years to compete with Ni-rich LiNi1-xMxO2 for a larger boost in energy density (both gravimetric and volumetric) at a lower cost. Unfortunately, it is still plagued by voltage fade during cell operation due to the intrinsic layered to spinel phase transition. Finally, polyanion-based compounds provide a futile chemical playground for searching novel cathode materials for advanced high-voltage Li-ion batteries. However, those developed thus far still suffer from an energy and power penalty owing to the heavier formula weight, inadequate intrinsic electronic/ionic transport, and lower powder packing density, making them uncompetitive for the portable electronic device and transportation applications. 7 There have been several reviews published recently concerning one or several classes of high-voltage positive electrode materials mentioned above.13-16, 18, 22-28 As the operation principles of these materials have been described in detail, here we focus on recent research progress on the understanding of key challenges for their commercialization and corresponding strategies. Notably, the electrode-electrolyte compatibility becomes a pivotal issue in this field, hence a particular emphasis on unwanted interactions of various high- voltage cathode materials with state-of-the-art electrolyte systems under highly oxidizing conditions. Some other aspects of material designs also pertinent to their electrochemical performance are also briefly mentioned in the present review, such as methods of synthesis and general composition-morphology-property relationships. Further, the presentation is organized in a way that a straightforward comparison of the various high-voltage cathode materials can be drawn.

2. The general implications of high-voltage electrochemical operation

Figure 2 displays the working voltage vs. the reversible capacity of various cathode materials

in relation to the energy levels of the electrolyte at an open circuit state. The negative

electrode (anode) is a reductant while the positive electrode (cathode) is an oxidant. The electrochemical stability window of an electrolyte is determined by its lowest unoccupied molecular orbital (LUMO) and highest occupied molecular orbital (HOMO). Ideally, rechargeable batteries operate within this window of LUMO and HOMO, in which the electrolyte simply serves as a chemically-inert medium enabling efficient ionic transport between the two electrodes to maintain charge balance. Yet in the case of Li-ion technology, the choice for the negative electrode (anode) is almost limited exclusively to graphite at the commercial stage, of which the reversible electrochemical operation crucially relies on the 8 formation of a stable passivation film (i.e., solid-electrolyte interphase, SEI) at the electrode/electrolyte interface.2, 29 This is because (i) the operating potential of graphite (~

0.2 V vs. Li/Li+, strongly reducing) lies well above the LUMO of almost any known organic

electrolyte solvents, and (ii) the layered graphite structure held by weak wan de Waals forces is prone to solvent co-intercalation along with lithium uptake. Thus, the SEI has spurred extensive research efforts to understand and improve its formation and functioning mechanisms. It is not an exaggeration to say that without this essential component, the mass production of Li-ion batteries in parallel with the rapid growth of the electronics markets over the past twenty-five years would not have been possible. As for the positive electrode (cathode), the commercially established conventional insertion compounds (LiCoO2, LiMn2O4, and LiFePO4) have a robust crystal structure and operate comfortably above the HOMO (~ 4.3 V vs. Li/Li+) of the nonaqueous electrolyte components.1, 2 Hence, there is no thermodynamic driving force for electrolyte oxidation; the stability of the electrode/electrolyte interfaces and interphases formed at the positive side for conventional cathodes are perceived not as crucial for cycle and calendar life compared to their negative counterparts. The high-voltage cathode materials for advanced Li-ion batteries with high energy and power densities, however, have their electrochemical potential beyond the electrolyte oxidation limits, with some approaching 5.0 V or above.14-19 As mentioned in the introduction, these cathodes include layered LiNi1-xMxO2 and Li1+xM1-xO2, spinel LiNi0.5Mn1.5O4, and phosphate-, sulfate-, silicate-based polyanionic compounds. Some of these materials have their reversible redox reactions occurring almost entirely outside of the thermodynamic stability window of the organic carbonate solvents, while others still operate largely within this window despite high upper voltage cut-RIIV • 4.3 V). Nevertheless, the electrode-electrolyte compatibility becomes very critical in generating stable interphases for 9 the practical deployment of these materials. The reasons for their capacity decline during cell operation are dependent on the specific chemical composition, surface chemistry, and operating voltages, but can mostly be categorized into several general degradation processes. Ultimately, these processes lead to a loss of active mass and Li ions as well as an increase in the internal resistance, due to interconnected irreversible changes in the bulk structure and modification of surface properties of the active material.30, 31 This section will discuss the general mechanisms of capacity decrease after repeated charge-discharge cycles for electrodes based on these emerging high-voltage materials before moving on to the specific aspects affecting their electrochemical properties and corresponding strategies for their commercialization.

2.1. Parasitic electrolyte oxidation at charged positive electrodes

One of the most pronounced and widely studied impacts of high-voltage electrochemical operation is the parasitic oxidative decomposition of electrolyte components.30, 31 In state-of- the-art Li-ion electrolyte systems, i.e., dilute LiPF6/carbonate-based nonaqueous solutions as the skeleton formulation, irreversible decomposition can be triggered by many factors, such as trace moisture, elevated temperature, and strongly oxidizing/reducing environments. Since the emerging high-voltage cathode materials have their operating potential in part or entirely above their oxidation stability limit (> 4.3 V vs. Li/Li+), the nonaqueous electrolytes become thermodynamically unstable. Moreover, electrolyte breakdown and subsequent polymerization of the oxidized products can be catalyzed by (i) localized self-discharge and surface oxygen nucleophilic attack of the active cathode material at highly delithiated states, and (ii) attack of acidic species produced by LiPF6 hydrolysis (see section 2.2). This process produces a variety of highly complex organic and inorganic deposits at the cathode surface as well as gaseous species. As shown in Figure 3, the interphasial species include 10 semicarbonates (ROCO2M, M = Li and transition metal), polycarbonates, alkoxides, ethers, LiF, Li2CO3, LixPOyFz, RCFx, etc.,32-36 which are strikingly similar to those identified in the graphite SEI.29 To give a few examples, several studies on the stability of carbonate-based electrolyte solutions (ethylene carbonate (EC), diethyl carbonate (DEC), dimethyl carbonate (DMC), etc.) on the high-voltage spinel oxide (LiNi0.5-xMn1.5+xO4) beyond 4.5 V discovered significant surface organic/inorganic species buildup through ex situ Fourier transform infrared (FTIR) spectroscopy and X-ray photoelectron spectroscopy (XPS),37, 38 and in situ neutron reflectometry (NR).39 The interphases were estimated to be around 3 nm thick by in situ NR on cycled Li1-xNi0.5Mn1.5O4

39 and by in situ synchrotron XPS and X-ray absorption

spectroscopy (XAS) on cycled Li1-xNi0.2Co0.7Mn0.1O2,40 close to the calculated value (~ 0.3 ² 4 nm) based on ex situ XPS analysis.41 At a first glance, it may appear that the cathode interphasial species and their thickness are often not notably different from the graphite SEI. In fact, the interphases formed on positive and negative electrodes do differ drastically in many aspects. The electrolyte oxidation contributes little to the commonly observed first- (~ 36 mA h g-1) of LiNi1/3Co1/3Mn1/3O2 upon the initial charge to 4.6 V vs. Li/Li+ is due to kinetic limitations and can be mostly recovered by applying a constant-potential step at the end of discharge.42 In stark contrast, the reductive process of SEI-forming cyclic carbonates (e.g., EC) during the initial discharge of graphite is always associated with a moderate irreversible capacity.43 With likely insufficient formation, passivation of the interphasial species on cathodes is typically far less effective than the graphite SEI. It has been revealed that the oxidative breakdown of electrolyte components persists (albeit slowly) throughout by an incomplete coverage of these interphasial species that permeate electron transfer (in 11 contrast to an ideal SEI). As a result, sustained parasitic reactions at high potentials lead to consumption of active Li ions, large impedance buildup, excessive gas production, and even electrolyte depletion in extreme cases that lead to abrupt, premature cell death.47-49 Importantly, the onset potentials for noticeable parasitic reactions to occur are strongly related to the varying surface reactivity of different cathodes at highly delithiated states (e.g., oxygen nucleophilic attack). For example, Dahn·V JURXS44-46 have systematically investigated this phenomenon for a series of nickel-based layered oxides (e.g., LiNi0.4Co0.2Mn0.4O2 and LiNi0.8Co0.1Mn0.1O2) in LiPF6/EC-based electrolyte systems via isothermal microcalorimetry and have detected notable parasitic electrolyte oxidation at • 4.3 V vs. Li/Li+ as cycling proceeds (Figure 4a). On the other hand, parasitic reactions only becomes notable at 4.7 V for LiNi1/3Co1/3Mn1/3O2 (Figure 4b)42 and 4.8 V for a PVdF-bonded conductive carbon electrode free of active components (Figure 11d),50 respectively, via on-line electrochemical mass spectrometry (OEMS). It is also worth mentioning that the negative impact of parasitic electrolyte oxidative decomposition alone on the cycling stability of many high-voltage active Li ions and electrolyte is essentially unlimited. For example, LiNi0.5Mn1.5O4 in half cells with an upper cut-off voltage of 4.8 ² 5.2 V often shows good cycle life in simple LiPF6/EC- based solutions, in spite of severe parasitic reactions indicated by low Coulombic efficiencies during cycling.51 In practical Li-ion cells paired with carbon anodes, however, very poor cyclability of the same sample is observed (see section 4). The nucleophilic attack refers to the tendency of oxide ions at the surface of positive electrode materials to form chemical bonds with electrophilic carbonate solvent molecules in electrolyte solutions. Figure 5a shows the ring opening of EC and the formation of surface deposits mentioned above, including Li2CO3, semicarbonates, polycarbonates, alkoxides and 12 more. As a rule, the surface oxygen of layered oxides (Li1-xMO2, M = Co, Ni and Mn) has stronger nucleophilicity (or Lewis basicity) than that of spinel oxides (Li1-xM2O4, M = Mn and Ni), as demonstrated by their impedance increase after cycling.52, 53 For oxides of the same structural configuration, the nucleophilicity of surface oxygen becomes stronger with transition metals of greater electronegativity (e.g., Ni > Co > Mn) and a greater covalent component of the M-O bonds. Thus, layered Li1-xNiO2, compared to Li1-xCoO2 and Li1- xMn2O4, is most reactive towards electrolyte components and shows the largest impedance

increase after the high-voltage electrochemical operation (Figure 5b).34, 35, 53, 54 Gauthier et al.32

proposed that, in the layered LiMO2 system, with the O 2p band moving to higher energies with an increase in the atomic number of the 3d transition metals, the tendency of surface oxygen nucleophilic attack on carbonate solvents increases simultaneously, as seen in Figure

5a. This theory is corroborated by experimental data of the nucleophilicity of a series of

metal oxides (MgO, SrO, CaO, etc.).55 More complicated and pronounced parasitic electrolyte oxidation arises in the case of another attractive high-voltage cathode material, Li-rich layered oxides (Li1-xM1+xO2). During the first charge at • 4BD 9, Li1-xM1+xO2 tends to release molecular oxygen and/or to form surface peroxo-like (O2)n- species. This phenomenon is related to their unusual anionic redox reactions (2O2- ȼ 22)n-), in addition to the common cationic processes (Mn+ ȼ 0(n+1)+) for conventional Li-ion battery cathodes. The origin for these anionic reactions is ascribed to penetration of the metal d band into the O 2p band in the electronic structure of Li1-xM1+xO2,56-59 which will be discussed in details in section 5.1.2. It is worth noting that overcharging conventional layered Li-stoichiometric Co- or Ni-based LiMO2 to ~ 4.3 V or above also causes the release of oxygen gas from the host lattice and poses safety concerns.2, 60, 61 The evolution of molecular O2 and surface (O2)n- species for Li- rich Li1-xM1+xO2, however, is necessary upon the initial charge to access their large capacity, 13 which catalyzes nucleophilic attack of the carbonate-based electrolytes and formation of additional interphasial species.62-65 Moreover, proton-abstraction-induced oxidation66, 67 of certain electrolyte components also becomes possible through the involvement of these oxygen species. Still, this is a relatively underexplored field and much work is needed to understand the role of these surface oxygen species during the interfacial evolution and their influences on the electrochemical properties during long-term cycling.

2.2. Dissolution of active positive electrode materials

Dissolution of transition-metal cations from active positive materials in rechargeable Li-ion batteries has long been recognized as one of the primary causes for their capacity fade and limited cycle life.30, 31 Hydrolytic and/or thermal decomposition of LiPF6 (LiPF6 Ⱥ IL) Ą PF5 and LiPF6 + H22 Ⱥ IL) Ą 32)3 + 2HF) produces acidic species (PF5 and HF) that are critical for the passivation of aluminum substrate current collectors for cathode electrodes at >

3.5 V vs. Li/Li+.68-70 Unfortunately, these acidic species also aggressively attack the active

cathode materials, leading to deterioration of their interfacial stability towards the electrolyte. In addition, they catalyze electrolyte breakdown and subsequent polycarbonate generation.38,

71, 72 Aurbach and co-workers34, 35 systematically investigated the susceptibility of a series of

positive electrode materials to acid leaching in LiPF6-containing solutions contaminated with water, including LiCoO2, LiNiO2, LiNi0.5Mn0.5O2, LiMn2O4, LiNi0.5Mn1.5O4, V2O5, and LiFePO4, and concluded that active mass dissolution is a common phenomenon regardless of their crystal structure and chemical composition. The disproportionation reactions that produce soluble divalent transition-metal cations (e.g., 2Mn3+ Ⱥ 0Q2+ + Mn4+) explain the particular susceptibility to dissolution of manganese-based materials, such as LiMn2O4.73 As a result, continuous accumulation of active mass dissolution products at the surface of composite electrodes and separator pores hinders Li-ion transport, resulting in an increase in 14 internal impedance and capacity fade. Indeed, compared to LiAsF6 that does not produce HF, Li1-xNiO2 and Li1-xMnO4 in LiPF6 solutions show a much larger increase in impedance after cycling (Figure 5b).53 Active mass dissolution is exacerbated with increases in temperature, storage time, states of charge, and not surprisingly, operating voltages.74 While generation of acidic species is inevitably accelerated under increasingly oxidizing environments, growing evidence has also connected parasitic electrolyte oxidation and production of surface divalent transition-metal cations prone to dissolution. Self-discharge of delithiated cathodes during storage with electrolyte oxidation in close proximity has long been recognized to promote dissolution.74, 75 catalysis mechanism for the dissolution of delithiated transition-metal oxides during cellquotesdbs_dbs46.pdfusesText_46

[PDF] 2016 j&k rowling new book

[PDF] 2016 jazz at the white house

[PDF] 2016 jazz blue challenger

[PDF] 2016 jazz blue chrysler 300

[PDF] 2016 jazz fest line up

[PDF] 2016 jazz grammy winners

[PDF] 2016 literature awards

[PDF] 2016 literature gonference in sweden

[PDF] 2016 literature jamb past question/answer

[PDF] 2016 literature nobel prize winner

[PDF] 2016 literature nobelist

[PDF] 2016 lyon county fair emporia ks schedule

[PDF] 2016 nancy drew midnight in salem

[PDF] 2016 nancy kerrigan pictures

[PDF] 2016 nice english song