Readiness Level of Sodium‐Ion Battery Technology: A ...mse2.kaist.ac.kr/~ncrl/pub/2018_Readiness...

REVIEW

1700153 (1 of 37) © 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advsustainsys.com

Readiness Level of Sodium-Ion Battery Technology: A Materials Review

Lin Chen, Michele Fiore, Ji Eun Wang, Riccardo Ruffo, Do-Kyung Kim, and Gianluca Longoni*

DOI: 10.1002/adsu.201700153

1. Introduction

As renewable energy sources are taking a wider share of world-wide energy production,[1] grids reliability and utilization efficiency are plummeting.[2,3] This is mainly due to intermit-tency and discontinuity of power sources, such as wind and solar, which determine uncertainties in energy production capability and in fulfilling the instantaneous energy demand. This aspect, in turn, sensibly increases the unpredictability of energy prices spikes and marginal and maintenance costs of traditional fossil fuel plants, which are demanded

Since the breakthrough achieved in the research around material intercalating lithium, almost a decade has passed before the commercialization of the first lithium-ion battery (LIB). On the brink of an energy voracious future, convergence of scientific efforts over efficient and low-cost energy production and storage would be advantageous and beneficial. The research hovering around sodium-ion rechargeable batteries (SIBs), a more sustainable alter-native to LIBs, has been observing a positive momentum for ten years now, and chemically stable and electrochemically performing anode and cathode materials represent important milestones on the path toward a commercial full-cell. Material science breakthroughs achieved in carbon and graphite based matrices, layered and open framework structures, and sodium storing alloys, disclose new full-cell set up opportunities going beyond traditional “rocking chair” configuration. In this contribution an in-depth analysis of chemical and physical principles lying beyond the energy storage provided by SIBs most recently investigated active materials is given. In the second half of the review, challenges, opportunities, and state-of-the art description of full-cell SIBs lab scale prototypes are discussed. The latter, indeed, stands for a technological validation of a low-cost alternative to lithium-ion batteries guaranteeing energy densities close to 150 Wh kg−1.

L. Chen, Dr. G. LongoniNanochemistryIstituto Italiano di Tecnologiavia Morego 30, 16163 Genova, ItalyE-mail: [email protected]. ChenDepartment of Chemistry and Industrial ChemistryUniversità degli Studi di Genovavia Balbi 5, 16162 Genova, Italy

The ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/10.1002/adsu.201700153.

M. Fiore, Prof. R. RuffoMaterials Science Department (DSM)Università di Milano-Bicoccavia Cozzi 55, 20125 Milan, ItalyJ. E. Wang, Prof. D.-K. KimDepartment of Material Science and EngineeringKorea Advanced Institute of Science and Technology (KAIST)291 Daehak-ro, Yuseong-gu 34141, Daejeon, Republic of Korea

Sodium-Ion Batteries

discontinuously to provide for power unbalances between demand and supply. In general, resilience to these drawbacks would be higher providing an incre-mented flexibility from demand response resources, interregional energy transmis-sion, and energy storage. Plenty of energy storage systems (ESS) are being utilized to curb renewable energy sources intermit-tency. Among them, worth to be listed are hydroelectric (pumped hydro), mechanical (flywheels and compressed air), and elec-trochemical (lead-acid, Na-S, Na-NiCl2, and Li-ion batteries).[4] Nevertheless not all the previously cited energy storage tech-nologies are comparable one with another in terms of scalability, environmental impact, investment costs, maintenance, and moreover, responsivity to energy needs. Batteries energy storage, directly suppling electric energy without requiring any mechanical-to-electrical energy trans-ducer, surely represents the most versatile appliance. In particular, intrinsic flexibility of modern Li-ion batteries coupled to

renewable power plants, ensures the proper response not only to grid on-peak condition but also to short term distribution overload situations. Energy storage systems exploiting sec-ondary battery technology revealed to be versatile, scalable, and cost-effective solutions in many stationary applications. Apart from the so called front-of-the-meter storage, namely ESS cou-pled to centralized energy source plants (i.e., solar or wind, with a nominal power above 1–2 MWp), an increasing relevance is being gained by the smaller behind-the-meter storage, or in other words on the customer side of the meter (private houses and small enterprises).[5] This aspect is communicating the

Adv. Sustainable Syst. 2018, 1700153

www.advancedsciencenews.com

© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim1700153 (2 of 37)


high readiness level of battery technologies in effectively hitting the mass market of private energy storage. Such a fragmented electric energy storage, intended as an extended energy net-work made of thousands of small-scale storage nodes, could be further extended in its flexibility by integrating the energy storage capabilities of the expanding hybrid electric vehicles (HEVs) and full electric vehicles (EVs) fleet. Some countries started experimenting vehicle-to-grid energy dispatching in order to take advantage of fast energy supply coming from elec-tric vehicles connected to the grid.[6] In this case the benefit is mutual, since a small amount of energy (1–2% of the vehicle battery capacity) is collected by the grid during on-peak periods, avoiding fossil fuel based power plants to be started, while it is returned during off-peak moments, restoring the original capacity. In such way, electric vehicle owners would derive an indirect advantage buying electricity from the grid when its price is lower (low demand and high production from renewa-bles) and sell it back to the grid when its price inflates due to the higher demand (on-peak moments) deriving an overall trade benefit. From the energy supplier’s point of view, intrinsic advantages connected to plant’s maintenance, start-and-stop moments, and production inefficiencies avoidance have to be considered. Such performing albeit complex infrastructure will require its main actor, namely the battery, to be reliable, safe, sufficiently cheap, and environmentally friendly. Lithium-ion battery since its first launch in the market by Sony in 1990 in the LiCoO2 cathode–graphite anode configuration,[7,8] has been observing a booming development in terms of active materials involved and increased performances, which progressively disclosed new applications.[9] In more or less two decades, lith-ium-ion based batteries’ volumetric energy density has indeed increased from 250 to 570 Wh L−1, thanks to materials and cell design optimization,[10] in the face of a more general energy density and power density of 186 Wh kg−1 (for LiCoO2-based lithium-ion batteries (LIBs))[11] and 1000 W kg−1,[12] respectively. Lithium-ion batteries’ pricing has continuously been dropping since their first commercialization. This effect is ascribable to multiple factors such as expansion of worldwide production capacity, increased by 11% between 2015 and 2016,[13] accompa-nied with momentary oversupply, since only 40% of available 53 GWh worth lithium-ion cell produced in 2015 was uti-lized.[14] Significant improvements in materials supply chain evolution, battery manufacturing, and materials handling and processing are contributing in further reducing battery pack cost, which might be hitting values well below 300 US$ kW h−1 within the year 2020.[15] Apart from portable electronics, HEVs plug-in hybrid electric vehicles (PHEVs) and EVs, lithium-ion battery penetration in the on-grid storage market fostered the lithium rush as well. In this landscape, lithium price, and its commercially equivalent LiCO3, started to rise significantly from the beginning of 2010s. Simultaneously, concerns related to lithium elemental abundance and thus the sustainability of a lithium-economy made their appearance all over the scientific community.[16–18] Calculations revealed that at the average 5% yearly growth rate in lithium mining necessary to pace up the demand steady increase, lithium reserves (esteemed to be around 14 million metric tons[13,19]) will encounter a severe shortage in less than 65 years.[8,20] Together with a possible raw material shortage, also the political sensitivity of countries

within whom borders are included the most readily available lithium sources, is casting shadows on future sustainability of lithium market.[21,22] Although these forecasts are affected by uncertainties and conclusions about new lithium reserves depletion might appear premature, the scientific community, embracing a proactive attitude, started investigating alternative energy storage technologies based on more naturally abundant

Lin Chen received her B.S. degree in General Chemistry and Materials in Liaocheng University (China) in 2010. She is currently a Ph.D. can-didate at the Istituto Italiano di Tecnologia (Genova, Italy) under the supervision of Prof. Liberato Manna. Her main research interests focus on the development of nanomaterials for electro-

chemical energy storage devices, in particular Na and Li ion batteries.

Michele Fiore received his M.S. degree in Chemistry in the University of Milan-Bicocca (Milan, Italy). He is currently a Ph.D. can-didate in the Department of Materials Science at University of Milano Bicocca under the supervision of Prof. Riccardo Ruffo. His research and study activi-ties focus on the synthesis

and characteri zation of novel materials for alkaline ion secondary batteries with particular attention to structure–electrochemical property correlations.

Gianluca Longoni received his Ph.D. degree in Chemistry from University of Milan-Bicocca, conducting a systematic study of morphological– electrochemical properties correlation of functional inorganic materials for sodium-ion batteries. Among his collaborations, the Material Science and

Engineering Department of Stanford University and Korea Advanced Institute of Science and Technology are the most relevant. He is currently holding a postdoctoral position at the Istituto Italiano di Tecnologia (Genova, Italy), conducting research on graphene-based materials for Li-ion and Na-ion batteries and studies of material optimization for RedOx flow battery.





alkaline metals, such as sodium,[23] potassium,[16] magne-sium,[24] and calcium.[25] Among the cited, sodium chemistry, thanks to chemical–physical similarities with lithium, is the one that ideally would require less technological efforts in oper-ating such a transition. In terms of raw materials supply costs, moreover, sodium outpaces lithium as one of the cheapest ele-ment to be extracted and processed to high purity (about 135–165 $ per ton against 5000 $ per ton for lithium).[20] Together with the supply cost reduced, a further advantage in exploiting a sodium-based energy storage technology would derive from the possible replacement of copper current col-lector with the lighter and less costly aluminum. This very last element, contrarily to what happens with lithium, does not alloy sodium, and thus can be safely employed as both the anode and cathode material substrate.[26] At the dawn of lith-ium-ion based energy storage at the end of 70s and beginning of 80s, sodium was being considered as a candidate equally promising as lithium for rechargeable battery applications, and sodium intercalating compounds started to draw research efforts especially involving disulfides and diselenides com-pounds. Well-known are the exploratory works by Whit-tingham[27] and Abraham,[28] in which thorough descriptions of mechanistic and structural alkaline cations intercalation in chalgogenide hosts are provided. It was the ground-breaking discovery of LiCoO2 as a high energy density lithium interca-lating cathode made in 1980 by Goodenough and co-workers,[29] and other ancillary studies,[30,31] which paved the way for a high voltage full lithium-ion cell, and that momentarily sank the enthusiasm for sodium and other alkaline metals battery. Thanks to recent renaissance of the matter, encouraged by, as previously stated, sustainability and battery market speciation and expansion, sodium-ion based rechargeable battery research is knowing a positive momentum. Efforts put across last decade in sodium-ion rechargeable batteries (SIBs) chemistries optimi-zation, contributed in reducing the technological gap with lithium counterpart. Commercial SIBs started to appear on the market as well, with chemistries based on aqueous as well as nonaqueous electrolytes.

Among the most puzzling difficulties that sodium-ion rechargeable batteries present, the vast majority are related to physical discrepancies between sodium and lithium ion. The fact that sodium naked ion is 34% larger than lithium one (1.02 Å vs 0.76 Å) entails limitation of intercalation capabili-ties in host crystalline structures, affects transport properties in solid and liquid phase and mines interphase formation and stability. Nevertheless, in the recent times several classes of crystalline host structures tailored on sodium ionic radius made their appearance.[32] Provided the plethora of promising anode and cathode materials so far proposed for nonaqueous sodium-ion rechargeable batteries, many research groups started inves-tigating whole sodium-ion cell (full-cell). Due to the higher reactivity of sodium, its tendency in forming dendrites and its low melting point (97 °C), several safety issues are connected to the development of a metallic sodium-based full battery. On the contrary fully ion-dependent cells, inspired by lithium-ion “rocking chair” set up,[33] are preferred. In this configura-tion anodic materials whose electrochemical activity is based on intercalative or pseudocapacitive[34] behavior, are employed in place of metallic sodium. Very recently a large number of both

cathode and anode materials has been thoroughly examined and characterized. Among the compounds suitable as cathodes, layered O3 and P2-type transition metal oxides[35,36] and transi-tion metals polyanion[37–39] obtained remarkable attention, and so far efforts have been focusing on crystalline phases stability, achieved via doping[40] or selecting the most suitable final crys-talline phase by finely tuning the synthetic conditions.[36] On the anode side, instead, more than one single chemistry became appealing and worth to be investigated. The direct transfer of graphitic anode of lithium batteries to sodium one revealed, in fact, as a poor choice. The insufficient interlayer spacing (≈0.34 nm) of standard commercially employed graphite for sodium intercalation, which has been demonstrated to require at least a 0.37 nm gap of a chemically expanded graphite[41] revealed indeed as the most important limitation. Other carbo-naceous materials with remarkable sodium storage capabilities include soft carbons,[42] hard carbons,[43] hydrogen and nitrogen-doped carbons[44] and carbon nanostructures, such as nano-tubes[45] and nanowires,[46] nanospheres,[47] and reduced gra-phene oxide (rGO).[48] While for carbon-based anode materials applied to sodium environment the achievable specific capaci-ties hardly exceed 300 mAh g−1, alloying[49] and conversion[50] types of materials, would represent surpassingly more advantageous alternatives thanks to their theoretical specific capacities abun-dantly exceeding 600 mAh g−1. Among the elements effectively alloying sodium, recent studies focused on 14 group elements (Ge,[51] Sn[52]) and 15 group elements (Sb,[53,54] Bi[55])). Their most prominent drawback, namely the dramatic volume expan-sion during sodium alloying and consecutive mechanical dis-ruption, has been tackled by proposing hybrid structures with carbon matrices[56–59] able to buffer volume changes. The mate-rial classes here only introduced, will be thoroughly and widely analyzed in the following paragraphs.

2. Full-Cell Preparation Challenges

The present work is intended to give an in-depth review over the most significant progresses, in terms of practical perfor-mances, reached by SIBs in full-cell configuration. Indeed while plenty of review articles have been published in the last years providing a comprehensive view over functional materials for sodium rechargeable batteries,[20,23,32,60–64] full-cell configu-rations have been often neglected. The final aim would be that of providing a comprehensive tool to refer to in order to system-atically evaluate and critically asses each new sodium-ion full-cell chemistry proposed in the literature, in order to probe its technological readiness level. This is indeed a compulsory pro-cess to validate sodium ion technology capability in replacing lithium technology in various energy storage applications, and primarily large-scale stationary energy storage, which requires considerable manufacturing investments. Not all the applica-tions field would work for a sodium-ion technology. For electric mobility applications, indeed, (i.e., EV, HEV, and PHEV) impor-tant figures of merit are both volumetric (kWh dm−3) and gravi-metric energy density (kWh kg−1) since hindrance and weight of the whole battery pack would heavily influence car overall performances. On the other hand for stationary energy storage systems (i.e., power plant backup systems, domestic storage and





renewable energy sources continuity units), cell final requisites in terms of volumetric and gravimetric energy density are less stringent. Portable electronics, eventually, demand improved capabilities mainly in terms of amount of capacity constrained in a sufficiently small volume and reliability during charge/discharge cycles. When the assembling of a prototype full-cell is considered on a laboratory scale, critical factors ought to be carefully weighted in order to accurately mimic practical cell conditions and thus produce sensible results. Criticalities are (i) the greater sensitivity of cell performance to the coulombic efficiencies of the electrodes, (ii) the irreversible capacity con-nected to solid electrolyte interphase (SEI) formation (both ini-tially and ongoing during cycling), and (iii) any side reactions occurring within the cell. The latter aspect, in particular, poten-tially can lead to sensible modification in charge/discharge electrode profiles. This, in turn, will result in an unreliable full-cell potential output. SEI layer is defined as the polymeric film forming during the very first charge cycle at anode–electrolyte interface. It directly derives from the electrochemical decom-position of electrolyte solvents and salts producing different species. This process entails the irreversible segregation, inside the polymer matrix, of a varying quantity of lithium ions (either complexed or strongly bonded to polymer branches and func-tionalities). In full-cell configuration this amount of lithium ions it is no more available for further reactivity and thus will set an irreversible contribute to the capacity of the first cycle. This contribute can range from 5 to 15% of the total capacity of the cell and it needs to be properly balanced with an excess of cathode material. Coulombic efficiency (CE), defined as the ratio between the charge supplied during charge and charge obtained during discharge, is moreover an important indi-cator of the irreversible contributes influencing each charge/ discharge loop. A CE of 99.95% (0.05% irreversible capacity per cycle), although might seem contained and not not particularly significant, would determine an overall 10% capacity loss in 200 charge/discharge cycles. Such value is incompatible with commercial products which are supposed to last, according to the utilization, between 500 and 10000 cycles. Porosity of the electrodes, intrinsic volume changes during charge/discharge can affect both the volume of the cell and the capacity or rate performances of the electrodes, and therefore careful optimization of the components is required for optimum cell performance.[65] Delicate aspects also include the accurate bal-ancing between anode and cathode materials, in order to pair up capacities Practical specific capacities of anode and cathode species have to be taken into careful consideration. This values often differs from theoretical specific capacity due to limitations in exploiting completely the RedOx and capacitive mecha-nism involved in the material energy storage. For instance, P2–O2 phase transition of certain layered cathode materials, P2-Na0.7[(Fe0.5Mn0.5)1-xCox]O2 is an example,[66] determines a progressive capacity reduction during cycling connected to mechanical stresses related to considerable unit cell volume change (±15%). For this reason, in order to achieve a prolonged operation life of the full-cell, excluding such phase transition by reducing operational voltage window and thus the capacity practically achievable, would be appropriate. Differently from half-cell set-ups, in which metallic sodium counter electrode behaves as an almost infinite sodium reservoir, for full-cell

configurations irreversible sodium consumption connected to SEI formation at the anode, or electrode activation processes, need to be properly addressed. A selected excess of cathode material has thus to be loaded, but its precise quantification is not an easy task. In research field, where the main focus of the investigation is most of the time the gravimetric energy density of either cathode or anode material, a counter electrode mate-rial excess is considered. In reporting the latest results achieved in this field, consequently, it will be always indicated if energy density and specific capacity refer to a cathode or anode limited full-cell. Recently the research around high capacity anodes for SIBs (based on 14th and 15th group elements alloys or oxide based conversion materials) has been knowing a positive momentum, thanks to the hype also deriving from lithium-rich full battery counterpart (silicon[67,68] and sulfur[69] based LIBs). The issue represented by balancing a full cell based on these classes of materials is even more severe. Due to the high gravi-metric energy densities of the above mentioned elements and compounds, concerns related to the capacity balancing using a low gravimetric energy densities cathode arise. The required amount of cathode material might result unseemly and eventu-ally can create a dramatic kinetic bottleneck due to resistivity imposed by electrode thickness. The attention will be directed to the description of the most promising full-cell architectures and chemistries proposed so far on a laboratory scale. Full-cell characteristics and electrochemical performances will be eventually collected in capitulatory tables and graphs, in order to facilitate the comparison with lithium chemistries as well, and highlight valuable patterning and clustering of full-cell SIBs architectures on a Ragone-like plot.[70] To better report the recent literature findings, it has been chosen to differen-tiate the sodium full-cell architectures in two main families: nonsymmetric and symmetric full-cells. In the first category, the adopted configuration resembles the one utilized nowadays by lithium-ion technology in which anode and cathode mate-rials belong to two distinct families. For symmetric devices, the same material will be addressed both as the anode as well as the cathode inside the cell, exploiting its extended electrochemical stability and wide potential window operation and multiredox centers. This last class of device surely represent an interesting and costly efficient configuration albeit often affected by lower energy and power density content.

3. Sodium-Ion Battery Materials

Before diving in the thorough description of full-cell SIBs chemistries and configurations an overlook of the most studied classes of active anode and cathode materials is provided. Spe-cific attention will be addressed to energy storage principia description, and fundamental notions will be accompanied by state-of-the-art research examples. Active materials for SIBs are being continuatively and extensively studied for more than a decade now and many of them are proven as promising cathode and anode compounds ensuring stability and reliability at high currents.[23,64,71–74] The recent discovery and optimization of high-voltage cathode materials for SIBs,[75–79] indeed contributed in reviving the interests for a full sodium ion battery, contrarily to the downing of sodium ion rechargeable batteries in the 1980s.





3.1. Anode Materials for SIBs

Research about anode material of a sodium-ion cell has recently drawn the attention of the scientific community. Despite this, a net numerical discrepancy between published works on cathode material and anode ones has been registered in favor of the former. Among the difficulties experienced by people approaching the negative-side topic worth to be mentioned are the operational potential of the material in a sodium environ-ment, often too high to guarantee an appreciable final energy density, overall stability and frequent disruption of the material throughout cycling and low coulombic efficiencies. A worldwide accepted criterion in investigating this type of species has been so far, mimicking the modern research trend in LIBs research, the proposal of negative material with exceptionally high gravimetric capacities, well above the average values of commercial devices (≈300 mAh g−1).[80] This approach, albeit source of extraordinary results often limited to first cycles, is short-sighted toward a pos-sible coupling with a cathode material, in a full cell assembly. It must be in fact considered that nowadays research on cathode materials for SIBs, and it will become clear in following sections, is limited to insertion compounds with gravimetric capacities limited to only few hundreds of mAh g−1. Managing materials quantity balance inside the cell and thicknesses of active deposi-tions will then become a critical aspect when full-cell SIB is put under study. The overwhelming number of compounds proposed in literature will be, to facilitate navigation through the present review, organized in four categories, namely (i) carbon-based materials, (ii) oxides based on conversion reaction, (iii) p-block alloying elements (metals, alloys, and phosphorous) showing reversible sodiation/desodiation reaction, and (iv) oxides as Na-insertion topotactic materials.

3.1.1. Carbon-Based Materials

Carbonaceous materials were the most immediate and trivial answers to the need of an anode material for sodium ion bat-teries. Many aspects can be listed among motivations beyond this, such as lithium-ion batteries heritage connected to graphite, low costs, environmental friendliness, and electro-chemical stability. Most likely, studies of graphite-based anode material intercalating sodium started together with those about lithium and potassium, and confirmed their revers-ible intercalation forming LiC6 and KC8 graphite intercala-tion compound (GIC) respectively. Despite Na having a ionic radius (≈1.05 Å) situated in the middle between that of Li (≈0.76 Å) and K (≈1.5 Å), and thus a higher chance to trigger staging into layered material, only 12 mAh g−1 are delivered by its insertion. Difficulties in sodium intercalation refers to incompatibility of the small interlayer distance of graphite (d(002) = 0.334 nm)[83–85] with ionic radius of solvated sodium. Furthermore, Stage-1 Na-GIC formation (occupation of every interlayer space in graphite layered structure) has been demon-strated to be thermodynamically unfavorable, due to excessively high sodium RedOx potential which makes the sodium plating upon graphite, rather than intercalation in it, more likely.[86] Other factors that undermine the proper staging are the large elongation of CC bond length and the scarce binding energy

between carbon and sodium atoms. Recently it has been dem-onstrated how expanding along c axis the single layers stacking of graphite, is possible to achieve reversible capacities hovering around 284 mAh g−1 at 20 mA g−1, with the superior capacity retention of 73% after 2000 cycles if cycled at 100 mA g−1. In the work of Wen et al., indeed, the importance of main-taining a long-range vertical ordering is stressed, and moreover, concentration of oxidized carbon sites, namely surface oxygen concentration, is considered a fundamental aspect.[41] Jache and Adelhelm claimed that also the electrolyte solution plays a relevant role in allowing sodium intercalation, and they have been able to obtain a reversible insertion of sodium ions using an diglyme-based electrolyte solution.[87] In their work, the co-intercalation of solvent molecules chelating sodium ions, according to the reported equation, is emphasized

( )+ + + ↔− + + −C e Na solv Na solv Cyn y n (1)

Low potential intercalation occurs, in this case, with a modest overpotential and extending for ≈100 mAh g−1. Structural X-ray diffraction (XRD) analysis confirms graphite layers expansion along c axis, leading to a total volume expansion of 15%. Theo-retical study hypothesized an ether co-intercalation of —one to two molecules of diglyme, considering that alkaline atoms are preferentially coordinated by six oxygen atoms.[88] Further-more, Kim et al.[89] reported unusual Na storage behavior in natural graphite through Na+-solvent co-intercalation combined with pseudocapacitive behavior using ether-based electrolyte. Wen et al.[41] proposed chemically modified graphite[41,87] as anode for SIBs with an enlarged interlayer lattice distance. The expanded graphite is a graphite-derived material formed by a two-step oxidation–reduction process that retains the long-range-ordered layered structure of graphite, yielding a generally larger interlayer distance (>0.34 nm) as schematized in image (a) in Figure 1. These features provide favorable conditions for electrochemical intercalation of Na ions.

As the utilization of pure graphite has been discouraged, other carbonaceous material, like hard-carbon, soft-carbon, and amorphous carbon started being extensively investigated as alternatives. A large number of nongraphitic carbons have demonstrated to reversibly interact with sodium with mecha-nisms other than pure intercalation. This class of material includes partially crystalline hard carbons and soft carbons. These two materials differ not only for the synthetic methods and precursors involved but also upon crystalline structure and stability to thermal treatments. Hard-carbons, defined as nongra-phitic nongraphitizable carbons, are obtained from solid-phase pyrolysis of phenol-formaldehyde resins, cellulose, charcoal, coconut and sugar,[90] and other biomass wastes.[91,92] Soft-carbons, termed as nongraphitic graphitizable carbons, are obtained from liquid or gas-phase pyrolysis. The term nongraphitic refers to the absence, in the carbon structure, of crystalline graphitic domains with long-range periodicity. The graphitizable concept, instead, involves the possibility of obtaining such long-range periodicity upon high-temperature (>1000 °C) thermal treatment of the carbon. As hard-carbons appear more stable if subjected to high temperature thermal treatments and oxidative environments, soft-carbons have the advantageous property of graphene sheets interlayer spacing tunability upon appropriate heat treatment.





As thoroughly reported by Luo et al., increasing heat treatment temperatures induces a denser stacking of graphene layers inside micrometric graphitic domains, as confirmed by shift toward higher angles of (002) XRD diffraction peak.[93] Never-theless, compared to reversible capacity extracted from hard-carbons, soft-carbons deliver lower capacities (≈120 mAh g−1 at 20 mA g−1) at slightly higher potential (0.6 V vs Na/Na+). As previously quoted, for nongraphititc carbon and hard-carbon in particular, multiple processes contribute to sodium storage and they are inevitably connected to carbon microstructure. Hard-carbons show the higher heterogeneity in structure, namely micrometric-size graphitic domains (image (b) in Figure 1) whose long-range continuity is interrupted by randomly arranged single and few-layer graphene sheets as well as amor-phous areas. This casual arrangement, intuitively simplified by depicting it as a “falling cards model,” confers a porous struc-ture whose degree depends on pyrolysis conditions such as temperature and gas fluxes. As an explicative case, using sugar as carbon precursor, a steep decrease in micropore surface has been observed if pyrolysis temperature is raised from 900 to 1000 °C, while a smoother decrease occurs above 1200 °C. The specific surface area (SSA) changes accordingly from hundreds to <10 m2 g−1 if pyrolysis takes place above 1000 °C due to micropores closure.[94] Worth to be noted is that porosity and surface area extension can be also tuned by precursor pretreat-ments as well as by postpyrolysis heat treatment in particular flowing atmospheres.[95] Analyzing in more details the involved

mechanisms in sodium uptake, two main processes are found to be responsible for the majority of the stored capacity. They can be easily spotted by looking at the potential versus capacity curves of a galvanostatic cycling (right side of image (a) in Figure 2), and consist in a sloping potential region extending to ≈0.2 V and a low potential plateau that gently glides toward 0.0 V versus Na/Na+. The former of the two is widely recog-nized to be due to alkaline ion insertion between layers in pseu-dographitic domains. These process is not limited and does not coincide with a single potential plateau, as in the case of lithium insertion into graphite, since the pseudographitic domain dis-order produces sites with different chemical environments and thus different insertion energies. The process occurring at lower potentials is instead related to ions adsorption in micropores as backed up by recent density functional theory (DFT) calculation.[96] These two processes, also active for lithium, can be practically monitored by ex situ XRD and small-angle X-ray scattering (SAXS) analysis. During the first step an expansion of the interlayer distance between graphene sheets from 3.8 to ≈4.15 Å can be detected, while in connection with the second step, a reduction of the scattering intensity around 0.03–0.07 Å−1 in the SAXS spectra well agrees with the filling of micropores via adsorption mechanism.[97] A common drawback shared, in different extents, by nearly all hard-carbons investi-gated as anode for sodium batteries, is the non-negligible irre-versible capacity correlated with the first sodiation. This is mostly due to SEI formation and only partially to irreversible


Figure 1. Schematic and clarifying representation of sodium storage mechanisms in carbon-based materials. a) Na+ is scarcely intercalated into graphite because of the incompatible ionic radius-graphite interlayer spacing ration. Electrochemical intercalation of Na+ into highly chemically oxidized graphite (graphite oxide) is instead enabled by the widened interlayer distance derived from interposed oxidized functionalities. Nevertheless, the intercala-tion is limited by steric hindering from large amounts of oxygen-containing groups. A significant amount of Na+ is, on the contrary, electrochemically intercalated into expanded graphite, obtained via mild reduction of graphite oxide, owing to suitable interlayer distance and reduced oxygen-containing groups in the interlayers. Reproduced with permission.[41] Copyright 2014, Macmillan Publisher Limited. b) A schematization of the “falling cards” model for sodium filled hard-carbon, where expanded graphitic domains are randomly arranged conveying a microporous structure to the carbon structure. Reproduced with permission.[81] Copyright 2000, The Electrochemical Society. c) The rocking chair type carbon based Na-ion battery proposed by Komaba et al. Reproduced with permission.[82] Copyright 2011, America Chemical Society.




sodium trapping inside carbon structure. In general massive capacity losses related to first cycle are deleterious and highly problematic, since they are hardly quantifiable sodium-con-suming mechanisms overcomplicating anode to cathode mass balance in full-cell assembly. An overestimation of cathode excess required to counterbalance the irreversible capacity might also lead to hazardous consequences such as sodium plating upon overcharging or at low temperature operation con-dition. Irreversible capacity is unfortunately a common feature for anode materials, and the highest coulombic efficiency upon the first cycle reported to date is 85% for hard-carbon samples with low SSA (<10 m2 g−1). A detailed review of carbon mate-rials, so far the most cost effective and easy to synthesize candi-dates, as suitable anode for SIBs has been recently provided by Balogun et al.[42] Together with concentration of surface func-tional groups, surface area extension has been demonstrated to be a crucial parameter in defining the SEI layer formation and thus irreversible capacity build up, and a direct linear relation between Brunauer–Emmett–Teller (BET) specific surface area (SSABET) and irreversible capacity is often observed. Some

studies conducted with lithium suggested that more than the SSABET accessible to the electrolyte, a more indicative param-eter of the surface reactivity with alkaline atoms and salts would be the active surface area.[98] It can be effectively determined via oxygen chemisorption to form oxygenated complexes and the following quantification of oxygen outgassing by mass spectros-copy. Direct truthfulness of this aspect for SIBs is still to be confirmed. Anyway, as a general rule of thumb, low surface specific area hard-carbons have to be preferred to higher area ones in consuming less material during SEI formation. Since SEI layer composition strongly depends on sodium salt and solvent employed in electrolyte formulation, many efforts have been addressed to find an optimum electrolyte formulation. Komaba et al. successfully achieved a stable capacity of 230 mAh g−1 for more than 100 cycles with 1 m NaClO4 in propylene carbonate (PC) or ethylene carbonate (EC):diethylene carbonate based electrolytes.[97] Under a structural point of view, a progressive reduction in reversible capacity connected to Na+ insertion in hard-carbon graphitic layers is observed if the layers spacing is reduced upon increased pyrolysis


Figure 2. a) Cartoons depicting different types of carbonaceous material according to their microstructure; hard-carbon nongraphitizable carbons in the “falling cards” model possess a hierarchical structure in which insertion domain and adsorption microporous and mesoporous sites are both present (left); on the right the two sodium storage mechanisms are illustrated paired up with observed charge/discharge curves. Reproduced with permission.[90] Copyright 2015, Electrochemical Society. b) Evolution of Raman signals for a hard-carbon obtained from PAN pyrolized at different temperatures. Reproduced with permission.[99] Copyright 2014, Elsevier. c) Sodiation profile evolution with cycles of a O-doped carbon obtained from pyrolysis of perylene-3,4,9,10-tetracarboxylic acid dian-hydride (PTCDA). Reproduced with permission.[101] Copyright 2016, Wiley-VCH.




temperature.[99] In other words, by increasing heat treatment temperature, an ordering process is induced and eventually a more graphitic structure is obtained as showed by ID/IG bands intensity in Raman spectra (image (b) in Figure 2). The latter has been recognized as a further diagnostic parameter of the ordering level of carbon structure, being D-band and G-band in Raman spectra defect-induced and E2g graphitic induced modes, respectively. D-band typically falling at ≈1350 cm−1 is associated with defects of disordered structures. G-band instead, peaking at ≈1580, corresponds to ordered graphitic domains.[100] ID/IG values thus give a qualitative representation of carbon microstructure, and as they drop below 1, more ordered structure and graphitic-like structures are achieved. Typical values range from 1.05 to 0.94 as the heat treatment temperature is increased from 800 to 1500 °C. On the other hand, reversible capacity associated to low potential plateau, and thus to sodium adsorption in micropores, was found to increase with temperature. This means that even if micropores close as the temperature is raised, they are still accessible for Na+ adsorption and the only parameter controlling low poten-tial capacity is the average pore size. A mounting attention has also been directed to O-rich carbons. Their oxygen containing functionalities, such as CO and CO, actively participate to Na+ storage mechanism by reversible O reduction.[101] In these cases a completely new RedOx process taking place at 1.5 V versus Na+/Na progressively appears in the charge/discharge

profile as CO and CO undergo activation. Previously described carbonaceous materials for SIBs and many other are collected in Table 1. Important parameters are included in the table: synthesis conditions, state of graphitization by reporting ID/IG ratio (when Raman data are available), SSABET, and Coulombic efficiency related to the first cycle.

An effective strategy to improve the electrochemical properties of carbonaceous materials has been so far heteroatom (such as N, B, S, and P) doping. Doping hard-carbon or graphene-like structures creates a defect sites to absorb sodium ions and improves the elec-trode–electrolyte interaction.[20] Hwang et al. recently proposed a 3D interconnected structure of free-standing flexible films composed of nitrogen-doped porous nanofibers that exhibited 212 mAh g−1 at 5 A g−1 with a capacity retention of 99% after 7000 cycles.[107] In another work, nitrogen rich hard carbon derived from biomass with improved capacitance exhibited stable 204 mAh g−1 for over 1000 cycles at 1 A g−1 but a Coulombic efficiency of 34% in the first cycle.[108] Biomass derived carbons, which have been widely investigated due to their low production cost and low energy con-sumption during the synthesis procedure, indeed often exhibit very low Coulombic efficiency.[109] Dealing with such problem requires properly selecting the biomass resource and optimizing synthesis procedure.[110] Zhang et al., for example, successfully obtained higher Coulombic efficiency in the first cycle (85%) and stable capacity of 334 mAh g−1 after 120 cycles at 0.1 C, optimizing the preparation method of hard-carbon derived by pinecone.[111]


Table 1. Summary of the most relevant carbonaceous matrices and microstructures proposed as anode materials for sodium ion batteries.

Precursor and synthesis Pyrolysis conditions

ID/IG or carbon type SSABET [m2 g−1] Coulombic efficiency (1st cycle)

Reversible capacity

N-doped carbon sheet Dopamine 800 °C, 2 h 0.93 76, mesoporous 26.4% 165 mAh g−1 at 200 mA g−1

Templated carbon[102] Mesophase-pitch

infiltrated silica monolits700 °C Mostly nongraphitic 346 mesoporous

and macroporous≈17% 110 mAh g−1 at 372 mA g−1

decreasing

Nanospheres[103] Controlled thermal

decomposition of

mesitylene

700 °C, 5 min 1.1 4 mesoporous ≈60% 92 mAh g−1 at 150 mA g−1

stable

Hollow nanospheres[104] Glucose carbonization

on latex templating

spheres

1000 °C Weakly ordered

turbostratic carbon

structuring

410 microporous 41.53% 150 mAh g−1 at 100 mA g−1

stable

Highly disordered

carbon[105]

Self-assembling of

polyelectrolytes (PDDA

and PSA) in water

800 °C, 2 h <1.0 2.5 m2 g−1

macroporous

57.61% 225 mAh g−1 at 100 mA g−1

after 180 cycles

Hollow nanowires[45] Pyrolisis of hollow

polyaniline1150 °C, 6 h Disordered graphite

nanocrystallites

50.5% 206 mAh g−1 at 50 mA g−1

after 400 cycles

Electrospun

nanofiber[46,99]

Electrospun PAN fiber 1250 °C,

30 min, N2

0.97 14.6 mesoporous 72.0% 271 mAh g−1 at 100 mA g−1

stable after 100 cycles

H3PO4-activated porous

carbon[43]

Pyrolysis of H3PO4

treated pomelo peels700 °C, 2 h, N2 >1.0 1272 mesoporous 27% 181 mAh g−1 at 50 mA g−1

after 200 cycles

Microspheres[47] Sucrose microwave-

assisted reaction500 °C, 2 h Disordered carbon 393.1 microporous ≈40% 183 mAh g−1 at 30 mA g−1

after 50 cycles

Expanded graphite[41] Modified Hummer’s

method600 °C, 1 h, Ar Graphitic 49.5% 280 mAh g−1 at 100 mA g−1

after 2000 cycles

Rape seed shuck hard

carbon[106]

Rape seed shuck HTC

followed by pyrolysis700 °C, 2 h, Ar 0.53 (IG/ID)

amorphous

11.93 mesoporous ≈80% 143 mAh g−1 at 0.1 A g−1

after 200 cycles

O-doped 3D

interdigital carbon[101]

Polymerization of PTCDA 500 °C, 4 h, Ar Disordered carbon 63.1% 223 mAh g−1 at 1000 mA g−1

after 1200 cycles




3.1.2. Conversion Materials

Conversion Transition Metal Oxides: Conversion transition metal oxides are materials worth of interest as anode for SIBs due to their high theoretical specific capacity (≈900 mAh g−1). They are also thermodynamically favored among other transition metal halides and chalcogenides, thanks to the surpassingly advantageous potential gain in sodium environment if com-pared with lithium.[50,113] Considering a full conversion reac-tion leading to metallic transition metal and sodium oxide, the theoretical specific capacity increases with the initial oxidation state of the transition metal (715 mAh g−1 for CoO where CoII and 891 mAh g−1 for Co3O4 where CoIII) and decreases with its atomic number. One of the major challenges that conver-sion oxides materials carry with themselves is the important volume expansion occurring during conversion mechanism, reported below[50]

bc a ba b c r r rM X Na M Na X G H T S( )+ ↔ + ∆ = ∆ − ∆ (2)

It accounts for 150–250% volume change for oxides and is second only to transition metal phosphide for which it can reach the extraordinary value of 400%. The induced structural strain connected to expansion and contraction is thus considerable especially at the interface between metal and Na2O, due to lat-tice mismatch. Using material engineering, many solutions have been proposed in order to minimize this drawback and end-less efforts have been addressed to investigating mechanism in order to provide a more detailed analysis of the tricky conversion process. A list of the most emblematic conversion oxides inves-tigated as anode for SIBs is provided below, together with an in-depth discussion of their mechanisms and chemistries.

Iron oxide, in its multiple crystalline phases and allotropes, represents a well investigated conversion material for both lithium and sodium batteries. Hematite α-Fe2O3 as well as magnetite Fe3O4 have been taken into consideration and their conversion reactivity has been widely characterized. Hariharan et al.[114] proposed a material design approach in trying to curb material active disruption induced by mechanical stresses upon cycling. In a one-pot hydrothermal synthesis they have been able to synthesize nanometric Fe3O4 particles embedded in a carbon matrix, the latter expected to take upon itself the sluggish elec-tronic conduction of magnetite and provide a mechanical rein-forcement capable of withstanding volume expansion stresses. Despite the promising first sodiation specific capacity of 643 mAh g−1, a fast decay within the following ten cycle, symptom of progressive material degeneration, is observed. Selected area electron diffraction (SAED) patterns of desodiated (charged) electrodes were selectively used to study the reaction path. Traces of metallic Fe and crystalline Na2O have been observed as a clear footprint of an incomplete oxidation to Fe3O4 during Na removal. Magnetite Fe3O4 phase diffraction rings are how-ever present, and no intermediate iron oxide phases have been spotted. Even though a thorough description of the dynamic evolution of phases upon sodiation and desodiation is still lacking, the general conversion mechanism is widely accepted as the main occurring process for iron oxide. Fe2O3@GNS (gra-phene nanosheets) composite has also been proposed[115] as a viable solution to two of the major issues pestering iron oxides,

namely the (i) sluggish kinetic properties due to low elec-tronic conductivity and (ii) limits to sodium intercalation due to a 34% excess of sodium ionic radius compared to lithium. The gravimetric GNS content achieved via simple nanocasting technology was considerable and quantified, by thermal gravi-metric analysis (TGA), to be around 37.4% and thus no sup-porting carbon matrix has been added in electrode preparation. Despite the conspicuous irreversible capacity registered during the first cycle (≈50%), possibly connected to the high content of a carbonaceous matrix, electrochemical performances are quite impressive and result in a stable specific capacity of nearly 400 mAh g−1 obtained at 100 mA g−1 and maintained for more than 200 cycles. The operational potential is 0.75 V versus Na/Na+, an interesting value for an anode material able to guar-antee a satisfactorily energy density. Recently using as template a metal–organic frameworks a new Fe2O3 carbon composite has been proposed by Li et al.[116] Thanks to its stable high porous structure that can shorten ionic diffusion length and provide efficient channels for mass transport, the as prepared material manifests outstanding electrochemical performance. With high coulombic efficiency, even in the first cycle (87.6%), the com-posite has been proved to deliver 710 mAh g−1 with a capacity retention of 93% after 200 cycles. Even at high current rates (4000 mA g−1) the capacity provided is 421 mAh g−1.

rGO is often suggested as a submissive element to improve both capacity and kinetic properties of conversion transition metal oxides, and frequently in modest gravimetric amount. Among its exploited positive features can be enumerated the excellent electronic properties, interesting mechanical prop-erties in accommodating volume changes and providing a uniform substrate for an evenly distribution of electrochemi-cally active particles. Worth to be noted is that nanocasting or hydrothermal impregnation methods are required in order to achieve an intimate adhesion (often resulting in a covalent bonding) between graphene oxide sheets and electrochemically active particles and that a simple mechanical mixing will not suffice. Nevertheless this approach does not allow to accurately distinguish the capacity directly deriving from conversion mech-anisms from that related to carbon based additive which can be notable, as previously illustrated. A significantly different mate-rial design has been proposed by Yang et al. in investigating spinel Co3O4 conversion material.[112] A highly ordered 3D porous structure has been obtained by a nanocasting synthesis using nanoporous silica (KIT-6) as a templating agent (image (a) in Figure 3). All the diffraction peaks from its XRD analysis indexed to Co3O4 spinel phase progressively disappear, as can be spotted in image (b) in Figure 3, as potential reaches the first sodiation low potential values pseudo plateau. Only when a further downward bending of potential curve occurs, close to 0.0 versus Na/Na+, Na2O peaks appear in the diffractogram. Co3O4 is partially recovered upon desodiation. Impurities for-mation or other CoxOy phases evolution cannot be excluded since weaker diffraction peak can be clearly distinguished at the end of the first cycle. A distinct modification of charge/discharge profile occurs after the very first cycle and recent XRD ex situ investigation of bulk porous Co3O4 electrodes, success-fully confirmed the recovery, after first sodiation, of an interme-diately oxidized CoO phase.[117,118] Under the electrochemical performances point of view, nanoporous Co3O4 shows upon





long cycling a specific capacity of nearly 500 mAh g−1, if com-pared to bulk Co3O4 particles. If 5% fluoroethylene carbonate (FEC) is added to the electrolyte solution, a net increment of the capacity stability can be achieved (image (c) in Figure 3). FEC is recognized to be a forming agent of particularly stable SEI layers, and its use, in low gravimetric concentrations (2–5 wt%), is often encouraged in investigating anode materials for SIBs.[63,72,119] Practical specific capacity is considerably lower than the theoretical value (891 mAh g−1) but still an inter-esting value if compared to carbonaceous anode materials. The almost halving of the theoretical value has to be attributed to high irreversible capacity involved in first cycle as well as to the large size and thus poor mobility of Na+ inside the oxide lattice. In comparison Li, giving a pure conversion mechanism with Co3O4, is a lot more performing giving capacity closer to or higher than 800 mAh g−1 at 1 C.[120,121] Other solution have

been proposed to accommodate electrode expansion such as homogeneous dispersion of cobalt oxide particles on graphene oxides sheets using a wet method[122] as well as on a carbon nanotubes network via pulsed plasma technique.[123] In the latter case ex situ SAED analysis taken after the tenth charge to 0.01 V versus Na/Na+ revealed how only Na2O and metallic Co are present, symptom of a complete conversion reaction. On the other hand, after each discharge back to 3.0 V only one component is identified, whose SAED pattern well matches with crystallographic parameters of Co3O4. These results seem to be backed up by TEM and ex situ XRD analysis of cycled elec-trodes. Same way as iron oxide, tin oxides, SnO and SnO2, have been also proposed as safer anodes for SIBs instead of cobalt-based compounds. Flower-like SnO particles and nanometric sized SnO2 (carbon coated) particles have been hydrothermally synthesized by Lu et al., and their electrochemical properties in


Figure 3. a) Synthetic approach proposed by Yang et al. based on a templated or nanocasting method for the preparation of a porous Co3O4 3D-structure. b) Ex situ XRD studies conducted on m-Co3O4 sample in order to monitor phases evolution during material charge and discharge (left) and transmis-sion electron microscopy (TEM) micrograph (right): the highly ordered microporous patterning is clearly visible. c) Galvanostatic performances of the prepared nanoporous conversion materials. Reproduced with permission.[112] Copyright 2016, Wiley-VCH.




sodium environment thoroughly tested in the potential window with different sodiation cut-off values.[124] SnO particles show, at a first galvanostatic screening analysis, the best electro-chemical behavior, guaranteeing a stable specific capacity of 530 mAh g−1 after 50 cycles. On the other hand a decreasing trend is obtained from carbon coated SnO2, with a capacity retention of 58% after 50 cycles. From galvanostatic cycling tests adopting different sodiation cut-off potentials (0.3 and 0.05 V vs Na/Na+) progressively increasing Coulombic efficiencies, 49.4% to 61.7% for SnO and 38.5 to 52.5% for SnO2/C, are recorded. The even lower Coulombic efficiencies recorder for a higher cut-off (0.5 V) has to be than related to irreversible SEI formation mechanisms taking place at relatively higher potentials during the first sodiation, as observed for other oxide materials. By fol-lowing reaction mechanism using ex situ XRD analysis, metallic Sn phase starts being detected from 0.5 V for SnO and is not fully reoxidized when electrode potential jumps back to 2.0 V. It is worth noting that by looking at SAED patterns, SnO2 rather SnO is spotted when the electrode is desodiated, this is presumably due to the higher thermodynamic stability of SnO2 phase com-pared to SnO. If this would be completely true, relative perfor-mances of the two materials, in term of charge/discharge profiles and specific capacities, should tend to uniform upon cycling resulting in an overlap, which is not achieved. The discrepancies in material morphologies, which are remarkable, must thus have a role. In the case of SnO, theoretical specific capacity coinciding with two Na ions per mole of SnO is 398 mAh g−1, nearly half of the capacity extracted during first sodiation (720 mAh g−1). Authors claimed that this is a demonstration of the reaction mechanism occurring for 55% via conversion reaction and for 45% through the alloying reaction of sodium in metallic Sn. Added to this a lower reversibility in the reaction region is also called since Coulombic efficiencies are low for high cut-off poten-tials. The theoretical conversion capacity for SnO2 is instead con-siderably higher (711 mAh g−1) and therefore in compliance with the practical capacity value extracted from first sodiation. The conversion/alloying balance for SnO2 has been thus calculated to be 95 to 5. Having the conversion mechanism, affected by a large irreversibility, a major role in contributing to capacity harvesting, SnO2 performances progressively degrade. The authors then concluded that the outstanding SnO energy storage capabilities must be attributed to various degree of alloying reactions in the sodiation. Partial reoxidation of metallic Sn finds also a counter-check in the work by Su et al.,[125] where tetragonal metallic Sn is detected when the electrode is discharged to 3.0 V. Contrarily to what Lu et al., here SnO is not entirely converted to SnO2 upon following oxidations, on the other hand, SnO is restored with a change in its crystallographic phase that passes from a tetragonal symmetry to an orthorhombic one. Despite the widespread disa-greements still present in literature concerning the reversible mechanism involved in SnO energy storage, two main processes can be undoubtedly cited. The first one is the complete conver-sion reaction leading to metallic Sn, possibly (but not completely excluded) bypassing a potential sodium intercalation in SnO structure according to the equation

SnO 2 Na Na O Sn

theoretical capacity: 398 mAh g2

1( )+ → +

− (3)

The second process is sodium alloying in Sn at low poten-tials leading to NaSn2, as elucidated by more accurate XRD analysis

2 Sn Na NaSn theoretical capacity: 112.8 mAh g21( )+ → − (4)

The total amount of theoretical specific charge would thus amount to 511 mAh g−1, that well agrees with experimental reversible capacity values (403 mAh g−1 at 20 mA g−1). Formation of a NaxSn phase has been also recently confirmed by theoretical calculation by Wang and co-workers.[126]

Conversion Transition Metal Chalcogenides: Similar to transi-tion metal oxides, transition metal chalcogenide materials may also undergo two different types of reaction, such as intercala-tion followed by conversion reaction[127] or conversion followed by alloying reaction.[128,129] Transition metal chacogenides of general formula TMX (X = S, Se, Te and TM = Fe, Co, Ni, Sn, Mo, W, Sb, Mn, Zn, Cu, Ti, Ta, V, Bi, Nb, or a mixture of them), albeit bearing the advantage of improved specific capaci-ties, introduced new technological challenges. Sulfur-based TMX, for instance, required the shuttling of soluble lithium and sodium polysulfides, intermediate reaction products, to be carefully addressed, while more in general, volume variations remained as a major limit.[130–132]

Transition metal chalcogenides such as SnS and SnS2, which are subjected both to conversion-alloying mechanism, exhibit high theoretical capacity but bear with themselves the disad-vantage of low intrinsic diffusion kinetics and a large volume change. Various approaches have been adapted in order to overcome these drawbacks. Kim[128] successfully prepared via hydrothermal synthesis and sequential carbothermal reduction, SnS/rGO hybrid nanocomposites with SnS crystalline size of 2 nm. Excellent reversible electronic capacity of 1230 mAh g−1 and long cycle stability were achieved, confirming that the low intrinsic electronic conductivity and large volume expansion limits can be successfully overcome by integration with rGO matrix. The rGO matrix was employed for the ad hoc synthesis of acid-exfoliated MoS2-graphene self-standing electrode.[133] Despite the interesting mechanical properties, in terms of ten-sile strength and failure strain, of the MoS2/graphene-based paper, its specific capacity when assembled in a sodium half-cell, was limited to only 240 mAh g−1 obtained at 25 mA g−1. Recently, Jung and co-workers collected in a theoretical work the calculated electrochemical data of a large number of transi-tion metal chalcogenides, such as TiS2, VS2, CrS2, CoTe2, NiTe2, ZrS2, NbS2, and MoS2.[134] In their theoretical approach, worth to be noted are the efforts put in taking into account also phase transition information, often neglected. TiS2 and NbS2 were found to be the chalcogenides with the lowest energetic bar-rier to sodium diffusion (0.22 and 0.07 eV, respectively). These findings on NbS2 received recently an experimental confirma-tion thanks to the work by Yan and co-workers.[135] In this work also metal and Se-doped NbS2 nanosheets were found to be an excellent high-rate anode compound for SIBs, delivering a capacity of 260 mAh g/8 at 1 A g−1 during the 750th cycle.

In general, three different approaches were successful in tackling the most puzzling limitations of transition metal chal-cogenides. Nanostructurization and morphology engineering is one of the most relevant. As an example, chemically exfoliated





NbS2 nanosheets delivered the high reversible specific capacity of 205 mAh g−1 at 100 mA g−1, exhibiting high rate perfor-mance and excellent cycling stability.[136] Similarly, urchin-like CoSe2 single crystals,[137] NiSe2 nanooctahedra,[138] and CuS2 nanodisks[139] have also been reported. Recently, Shen and co-workers[140] proposed a novel strategy to successfully syn-thesize SnS2 nanosheets assembled hierarchical tubular struc-tures, by using cobalt–nitrilotriacetic acid chelate nanowires as soluble template. Owing to the hollow core and well separated nanosheets with maximally exposed edges, the SnS2 tubular structures exhibited excellent electrochemical performance. When evaluated as anode material for SIBs, they delivered a high discharge capacity of 708 mAh g−1 at 50 mA g−1 preserving up to 414 mAh g−1 after 50 cycles. Chemical stability can be improved by TMX hybridization with a conductive matrix, e.g., carbonaceous coating, conductive polymers, such as the case of FeTe2−rGO hybrid powders,[141] NiSe2/rGO,[142] and SnS/C.[143] CoS2 nanoparticles anchoring onto multichannel carbon nanofiber (MCNF) have been recently obtained by Zhang and co-workers via a simple solvothermal method.[144] Benefiting from the innovative structure, CoS2@MCNF electrode deliv-ered high capacity (537.5 mAh g−1 at 0.1 A g−1), superior rate capacity (537.5 mAh g−1 at 0.1 A g−1 and 201.9 mAh g−1 at 10 A g−1), and long cycle life (315.7 mAh g−1 at at 1 A g−1 after 1000 cycles). Very recently, sponge-like composite assembled by cobalt sulfides quantum dots (Co9S8 QD), mesoporous hollow carbon polyhedral matrix, and a rGO wrapping sheets were synthesized by Sun and co-workers via a simultaneous thermal reduction, carbonization, and sulfidation of zeolitic imidazolate frameworks on GO precursors. Benefiting from the hierarchical porosity, conductive network, structural integ-rity, and a synergistic effect of the components, the sponge-like composites used as binder-free anodes manifest outstanding sodium-storage performance in terms of excellent stable capacity (628 mAh g−1 after 500 cycles at 300 mA g−1) and exceptional rate capability (529, 448, and 330 mAh g−1 at 1600, 3200, and 6400 mA g−1).[145] In a very recent work, the fast exfoliation of few-layers SnS2 put Lee and co-workers in condition to prepare a SnS2–graphene sheets hybrid with extraordinary power capabili-ties (up to 8 kW kg−1).[146] In this case, sodium storage has been demonstrated to occur via full pseudocapacitive, allowing an excellent structural retention and limited electrode disruption.

Ultimately, heteroatom doping and transition metal substi-tution of TMX, forming binary (CuV2S4,[147] FeV2S4

[148]) or ter-nary (CuNiSnS[149]) structures, have also been considered as valuable alternative to improve SIBs anode performances. Kang and co-workers, for example, prepared a series of Co-doped FeS2 with different Co contents, to combine the high rate capa-bility of FeS2 with superior capacity of CoS2.[150] Unique lychee-like FeS2@FeSe2 core–shell microspheres (FeS2@FeSe2) were designed by Li and co-workers.[151] They have been fabricated through two consecuetive hydrothermal steps, and were fur-ther investigated as anode material for SIBs. They displayed superior sodium storage performance with a high reversible residual discharge capacity of 350 mAh g−1 after 2700 cycles at 1 A g−1, and an outstanding 301.5 mA g−1 after 3850 cycles at 5 A g−1 with a Coulombic efficiency close to 97%. Worth to be noted is that in this very last case no appreciable improve-ments, in terms of specific capacity, were achieved if compared

to previously discussed carbon-based anode for SIBs. The Na storage properties of metal chalcogenides are still not enough qualified to replace the commonly used electrode materials used in LIBs, in terms of specific capacities, cycling stability, cost, and safety issues. To rival the other well developed mate-rials, taking advantage of higher theoretical capacities of TMX, further development is still required to build high-efficient, cost-effective, and robust battery electrodes.

3.1.3. Alloying Materials

Vast computational efforts have been recently dedicated to predicting electrode potentials and stoichiometry for binary Na–Me intermetallic compounds formed with Group 14 (Si, Ge, Sn, and Pb) and Group 15 (P, As, Sb, and Bi) metallic and non-metallic elements. Differently from intercalation compounds, for which sodium uptake is limited by constraint induced by rigid frameworks, usually metals and metalloids can withstand multiple sodium atoms per single atom, resulting in capacities ranging from 300 to 2000 mAh g−1 with operational voltages sufficiently low (below 1.0 V vs Na/Na+) to be appealing for usage on SIBs anode side. Starting from the exploratory theo-retical investigation of sodium alloys with group 14, 15, and 16 elements (Si, Ge,[152] Sn,[153,154] Pb,[154] as well as P,[155,156] As and Sb[157,158]) conducted by Chevrier and Ceder[159] in 2011, the research activity has flourished. As already highlighted for conversion material, the uptake of more than a sodium atom has dramatic consequences on structural stability of the host material. If this aspect is mild for lithium compounds, it is overwhelmingly severe for sodium, due to the larger ionic radius, and leads to irreparable consequences if not prop-erly addressed. Contrarily to lithium, for which silicon is representing a true game-changer for what concern achievable energy densities,[160–162] sodium does not form intermetallic compounds, being the formation of the amorphous com-pound NaxSi energetically unlikely starting from crystalline silicon.[163,164] Nevertheless recent theoretical atomic-level assessment[164] revealed how pure amorphous silicon would allow the uptake of 0.76 Na per Si atom guaranteeing capacity of 725 mAh g−1 comparable with other Group 14 elements. Such interesting property is supposed to derive from the relatively strong ionic bond between Si and Na atoms, connected to a modest decrease in Na charge state with x increase, before clus-tering of Na in Si bulk phase leading to neutral Na atoms. A very recent work aimed at finding an experimental confirmation failed at obtaining the forecasted theoretical capacities, and only a stable capacity of almost 150 mAh g−1 has been achieved at 0.05 C, after a poor first cycle Coulombic efficiency of 17%.[165]

Baggetto et al. took a basic but thorough electrochemical analysis of sodium alloying in a germanium thin film sputtered on roughened copper foil.[152] Throughout galvanostatic inter-mittent titration technique measurement at low current (C/60), calculated thermodynamic potential values could be closely approached during sodiation. A larger gap is instead observed when the desodiation is ongoing due to a high polarization con-nected to sodium removal. Capacity retention, despite prom-ising values extracted during first cycles (≈300 mAh g−1) quickly drops, within 15 cycles, to few tens of mAh g−1. Recent in situ





TEM analysis of amorphous germanium nanowires revealed fast sodiation/desodiation kinetics associated to sodium richer stoichiometry with x = 1.56.[51] Experimental studies conducted on metallic tin have showed a general inconsistency with DFT calculation aimed at finding the stable intermediate phases in Na–Sn system. Despite calculation would predict a sodia-tion proceeding through the steps NaSn5, NaSn, Na9Sn4, and Na15Sn4,[166] ex situ characterization attempted by Ellis et al. on an experimentally sodiated tin electrode, identified four alloying steps with different composition and phases, namely amor-phous NaSn3, α-NaSn, Na9Sn4, and crystalline Na15Sn4. The conclusion of the authors has been that the correspondence between experimental and calculated plateaus is only apparent and conceals the intrinsic thermodynamic limits in achieving complex crystalline intermetallic phases, even though stable, at room temperature. The formation of amorphous phases,[52] not predicted by DFT calculation, can be thus motivated by low mobility of sodium and tin particle morphological specific fea-tures that deviate the system from calculated equilibrium crys-talline states. Volume expansion evaluations derived from in situ TEM analysis suggested how drastic measures were due to guarantee the electrode survival to cycling and overcome rup-ture. Fine dispersion of tin particles on carbonaceous matrix, achieved by high energy milling method, has been then pro-posed to avoid mechanical stress build up with discrete results in terms of Coulombic efficiency and rate capabilities.[56] More exotic approaches, using cellulose fiber as a mechanical buffer, have been recently proposed.[167] According to Sb phase dia-gram, two intermetallic phases are suitable for sodium storage: NaSb and Na3Sb. Qian et al. first demonstrated a Sb–carbon nanocomposite with the initial capacity of 610 mAh g−1.[58] Both rate capabilities and cyclability were astonishingly impres-sive, since 300 mAh g−1 could be achieved at 2000 mA g−1 and capacity fade was limited to few percent within the first 100 cycles. Nonetheless, the latter feature was achieved only if FEC was employed as electrolyte additive and SEI-forming agent. This aspect strongly bonds the reversibility of sodium alloying mechanism to the stability and mechanical robustness of solid electrolyte interface, and can be reasonably extended to other intermetallic compounds. As for other metals also for antimony information on charge storage mechanism is nowa-days limited. XRD patterns of the ongoing alloying process are rather featureless, symptom of intermediate amorphous phases between Sb and Na3Sb.[54] Added to these sensible differences between first cycle potential profile and following ones are reg-istered, and appreciable discrepancies in mechanisms are also collected if Sb particles with different morphologies and dimen-sion are investigated. As an insight derived from in situ syn-chrotron XRD measurements taken on Sb/C composite (7:3), Ramireddy et al. failed in the recovery of crystallinity of Sb after first desodiation.

While binary systems involving Sn–Sb,[168] Sn–Cu,[169] and Al–Sb[53] have been studied as potential SIBs anode, a recent trend is to inspect ternary Sn–Sb–Ge phases. Addition of a third element to a binary system further complicates phases stability and equilibria, marking the appearance of amorphous phases or modified lattice parameters due to substitutional solid solubility of Sn into Ge, for example. Many ternary phases have been investigated, and among them Sn80Ge10Sb10,

Sn60Ge20Sb20, Sn50Ge25Sb2, and Sn33Ge33Sb33 demon-strated to be valid candidates as anode material for SIBs with reversible capacities of 728, 829, 833, and 669 mAh g−1 respectively. The highest stability is achieved by Sn50Ge25Sb25 composition, which guarantees 662 mAh g−1 after 50 cycles. Interestingly all the experimental values, except for Sn80Ge10Sb10, are well above the weighted average combina-tion of the theoretical capacities of a single element, 800, 743, 714, and 664 mAh g−1. A possible explanation for this advanta-geous behavior has been put fourth thanks to TEM and high-resolution transmission electron microscopy (HRTEM) micro-structural analysis and concerns the optimum system consti-tuted by 10–15 nm Sn and Ge(Sn) crystalline nanoparticles dis-persed in an amorphous matrix. The enhanced capacity would be a consequence of a unique ability of Ge nanocrystallites, that are heavily alloyed with Sn, to sodiate beyond 1:1 Ge:Na (369 mAh g−1) ratio.[170] An hybrid conversion–alloying mech-anism has been recently proposed to explain SnSe/rGO com-posite electrochemical behavior.[59] Overlapping potential pro-files and CV scans have been obtained in this case, symptom of highly reversible processes. Sorting all the ongoing process out is however a tough task that requires further investigation. Composites made of nanostructured alloying elements depos-ited on suitable carbonaceous substrates, positively contributed in extending cycle life of alloying materials in sodium cells. Sn@C-nanosphere,[171] Sb/C,[58] Se/C,[172] Sb/C-nanofibers[173] allowed to achieve promising reversible capacities of 200, 575, 485, 446, and 435 mAh g−1, respectively.

3.1.4. Phosphorus

Phosphorous, belonging to the 15th group of the periodic table, has three widely known allotropes commonly named red, black, and white phosphorus. The latter, whose fundamental forming unit is the P4 tetrahedron, is highly reactive and toxic. Red phos-phorus electrochemical reactivity with sodium has been inves-tigated before but its poor electronic properties make it only a modest and not reliable anode material for SIBs.[155] Black phos-phorus, with its unique crystalline structure, is the element cur-rently addressed as the “holy Grail” in many fields, including the energy storage one. It is a layered material resembling graphite with an unusual puckered structure made of cova-lently bonded P atoms.[174,175] It has an interestingly high elec-tronic conductivity (≈300 S m−1) and its cycling performances are known to be promising with lithium,[176] but sluggish with sodium. Cui and co-workers overcame the limitation imposed by the conversion mechanism by preparing a sandwiched graphene–black phosphorus hybrid material with impressive capacity retention and rate capabilities.[156] As claimed by the authors the graphene sheets provide an elastic buffer accom-modating the expansion of exfoliated black phosphorous (phos-phorene) layers along y and z axes during alloying to Na3P. This guarantees capacities abundantly above 2000 mAh g−1, stable over 100 cycles, at 0.02 C. The sandwiched structure has been obtained by a relatively simple self-assembly approach conducted in N-methyl-2-pirrolydone media. Recently, Komaba and co-workers tried to investigate the mechanism taking place during the three-electron reduction of P to Na3P, and a complete





conversion from orthorhombic black P to hexagonal red P was found.[175] Crystalline black P is not obtained upon reoxida-tion and this has been addressed to the metastable nature of the allotrope. In the same research, the electrolyte solutions and additives have been recognized to have a fundamental role in stabilizing conversion reaction. Amorphous phosphorous has been recognized to be favorable under the volume change point of view and in recent studies this route is encouraged rather than the crystalline phosphorus one. Amorphous red P/carbon composites in which the PC bonds formation is intentionally sought to firmly anchor P to graphene layers are nowadays able to ensure ultrastable efficiencies (0.002% decay per cycle within 400 cycles) and excellent rate capabilities (809 mAh g−1 at 1500 mA g−1).

3.1.5. Insertion Materials: Titanate and MXenes

Some inorganic compounds are known to be active toward sodium ions intercalation at low potentials. This feature makes them attractive to be used as anodes in SIBs. The study of such compounds have been focusing in the recent years on TiO2-based materials and layered transition metal carbides or as generally called MXenes layered compounds. Titanium oxide in all its polymorphs (anatase, rutile, brookite, and TiO2-B[177]) has repeatedly fascinated the scientific community thanks to its interesting physical properties that worthed it the honor to be studied in many different technological fields such as photocatalysis applied to pollutant degradation and water split-ting, sensors technology, medicine, and ultimately energy storage facilities. Added to this a further drive to its fortune has been the natural abundance, environmental friendliness, tunable synthesis, and nontoxicity for human beings. Ti3+/Ti4+ couple exhibits the relatively high RedOx potential around 1.5 V if cycled against lithium as widely demonstrated in literature.[178] This aspect makes it unsuitable to be used both as high energy anode and cathode for LIBs. In a sodium environ-ment, Ti-based compounds show an electrochemical activity centered at lower potential values, from 0.5 to 1.0 V versus Na/Na+, interesting for a possible development of anodes not shadowed by sodium plating danger. Both amorphous and crystalline TiO2 have been tested against sodium. The former, shaped in nanotubes directly grown on a Ti substrate, manifests a slow and undesirable activation process that progressively sets the capacity up to ≈100 mAh g−1. In this case, the energy storage mechanism seems to occur almost exclusively via double layer capacitance instauration.[179] Mechanism involved in sodium interaction with crystalline TiO2 is controversial still today. Some studies revealed that the crystalline structures of nano-sized anatase TiO2 are well retained upon sodiation, conversely, other research groups claim that a conversion-like process that in the end leads to metallic Ti is involved.[180] A carbon coating sensibly increases electrochemical performances of anatase nanocrystals (NCs).[181] In a recent investigation, different crys-talline facets exposition of ad hoc synthesized TiO2 anatase nanocrystals confirmed as a fundamental factor in extending capacity and stability.[182]

Sodium titanate with the unit formula Na2Ti3O7, prepared using a ball milling procedure, has been recently investigated.[183]

It has been found to be active toward insertion of two sodium ions per unit formula (Na4Ti3O7) at an incredibly low potential (<0.5 V vs Na/Na+) through a two-phase process to which correspond stable plateaus with low charge/discharge hysteresis. Structural analysis of intercalated titanate disclosed interesting properties of crystalline lattice in varying Na site through vari-ation of joint angles between TiO block, demonstrating an overall structural flexibility never observed in TMOs. Never-theless, few days storage of the sodiated compounds in an inert environment gradually makes it reverting back to the Na-poor phase, unveiling a self-discharge mechanism poten-tially bottlenecking the use of this material as an anode in a full-cell assembly.

MXenes is a class of 2D transition metal carbides and car-bonitrides first proposed by Barsoum and co-workers.[184] They are obtained by chemical etching in a hydrogen fluoride (HF)-containing environment, of the A layers from Mn+1AXn (n = 1,2,3), being M an early transition metal, A a 13 or 14 group element, and X carbon or nitrogen. Since when they were pro-posed for the first time as anode materials for LIBs the debate focused on the type of mechanism involved in the energy storage process, possibly capacitive/pseudocapacitive involving material surface[185] or more based on a pure intercalation of desolvated Na ions.[186] In both cases, in which the investigation of Ti2CTx was perpetrated (where Tx represents the different surface terminations introduced by the HF treatment[187]), high energy performances resembling supercapacitor ones could be achieved. Wang et al. shed light onto the surface and intercala-tion processes using DFT and scanning transmission electron microscopy (STEM) and they emphasized the topotactic locali-zation of functional groups (e.g., OH−, F−, O2−) and intercalated Na atoms on the top site of the central Ti atoms and C atoms on the Ti3C2Tx monolayer.[188]

To explore their unique properties, single/few-layer MXenes have been prepared by intercalation of water or organic mole-cules in aqueous medium followed by sonication.[189] In par-ticular, Ti3C2Tx/CNT free standing composite electrodes were realized, thanks to the negatively charged nanosheets and the positively charged CNT, which partially prevented the restack of the MXene layers and revealed high volumetric capacity of 420 mAh cm−1 at 20 mA g−1 and good cycling stability.[190] The importance of MXene electrode architecture in energy storage applications has been further confirmed by Zhao et al. The free standing electrode made of hollow MXene spheres with a 3D macroporous structure prepared via polymethil-methacrylate spherical templates exhibited indeed improved performances compared to the multilayer MXene in terms of capacity and rate capability.[191] Recently Wu et al. reported another delamination method for Ti3C2Tx by organic solvent assisted high energy ball milling, which may prevent the par-tial spontaneous oxidation of the MXenes sheets in the aqueous medium. The as prepared few-layer nanosheets structure shows improved rate performance compared to the pristine MXene when tested as anode in sodium ion batteries. Thanks to their layered structure, hydrophilic terminal groups and excellent electrical conductivity hybrid nanomaterials with MXenes can be easily prepared and show improved reversible capacity in SIB. Sb2O3/Ti3C2Tx and MoS2/Ti3C2Tx composites have been reported.[192,193] The former in particular inspires possible





further development of novel MXene materials, showing high rate performance and good reversibility as well (470 mAh g−1 almost constant during 50 cycles at 50 mA g−1 compared to a capacity retention of only 43% in the case of the bare Sb2O3), suggesting that the MXenes can provide fast electron pathways and help buffering the volume variation during cycling.

3.2. Cathode Materials for SIBs

In this section, an overview of the electrochemical and phys-ical principles lying beyond cathode materials so far synthe-sized and characterized will be provided. Going through the materials, layered transition metal oxides, whose study has recently regained enthusiasm, will be described at first place. Additionally most recently proposed advancements, such as polyanaionic and Na superionic conductor (NASICON)-type compounds, will be discussed as they tried to enhance the cycle stability, environmental friendliness, reliability, and energy densities. Switching from lithium to sodium intrinsically bares with itself the drawback represented by a lower standard reduc-tion potential of Na/Na+ RedOx couple, this is the reason why efforts have to be put in extending reversible capacity in order to reach comparable energy densities. It can be calculated that, in order to overcome the ≈0.5 V discrepancy between lithium and sodium based systems, a 12.5% specific capacity excess should be kept in mind in designing an innovative active material. As previously stated, studies of sodium intercalation in inorganic layered material started at the end of the 1980s, in conjunc-tion with lithium ones. They were both referred to cathode materials, TiS2

[196] and P2-NaxCoO2,[197,198] for which sodium intercalation was confirmed and quantified, as well as anode sodium–lead alloys. Prototypes of full sodium-ion cells were also assembled demonstrating surprisingly high cyclability for more than 300 cycles at an operational potential slightly lower than 3.0 V. Precisely this latter feature, did not particularly draw the attention if compared to high voltage LiCoO2/graphite sys-tems. After the sodium renaissance started due to mounting sustainability concerns back in 2010, a plethora of materials have been proposed as cathode, including layered and tunnel-like transition metal oxides,[199,200] transition metal sulfides and fluorides,[201] polyanionic compounds,[39,202,203] Prussian blue analogues,[204] as well as organic carboxylates and polymers.[205] According to the classification, Delmas et al. conceived based on (i) coordination configuration that sodium ions assume once inserted in the layered structure and (ii) layers stacking sequence,[206] different layered oxides have been here listed, and the most relevant electrochemical and structural features as cathode material for sodium batteries have been provided.

3.2.1. Layered Transition Metal Oxides (TMO)

P2-type TMOs: NaxCo2 and NaxMnO2 stand as the most widely investigated P2-type layered material in sodium environment. NaxCoO2 can be prepared relatively easily in almost every lay-ered configuration according to synthetic temperature and sodium precursor content.[194] P2 phase NaxCoO2 has a well-defined stability window related to a sodium composition (x)

falling in the range 0.46 < x < 0.83, while for higher content of Na (0.83 < x < 1.00) a conversion to O3-NaxCoO2 occurs. All the above and other equilibrium phases can be appreciated in the phase diagram reported in Figure 4 (image (a)). P2-NaxCoO2 phase potential profile shows a substantially different trend if compared to lithium analogue one. While in the latter case a single and stable plateau at moderately high potential (4.0 V) with low polarization is observed; the former compounds show a stepwise potential change, spread over a wider window of potential values. As a consequence, energy densities of con-nected full-cell will be considerably lower. In sodium analogue the potential profile originates from both single-phase, referred to solid solution like mechanism and two-phase processes. In the second case, a two-phase like behavior might manifest inside the same stability range of a P or O, namely the same phase, due to sodium sub-lattice pattern formation. Inter-estingly, together with phase transformation also a sensible variation in sodium diffusion is obtained. O3-NaxCoO2 outper-forms P2-NaxCoO2 at high sodium content, while the trend is reversed for lower sodium concentration in the lattice due to fast Na diffusion in a honeycomb-like sub-lattice. In general, for NaxCoO2, diffusion coefficients are higher by an order of magnitude[207] (≈0.5–1.5 × 10−10 cm2 s−1 vs 1 × 10−11 cm2 s−1) than LixCoO2, and this will result also true for other insertion materials. A plausible explanation to this effect might be related to the weaker nature of Na+ as a Lewis acid compared to Li+ that would allow a faster diffusion between layers. Speaking about electrochemical performances, P2-Na0.74CoO2 synthesized by a solid-state route exhibits a reversible discharge capacity of 107 mAh g−1 at 0.1 C, over a potential range spanning from 2.0 to 3.8 V.

P2-NaxMnO2, despite the electrochemical activity compa-rable to those of cobalt analogue in the 0.45–0.85 composi-tion range, showed low sodium diffusion and poor structural reversibility due to a strong Jahn–Teller effect of Mn3+.[208] In recent investigations, the cations substitution revealed as an efficient method to enhance structural stability of NaxMnO2. Ni/Mn, Fe/Mn, and Co/Mn substitutions have been consid-ered and some research groups succeeded in suppressing the multiple phase transitions improving reversibility. For example, P2-Na0.67Ni0.33Mn0.67O2 has been claimed by Lu and Dahn to provide 170.7 mAh g−1 by the reversible extraction of 0.67 of sodium. Further XRD analysis revealed how P2 phase is maintained until 0.33 stoichiometry is reached while a coex-istence of P2 phase, O2-type stacking faults, and O2 phase is achieved below x = 0.33.[209] 20 at% Mg cation substitution in Na2/3Mn1-xMgxO2,[210] coupled to a slow cooling step during solid state synthesis, allowed Komaba and co-workers to achieve pure P2 phase coupled to an anomalous initial capacity of 200 mAh g−1, quickly fading toward lower values.[211] Further doping of Na–Ni–Mn–O phases, disclosed as a valu-able strategy in improving electrochemical performances, and in particular, in suppressing deleterious P2–O2 phase transition occurring at potentials above 4.1 V versus Na+/Na. In P2-Na0.83Li0.07Ni0.31Mn0.62O2

[212] substitutional Li+ into TM sites is accountable for the residual persistence of sodium atoms in the structure even in deep charge conditions, thus stabilizing the layered stacking in the P2 configuration. P2-Na2/3Ni1/3Mn2/3−xTixO2, with an optimum composition of






Figure 4. a) Temperature/composition phase diagram of layered CoO2 compound in which single phase stability zones and equilibrium lines of different structures are indicated. Reproduced with permission.[194] Copyright 2014, American Chemical Society. b) Lattice parameters evo-lution of P2-Na0.7Fe0.4Mn0.4Co0.2O2 electrode from in situ synchrotron analysis, potential profile of capillary cell is also reported at the top of the image. c) Schematic illustration of the cell experimental set-up used in synchrotron analysis and in situ XRD patterns evolution for P2-Na0.7Fe0.4Mn0.4Co0.2O2. Reproduced with permission.[66] Copyright 2015, Wiley-VCH. d) Material design of spherical particles with a radially aligned gradient composition and connected charge/discharge profile (left) and GCPL electrochemical characterization at different tempera-ture (right) of the radially aligned particle and a control bulk sample. The core of the particles is mainly composed by Na[Ni0.75Co0.02Mn0.23]O2 while outer shell gets progressively poorer in Ni concentration, transitioning to Na[Ni0.58Co0.06Mn0.36]O2. Reproduced with permission.[195] Copyright 2015, Macmillan Publishers Limited.




x = 1/3, has been investigated by Komaba and co-workers.[213] It delivered a specific capacity higher than 100 mAh g−1 with a good capacity retention, thanks to the stabilizing effect of Ti atoms which sensibly reduces, by nearly 50%, the crystal-line cell volume shrinkage (23% of undoped sample versus 12–13% of the doped one). Na0.67Mn0.67Ni0.33−xMgxO2

[214] with x = 0.05 represents, among the stoichiometries investigated (0 <x < 0.15), a sensible tradeoff between good specific capacity (≈120 mAh g−1) and improved capacity retention. The coex-istence of P2 and O2 phase was confirmed, for the undoped sample, on the microscale by STEM analysis. This implies stacking faults formation which negatively influences stability and reversibility. Together with P2–O2 transition suppres-sion, different levels of TM substitutions revealed to be effec-tive in containing Mn leaching with cycling. Passerini and co-workers[215] systematically studied the Ni-to-Fe ratio effect in NaxMO2 (with M = Ni, Fe, and Mn) in suppressing Mn dis-solution and improving material stability. Worth noticing is that the Ni-rich phase Na0.6Ni0.22Fe0.11Mn0.66O2 shows significantly improved electrochemical characteristics.

Apparently the only efficient way to increase reversibility excluding deleterious phase transitions is to narrow the opera-tive potential window. An important contribution to the under-standing of phase transition mechanisms was given by Jung et al.[66] by in situ synchrotron X-ray diffraction of the com-pound P2-Na0.7[(Fe0.5Mn0.5)1−xCox]O2 (x = 0, 0.05, 0.10, and 0.20). Lattice parameters have been observed varying according to multiple effects occurring during charging and discharging (images (b) and (c) in Figure 4). During desodiation (charging), a general contraction on the ab plane (aligned TMO2 slabs) is observed due to a decrease in transition metal atoms radii connected to oxidation. Along c direction, instead an expan-sion is observed due to a mounting electrostatic repulsion between oxygen atoms, left unshielded by Na+ cations. Since no phase transitions are so far involved, at least below 4.1 V versus Na/Na+, the general behavior of the material is the one of a solid-solution, with a slight cell volume change (2.1%) and a sloping operational potential. Above 4.1, a phase transition from P2 (P63/mmc) to a O2 (P63mc) structure occurs and it is underpinned by two-phase plateau region in potential profile and P2 phase XRD diffraction peaks progressively fading, as clearly visible in associated XRD pattern evolution in image (c) in Figure 4. The phase transition finds its explanation in the gliding of TMO2 sheets, that modifies the sodium coordina-tion sites; this structural modification induces a severe volume contraction of ≈15%, accompanied by mechanical stresses that progressively lead to a capacity loss with cycling. Reduc-tion in lattice cell volume has to be specifically connected to shrinking of sheets spacing along c axis due to deep desodia-tion, process already observed for P2-Na2/3[Ni1/3Mn2/3]O2.[216] Nevertheless, this mechanism shows an interesting revers-ibility even if accompanied with a conspicuous hysteresis due to dramatic cell parameters change. This aspect finds its expla-nation in the fact that no covalent bonds breaking or forming is involved during P2–O2 phase rearrangement, as would be the case of a P2–O3 transition. In this latter case, together with a lattice plane gliding, a 60° rotation of all the MO6 octahedra would be involved with an inevitable braking of MO bonds. Electrochemical performances account for a starting specific

capacity of 170 mAh g−1, averagely equal for all the Co level of doping. Starting capacity moderately decreases during the first 60 cycles reaching 100 mAh g−1. It has recently thoroughly examined the coexistence of different crystalline phases in NaxNi1/3Co1/3Mn1/3O2 and their interplay mechanism in sta-bilizing the layered structure.[36] NaxNi1/3Co1/3Mn1/3O2 with intergrowth P2/O3/O1 structures showed a better stability and a suppression of interfacial microstrain during high voltage charging. This eventually allowed to achieve a high initial reversible capacity of 142 mAh g−1 and an outstanding capacity retention of 93% after 50 cycles.

O3-Type TMOs: As an excess of sodium precursor is employed during layered oxide preparation, O3 phase is more likely to be obtained over a wide range of synthesis tempera-tures. O3-TMOs (M = Mn, Cr, Fe, and Ni) are thus intrin-sically provided with a higher starting content of sodium and could potentially deliver stable capacities suitable for commercial rechargeable cells. O3- NaCrO2 has been widely investigated and the most promising results line up on capaci-ties between 113 and 120 mAh g−1, stable for 50 cycles[217] or more, depending on the application of a carbon coating.[218] The carbon coating in the latter and other cases has proven to be extremely beneficial, apart from increasing electrode rate capa-bilities, also in preventing the transition metal ions leaching from the structure during cycling, a drawback responsible for capacity fading. Komaba et al., working on O3-NaNiO2 pure and Mn-doped compounds, highlighted how as in P3-TMOs, like-wise in O3-TMOs, the potential window amplitude has a crucial role in limiting deleterious phase transitions. By Mn doping of the structure can be obtained better performance in terms of rate capabilities and stability: stable specific capacities between 105 and 125 mAh g−1 can be obtained over a large range of current densities (240–4.8 mA g−1) in the 2.2–3.8 V versus Na/Na+ window. Capacities drastically improve if the desodiation is pushed further up to 4.5 V, inevitably involving additional phase transition that for O3-NaNi0.5Mn0.5O2 amount to 4: O3→O′3→P3→P′3→P3″, just as much the steps appearing in the potential profile.[35] Despite these multiple steps involved, the material ensures a good reversibility, symptom of the high tolerance to phase transition of a layered structure in which cova-lent bonds are not involved in vertical ordering. The equilibria Ni2+/Ni3+/Ni4+ are responsible for the charge compensation inside the structure during desodiation while Mn ions behave as an electrochemically inactive species. A benefit of Ni-rich layered phases resides also in the great difference in the ionic radii of Na+ (1.02 Å) and Ni2+ (0.69 Å), which likely avoids ion exchange between Na+ and Ni2+ in the layers, extending stability and the high-capacity delivery. An outstanding contribution to this topic has been given by Hwang et al.[195] which in 2015 proposed a O3 cathode constituted by spheres made of radially aligned columnar structures with a gradient composition tran-sitioning from an inner part richer in Na[Ni0.75Co0.02Mn0.23]O2 to an outer one where Na[Ni0.58Co0.06Mn0.36]O2 was the major phase (image (d) in Figure 4). By scanning electron micro-scopy (SEM) probe analysis, the Ni concentration reduction, from 80 to 58% by weight, has been verified to occur along all the 6 µm long columnar structures. In a sodium half-cell con-figuration, this material delivered reversible specific capacity of 157 mAh g−1 at 15 mA g−1, while in a whole-cell set up,





coupled with a hard-carbon anode, it demonstrated a high capacity retention of 80% after 300 cycle. The improved stability has been addressed to the higher resistance of a Ni poorer phase to electrolyte corrosion compared to a richer phase, which, by many others, has been demonstrated to be more electrochemi-cally active toward sodium intercalation. Among O3 phases, even Li-doped compounds such as Na0.95Li0.15(Ni0.15Mn0.55Co0.1)O2,[219] Na0.78Li0.18Ni0.25Mn0.583O2,[220] as well as quaternary phases Na(Mn0.25Fe0.25Co0.25Ni0.25)O2,[221] have been recently pre-pared and suggested as high energy cathodes, delivering ≈200 mAh g−1 and improved reversibility compared to NaNi0.33Mn0.33Co0.33O2. Despite being highly promising for what concerns their electrochemical behavior, they heavily suffer stability issues connected to moisture and CO2 sensi-tiveness. As synthesized O3- NaNi0.33Mn0.33Co0.33O2 indeed progressively converts into O1-phase to end up to P3 phase if aged in air for 300 d.[222] Intercalation of water, similarly, com-petes with Na insertion and progressively excludes the latter and reduces Na uptake capabilities below 0.33 per formula unit upon cycling as demonstrated for NaxNi0.22Co0.11Mn0.66O2. Also carbonated insertion, aided by water, has been discussed as a relevant factor that oxidizes Mn3+ to Mn4+ and decreases revers-ibility and increases overpotential. Further improvements are thus required to make them not sensitive toward air for their mass production, effort recently put in developing a water-proof cathode material O3-Na0.9[Cu0.22Fe0.30Mn0.48] O2 are heading in the right direction.[222] Quaternary O3 structures, namely Na[Ni0.4Fe0.2Mn0.4-xTix]O2,[223] NaTi0.25Fe0.25Co0.25Ni0.25O2,[224] showed how titanium doping is a wise approach in stabilizing such heterogeneous layered structures. A structure stabilization effect has been called upon Ti positive effect. In particular, Fe4+ ions migration into interlayer sites and Na-Fe ions exchange would be prevented during deep desodiation phase and following sodiation, respectively.

3.2.2. Polyanionic Compounds

This class of compounds derives its name from the tetrahedral XO4 and trigonal XO3 groups that partially replace MO6 octa-hedral metal clusters. A considerable quantity of transition metal phosphates,[225,226] fluorophosphates,[39] sulfate,[38] and pyrophosphates,[37,227,228] just to cite a few, offers the opportu-nity to exploit their peculiar 3D open-framework structure cre-ated by edge or corner sharing polyhedrons, to host alkaline ions. Instead of vertically aligned slabs of covalently bonded atoms, as in layered transition metal oxides, the structure of polyanionic compounds is permeated by straight or tangled channels with relatively low Na+ diffusion energies. Due to the extensively higher rigidity of a covalent 3D network, polyanionic compounds show thermal stability and impressive resistance to oxidation without phase transitioning connected to sodium uptake.[37,229–232] Added to this an overall higher operational voltage has been observed for polyanionic compounds either for lithium and sodium batteries. Since the introduction of olivine LiFePO4 compound by Goodenough and co-workers in 1997 as a high energy cathode for lithium batteries,[233] it appeared clear that large PO4

3− anion, thanks to its strong inductive effect and high electronegativity, has the power of sensibly increasing

the operational potential. Nevertheless NaFePO4, in its most thermodynamically stable maricite phase, does not show any activity due to the absence of a sufficiently wide pathway for sodium insertion. Interestingly, an electrochemically active triphylite-type phase obtained from alkaline cation exchange in olivine LiFePO4, delivered 125 mAh g−1 with capacity reten-tion of 88% after 50 cycles.[234] Interestingly, NaFePO4 charge/discharge mechanism occurs with different potential profiles. If during discharge two plateaus are observed, during charge only one is present, indicating that sodiation and desodiation might occur through different electrochemical routes. The keystone has been demonstrated to be the intermediate phase Na0.7FePO4 and the difference in volumetric mismatch between sodium-free phase FePO4 and Na0.7FePO4 (13.48%) compared to Na0.7FePO4 and NaFePO4 one (3.62%). This means that during charge, before FePO4 starts to form, a complete conver-sion of NaFePO4 to Na0.7FePO4 must occur. The two-step pro-cess is clearly visible in potential curve even at high current rates. During discharge, instead, due to the low energetic bar-rier between sodium-containing phases, once the lattice mis-match between FePO4 and Na0.7FePO4 is overcome, sodiation of FePO4 and Na0.7FePO4 can happen concurrently and a one-step curve is achieved.[235]

An interesting and handy way to induce sensible lattice modification and create new sodium pathways in a phosphate structure is to insert another anion as a charge-balancing spe-cies. It has been done including fluorine atom in the new fluo-rophosphates class of material, recently investigated as cathode for both lithium and sodium-ion batteries. Na2FePO4F[236] and its isostructural Na2CoPO4F[237] have been recently proposed thanks to their interesting electrochemical properties as lay-ered insertion materials in which intercalation mechanism proceeds with a solid-solution RedOx process rather than with a two-phase mechanism. The positive effect induced by F− ion emerges from its pronounced electronegativity, which further enhances the inductive effect and thus the potential. Added to this, the higher ionicity of NaF bond substantially modi-fies the coordination geometry of sodium as pseudooctahedral. The exclusive layered structure of fluorine-enriched phosphates results in the uncommon coordination of transition metal ion (Fe, Co, or Mn) which for Na2FePO4F consists in octahe-drons whose vertex positions are occupied by four O and two F atoms. The octahedrons are then arranged in face-sharing pairs (Fe2O6F3). These units are corner-sharing connected (via fluorine atoms) to other pairs, forming chains along a direction. Corner-sharing PO4

3− tetrahedra complete the stacking along c direction. Further improvements were achieved by carbon-coating Na2CoPO4F.[238] Carbon coating inorganic particles is a procedure that, carried out by pyrolysis of a previously adsorbed organic molecule onto electrochemically active material surface, improves its electronic conduction and then the kinetic prop-erties. In the latter case, Na2CoPO4F/C nanoparticles showed interesting properties for what concerns specific capacity of the first cycle, around 120 mAh g−1, and a flat plateau at high potential, namely 4.5 V versus Na/Na+, although further effort should be put in extending its electrochemical stability. Among fluorosulfates, NaFeSO4F had been found, thanks to theoretical calculation, to be potentially active toward sodium intercalation thanks to Fe2+/Fe3+ couple.[239] Experimental data, however, are





lacking and no limitations to material synthesis have to be over-come yet.

Silicates: Transition metals orthosilicates, Na2MSiO4 (M = Fe, Mn, Co), have attracted a lot of attention due to their superior environmental sustainability and the possibility of exchanging more than one sodium atom per unit formula. This aspect significantly extends their theoretical capacity to value higher than 250 mAh g−1. Clear electrochemical performances have recently reported for Na2CoSiO4.[240] Corner-sharing SiO4 and CoO4 tethraedra constitute the backbone of the crystalline structure, while sodium ions occupy the tetrahedral empty spaces created consequently. Specific capacity is however lim-ited to 100 mAh g−1, with an average practical voltage of 3.3 V versus Na/Na+. Better results were obtained for Na2FeSiO4/C almost phase-pure composite.[241] Thanks to Na-rich phase, 181 mAh g−1 could be reversibly extracted from the structure, at a current density of 27.6 mA g−1. Capacity retention was however limited to 80% after 100 cycles, symptom of a pro-gressive degeneration of the structure. Energy density of such compound is not supposed to be exceptionally high, owing the relatively low operational voltage plateau of 2.3 V versus Na/Na+. Notably, silicates demonstrate an outstanding thermal stability up to several hundreds degrees celsius (>800 °C for Na2FeSiO4). Finally, Na2MnSiO4/C/graphene has been recently synthesized, via a facile assisted sol–gel method, by Zhu et al.[242] Due to the electronic insulation nature of silicates, heavy addition of carbon matrices is required. Despite the effec-tive graphene wrapping of Na2MnSiO4/C particles, the inter-esting initial capacity of 182.4 mAh g−1 is hardly retained during the following tens of cycles. Sodium silicates are extremely interesting compounds, but their sluggish stability still requires research efforts.

NASICON Materials: NASICON (Na+ superionic conductor)-type materials have been largely studied as solid electrolytes in high-temperature Na-S batteries. Since the exploratory work by Delmas et al.,[243] dated back to 1988, NASICON compounds with the general formula AxM2(XO4)3 (where A = Li, Na; M = V, Fe, Mn, Co; and X = P, S, W) have been investigated to be applied as electrode material for both LIBs and SIBs. The most rep-resentative compound of the class, Na3V2(PO4)3, acquires a 3D framework of PO4 tetrahedra and VO6 octahedra. Electro-chemical investigating of this composite showed two primary RedOx processes occuring at 3.4 and 1.6 V versus Na/Na+ highlighted by sharp peaks in cyclic voltammetry and corre-sponding to V3+/V4+ and V2+/V3+ RedOx couples. Galvanostatic cycling analysis with potential limitation (GCPL) revealed a reversible capacity of 90.9 mAh g−1 after 10 cycles within the 2.7–3.8 V potential window, unfortunately the Coulombic efficiency as well as capacity retention remain a major limit. Even though NASICON-type materials show an outstanding ionic conductivity, they are scarce electron conductors, for this reason electrochemical properties are severely hindered. Liu et al. developed a ball-milling based method, followed by Ar-annealing, to deposit a thickness-tunable carbon coating, extremely beneficial for the material performances. Nanostruc-turing and carbon-coating have been many times demonstrated to be valid approaches to overcome poor cycle stability and rate capability. Following is reported to be a recent and particularly clever method employed to extend NASICON performances

by means of electrospun carbon fiber embedding Na3V2(PO4)3 nanoparticles. The extremely simple preparation was made of two steps: (i) electrospinning of an aqueous formulation con-taining both the carbon fiber precursor and NASICON pre-cursors (NaH2PO4 and NH4VO3), and (ii) an annealing step in Ar atmosphere.[226] The outcome has been a homogeneous dispersion of nanometric particles confined in the central axis of structurally stable carbon fibers. Outstanding cycling performances at relatively high currents (2C), guaranteed more than 70 mAh g−1, provided at a stable potential of 3.4 V versus Na/Na+. As for general phosphate compounds, also NASICON-type materials can be subjected to fluorination with some beneficial effects. An example is NASICON analogue Na3V2(PO4)2F3, interesting for its high average voltage of 3.9 V and single-phase behavior with modest volume change (2%). Other examples include Na2FePO4F and Na1.5VPO4.8F0.7.[62] The latter revealed excellent cycling performances, with capaci-ties retention of 95% and 84% after 100 and 500 cycles, respec-tively, at 1 C. Such a promising feature has been recognized to be related to small volume change, fast Na diffusion (Ea ≈ 0.35 eV) in ab plane and lack of ordered compositions within the RedOx window.[244] Other NASICON-type materials include Na3Ti2(PO4)3. Unlike previously cited NASICON materials, and however similarly to Ti based compounds, Na3Ti2(PO4)3 has been proposed as an high performing anode for SIBs, due to its low operational voltage plateaus (<1.0 V vs Na/Na+). The clear Faradaic process occurring at such low potential is attrib-uted to Ti3+/Ti2+ RedOx couple, made available by the strong inductive effect of PO4.[245] Recently, transition metal pyrophos-phate arose as promising material proposed by Barpanda et al. Na2FeP2O7 was the first of its kind, exhibiting a modest revers-ible capacity of 82 mAh g−1 at C/20 in the potential window between 2.0 and 4.0 V versus Na/Na+. In Na2FeP2O7 lattice Fe2+ ions reside in two different sites and subsequently undergo oxidation to Fe3+ upon charge.[229] Particularly, Na2FeP2O7 supports quasi-3D sodium diffusion network with activa-tion energy that slightly varies according to sites accessibility and transition metal substitution. Energy barriers are indeed lower for Na2FeP2O7 (<0.49 eV) and higher for Na2MnP2O7 (0.58 eV).[246] Other polymorphs and analogues have been pre-pared, including β-Na2MnP2O7

[247] Na2CoP2O7[230] but they all

showed similar modest capacities (≈80 mAh g−1), poor cycla-bility, stability, and rate capabilities. The unique structure of pyrophosphate, despite some of them show a layered struc-ture,[230] arises from large tunnels formed by alternate stacking of layers of FeO6 octahedra and layers containing P2O7 units. Interestingly, together with a promising thermal stability up to 600 °C, they show also a polymorphic transition happening at 560 °C with no oxygen evolution nor structural rupture, but with a phase change from a triclinic (P-1) to a monoclinic (P21/c) spatial group.[232] This stability by far exceeds that of lay-ered oxides and is comparable to other polyanionic compounds. Intrinsic stability of pyrophosphate building block is respon-sible for its inertness and represents an added value to safety. More exotic polyanion compounds are those based on nitri-dophosphate group such as Na3TiP3O9N.[248] Na3TiP3O9N was the first member of cubic ionic conductor (CUBICON) family of materials that showed reversible Na+ insertion at room temperature. Despite the sodium diffusivity inside CUBICON





framework is considerably lower if compared to lithium ana-logues, the ionic conductivity is good, as well as thermal stability. Inside CUBICON structure different sites and sodium paths are possible, and substitution of Ti with other aliovalent transition metals with more than one RedOx pair achievable might extend the practical capacities, limited for Na3TiP3O9N to 80 mAh g−1. As previously reported, P2-Nax[Fe0.5Mn0.5]O2

[249] and O3-NaFeO2

[250] layered transition metal oxides suffer from low operating voltage, usually around 3.0 V. Yamada and co-workers recently reported an alluaudite-type structure cathode compound based on (SO4)2− anion, Na4Fe3(SO4)3.[251] The stable 3D structure made of Fe2O10 dimers interconnected via (SO4)2− groups ensures relatively fast sodium diffusion happening at potential of 3.8 V versus Na/Na+, extraordinarily high for Fe3+/Fe2+ RedOx couple. Theoretic calculation suggested that a partial Co doping of the above mentioned compound could further raise the operational potential to 4.76 V, when a par-tial desodiation would occur. Full desodiation would have been predicted to happen at 5.76 V. Experimental details are still lacking due to the difficulties in finding electrolyte solutions stable enough at such high potential values.[38]

4. Sodium-Ion Full-Cell

In the light of what previously introduced, we propose in the following chapters, a comprehensive depiction of the most representative sodium-ion full-cell architectures. The reader will be guided through the most recent advancements in the field. They have been here organized according to the overall symmetric or nonsymmetric assembly first, and subsequently, according to the material classes meaningful for the SIB full-cell architecture.

4.1. Nonsymmetric Full Battery

4.1.1. Graphite-Based SIB Full-Cells

The most widely studied anodes for SIBs are carbon-based materials. This is directly related to electrochemical peculiari-ties of carbonaceous matrix, as previously stated, in terms of operational intercalation and pseudocapacitance potentials. Added to this, natural abundance, low preparation cost and environmental friendliness add value to this class of mate-rials. Despite solvent co-intercalation mechanism is not fully understood, yet it acquires a foremost importance since it is associated with a complete recovery of the intercalated solvent molecules during discharge, as proven by the stability tests con-ducted on the full-cell. The practical validity of the graphite in SIBs full cells was examined by combing the graphite anode with different cathode materials. A Na1.5VPO4.8F0.7//graphite full cell[89] with the NaPF6 in diethylene glycol dimethylether electrolyte is reported and exhibits an average discharge voltage of 2.9 V versus Na+/Na. The energy density obtained from this system was about 120 Wh kg−1, based on the total electroactive materials, with a retention of the 70% of the initial discharge capacity after 250 cycles. By using laser drilling technique, Han et al.[252] explored a novel porous graphite film for an integrated

SIBs anode. When it is assembled with Na2V2(PO4)3 cathode, the full-cell showed an energy density of 130.7 Wh kg−1 with a capacity retention as high as 92% after 500 cycles. The possibility of practical use of graphite in SIBs has also been evidenced by Chen and co-workers,[253] by constructing a carbon coated Na3V2(PO4) cathode paired full cell with opera-tional voltage above 2.2 V and capacity retention over 80% for 400 cycles. Solvent molecules co-intercalation mechanism leading to a stage-1 GIC formation is considered the key step guaranteeing fast intercalation kinetics and reversibility. Scro-sati and co-workers[254] reported a novel high power SIB based on layered P2-Na0.7CoO2 cathode and the graphite anode in an optimized ether-based electrolyte. In this work significant is the graphite electrochemical presodiation process performed in a sodium half-cell, in order to precondition graphite and reduce the irreversible contribution deriving from activation of both anode and cathode material in the full-cell. Following a sodium rocking chair process, this system delivers an overall stable capacity of 80 mAh g−1 over more than 100 cycles, maintaining, at 10 C (i.e., 1750 mA g−1), 45% of its maximum capacity. The low specific capacity of the graphite is still an issue even though the cycling stability and rate capability of the graphite have been significantly improved over the years.[255]

4.1.2. Hard Carbon-Based SIBs Full-Cells

Exploring alternative anode materials for SIBs, more attention is payed to hard-carbon, which has a highly distorted structure and large interlayer distance.[81] Intercalation mechanism, as we know it at the basis of the mass-market lithium recharge-able batteries, is not the only process contributing to sodium energy storage. Sodium was found to be able to favorably interact with noncrystalline or partially crystalline carbona-ceous materials. As previously stated, disordered layers of hard-carbon are accompanied by large specific surface areas. Sodium storage in such structures takes place via different mechanism. They are: (i) direct intercalation in expanded-graphite domains,[96,256] at potential above 1 V versus Na+/Na, the (ii) adsorption on the surface and in correspondence of defects (i.e., surface defects as well as carbon flakes edges), and the (iii) mesopores filling by solvated Na atoms, forming a capacitive double layer, occurring at potentials below 1 V. This last process has been demonstrated to eventually lead to a mild volume expansion.[159] With all these mechanisms concurring in increasing material specific capacity, it becomes impor-tant the tracking of possible source of irreversible capacity as well as potential profile evolution with cycling. Despite the numerous studies directed to the disclosure of mechanism of sodium storage in hard-carbons, great discrepancies between results appear. Tarascon and co-workers[256] synthesized carbon nanofibers (CNFs) via electrospum polyacrylonitrile (PAN) fibers carbonization and showed that their electrochemical performances toward Na+ intercalation strongly depend on their structural and morphological characteristics induced by different carbonization temperature (Figure 5, image (a)). Low temperature (stage I) made CNFs contain copious amounts of O and N heteroatoms, a highly disordered structure and high amounts of ultramicropores and surface defects. Carbonization





at T > 1000 °C (stage II) gradually removes the heteroatoms and increases crystallinity as well as a mesoporous volume. For temperature higher than 2000 °C (stage III), a mas-sive removal of the pores takes place together with a further ordering of carbon crystalline planes (pairing of graphene sheets and progressive reduction of d-spacing). Komaba et al.[97,82] first reported a high capacity and excellent revers-ibility sodium ion full battery based on a hard-carbon anode and a layered NaNi0.5Mn0.5O2 cathode (Figure 5, image (b)) in propylene carbonate electrolyte solutions containing different sodium salts in concentration 1 mol dm−3: NaClO4, NaPF6, and NaN(SO2CF3)2 (sodium bis(trifluoromethanesulfonyl)amide, NaTFSA). The NaNi0.5Mn0.5O2// hard-carbon with 1 mol dm−3 NaTFSA propylene carbonate electrolyte exhibited a capacity higher than 200 mAh g−1, based on anode mass, and an average operating voltage of 2.8 V. Furthermore, no signifi-cant decrease in capacity was observed when the full-cell was

tested at the high rate of 300 mA g−1. Carbonate-based electro-lytes coupled with hard-carbon anodes introduce low first-cycle Coulombic efficiency (FCCE). In recent times, a lot of efforts have been directed to the synthesis of high SSA hard-carbons nanostructures (>400 m2 g−1)[104] as the external high surface area demonstrated to positively affect stored capacity, being directly connected to double-layer storage, kinetics properties of the materials, and electrode electronic connectivity. Never-theless, a large surface area underpins also a prominent SEI formation which massively irreversible segregates sodium atoms. This effect ultimately will decrease the total reversible capacity of the full-cell as the sodium atoms stored in SEI layer will systematically derive from the cathode material.[94,95] Luo et al.[258] employed GO as a 2D dehydration agent to prevent foaming process during caramelization of sucrose used for hard-carbon synthesis. The specific surface area of the resulting carbons reduces from 137.2 to 5.4 m2 g−1, and the FCCE result


Figure 5. a) The schematic representation of sodium ion storage mechanism in thermally treated carbon structures: below 1000 °C (I), between 1000 and 2000 °C (II), and above 2000 °C (III). Reproduced with permission.[256] Copyright 2016, Wiley-VCH. b) Full-cell performances (charge/discharge profiles and capacity vs cycle plots) of hard-carbon/NaNi0.5Mn0.5O2 battery assembly. Reproduced with permission.[97] Copyright 2011. Wiley-VCH. c) Full-cell charge/discharge profiles (upper image) and anode and cathode separate voltage profiles (lower image) of a starch-derived hard-carbon//P2/P3/O2-Na0.76Mn0.5Ni0.3Fe0.1Mg0.1O2. Na metal has been used as reference electrode and 1 m NaPF6 solution in PC as electrolyte. Reproduced with permission.[257] Copyright 2016, Wiley-VCH.




improved from 74% to 83%. By pretreating cellulose fiber with 2,2,6,6-tetramethylpiperidine-1-oxyl, before thermal car-bonization, Hu and co-workers[259] were able to reduce the as-synthesized hard-carbon surface area from 586 to 126 m2 g−1 with even in that case a net curtailment on the irreversible capacity related to the very first cycle. Electrochemical[260] as well as chemical[261] presodiation processes, i.e., sodium enrich-ment of anode material before full-cell assembling, confirmed to be another effective way to enhance the hard-carbon FCCE, as reported by Ai and co-workers.[260] The authors fabricated a full-cell with a layered NaNiTiO2 cathode and a presodiated hard-carbon anode, via a facile approach in a three-electrode battery. The FCCE of this cell, with respect to a control cell, was enhanced from 64.7% to 73%, with a specific capacity of 93 mAh g−1. Passerini and co-workers[257] recently assembled an optimized Na full battery by combining starch-derived hard-carbon anode with a mixed layered oxide cathode, having the overall composition P2/P3/O2-Na0.76Mn0.5Ni0.3Fe0.1Mg0.1O2. The full Na-ion cell exhibited a high operating discharge volt-ages of 3.3 V, high energy density of 240 Wh g−1 normalized on both cathode and anode materials, and a capacity retention extended to 200 cycles. In this particular case, carbon anode electrochemical activation, by means of a sodium metal sac-rificial electrode, was investigated as well. Notably, the Cou-lombic efficiencies were found to be close to 100% after the first few cycles. Huang and co-workers,[262] finally, investigated how a composite system made of monodispersed hard-carbon spherules coated with soft-carbon layer can lead to a sensible improvement in first cycle Coulombic efficiency from 54% to 83%. Correct mass balancing of carbon anode and cathode recently arose as one of the trickiest point in sodium full-cell preparation. The main reason is that carbon is fully desodi-ated while the cathode is often far from being fully sodiated, as Tarascon and co-workers wisely pointed out.[263] A sodium reservoir, namely a cathode material enriched in sodium, would be thus required to overcome this limitation.[264]

4.1.3. Transition Metal Oxides-Based SIBs Full-Cells

Ti and V-Containing Oxides: That of transition metal oxide rep-resents a wide material class worth of interest as anode for SIBs due to their high theoretical specific capacity and sodium insertion reversibility. This class includes conversion species, insertion/intercalation compounds, and pseudocapacitive mate-rials. The former, in particular, is thermodynamically appealing among other transition metal halides and chalcogenides, thanks to the surpassingly advantageous operational potential in sodium environment if compared to lithium.[50]

Among metal oxides, Ti-based electrodes are widely recog-nized as potential anode materials for SIBs thanks to the low cost, low toxicity, negligible strain, and excellent structural sta-bility during the insertion/extraction of sodium ions. Huang and co-workers[268] demonstrated that the spinel Li4Ti5O12, a widely investigated “zero-strain” anode for LIBs,[269] is able to withstand sodium insertion, exhibiting an average voltage of 0.91 V with dis-tinguished cycling stability. The same group introduced another layered material, P2-Na0.66[Li0.22Ti0.78]O2,[270] as the negative elec-trode, which shows a negligible volume change, attested around

0.77%, during sodium insertion/extraction. Both compounds have been tested as anodes in a full-cell configuration, while Na3V2(PO4)3 was used as cathode. The Na3V2(PO4)3//Li4Ti5O12 and Na3V2(PO4)3//Na0.66[Li0.22Ti0.78]O2 full cells give rise to a 2.4 V plateau at moderate rate and good cyclic performance. Na2Ti6O13 has been reported to be another promising anode material with a plateau around 0.8 V. However, the specific capacity of the Na2Ti6O13-based full-cell was proved by Balaya and co-workers[271] to be very low, thus seriously limiting its further development. Since the high theoretical capacity of 310 mAh g−1 and the reversible insertion of two sodium ions in Na2Ti3O7 were demonstrated to occur at a voltage as low as 0.3 V versus Na+/Na, there are several attempts to use this mate-rial as anode for the sodium ion full batteries. Li[265] reported a Na2Ti3O7 anode with high electrochemical Na-storage activity by growing a surface engineered Na2Ti3O7 nanotube arrays directly on a Ti substrates, as shown in Figure 6, image (a). The assem-bled Na2Ti3O7//Na2/3(Ni1/3Mn2/3)O2 full cell is capable to work at an average voltage of 2.7 V, delivering a reversible capacity of 210 mAh g−1, and thus a high specific energy of 110 Wh kg−1 (based on the mass of both electrodes). Moreover, the layered Na2Ti3O7 nanotubes and VOPO4 nanosheets have been systematically investigated as anode and cathode materials, respectively, to build high performance sodium ion full-cells by Yu and co-workers,[267] as shown in image (c) in Figure 6. A stable reversible capacity of 76 mAh g−1 at 0.1 C is obtained by normalizing the capacity value over the mass of both anode and cathode materials, with a maximum energy density of 220 Wh kg−1.

As a versatile and cost-effective material, used in numerous important technological areas, TiO2 has also been investigated in sodium ion full batteries. Xiong et al.[272] proposed an all-oxide sodium-ion full battery with amorphous TiO2 nanotubes anode coupled with Na1.0Li0.2Ni0.25Mn0.75Oδ cathode. Amor-phous TiO2 high specific surface nanostructures are here inves-tigated as a low cost anode material providing a sodium storage which is predominantly capacitive or pseudocapacitive.[273] The reported battery showed an operation voltage of 1.8 V and good rate capabilities with 70% of the original capacity retained at 11 C. Moreover, Yang and co-workers built a double-wall Sb@TiO2−x structure[274] by the calcination of Sb2S3@TiO2 nanorods, in a Ar/H2 atmosphere, produced by the hydrolysis of tetrabutyl titanate on Sb2S3 nanorods. The schematic repre-sentation of the synthetic procedure is shown in image (b) in Figure 6. In this compound the excellent electrochemical sta-bility of TiO2−x is combined with the high theoretical capacity of Sb as an alloying material. Such structure accommodates the volume variation upon cycling and then keeps the structure stable. In Sb@TiO2−x//Na3V2(PO4)3-C full battery set up, the energy density turned out to be 151 Wh kg−1 with a delivered overall power of 21 W kg−1. The possibility of using V2O5 as host structure for Na+ was initially explored in the 1980s and con-tinues to raise interest.[275] The amorphous V2O5 aerogel as anode material for sodium ion full batteries was explored by Passerini and co-workers.[276] When evaluated as Na3V2(PO4)3//V2O5, the full-cell displays the capacities of 113 and 60 mAh g−1 based on the anode and cathode materials, respectively. The cell capacity retention after 150 cycles was about 52% with respect to the maximum value obtained in the 28th cycle. Although





oxide-based anode materials have been extensively investigated in sodium half-cell with good chemical performance, their per-formances in full-cell still remain insufficient and unsatisfactory.

Iron Based Conversion Oxides: Metal oxides undergoing con-version reaction in presence of sodium have been thoroughly explored in full-cell set-ups. Despite their appealing theoretical capacity connected to the many-electrons conversion reaction (MxOy + 2yNa+ + 2ye− ↔ xM + yNa2O, M = transition metal), they suffer from poor cycle ability due to difficulties in withstanding volume expansion and SEI disruption. Molybdenum oxide (MoO3) has attracted great interest as anode material, due to

a superior theoretical specific capacity of nearly 1111 mAh g−1, which is about three times of that for graphite.[277] The revers-ible sodium storage in orthorhombic α-MoO3 via conversion mechanism was confirmed for the first time by Balaya and co-workers.[278] Despite the poor Coulombic efficiency of 53%[279] associated to the first cycle, the performance of a half-cell set up sets at 240 mAh g−1 at 0.1 C, manifesting a capacity retention of 55% of the initial capacity after 500 cycles at 0.2 C. A MoO3//Na3V2(PO4)3 full-cell[278] operating at 1.4 V in nonaqueous environment was further assembled, and an initial discharge capacity of 164 mAh g−1, based on anode mass, achieved. The


Figure 6. a) Schematic illustration of the fabrication of surface engineered Na2Ti3O7 nanotube arrays grown on Ti foil. Reproduced with permission.[265] Copyright 2016, Wiley-VCH. b) The preparation process of double-walled Sb@TiO2−x nanotubes. Reproduced with permission.[266] Copyright 2016, Wiley-VCH. c) Electrochemical characteristics of Na2Ti3O7 and VOPO4 electrodes in the half-cell format versus Na+/Na: (a) the CV curve of VOPO4 in the voltage range of 2.5–4.3 V for V5+/V4+ (orange) and Na2Ti3O7 in the voltage range of 0.01–2.5 V for Ti4+/Ti3+ (blue) versus Na+/Na. (b) Schematic of potential of the V5+/V4+ RedOx couple versus Na+/Na in layered Na2Ti3O7 and Ti4+/Ti3+ RedOx couples versus Na+/Na in layered VOPO4 materials. The insets display the average potentials of the VOPO4 and Na2Ti3O7/Na half cells as the cathode and the anode, respectively, and further predicted the average charge voltage of 3.2 V in the Na2Ti3O7//VOPO4 full cells. (c) The typical half-cell charge/discharge profiles of the VOPO4 in the voltage range of 2.5–4.3 V for V5+/V4+ (orange) and Na2Ti3O7 in the voltage range of 0.01–2.5 V for the Ti4+/Ti3+ (blue) versus Na+/Na. Reproduced with permission.[267] Copyright 2016, The Royal Society of Chemistry.




low-cost nanostructured Fe3O4, synthesized in round-shaped nanoparticles[280] with an average size of less than 10 nm, has been proposed as an attractive candidate as negative electrode material for SIBs, delivering 130 mAh g−1,[281] Nevertheless, Fe3O4-based anodes are affected by poor overall stability[282] and among the conversion materials so far investigated are the less performing if not properly engineered. Full battery made of Na3V2(PO4)3/graphene cathode and precycled (i.e., presodi-ated) Fe3O4 anode, indeed, quickly loses one third of its orig-inal capacity (92 mAh g−1) in the first 20 cycles. The precycling step, performed against a Na metal electrode, has been demon-strated to positively affect further cycling in full-cell system by overcoming the large irreversible capacity of Fe3O4. Mechanistic principles lying beyond this effect are however obscure still today and require more studies. The full-cell delivers an initial discharge capacity of 92 mAh g−1, calculated in respect to the cathode mass, at 0.2 C. Hydrothermally prepared carbon-coated Fe3O4 particles[283] have demonstrated to have better retain capacity, thanks to the conversion reaction occurring only par-tially (22% of the total mass of the material). The sodium full-cell prepared assembling C-Fe3O4 anode and O3-type layered Na[Ni0.25Fe0.5Mn0.25]O2 as cathode material operates reversibly around 2.4 V, delivering a capacity of about 130 mAh g−1 (based on cathode mass), and offers a good capacity retention for extensive cycling: 82.8% at 100th cycle and 76.1% at 150th cycle with a Coulombic efficiency approaching 100%. NaClO4 salt in a mixture of ethyl methanesulfonate electrolyte and 2 vol% FEC additive moreover increased the stability and conductivity of the system. Recently, bearing in mind that conversion pro-cesses are accompanied by chemical bond breaking and radical structural rearrangement, Ye et al. prepared ultrasmall, poorly crystalline FeOx nanoparticles supported on carbon nanotubes, to reduce the activation barriers toward the conversion reaction. When combined with Prussian blue cathode materials, it leads to iron-based full batteries with a working voltage of about 2 V and energy density of around 136 Wh k−1 g and outstanding cycle life.

4.1.4. Alloy-Based SIBs Full-Cells

A major limitation to sodium–metal alloys implementation resides in the tremendous molar volume increase achieved as higher sodium contents that are reached inside the alloys.[63] This aspect, together with the technological challenge repre-sented by pairing up materials, in a full-cell assembly, with con-siderably different theoretical specific capacities, contributed in discouraging the investigation of alloy-based sodium full-cells. A successful attempt in doing so has been conducted by Chen and co-workers.[153] They successfully obtained tin nanodots (average size between 1 and 2 nm) simultaneously achieving a homogeneous dispersion on an electrospun N-doped porous carbon matrix (PNC). The absolute strain generated during sodiation/desodiation resulted mitigated, and fracture and decrep-itation of electrode active materials delayed. The as-prepared Sn NDs@PNC free-standing anode has been conveniently paired with NaVPO4F/C cathode, with the full-cell exhib-iting an initial capacity of 540 mAh g−1, based on the mass of the anode material. After 100 cycles, the reversible residual

capacity was 460 mAh g−1, proving a good capacity retention (85.2%) with Columbic efficiency hovering around 99%. Xie et al.[284] developed a facile soaking-chemical vapour depo-sition method to grow core–sheath structured Sn@CNT nanopillar arrays on carbon paper, with a unique 3D hier-archical architecture as free-standing electrodes. Coupled with Na0.80Li0.12Ni0.22Mn0.66O2 cathode, the Na-ion full cell was assembled, providing sufficient energy to power a light-emitting diode (LED) light. Morever, after the homogeneous embedding of Sb and Na3V2(PO4)3 nanoparticles in an inter-connected framework of rGO nanosheets, which acts as a structurally stable and electronically conductive housing, Zhang et al.[158] tested the Na3V2(PO4)3/rGO//Sb/rGO full-cell set up. This configuration delivered highly reversible capacity of 400 mAh g−1 at a current of 100 mA g−1 after 100 charge/ discharge cycles. Similarly, Hassoun and co-workers[285] assem-bled a rechargeable sodium-ion full-cell based on nanostructured Sb–C anode and P2-layered Na0.6Ni0.22Fe0.11Mn0.66O2 cathode. The full-cell showed an average working voltage of about 2.7 V and delivered a reversible capacity around 120 mAh g−1.

Construction of a hybrid–alloy phase is also a valid method to improve the structural stability of the anode material, since two different metal phases can operate as mutual buffers with each other to reduce the volume changes.[129,170,286–288] Walter et al.[289] reported a simple and inexpensive colloidal synthesis of SnSb NCs notably without involving surfactants molecules, and dem-onstrated their utility as lithium-ion and sodium-ion anode mate-rials. Full-cells were constructed as well, with Na1.5VPO4.8F0.7 as cathodes, achieving anodic capacity of 400 mAh g−1 with an average discharge voltage of 2.7 V for sodium-ion cell. Inter-estingly, the cycle life and energy density of the alloy-based sodium-ion full batteries exhibit competitiveness in regard to carbon-based full-cells. However, it is a major challenge to develop the alloy-based sodium ion full cells with promoted oper-ating voltage and extended cycle-life in a cheap way.[255]

4.1.5. Other Nonsymmetric Assemblies

Metal sulfides (Sb2S3,[290] SnS2,[291,292] FeS2,[293] WS2,[294] MoS2,[295] Bi2S3,[55] CoS[296]) have been recently drawing atten-tion among the scientific community as anode material for lithium and sodium energy storage. Potentially, in some of the cases cited above like Sb2S3, SnS2, and SnO2,[57] to the con-version reaction previously described, also the alloying in the reduced metal contributes in further extending available theo-retical capacity. For such materials have been thus proposed the following tandem mechanisms[290]

Conversion reaction: Sb S 6 Na 6 e 2 Sb 3 Na S2 3 2+ + → ++ − (5)

Alloying reaction: 2 Sb 6 Na 6 e 2 Na Sb3+ + →− − (6)

Theoretical capacity involved in the 12 electrons charge transfer would be in this way increased to 946 mAh g−1. Yu et al.[290] reported a uniform coating of antimony sulphide (stibnite) on graphene. The composite preparation was entirely conducted in aqueous environment, starting from graphene oxide stable suspension and Sb hydroxide as antimony sulfide





precursor. Na2/3Ni1/3Mn2/3O2 was employed as cathode mate-rial to form a SIB full-cell. The latter demonstrates an energy density of 80 Wh kg−1 and shows good cycle performance. Similarly to oxide-based anode materials, sulfides usually suffer from severe volume expansion during sodiation and sluggish kinetics for related Na+ diffusion. SnS2 possesses a CdI2-like crystalline layered structure (lattice parameters: a = 0.3648 nm, c = 0.5899 nm, space group P3m1) consisting of alternating layers of tin atoms sandwiched between two layers of hexago-nally close-packed sulfur atoms. A SnS2-rGO hybrid structure was designed for reversible storage of Na+ by Qu et al.[297] The electrons can be fast collected and conducted through the highly conductive rGO network. The SnS2-RGO anode was paired up with P2-Na0.80Li0.12Ni0.22Mn0.66O2 cathode, which holds a theoretical capacity of 118 mAh g−1. The full-cell dis-played relatively good stability, with a capacity retention of 74% after 50 cycles. Recently, Peng et al.[298] employed a simple sol-vothermal method for in situ decoration of rGO nanosheets with 2D CoS nanoplatelets. Based on CoS@rGO hybrid anode and electrospun Na3V2(PO4)3@carbon cathode, a full cell has been additionally assembled and its electrochemical characteri-zation manifested high capacity and promising cycle stability, displaying an initial charge and discharge capacities of 498 and 381 mAh g−1 respectively (based on the mass of anode) with 75% capacity retention after 100 cycles.

4.1.6. Aqueous Nonsymmetric Sodium Full Battery

Some full-cell configurations have shown exceptional performance with aqueous electrolytes instead of nonaqueous aprotic organic electrolytes. Sodium-ion battery with aqueous electrolyte has the advantage in perspective of lower cost, faster ion transportation, and nonflammability, and therefore it has received much atten-tion for large-scale energy storage application. Full-cell commer-cial products, directly exploiting aqueous-based sodium storage batteries, recently made their appearance on the market. Low cost, low toxicity, and versatility for stationary storage are among the most notable strengths of manufacturer such as AQUION energy (USA)[34] and FARADION (UK)[20] companies. The former, which recently considered to move manufacturing facilities to foreign countries, made a spinel MnO2/Activated carbon full-cell the core of its business. The excellent cycling performance ensuring more than 10 000 cycles coupled with ≈20 Wh kg−1 of specific cell capacity, thanks to a sulfate-based aqueous electrolyte and pseudocapacitive sodium storage. The most renown drawbacks of SIBs aqueous technologies are however the limited performance, due to narrow voltage window available, and the limited number of candidate materials. The Prussian blue analogues, i.e., TM hexacyanoferrates and TM hexacyanomanga-nates, have been studied as aqueous sodium-ion battery cathode materials. Many of them have demonstrated as valuable mate-rials in terms of facile synthesis and stable electrochemical perfor-mances,[299,300] and some others promising for full-cell aqueous assemblies.[301,302] The rigid metal–organic framework of this class of materials characterized by large nanometric-sized voids pro-vides a stable scaffold in which a large variety of alkaline cations can be easily and reversibly housed.[204,303] Charge neutrality is ensured by RedOx active couples constituting the metal–organic

framework itself. Kim et al.[302] reported an aqueous SIB hybrid full-cell with Prussian blue analogue, namely KCu0.5Cu0.5Fe(CN)6, and a novel carbonyl-based organic salt, disodium nahthalenedi-imide (SNDI), which had high operational voltage of 1.1 V and excellent cycle performance. Polyanion compounds have also been considered as electrode materials for aqueous sodium-ion battery. Kumar et al.[304] demonstrated a high energy den-sity sodium-ion battery full-cell with aqueous electrolyte with Na3V2O2x(PO4)2F3−2x/multi-walled carbon nanotubes (MWCNT) composite as cathode material and NaTi2(PO4)-C composite or Zn metal as anode material. The as obtained full-cell achieved good cycle stability even under high current density of 10 C and reached the outstanding energy of 84 Wh kg−1. Ultimately, vana-dium-based compounds are of particular interest as well due to the different oxidation states accessible realizing multielectrons RedOx reactions. 1D nanostructured sodium vanadium oxide Na2V6O16·nH2O was introduced as a novel anode material for aqueous Na-ion batteries by Shang and co-workers.[305] A full aqueous Na-ion battery was built using Na0.44MnO2 as cathode and Na2V6O16·nH2O as anode. However, the charge capacity of the full-cell fades quickly in the initial few cycles, which is in agreement with the fast fading of the sodium-ion intercalation capacity of Na2V6O16·nH2O.

4.2. Symmetric Full Battery

It has been already extensively discussed in literature how partial metal substitution (Li, Mg, Zn, and Ti) proved to be surpassingly effective in incrementing structure stability of lay-ered transition metal oxides when used as SIBs cathode mate-rials. For example, Li doping of P2-type β-Na0.7[Mn1−xLix]-O2+y (x = 0.07) effectively and considerably suppresses Jahn–Teller distortion connected to Mn high oxidation states.[40] Similarly, Mg substitution revealed to be a promising strategy to suppress irreversible P2–O2 transition in Na0.67Mn0.67Ni0.33O2.[214] Explo-ration of diverse stoichiometry of mixed transition metal oxides recently led to meaningful results concerning dual-RedOx materials. With this term are addressed compounds in which two red-ox processes can occur at significantly different poten-tial values, this being related to thermodynamics of involved transition metal atoms and their chemical surroundings.[306] This feature, coupled with a crystalline structure capable of sodium hosting, discloses a full-cell set up in which the same material can be used as both anode and cathode. Nonsym-metric sodium ion full batteries, previously described, bear with themselves potential safety concerns related to possible thermal runaway connected to operational potentials close to sodium plating when certain anode materials, especially carbon-based materials, are employed.[20] This kind of SIBs full-cell design unlikely meets long-life, low cost, and safety requirements for large-scale energy storage applications. Safer and easy-to-assemble configuration are thus urgently needed. For this reason, symmetric SIBs became very recently an attractive and promising alternative from a commercial standpoint. The same active material, exploiting different red-ox sites and acting as both cathode and anode, can enable cells overcharging to some extent, buffer the large volume expansion (cathode expanding accompanied by anode shrinking, and vice versa), and greatly





reduce the manufacturing costs and simplify the fabrication process.[255,307–310] Up to now, the full symmetric vanadium and titanium-based SIBs in terms of two TM4+/TM3+ and TM3+/TM2+ (TM = Ti or V) RedOx couples in the active mate-rials have been proposed.

4.2.1. Sodium-Ion Full-Cells Based on NASICON Structures

NASICON-type phosphates Na3V2(PO4)3 (NVP) possesses two RedOx active centers, coinciding with vanadium atoms occu-pying different crystalline sites. Standard potentials, connected to RedOx couples V4+/V3+ and V3+/V2+, are 3.4 and 1.6 V versus Na+/Na, respectively. Such properties allow NVP to be used not only as cathode but also as anode. The cell reactions in the con-sidered symmetric cells can be described as follows[311]

Anode A V (PO ) A e

A V V (PO )3 2

34 3

3 23 2

4 3

x x

x x x

( ) + + ↔+ + −

+ −+ + (7)

Cathode A V (PO )

A V V (PO ) A e3 2

34 3

3 23 4

4 3 x xx x x

( ) ↔+ +

+

− −+ + + − (8)

Overall cell reaction 2A V (PO )

A V V (PO ) A V V (PO )3 2

34 3

3 23 2

4 3 3 23 4

4 3x x x x x x

( ) ↔+

+

+ −+ +

− −+ + (9)

Where A stands for either Li or Na centers. The NVP//NVP symmetric cells were first fabricated and examined by Plash-nitsa et al.,[311] and a moderate dependence of rate capabilities and cycle abilities with temperature was highlighted. Based on the quasi-open-circuit voltage measurements, full-cell charge cut-off voltage was optimized at 1.85 V. To improve the thermal sta-bility of sodium-ion batteries, the room-temperature ionic liquid (RTIL) NaBF4/1-ethyl-3-methyl imidazolium tetrafluoroborate (EMIBF4) was used as electrolyte instead of electrolyte solution based on flammable carbonate organic solvents (EC and dym-ethylcarbonate). The charge and discharge capacities of the NVP symmetric full battery were found to be 102 and 64 mAh g−1, respectively, during the first cycle. The substitution of the organic electrolytes with the safer RTIL resulted in the decrease of the first discharge capacity. However, RTIL-based cells revealed better cyclability and a more stable behavior at elevated temperatures.

Noguchi et al.[307] assembled an all solid-state NVP/NASICON(Na3Zr2Si2PO12)/NVP symmetric cell.[307] In this case, a NASICON-type material, Zr-based, was employed as solid-state electrolyte. The typical solid-state half-cell was cycled at 2.0–3.6 V and 1.5–2.0 V versus Na. The voltage of the sodium extraction from the cathode falling at about 3.4 V was associated with the V4+/V3+ RedOx couple, while the plateau registered at 1.6 V was associated with the V3+/V2+ RedOx couple and indicated electrochemical insertion of Na+ into the NVP anode (see left side of image (a) in Figure 7). Considering the absolute capacity (mAh of each electrode) balance of the negative/positive electrodes,[314] it was sug-gested that a cathode-to-anode mass ratio of 1:3 is appropriate for the construction of symmetric-type cells. Parasitic reac-tions leading to NVP and Zr-NASICON phase decomposi-tion have been detected at NVP/Zr-NASICON interface, and inevitably worsen the capacity extracted from active materials.

Right-hand side of image (a) in Figure 7 shows the voltage profiles of the symmetric solid-state cell cycled at 25 °C under 1.2 µA cm−2 current areal density in the voltage range 0.01–1.9 V. The first discharge capacity was about 68 mAh g−1, which is around 80% of the first discharge capacity for the liquid NVP/NVP symmetric sodium-ion cell[311] using 1 m NaClO4 in PC as electrolyte and around 58% of the NVP theoretical capacity (117 mAh g−1). Li et al.[312] studied the effect of carbon matrix dimensionality on the electrochemical properties of the NVP nanograins for the symmetric sodium-ion battery. NVP nanograins were in this case dispersed in dif-ferent carbon matrices from 0D (acetylene carbon), 1D (carbon nanotubes) to 2D (graphite nanospheres) nanostructures. The NVP/AC showed the best electrochemical properties in half-cell configuration as well as in symmetric full battery. In the latter case, it exhibited 80% capacity retention after 200 cycles at 1 C and could tolerate high current rates (up to 10 C) as shown in image (b) in Figure 7. The ≈1.8 V voltage differences observed between V4+/V3+ and V3+/V2+ RedOx couples within Na3V2(PO4)3 encouraged the investigation over other transi-tion metals sodium phosphates. An example is represented by Na3Ti2(PO4)3. Senguttuvan et al.[312,313] have validated the con-cept of full Ti-based sodium ion cells through the assembly of symmetric cells involving Na3Ti2(PO4)3 (NTP) operating at an average potential of 1.7 V. V4+/V3+ and V3+/V2+ RedOx couples were effectively replaced by Ti4+/Ti3+ and Ti3+/Ti2+ couples, operating at a voltage of 2.1 and 0.4 V versus Na+/Na, respec-tively. Experimental results agreed well also with theoretical investigation of operational potentials based on ab initio calcu-lation method originally proposed by Ceder and co-workers.[315] In Figure 7, image (c), the potential versus composition pro-file for NTP half-cell, in galvanostatic mode at C/25 between 2.5 and 0 V, is reported. The voltage window of the symmetric cell was controlled by the positive electrode cut-off voltage. As can be noted from charge/discharge three-electrodes cell profiles, the complete excursion of positive electrode potential (red curve in image (c), on the right side, in Figure 7) within 2.6–1.9 V versus Na+/Na potential range, is achieved. The neg-ative side of the cell (green curve on the right side of image (c) in Figure 7), instead, experiences a progressive shift toward higher values. This is related to the cathode-limited mass bal-ancing operated in full-cell preparation.

4.2.2. Sodium-Ion Full Battery Based on Layered Oxides

Apart from the NASICON structure-type based symmetric full battery, the bipolar materials with layered structure and the general formula AxMO2 (A = alkali and M = transition metal), containing two electrochemically active transition metals with RedOx couples such as Ni4+/Ni2+ and Ti4+/Ti3+, are also attractive and promising alternatives for the imple-mentation in Na-ion symmetric batteries. Typical Na lay-ered oxides can be defined as P2 and O3 according to the geometry of Na+ surroundings (P: prismatic and O: octa-hedral) and the number of unique oxide layer forming the vertically aligned repetitive unit, as extensively dis-cussed.[255] Guo et al.[309] designed and synthesized the O3-type Na0.8Ni0.4Ti0.6O2 material, the stoichiometry between





nickel and titanium, defined here as a “golden pair,” not only greatly stabilizes electrochemical sodiation/desodiation process but also exhibits the unique double RedOx couples of Ni4+/Ni2+ (3.5 V vs Na+/Na) and Ti4+/Ti3+ (0.7 V Na+/Na). The proposed symmetric SIB, as shown in Figure 8, exhibits a reversible discharge capacity of 85 mAh g−1 with the average voltage of 2.8 V and presents a superior long life exceeding 150 cycles with capacity retention of 75%. Two distinct RedOx peaks set at 3.6/3.5 V (Ni4+/Ni2+) and 0.95/0.7 V (Ti4+/Ti3+) can be observed in the cyclic voltammetry reported in image (a) (right-hand side) in Figure 8, and they appear separated by a large potential difference.

In case of the P2-type structure, Wang et al.[316] realized that because of the strong Na+–Na+ interaction in the alkali metal layer and charge ordering in the transition metal layer, most P2-type layered oxides exhibit Na+/vacancy-ordered superstructures, which are supposed to lead to rapid capacity fading during cycling. They found the transition metal ordering is mainly controlled by the difference in the ionic radii of M1 and M2 in P2-Nax[M1,M2]O2-layered oxide. Through selecting transition metals as Cr3+ and Ti4+ which show similar ionic radii (0.615 and 0.605 Å, respectively[316])

but a substantial difference in RedOx potentials, they designed a P2-Na0.6[Cr0.6Ti0.4]O2-layered oxide, which is highly Cr/Ti and Na+/vacancy-disordered at any sodium content and over a wide range of temperature. The full-cell delivers an average operating voltage plateau at 2.53 V and extraordinary rate and superior cycling performance. Even at a very high rate of 12 C, the capacity retention is 75% if compared to capacity extracted at 1 C, as shown in image (b) in Figure 8. It was demonstrated that this cation-disordered material can function as both posi-tive and negative electrodes with average operation voltages of 3.5 and 0.8 V, corresponding to the RedOx couples Cr4+/Cr3+ and Ti4+/Ti3+, respectively. Titanium substitution with manganese in layered oxide chemical formula leads to the com-pound P2-Na0.5Ni0.25Mn0.75O2 with dual-electrode characteris-tics.[310] Contrarily to what has been observed for other man-ganese containing layered structures, Na0.67Mn0.67Ni0.33O2

[317] and Na0.78Ni0.23Mn0.69O2,[318] in which Mn4+ remains substan-tially inert due to potential values above 2.5 V versus Na+/Na, Mn changes its oxidation state when P2-Na0.5Ni0.25Mn0.75O2 Mn is cycled between 1.5 and 2.6 V versus Na+/Na. The dual-electrode behavior is here imputable to Ni2+/4+ RedOx couple (operating at 3.72/3.51 V versus sodium) and Mn3+/4+


Figure 7. a) The charge–discharge profiles of the metallic sodium NVP||NASICON||Na half-cells cycled within 2.0–3.6 and 1.5–2.0 V versus Na at 10 µA and 80 °C (left) and the charge–discharge profiles of the symmetric NVP||NASICON||NVP solid-state cell tested at 1.2 µA cm−2 between 0.01 and 1.9 V versus Na at room temperature (right). The specific capacities are plotted based on the cathode weight. Reproduced with permission.[307] Copyright 2013, Elsevier. b) The schematic illustrations and the rate performance of Na3V2(PO4)3/AC, Na3V2(PO4)3/CNT, and Na3V2(PO4)3 /graphite at various current rates ranging 0.5–10 C. Reproduced with permission.[312] Copyright 2014, Wiley-VCH. c) Galvanostatic cycling of Na3Ti2(PO4)3 full cell versus Na between 2.5 and 0 V for the first charge/discharge and subsequent cycles (left). Potential profile as a function of time for symmetric Na3Ti2(PO4)3||NaClO4 in PC||Na3Ti2(PO4)3 cells with a three electrode configuration cycled at C/20 at room temperature (right). The monitoring of operational potentials of both electrodes (blue curve for the cathode and green curve for the anode) is also reported. Reproduced with permission.[313] Copyright 2013, America Chemical Society.




oxidation taking place at much lower potentials values (2.15/1.76 V versus Na+/Na). Despite the discrepancies between reversible capacities extracted from half-cell config-uration (75 vs 100 mAh g−1), mass ratio between anode and cathode for full-cell fabrication was kept equal to 1. Based on this configuration the cell ensured a long cycle ability extended over 100 cycles with a capacity retention of 90% after 50 cycles and an output voltage of 2.0 V.

5. Conclusions and Outlook

This review comprehensively collected the most recent advancements in sodium-ion aprotic full-cell research. Motiva-tion behind this effort can be summarized in a readiness level first assessment of the sodium full-cell technology by reporting the most useful physical quantities to compare battery tech-nology such as energy and power densities and cell stability through cycling. Provided these information, comparison with

the affirmed lithium technology performances will be more agile. The final aim, however, is not the encouragement in identifying a sodium chemistry paralleling lithium counter-part performances, a hard task considering physical limitation of Na as stated in the introduction. Instead, the evaluation of suitability of sodium cells for energy storage applications dif-ferentiation and speciation would be more appropriate. There are several aspects that might contribute in making sodium technology competitive, in particular for stationary energy storage applications and other low-power utilizations with low-to-null restriction in terms of volumetric occupancy. Among these the most relevant are: (i) the advantages con-nected to more abundant raw materials, (ii) Co-free active elec-trode materials, (iii) manufacturing cost reduction deriving from aluminum utilization as anode current collector, and (iv) symmetric configuration. A considerable number of recent works proposing full-cell set-ups have been thoroughly revised and the most significant contributions have been spe-cifically sorted so to highlight unique correspondences between


Figure 8. a) Schematic illustration of the symmetric SIBs design via the two RedOx couples of nickel and titanium in the layered materials Na0.8Ni0.4Ti0.6O2. A diagram of the proposed symmetric cell based on O3-type Na0.8Ni0.4Ti0.6O2 (left) has been placed side by side with the schematic of energy versus density of states plot (center), showing the relative positions of the Fermi energy level of Ni4+/Ni2+ (Ni4+/Ni3+ and Ni3+/Ni2+) and Ti4+/Ti3+ RedOx cou-ples for O3-type Na0.8Ni0.4Ti0.6O2. On the right of the same image, the CV curve of Na0.8Ni0.4Ti0.6O2/Na extended over the whole voltage range of 0–4 V versus Na+/Na is reported while on the background un underlying picture of LED bulb driven up by the designed bipolar Na0.8Ni0.4Ti0.6O2-based symmetric cells appears. Reproduced with permission.[309] Copyright 2015, The Royal Society of Chemistry. b) Full-cell performance of P2-Na0.6[Cr0.6Ti0.4]O2 as both positive and negative electrodes in the NaPF6-based electrolyte. In the left the discharge profiles of Na0.6[Cr0.6Ti0.4]O2/ Na0.6[Cr0.6Ti0.4]O2 sodium-ion full cell at various rates are reported, while on the right appear the capacity and Coulombic efficiency versus cycle number of the full cell at 1 C rate. The capacity was calculated based on the mass of the negative electrodes only. Reproduced with permission.[316] Copyright 2015, Macmillan Publisher Limited.




chemistries and electrochemical features, namely cell voltage and specific capacities. In image (a) in Figure 9, a 3D plot collecting these findings is reported. Particularly, the collected specific capacities normalization has been made over both anode and cathode mass. This operation has often revealed as a challenging task, since electrode mass load-ings and/or anode to cathode ratio are not always discussed in the scrutinized papers. The sorting of the sodium chemistries, at the same time, purposely focused, in a first instance, on highlighting anode chemistry over cathode one. The former, indeed, has, at current knowledge, the more severe limitation in terms of stability and capacity retention. The classes of sodium full-cells have been thus organized in carbon-based, conversion/alloying material-based, insertion materials and organic based cells. To these groups, the symmetric cells have been set, fully based on insertion chemistry, and aqueous chem-istries have been added. It appears imme-diately evident the clustering of particular sodium full-cell chemistries within potential and capacity ranges. Interestingly, contrarily to what common sense and half-cell experi-ments would suggest, conversion or alloying based full-cell (red symbols in Figure 9) seems not to derive surpassing benefits from the superior theoretical capacity of the anode material. This is presumably due to the large excess of cathode compound required to overcome irreversible capacities connected to first cycle and kinetics bottlenecks connected to augmented cathode thickness. Cell poten-tials, for the above mentioned chemistries, are not surprisingly high as well, being con-version and alloying anode charge/discharge profiles shifted toward high potential values and affected by not negligible overpotential limitations. Contrarily, full-cells based on carbon anodes, either expanded graphite, hard-carbon, and soft-carbon or rGO ensure overall more intriguing performances, espe-cially in terms of cell potentials and capacity retention (z-axis of 3D plot). The involved sodium storage mechanisms (double layer capacitance and expanded graphitic layer intercalation) indeed constitute less damaging and compromising processes, even if occurring at high cell potential values. Data plotting in image (a) in Figure 9 allows also to appreciate the effect of sodium cell chemistries over capacity retention after several cycles. Once again, carbon-based sodium batteries ensure better per-formances, with capacity retention averagely higher than com-peting chemistries. Worth noticing are values above 100% for some chemistries, connected to a progressive activation of the full-cell with charge discharge cycles. In image (b) in Figure 9, the XY plane of the 3D plot is reported separately,

with references numbers inserted in each data point, in order to facilitate the tracking of the correspondent full-cell chem-istry listed in Table 2. Ultimately, in Figure 10, the energy density versus power density plot of investigated full-cell SIBs chemistries is reported. By comparing the listed examples so far explored with lithium technology (amber shaded area), emerges how the formers, despite the nonoptimized per-formances, overlap with lithium technology in many cases. Notably, in one case among the ones taken in consideration in this review, both power and energy densities of a full-cell SIB chemistry are considerably higher than those of the average


Figure 9. a) 3D plotting of full-cell potential versus specific capacity (normalized over the mass of both the anode and cathode materials) and, when reported by the authors, capacity retention after several cycles (at least 10). b) Cell potential versus specific capacity plot, inside each data circle the bibliographic number referred to the specific sodium-ion assembly is provided.





Table 2. Electrochemical and energy data for the studied full-cell sodium chemistries.

Anode Cathode Cell potential [V] Specific capacity [mAh g−1] Energy density [Wh kg−1] Power density [W kg−1]

Pyrolyzed carbon Na0.44MnO2[319] 2.7 56 160–200 72–80

Carbon nanofibers Na0.7MnO2.05/CNT[320] 1.8 35 90–110 60

NayC Na[Ni1/3Fe1/3Mn1/3]O2[321] 2.75 59 240–260 16–60

Hard carbon NaNi0.5Mn0.5O2[60] 2.7 81 210–230

Na[Ni0.6Co0.05Mn0.35]O2[195] 2.84 72 240–260 20

Na0.9[Cu0.22Fe0.30Mn0.48]O2[322] 3.2 86 220–220 138

Na0.61[Mn0.27Fe0.34Ti0.39]O2[323] 3.32 136 224 45

Na[Li0.05(Ni0.25Fe0.25Mn0.5)0.95]

O2[324]

3.4 107 360–380 36

NaFePO4[325] 2.6 86 190–210 45

Na3V2(PO4)3[326] 1.6 40 130–150 276

Na3V2(PO4)2F3[327] 3.65 73 270–290 53

Na3.5V2(PO4)2F3[264] 3.7 72 280–300 53

Na4Co3(PO4)2P2O7[328] 4.1 55 280–300 45

Na1.76Ni0.12Mn0.88[Fe(CN)6]0.98[329] 3.0 30 82 90

Na2MnFe(CN)6·zH2O[330] 3.2 72 230–250 164

Na1.92Fe[Fe(CN)6][331] 3.0 61 183 18

Na2C6O6[332] 2.0 89 220–240 27

NaNi0.5Mn0.5O2[97] 3 82 375 310

NaCrO2[333] 2.8 79 17

Hard carbon sucrose/

Na2/3Ni1/3Mn2/3O2[262]

3.5 60 200 21

Na0.9Cu0.22Fe0.30Mn0.48O2[92] 3.2 66 207 53

NaxNi0.22Co0.11Mn0.66O2[91] 2.4 76 182 15

NaCrO2@C[334] 2.95 71

Graphite Na1.5VPO4.8F0.7[89] 2.9 41 120 117

Na3V2(PO4)3[253] 2.2 35 77 85

SnS[292] Na3V2(PO4)3F3 2.4 36 86 57

SnS2@RGO[297] Na0.80Li0.12Ni0.22M0.66O2 2.3 66 152 83

SnS2[291] Na3V2(PO4)3 1.6 62.5 100 27

MoS2[295] Na3V2O2x(PO4)2F3−2x 1.8 46.6 189 92

Sb2S3[290] Na2/3Ni1/3Mn2/3O2 2.2 36.7 80 218

Bi2S3[55] Na3V2(PO4)3 1.6 41 66 19

CoS@rGO[298] Na3V2(PO4)3@C 1.7 54 92 121

WS2[294] NaVPO4F@C 1.6 73 117 36

FeSe2[335] Na3V2(PO4)3 1.7 98.5 156–179 425–515

CoSe2[137] Na3V2(PO4)3 1.2 58 69 92

CoSe carbon nanowires[296] Na3V2(PO4)3@C 1.7 69 115 38

Co0.5Fe0.5S2[150] Na3V2(PO4)3 1.6 83 132 78

CoS@rGO[336] Na3V2(PO4)3@C 1.6 54.4 92 114

SnS2@graphene[146] Na3V2(PO4)3 2.4 55 140 8300

FeS2@FeSe2[151] Na3V2(PO4)3@C 1.5 28 42 273

C8H5NaO4[337] Na0.75Mn0.7Ni0.23O2 3.6 42 151 22

Polyimide[338] Na3V2(PO4)3/C 1.2 75 90 63

Na4Fe(CN)6/C 1.2 70 84 59

Disodium terephthalate

(NaTP)[339]

N,N0-diamino-3,4,9,10-erylene-

tetracarboxylic polyimide

1.35 73 99 26




lithium battery.[146] In this case, the outstanding high-power capability was guaranteed by pseudocapacitive behavior of the composite. Ought to be noted is that in order to make compa-rable a technology which is already available on the market with a not commercialized and an early-stage one, some assump-tions had to be considered. Rather than providing what might have resulted in an unreliable esteem of the mass composition of a marketable sodium-ion battery cell, the resizing of energy and power density of commercial lithium technology, according to electrode mass composition alone, has been preferred. The emerging overall picture well agrees with originally hypothe-sized limitations of sodium based rechargeable battery, namely the lower energy density connected to discrepancy in standard electrochemical potentials between sodium and lithium. Nev-ertheless, sodium-ion technology is revealing as a technology capable of replacing lithium one in several applications. Once again, the patterning of SIBs chemistries, even in the Ragone plot-like graph, is evident. A further clustering of the reported data is overlapped in this very last graph in order to provide a classification based also on the cathode materials. This clever

plotting provides a quick tool that efficiently and rapidly allows the identification of carbon-based anode/TM-oxides cathode architectures as the most promising in terms of energy density, while the alloying-conversion anode/NASICON cathode geom-etry as the most performing in terms of power density.

Sodium ion battery technology is demonstrating as a valu-able alternative to the increasingly penetrating lithium-ion one. Differentiation of battery pack chemistries, according to appli-cation, will be a strategic asset of industrial manufacturers, extremely useful in curbing the deleterious effect deriving from raw materials depletion. A gargantuan effort has neverthe-less to be made in SIBs materials optimization and, most of all SIBs electrolytes formulation, topic intentionally neglected here in this review. Among the most delicate material science challenges that has to be faced in the optimization process, is recalled in this closing remarks the tackling of the still limiting irreversible capacity registered during the very first charge/dis-charge cycle, particularly severe for some anode material classes (alloying and high-surface carbon based materials). Stability of crystalline phases, to thousands of charge/discharge cycles, is


Anode Cathode Cell potential [V] Specific capacity [mAh g−1] Energy density [Wh kg−1] Power density [W kg−1]

Black phosphorus[340] Na0.66Ni0.26Zn0.07Mn0.67O2 1.4 42 59 8

Cu3P[341] Na3V2(PO4)3 2.7 47 104.5 273

CNT/FeOx[280] NaxFeFe(CN)6 2 78 136 12

Amorphous TiO2NT[272] Na1.0Li0.2Ni0.25-Mn0.75Oδ 1.6 38 61 7.6

Sb@TiO2−x[266] Na3V2(PO4)3 3.2 117 151

Fe3O4[283] Na[Ni0.25Fe0.5Mn0.25]O2)/C 2.4 59 142 14.2

Fe3O4[282] Na3V2(PO4)3 1.6 37 59 12

α-MoO3[278] Na3V2(PO4)3 1.4 54 75

V2O5[276] Na3V2(PO4)3 2.5 46.5 108–123 96–109

Na2Ti3O7[267] VOPO4 2.9 76 220 22

Na2Ti3O7[265] Na2/3(Ni1/3Mn2/3)O2 2.7 47 110 25

Na2Ti6O13[271] Na3V2(PO4)3F3/C 2.5 42 105 21

Li4Ti5O12[268] Na3V2(PO4)3/C 2.4 49 118 12

Na0.66[Li0.22Ti0.78]O2[270] Na3V2(PO4)3/C 2.5 61 153 19–24

Disodium naphthalenedi-

imide (SNDI)

KCu0.5Cu0.5Fe(CN)6[302] 1.1 34 26 53

NaTi2(PO4)3/C Na3V2O2x(PO4)2F3−2x/MWCNT[304] 1.45 39 54 78.3

Zn metal 1.65 53 84 138.6

Na2V6O16 nH2O[305] Na0.44MnO2 1.05 30 31

Na0.6[Cr0.6Ti0.4]O2[316] 2.5 34 83–88 83–88

Na0.8Ni0.4Ti0.6O2[309] 2.8 33 96 92

Na2.55V6O16·0.6H2O[342] 1.5 16 24 8

Na0.66Ni0.17Co0.17Ti0.66O2[343] 3.1 35 95 107

NaNi0.33Li0.11Ti0.56O2[344] 3.1 33 100 20–26

Na3V2(PO4)3[312] 1.7 25 43 21–25

Na3V2(PO4)3[345] 1.8 30 54 14

Na3Ti2(PO4)3[313] 1.7 –

Na2VTi(PO4)3[346] 1.2 31 37 34

Na0.5Ni0.25Mn0.75O2[310] 2.0 13 26 15

Table 2. Continued.





still a major concern on the cathode side, as revealed by the not-so-exceptional capacity retentions of the explored chemistries. Nevertheless in the very last years, SIBs commercial prototypes manufactured in standard formats (18 650 cylindrical cell and prismatic cell) made their appearance on the market. Claims of cyclability above 4000 cycles and improved power and energy densities, similar to lithium counterpart, are comforting and cast silver linings on future energy storage.

AknowledgementsItalian authors are grateful to the Ministero degli Affari Esteri e della Cooperazione Internazionale to support the bilateral Italian/ Korean project of international cooperation (contract no. KR16GR05) and Korean authors thank the National Research Foundation (NRF) of Korea (grant no. NRF-2016K1A3A1A25003532). The authors gratefully thank Prof. Liberato Manna (Istituto Italiano di Tecnologia, Genova, Italy) for the fundamental contribution in revising the manuscript and the precious advices provided.

Conflict of InterestThe authors declare no conflict of interest.

Keywordsalloying materials, conversion materials, full-cells, layered materials, sodium-ion batteries

Received: October 16, 2017Revised: November 20, 2017

Published online:

[1] T. L. Keiningham, B. Cooil, Renewables 2016 Global Status Report, http://www.ren21.net/wp-content/uploads/2016/06/GSR_2016_Full_Report.pdf, 11, 2017.

[2] A. Gilbert, How solar growth will wreck the economics of existing power markets http://energypost.eu/solar-growth-will-wreck-eco-nomics-existing-power-markets/, 11, 2017.

[3] Wind and Solar Power are Disrupting Electricity Systems, The Economist, 2017.

[4] I. Gyuk, M. Johnson, J. Vetrano, K. Lynn, W. Parks, R. Handa, L. D. Kannberg, S. Hearne, K. Waldrip, R. Braccio, Grid Energy Storage, US DOE, 2013, p. 67.

[5] GreenTechMedia Research, U.S. Energy Storage Monitor: 2016 Year in Review and Q1 2017 Full Report, 2016.

[6] Y. Zhou, X. Li, presented at Proc. 34th Chinese Control Conf., Hangzhou, China, July 2015.

[7] J. B. Goodenough, Energy Environ. Sci. 2014, 7, 14.[8] H. Pan, Y.-S. Hu, L. Chen, Energy Environ. Sci. 2013, 6, 2338.[9] J. B. Goodenough, Y. Kim, Chem. Mater. 2010, 22, 587.

[10] M. S. Whittingham, Proc. IEEE 2012, 100, 1518.[11] M.-S. Balogun, W. Qiu, Y. Luo, H. Meng, W. Mai, A. Onasanya,

T. K. Olaniyi, Y. Tong, Nano Res. 2016, 9, 2823.[12] T. Horiba, T. Maeshima, T. Matsumura, M. Koseki, J. Arai,

Y. Muranaka, J. Power Sources 2005, 146, 107.[13] U.S. Geological Survey, Mineral Commodities Summaries 2017, U.S.

Deprtment of the Interior, 2017.[14] D. Chung, E. Elgqvist, S. Santhanagopalan, Automotive Lithium-ion

Battery Supply Chain and US Competitiveness Considerations, 2016.

Figure 10. Ragone plot-like representation collecting aprotic sodium-ion full-cell assemblies sorted by either their anode and cathode chemistries (see legend). Data labels reporting stoichiometry, and shaded areas referring to material classes give information about the cathode side of the full-cell assemblies.

http://www.ren21.net/wp-content/uploads/2016/06/GSR_2016_Full_Report.pdf

http://www.ren21.net/wp-content/uploads/2016/06/GSR_2016_Full_Report.pdf

http://energypost.eu/solar-growth-will-wreck-economics-existing-power-markets/

http://energypost.eu/solar-growth-will-wreck-economics-existing-power-markets/





[15] E. Wesoff, How Soon Can Tesla Get Battery Cell Costs Below $100 per Kilowatt-Hour? 2016.

[16] D. Larcher, J.-M. Tarascon, Nat. Chem. 2014, 7, 19.[17] J.-M. Tarascon, Nat. Chem. 2010, 2, 510.[18] T. Hunt, Is There Enough Lithium to Maintain the Growth of the

Lithium-Ion Battery Market? 2015.[19] The Lithium Market, Fox-Davies, 2013.[20] J.-Y. Hwang, S.-T. Myung, Y.-K. Sun, Chem. Soc. Rev. 2017,

46, 3529.[21] A. Yaksic, J. E. Tilton, Resour. Policy 2009, 34, 185.[22] S. Fletcher, Bottle Lightning: Superbatteries, Electric Cars, and the

New Lithium Economy, Hill and Wang, NY 2011.[23] H. Kim, H. Kim, Z. Ding, M. H. Lee, K. Lim, G. Yoon, K. Kang, Adv.

Energy Mater. 2016, 6, 1600943.[24] D. Aurbach, Z. Lu, a. Schechter, Y. Gofer, H. Gizbar, R. Turgeman,

Y. Cohen, M. Moshkovich, E. Levi, Nature 2000, 407, 724.[25] A. Ponrouch, C. Frontter, F. Bardé, M. R. Palacìn, Nat. Mater.

2016, 15, 169.[26] A. W. Golubkov, D. Fuchs, J. Wagner, H. Wiltsche, C. Stangl,

G. Fauler, G. Voitic, A. Thaler, V. Hacker, RSC Adv. 2014, 4, 3633.[27] M. S. Whittingham, Prog. Solid State Chem. 1978, 12, 41.[28] K. M. Abraham, Solid State Ionics 1982, 7, 199.[29] K. Mizushima, P. C. Jones, P. J. Wiseman, J. B. Goodenough, Solid

State Ionics 1981, 4, 171.[30] J. B. Goodenough, K. Park, J. Am. Chem. Soc. 2013, 135, 1167.[31] L. Lu, X. Han, J. Li, J. Hua, M. Ouyang, J. Power Sources 2013, 226,

272.[32] C. Fang, Y. Huang, W. Zhang, J. Han, Z. Deng, Y. Cao, H. Yang,

Adv. Energy Mater. 2016, 6, 1501727.[33] B. Scrosati, J. Electrochem. Soc. 1992, 139, 2776.[34] M. R. Lukatskaya, B. Dunn, Y. Gogotsi, Nat. Commun. 2016,

7, 12647.[35] S. Komaba, N. Yabuuchi, T. Nakayama, A. Ogata, T. Ishikawa,

I. Nakai, Inorg. Chem. 2012, 51, 6211.[36] G.-L. Xu, R. Amine, Y.-F. Xu, J. Liu, J. Gim, T. Ma, Y. Ren, C. Sun,

Y. Liu, X. Zhang, S. M. Heald, A. Solhy, I. Saadoune, W. L. Mattis, S.-G. Sun, Z. Chen, K. Amine, Energy Environ. Sci. 2017, 10, 0.

[37] G. Longoni, J. E. Wang, Y. H. Jung, D. K. Kim, C. M. Mari, R. Ruffo, J. Power Sources 2016, 302, 61.

[38] D. Dwibedi, R. Gond, A. Dayamani, R. B. Araujo, S. Chakraborty, R. Ahuja, P. Barpanda, Dalton Trans. 2017, 2, 510.

[39] P. R. Kumar, Y. H. Jung, D. K. Kim, J. Solid State Electrochem. 2016, 21, 223.

[40] M. S. Kwon, S. G. Lim, Y. Park, S. M. Lee, K. Y. Chung, T. J. Shin, K. T. Lee, ACS Appl. Mater. Interfaces 2017, 9, 14758.

[41] Y. Wen, K. He, Y. Zhu, F. Han, Y. Xu, I. Matsuda, Y. Ishii, J. Cumings, C. Wang, Nat. Commun. 2014, 5, 4033.

[42] M. S. Balogun, Y. Luo, W. Qiu, P. Liu, Y. Tong, Carbon N. Y. 2016, 98, 162.

[43] K. Hong, L. Qie, R. Zeng, Z. Yi, W. Zhang, D. Wang, W. Yin, C. Wu, Q. Fan, W. Zhang, Y. Huang, J. Mater. Chem. A 2014, 2, 12733.

[44] F. Yang, Z. Zhang, K. Du, X. Zhao, W. Chen, Y. Lai, J. Li, Carbon N. Y. 2015, 91, 88.

[45] Y. Cao, L. Xiao, M. L. Sushko, W. Wang, B. Schwenzer, J. Xiao, Z. Nie, L. V. Saraf, Z. Yang, J. Liu, Nano Lett. 2012, 12, 3783.

[46] T. Chen, Y. Liu, L. Pan, T. Lu, Y. Yao, Z. Sun, D. H. C. Chua, Q. Chen, J. Mater. Chem. A 2014, 2, 4117.

[47] T. Chen, L. Pan, T. Lu, C. Fu, D. H. C. Chua, Z. Sun, J. Mater. Chem. A 2014, 2, 1263.

[48] Y.-X. Wang, S.-L. Chou, H.-K. Liu, S.-X. Dou, Carbon N. Y. 2013, 57, 202.

[49] M. Lao, Y. Zhang, W. Luo, Q. Yan, W. Sun, S. X. Dou, Adv. Mater. 2017, 1700622, 1.

[50] F. Klein, B. Jache, A. Bhide, P. Adelhelm, Phys. Chem. Chem. Phys. 2013, 15, 15876.

[51] X. Lu, E. R. Adkins, Y. He, L. Zhong, L. Luo, S. X. Mao, C.-M. Wang, B. A. Korgel, Chem. Mater. 2016, 28, 1236.

[52] J. W. Wang, X. H. Liu, S. X. Mao, J. Y. Huang, Nano Lett. 2012, 12, 5897.

[53] L. Baggetto, M. Marszewski, J. Górka, M. Jaroniec, G. M. Veith, J. Power Sources 2013, 243, 699.

[54] T. Ramireddy, N. Sharma, T. Xing, Y. Chen, J. Leforestier, A. M. Glushenkov, ACS Appl. Mater. Interfaces 2016, 8, 30152.

[55] W. Sun, X. Rui, D. Zhang, Y. Jiang, Z. Sun, H. Liu, S. Dou, J. Power Sources 2016, 309, 135.

[56] M. K. Datta, R. Epur, P. Saha, K. Kadakia, S. K. Park, P. N. Kumta, J. Power Sources 2013, 225, 316.

[57] Z. Li, J. Ding, D. Mitlin, Acc. Chem. Res. 2015, 48, 1657.[58] J. Qian, Y. Chen, L. Wu, Y. Cao, X. Ai, H. Yang, Chem. Commun.

(Cambridge, U. K.) 2012, 48, 7070.[59] X. Yang, R. Zhang, N. Chen, X. Meng, P. Yang, C. Wang, Y. Zhang,

Y. Wei, G. Chen, F. Du, Chem. - Eur. J. 2016, 22, 1445.[60] E. De La Llave, V. Borgel, K. J. Park, J. Y. Hwang, Y. K. Sun,

P. Hartmann, F. F. Chesneau, D. Aurbach, ACS Appl. Mater. Inter-faces 2016, 8, 1867.

[61] L. P. Wang, L. Yu, X. Wang, M. Srinivasan, Z. J. Xu, J. Mater. Chem. A 2015, 3, 9353.

[62] D. Kundu, E. Telaie, V. Duffort, L. F. Nazar, Angew. Chem. Int. Ed. 2015, 54, 3431.

[63] K. Kubota, S. Komaba, J. Electrochem. Soc. 2015, 162, A2538.[64] M. Dahbi, N. Yabuuchi, K. Kubota, K. Tokiwa, S. Komaba, Phys.

Chem. Chem. Phys. 2014, 16, 15007.[65] K. Smith, T. Treacher, D. Ledwoch, P. Adamson, E. Kendrik, ECS

Trans. 2017, 75, 13.[66] Y. H. Jung, A. S. Christiansen, R. E. Johnsen, P. Norby, D. K. Kim,

Adv. Funct. Mater. 2015, 25, 3227.[67] M. J. Loveridge, M. J. Lain, Q. Huang, C. Wan, A. J. Roberts,

G. S. Pappas, R. Bhagat, Phys. Chem. Chem. Phys. 2016, 18, 30677.[68] M. A. Rahman, G. Song, A. I. Bhatt, Y. C. Wong, C. Wen, Adv.

Funct. Mater. 2016, 26, 647.[69] W. Kang, N. Deng, J. Ju, Q. Li, D. Wu, X. Ma, L. Li, M. Naebe,

B. Cheng, Nanoscale 2016, 8, 16541.[70] T. Christen, M. W. Carlen, J. Power Sources 2000, 91, 210.[71] N. Nitta, F. Wu, J. T. Lee, G. Yushin, Mater. Today 2014, 18, 252.[72] M. D. Slater, D. Kim, E. Lee, C. S. Johnson, Adv. Funct. Mater.

2013, 23, 947.[73] N. Ortiz-Vitoriano, N. E. Drewett, E. Gonzalo, T. Rojo, Energy

Environ. Sci. 2017, 10, 1051.[74] X. Xiang, K. Zhang, J. Chen, Adv. Mater. 2015, 27, 5343.[75] P. Serras, V. Palomares, A. Goñi, I. Gil de Muro, P. Kubiak,

L. Lezama, T. Rojo, T. Rojo, C. M. Julien, H. Rosner, J. Mater. Chem. 2012, 22, 22301.

[76] V. Palomares, P. Serras, I. Villaluenga, K. B. Hueso, J. Carretero-González, T. Rojo, M. Paleologou, A. Yelon, C. Geibel, H. Rosner, Energy Environ. Sci. 2012, 5, 5884.

[77] I. Hasa, D. Buchholz, S. Passerini, B. Scrosati, J. Hassoun, Adv. Energy Mater. 2014, 4, 1400083.

[78] M. H. Han, E. Gonzalo, G. Singh, T. Rojo, Energy Environ. Sci. 2015, 8, 81.

[79] Y.-U. Park, D.-H. Seo, H. Kim, J. Kim, S. Lee, B. Kim, K. Kang, Adv. Funct. Mater. 2014, 24, 4603.

[80] C.-M. Park, J.-H. Kim, H. Kim, H.-J. Sohn, Chem. Soc. Rev. 2010, 39, 3115.

[81] D. a. Stevens, J. R. Dahn, J. Electrochem. Soc. 2000, 147, 1271.[82] S. Komaba, T. Ishikawa, N. Yabuuchi, W. Murata, A. Ito, Y. Ohsawa,

ACS Appl. Mater. Interfaces 2011, 3, 4165.[83] H. Kang, Y. Liu, K. Cao, Y. Zhao, L. Jiao, Y. Wang, H. Yuan,

A. Schober, Y. Lei, X. J. Huang, X. J. Huang, J. Liu, J. Mater. Chem. A 2015, 3, 17899.

[84] J. Qian, Y. Xiong, Y. Cao, X. Ai, H. Yang, Nano Lett. 2014, 14, 1865.





[85] M. M. Doeff, Y. Ma, S. J. Visco, L. C. De Jonghe, J. Electrochem. Soc. 1993, 140, L169.

[86] Y. Okamoto, J. Phys. Chem. C 2014, 118, 16.[87] B. Jache, P. Adelhelm, Angew. Chem., Int. Ed. 2014, 53, 10169.[88] W. A. Henderson, J. Phys. Chem. B 2006, 110, 13177.[89] H. Kim, J. Hong, Y. U. Park, J. Kim, I. Hwang, K. Kang, Adv. Funct.

Mater. 2015, 25, 534.[90] E. Irisarri, A. Ponrouch, M. R. Palacin, J. Electrochem. Soc. 2015,

162, A2476.[91] L. Wu, D. Buchholz, C. Vaalma, G. A. Giffin, S. Passerini, Chem

ElectroChem 2016, 3, 292.[92] Y. Li, Y.-S. Hu, H. Li, L. Chen, X. Huang, J. Mater. Chem. A 2016,

4, 96.[93] W. Luo, Z. Jian, Z. Xing, W. Wang, C. Bommier, M. M. Lerner, X. Ji,

ACS Cent. Sci. 2015, 1, 516.[94] E. R. Buiel, A. E. George, J. R. Dahn, Carbon N. Y. 1999, 37, 1399.[95] W. Xing, J. S. Xue, J. R. Dahn, J. Electrochem. Soc. 1996, 143, 3046.[96] P.-C. Tsai, S.-C. Chung, S.-K. Lin, A. Yamada, J. Mater. Chem.

A 2015, 3, 9763.[97] S. Komaba, W. Murata, T. Ishikawa, N. Yabuuchi, T. Ozeki,

T. Nakayama, A. Ogata, K. Gotoh, K. Fujiwara, Adv. Funct. Mater. 2011, 21, 3859.

[98] F. Béguin, F. Chevallier, C. Vix-Guterl, S. Saadallah, V. Bertagna, J. N. Rouzaud, E. Frackowiak, Carbon N. Y. 2005, 43, 2160.

[99] J. Jin, Z.-Q. Q. Shi, C.-Y. Y. Wang, Electrochim. Acta 2014, 141, 302.[100] H. Hou, X. Qiu, W. Wei, Y. Zhang, X. Ji, Adv. Energy Mater. 2017,

201602898, 1.[101] L. Fan, B. Lu, Small 2016, 12, 2783.[102] S. Wenzel, T. Hara, U. Janek, P. Adelhelm, Energy Environ. Sci.

2011, 4, 3342.[103] V. G. Pol, E. Lee, D. Zhou, F. Dogan, J. M. Calderon-Moreno,

C. S. Johnson, Electrochim. Acta 2014, 127, 61.[104] K. Tang, L. Fu, R. J. White, L. Yu, M.-M. Titirici, M. Antonietti,

J. Maier, Adv. Energy Mater. 2012, 2, 873.[105] X. Zhou, Y.-G. Guo, ChemElectroChem 2014, 1, 83.[106] L. Cao, W. Hui, Z. Xu, J. Huang, P. Zheng, J. Li, Q. Sun, J. Alloys

Compd. 2017, 695, 632.[107] J.-Y. Hwang, S.-T. Myung, Y.-K. Sun, Chem. Soc. Rev. 2017, 46, 3529.[108] R. R. Gaddam, A. H. Farokh Niaei, M. Hankel, D. J. Searles,

N. A. Kumar, X. S. Zhao, J. Mater. Chem. A 2017, 5, 22186.[109] E. M. Lotfabad, J. Ding, K. Cui, A. Kohandehghan, W. P. Kalisvaart,

M. Hazelton, D. Mitlin, ACS Nano 2014, 8, 7115.[110] K. Kim, D. G. Lim, C. W. Han, S. Osswald, V. Ortalan,

J. P. Youngblood, V. G. Pol, ACS Sustainable Chem. Eng. 2017, 5, 8720.

[111] T. Zhang, J. Mao, X. Liu, M. Xuan, K. Bi, X. L. Zhang, J. Hu, J. Fan, S. Chen, G. Shao, RSC Adv. 2017, 7, 41504.

[112] J. Yang, T. Zhou, R. Zhu, X. Chen, Z. Guo, J. Fan, H. K. Liu, W. X. Zhang, Adv. Mater. Interfaces 2016, 3, 1.

[113] S. H. Yu, S. H. Lee, D. J. Lee, Y. E. Sung, T. Hyeon, Small 2016, 12, 2146.

[114] S. Hariharan, K. Saravanan, V. Ramar, M. R. Palacín, P. Balaya, Phys. Chem. Chem. Phys. 2013, 15, 2945.

[115] Z. Jian, B. Zhao, P. Liu, F. Li, M. Zheng, M. Chen, Y. Shi, H. Zhou, Chem. Commun. 2014, 50, 1215.

[116] C. Li, Q. Hu, Y. Li, H. Zhou, Z. Lv, X. Yang, L. Liu, H. Guo, Sci. Rep. 2016, 6, 25556.

[117] S. Santangelo, M. Fiore, F. Pantò, S. Stelitano, M. Marelli, P. Frontera, P. Antonucci, G. Longoni, R. Ruffo, Solid State Ionics 2017, 309, 41.

[118] G. Longoni, M. Fiore, J.-H. Kim, Y. H. Jung, D. K. Kim, C. M. Mari, R. Ruffo, J. Power Sources 2016, 332, 42.

[119] D. Kundu, E. Telaie, V. Duffort, L. F. Nazar, Angew. Chem. Int. Ed. 2015, 54, 3431.

[120] D. Su, X. Xie, P. Munroe, S. Dou, G. Wang, Sci. Rep. 2014, 4, 6519.

[121] X. Xiao, X. Liu, H. Zhao, D. Chen, F. Liu, J. Xiang, Z. Hu, Y. Li, Adv. Mater. 2012, 24, 5762.

[122] Y. Liu, Z. Cheng, H. Sun, H. Arandiyan, J. Li, M. Ahmad, J. Power Sources 2015, 273, 878.

[123] M. M. Rahman, I. Sultana, Z. Chen, Nanoscale 2015, 7, 13088.[124] Y. C. Lu, C. Ma, J. Alvarado, T. Kidera, N. Dimov, Y. S. Meng,

S. Okada, J. Power Sources 2015, 284, 287.[125] D. Su, X. Xie, G. Wang, Chem. - Eur. J. 2014, 20, 3192.[126] M. Gu, A. Kushima, Y. Shao, J.-G. Zhang, J. Liu, N. D. Browning,

J. Li, C. Wang, Nano Lett. 2013, 13, 5203.[127] P. R. Kumar, Y. H. Jung, D. K. Kim, RSC Adv. 2015, 5, 79845.[128] J. H. Kim, Y. H. Jung, J. H. Yun, P. Ragupathy, D. K. Kim, Small

2017.[129] L. Wu, X. Hu, J. Qian, F. Pei, F. Wu, R. Mao, X. Ai, H. Yang, Y. Cao,

J. Mater. Chem. A 2013, 1, 7181.[130] Y.-P. Gao, X. Wu, K.-J. Huang, L.-L. Xing, Y.-Y. Zhang, L. Liu,

CrystEngComm 2017, 19, 404.[131] Y. Zhang, Q. Zhou, J. Zhu, Q. Yan, S. X. Dou, W. Sun, Adv. Funct.

Mater. 2017, 27, 1702317.[132] H. Zhang, Q. Yun, Q. Lu, X. Zhang, C. Tan, Angew. Chem., Int. Ed.

2018, 130, 614.[133] L. David, R. Bhandavat, G. Singh, ACS Nano 2014, 8, 1759.[134] E. Yang, H. Ji, Y. Jung, J. Phys. Chem. C 2015, 119, 26374.[135] J. Zhang, C. Du, Z. DAI, W. Chen, Y. Zheng, B. Li, Y. Zong,

X. Wang, J. Zhu, Q. Yan, ACS Nano 2017, 11, 10599.[136] X. Ou, X. Xiong, F. Zheng, C. Yang, Z. Lin, R. Hu, C. Jin, Y. Chen,

M. Liu, J. Power Sources 2016, 325, 410.[137] K. Zhang, M. Park, L. Zhou, G. H. Lee, W. Li, Y. M. Kang, J. Chen,

Adv. Funct. Mater. 2016, 26, 6728.[138] S. Zhu, Q. Li, Q. Wei, R. Sun, X. Liu, Q. An, L. Mai, ACS Appl.

Mater. Interfaces 2016, 9, 311.[139] N. R. Kim, J. Choi, H. J. Yoon, M. E. Lee, S. U. Son, H.-J. Jin,

Y. S. Yun, ACS Sustainable Chem. Eng. 2017, 5, 9802.[140] J. Zhao, X. Yu, Z. Gao, W. Zhao, R. Xu, Y. Liu, H. Shen, Chem. Eng.

J. 2018, 332, 548.[141] J. S. Cho, S. Y. Lee, J.-K. Lee, Y. C. Kang, ACS Appl. Mater. Interfaces

2016, 8, 21343.[142] X. Ou, J. Li, F. Zheng, P. Wu, Q. Pan, X. Xiong, C. Yang, M. Liu, J.

Power Sources 2017, 343, 483.[143] C. Xia, F. Zhang, H. Liang, H. N. Alshareef, Nano Res. 2017, 10,

4368.[144] Y. Pan, X. Cheng, Y. Huang, L. Gong, H. Zhang, ACS Appl. Mater.

Interfaces 2017, 9, 35820.[145] Z. Chen, R. Wu, M. Liu, H. Wang, H. Xu, Y. Guo, Y. Song, F. Fang,

X. Yu, D. Sun, Adv. Funct. Mater. 2017, 27, 1702046.[146] R. Thangavel, A. Samuthira Pandian, H. Vignesh Ramasamy,

Y.-S. Lee, ACS Appl. Mater. Interfaces 2017, 9, 40187.[147] M. Krengel, A.-L. Hansen, M. Kaus, S. Indris, N. Wolff, L. Kienle,

D. Westfal, W. Bensch, ACS Appl. Mater. Interfaces 2017, 9, 21283.[148] M. Krengel, P. Adelhelm, F. Klein, W. Bensch, Chem. Commun.

2015, 51, 13500.[149] S. Yuan, S. Wang, L. Li, Y.-h. Zhu, X.-b. Zhang, J.-m. Yan, ACS Appl.

Mater. Interfaces 2016, 8, 9178.[150] K. Zhang, M. Park, L. Zhou, G.-H. Lee, J. Shin, Z. Hu, S.-L. Chou,

J. Chen, Y.-M. Kang, Angew. Chem., Int. Ed. 2016, 55, 12822.[151] W. Zhao, C. Guo, C. M. Li, J. Mater. Chem. A 2017, 5, 19195.[152] L. Baggetto, J. K. Keum, J. F. Browning, G. M. Veith, Electrochem.

Commun. 2013, 34, 41.[153] Y. Liu, N. Zhang, L. Jiao, J. Chen, Adv. Mater. 2015, 27, 6702.[154] I. Hasa, J. Hassoun, Y.-K. Sun, B. Scrosati, ChemPhysChem 2014,

15, 2152.[155] W.-J. Li, S.-L. Chou, J.-Z. Wang, H.-K. Liu, S.-X. Dou, Nano Lett.

2013, 13, 5480.[156] J. Sun, H.-W. Lee, M. Pasta, H. Yuan, G. Zheng, Y. Sun, Y. Li,

Y. Cui, Nat. Nanotechnol. 2015, 10, 980.





[157] A. Darwiche, C. Marino, M. T. Sougrati, B. Fraisse, L. Stievano, L. Monconduit, J. Am. Chem. Soc. 2012, 134, 20805.

[158] W. Zhang, Y. Liu, C. Chen, Z. Li, Y. Huang, X. Hu, Small 2015, 11, 3822.

[159] V. L. Chevrier, G. Ceder, J. Electrochem. Soc. 2011, 158, A1011.[160] X. Liu, Y. Gao, R. Jin, H. Luo, P. Peng, Y. Liu, Nano Energy 2014,

4, 31.[161] D. S. Jung, M.-H. Ryou, Y. J. Sung, S. Bin Park, J. W. Choi, Proc.

Natl. Acad. Sci. USA 2013, 110, 12229.[162] N. Liu, K. Huo, M. T. McDowell, J. Zhao, Y. Cui, Sci. Rep. 2013,

3, 1919.[163] S. Komaba, Y. Matsuura, T. Ishikawa, N. Yabuuchi, W. Murata,

S. Kuze, Electrochem. Commun. 2012, 21, 65.[164] S. C. Jung, D. S. Jung, J. W. Choi, Y. K. Han, J. Phys. Chem. Lett.

2014, 5, 1283.[165] C. H. Lim, T. Y. Huang, P. S. Shao, J. H. Chien, Y. T. Weng,

H. F. Huang, B. J. Hwang, N. L. Wu, Electrochim. Acta 2016, 211, 265.

[166] L. D. Ellis, T. D. Hatchard, M. N. Obrovac, J. Electrochem. Soc. 2012, 159, A1801.

[167] H. Zhu, Z. Jia, Y. Chen, N. Weadock, J. Wan, O. Vaaland, X. Han, T. Li, L. Hu, Nano Lett. 2013, 13, 3093.

[168] L. Xiao, Y. Cao, J. Xiao, W. Wang, L. Kovarik, Z. Nie, J. Liu, Chem. Commun. (Cambridge, U. K.) 2012, 48, 3321.

[169] Y.-M. Lin, P. R. Abel, A. Gupta, J. B. Goodenough, A. Heller, C. B. Mullins, ACS Appl. Mater. Interfaces 2013, 5, 8273.

[170] B. Farbod, K. Cui, W. P. Kalisvaart, M. Kupsta, B. Zahiri, A. Kohandehghan, E. M. Lotfabad, Z. Li, E. J. Luber, D. Mitlin, ACS Nano 2014, 8, 4415.

[171] W. Chen, D. Deng, ACS Sustainable Chem. Eng. 2015, 3, 63.[172] C. Luo, Y. Xu, Y. Zhu, Y. Liu, S. Zheng, Y. Liu, A. Langrock, C. Wang,

ACS Nano 2013, 7, 8003.[173] L. Wu, X. Hu, J. Qian, F. Pei, F. Wu, R. Mao, X. Ai, H. Yang, Y. Cao,

Energy Environ. Sci. 2014, 7, 323.[174] C. Zhang, X. Wang, Q. Liang, X. Liu, Q. Weng, J. Liu, Y. Yang,

Z. Dai, K. Ding, Y. Bando, J. Tang, D. Golberg, Nano Lett. 2016, 16, 2054.

[175] M. Dahbi, N. Yabuuchi, M. Fukunishi, K. Kubota, K. Chihara, K. Tokiwa, X.-F. Yu, H. Ushiyama, K. Yamashita, J.-Y. Son, Y.-T. Cui, H. Oji, S. Komaba.

[176] J. Sun, G. Zheng, H. W. Lee, N. Liu, H. Wang, H. Yao, W. Yang, Y. Cui, Nano Lett. 2014, 14, 4573.

[177] Z. Yang, D. Choi, S. Kerisit, K. M. Rosso, D. Wang, J. Zhang, G. Graff, J. Liu, J. Power Sources 2009, 192, 588.

[178] G.-N. Zhu, Y.-G. Wang, Y.-Y. Xia, Energy Environ. Sci. 2012, 5, 6652.

[179] H. Xiong, M. D. Slater, M. Balasubramanian, C. S. Johnson, T. Rajh, J. Phys. Chem. Lett. 2011, 2, 2560.

[180] L. Wu, D. Bresser, D. Buchholz, G. Giffin, C. R. Castro, A. Ochel, S. Passerini, Adv. Energy Mater. 2014, 5, 1.

[181] M. N. Tahir, B. Oschmann, D. Buchholz, X. Dou, I. Lieberwirth, M. Panthöfer, W. Tremel, R. Zentel, S. Passerini, Adv. Energy Mater. 2015, 6, 1501489.

[182] G. Longoni, R. L. Pena Cabrera, S. Polizzi, M. D’Arienzo, C. M. Mari, Y. Cui, R. Ruffo, Nano Lett. 2017, 17, 992.

[183] J. Xu, C. Ma, M. Balasubramanian, Y. S. Meng, Chem. Commun. 2014, 50, 12564.

[184] M. Naguib, M. Kurtoglu, V. Presser, J. Lu, J. Niu, M. Heon, L. Hultman, Y. Gogotsi, M. W. Barsoum, Adv. Mater. 2011, 23, 4248.

[185] X. Wang, S. Kajiyama, H. Iinuma, E. Hosono, S. Oro, I. Moriguchi, M. Okubo, A. Yamada, Nat. Commun. 2015, 6, 6544.

[186] S. Kajiyama, L. Szabova, K. Sodeyama, H. Iinuma, R. Morita, K. Gotoh, Y. Tateyama, M. Okubo, A. Yamada, ACS Nano 2016, 10, 3334.

[187] M. A. Hope, A. C. Forse, K. J. Griffith, M. R. Lukatskaya, M. Ghidiu, Y. Gogotsi, C. P. Grey, Phys. Chem. Chem. Phys. 2016, 18, 5099.

[188] H. W. Wang, M. Naguib, K. Page, D. J. Wesolowski, Y. Gogotsi, Chem. Mater. 2016, 28, 349.

[189] O. Mashtalir, M. Naguib, V. N. Mochalin, Y. Dall’Agnese, M. Heon, M. W. Barsoum, Y. Gogotsi, Nat. Commun. 2013, 4, 1716.

[190] X. Xie, M. Q. Zhao, B. Anasori, K. Maleski, C. E. Ren, J. Li, Y. Gogotsi, Nano Energy 2016, 26, 513.

[191] M.-Q. Zhao, X. Xie, C. E. Ren, T. Makaryan, B. Anasori, G. Wang, Y. Gogotsi, Adv. Mater. 2017, 29, 1702410.

[192] X. Guo, X. Xie, S. Choi, Y. Zhao, H. Liu, C. Wang, S. Chang, G. Wang, J. Mater. Chem. A 2017, 5, 12445.

[193] Y. Wu, P. Nie, J. Jiang, B. Ding, H. Dou, X. Zhang, ChemElectroChem 2017, 4, 1560.

[194] Y. Lei, X. Li, L. Liu, G. Ceder, Chem. Mater. 2014, 26, 5288.[195] J.-Y. Hwang, S.-M. Oh, S.-T. Myung, K. Y. Chung, I. Belharouak,

Y.-K. Sun, Nat. Commun. 2015, 6, 6865.[196] G. H. Newmann, L. P. Klemann, J. Electrochem. Soc. 1980,

127, 2097.[197] R. Berthelot, D. Carlier, C. Delmas, Nat. Mater. 2011, 10, 74.[198] J.-J. Braconnier, C. Delmas, C. Fouassier, P. Hagenmuller, Mater.

Res. Bull. 1980, 15, 1797.[199] R. Ruffo, R. Fathi, D. K. J. Kim, Y. H. Jung, C. M. Mari, D. K. J. Kim,

Electrochim. Acta 2013, 108, 575.[200] M. D’Arienzo, R. Ruffo, R. Scotti, F. Morazzoni, C. M. Mari,

S. Polizzi, Phys. Chem. Chem. Phys. 2012, 14, 5945.[201] W.-H. Ryu, J.-W. Jung, K. Park, S.-J. Kim, I.-D. Kim, Nanoscale 2014,

6, 10975.[202] Y. H. Ung, C. H. Lim, D. K. Kim, J. Mater. Chem. A 2013, 1, 11350.[203] Y. H. Jung, C. H. Lim, J.-H. Kim, D. K. Kim, RSC Adv. 2014, 4, 9799.[204] H.-W. Lee, R. Y. Wang, M. Pasta, S. W. Lee, N. Liu, Y. Cui, Nat.

Commun. 2014, 5, 5280.[205] W. Deng, X. Liang, X. Wu, J. Qian, Y. Cao, X. Ai, J. Feng, H. Yang,

Sci. Rep. 2015, 3, 2671.[206] C. Delmas, C. Fouassier, P. Hagenmuller, Physica B+C (Amsterdam)

1980, 99, 81.[207] T. Shibata, Y. Fukuzumi, W. Kobayashi, Y. Moritomo, Sci. Rep.

2015, 5, 9006.[208] C. Delmas, A. Mendiboure, P. Hagenmuller, J. Solid State Chem.

1985, 57, 323.[209] Z. Lu, J. R. Dahn, J. Electrochem. Soc. 2001, 148, A1225.[210] R. J. Clément, P. G. Bruce, C. P. Grey, J. Electrochem. Soc. 2015,

162, A2589.[211] N. Yabuuchi, R. Hara, K. Kubota, J. Paulsen, S. Kumakura,

S. Komaba, J. Mater. Chem. A 2014, 2, 16851.[212] L. Ni Mn, J. Xu, D. Hoe Lee, le J. Cleent, X. Yu, M. Leskes, A. J. Pell,

G. Pintacuda, X.-Q. Yang, C. P. Grey, Y. Shirley Meng, C. Bernard Lyon, Chem. Mater. 2014, 26, 1260.

[213] H. Yoshida, N. Yabuuchi, K. Kubota, I. Ikeuchi, A. Garsuch, M. Schulz-Dobrick, S. Komaba, Chem. Commun. 2014, 50, 3677.

[214] P.-F. Wang, Y. You, Y.-X. Yin, Y.-S. Wang, L.-J. Wan, L. Gu, Y.-G. Guo, Angew. Chem., Int. Ed. 2016, 55, 7445.

[215] I. Hasa, D. Buchholz, S. Passerini, J. Hassoun, ACS Appl. Mater. Interfaces 2015, 7, 5206.

[216] D. H. Lee, J. Xu, Y. S. Meng, J. Xua, Y. S. Meng, Phys. Chem. Chem. Phys. 2013, 15, 3304.

[217] C.-Y. Chen, K. Matsumoto, T. Nohira, R. Hagiwara, A. Fukunaga, S. Sakai, K. Nitta, S. Inazawa, J. Power Sources 2013, 237, 52.

[218] C.-Y. Yu, J.-S. Park, H.-G. Jung, K.-Y. Chung, D. Aurbach, Y.-K. Sun, S.-T. Myung, Energy Environ. Sci 2015, 8, 2019.

[219] R. Kataoka, T. Mukai, A. Yoshizawa, T. Sakai, J. Electrochem. Soc. 2013, 160, A933.

[220] H. Liu, J. Xu, C. Ma, Y. S. Meng, Chem. Commun. 2015, 51, 4693.[221] X. Li, D. Wu, Y.-N. Zhou, L. Liu, X.-Q. Yang, G. Ceder, Electrochem.

Commun. 2014, 49, 51.




[222] M. Sathiya, K. Hemalatha, K. Ramesha, J.-M. Tarascon, A. S. Prakash, Chem. Mater. 2012, 24, 1846.

[223] X. Sun, Y. Jin, C.-Y. Zhang, J.-W. Wen, Y. Shao, Y. Zang, C.-H. Chen, J. Mater. Chem. A 2014, 2, 17268.

[224] P. Vassilaras, S. T. Dacek, H. Kim, T. T. Fister, S. Kim, G. Ceder, J. C. Kim, J. Electrochem. Soc. 2017, 164, A3484.

[225] R. Tripathi, S. M. Wood, M. S. Islam, L. F. Nazar, Energy Environ. Sci. 2013, 6, 2257.

[226] J. Liu, K. Tang, K. Song, P. A. van Aken, Y. Yu, J. Maier, Nanoscale 2014, 6, 5081.

[227] L. Tan, S. Zhang, C. Deng, J. Power Sources 2015, 275, 6.[228] D. Kundu, R. Tripathi, G. Popov, W. R. M. Makahnouk, L. F. Nazar,

Chemistry 2015, 4906, 5349.[229] P. Barpanda, T. Ye, S. Nishimura, S.-C. Chung, Y. Yamada,

M. Okubo, H. Zhou, A. Yamada, Electrochem. Commun. 2012, 24, 116.

[230] P. Barpanda, J. Lu, T. Ye, M. Kajiyama, S.-C. Chung, N. Yabuuchi, S. Komaba, A. Yamada, RSC Adv. 2013, 3, 3857.

[231] H. Kim, R. a. Shakoor, C. Park, S. Y. Lim, J. S. Kim, Y. N. Jo, W. Cho, K. Miyasaka, R. Kahraman, Y. Jung, J. W. Choi, Adv. Funct. Mater. 2013, 23, 1147.

[232] S. Chung, Y. Yamada, A. Yamada, Chem. Mater. 2013, 25, 4557.[233] A. K. Padhi, K. S. Nanjundaswamy, J. B. Goodenough, J. Electrochem.

Soc. 1997, 144, 1188.[234] M. Casas-Cabanas, V. V. Roddatis, D. Saurel, P. Kubiak,

J. Carretero-González, V. Palomares, P. Serras, T. Rojo, J. Mater. Chem. 2012, 22, 17421.

[235] S.-M. Oh, S.-T. Myung, J. Hassoun, B. Scrosati, Y.-K. Sun, Electrochem. Commun. 2012, 22, 149.

[236] Y. Kawabe, N. Yabuuchi, M. Kajiyama, N. Fukuhara, T. Inamasu, R. Okuyama, I. Nakai, S. Komaba, Electrochem. Commun. 2011, 13, 1225.

[237] B. L. Ellis, W. R. M. Makahnouk, W. N. Rowan-Weetaluktuk, D. H. Ryan, L. F. Nazar, Chem. Mater. 2010, 22, 1059.

[238] H. Zou, S. Li, X. Wu, M. J. Mcdonald, Y. Yang, ECS Electrochem. Lett. 2015, 4, 53.

[239] Q. Ni, Y. Bai, F. Wu, C. Wu, Adv. Sci. 2017, 4, 1600275.[240] J. C. Treacher, S. M. Wood, M. S. Islam, E. Kendrick, Phys. Chem.

Chem. Phys. 2016, 18, 32744.[241] W. Guan, B. Pan, P. Zhou, J. Mi, D. Zhang, J. Xu, Y. Jiang, ACS

Appl. Mater. Interfaces 2017, 9, 22369.[242] H. Zhu, J. Wang, X. Liu, X. Zhu, RSC Adv. 2017, 7, 14145.[243] C. Delmas, A. Nadiri, J. L. Soubeyroux, Solid State Ionics 1988,

28, 419.[244] Y.-U. Park, D.-H. Seo, H.-S. Kwon, B. Kim, J. Kim, H. Kim, I. Kim,

H.-I. Yoo, K. Kang, J. Am. Chem. Soc. 2013, 135, 13870.[245] D. Wang, Q. Liu, C. Chen, M. Li, X. Meng, X. Bie, Y. Wei,

Y. Huang, F. Du, C. Wang, G. Chen, Appl. Mater. Interfaces 2015, 8, 2238.

[246] J. M. Clark, P. Barpanda, A. Yamada, M. S. Islam, J. Mater. Chem. A 2014, 2, 11807.

[247] P. Barpanda, T. Ye, M. Avdeev, S. Chung, A. Yamada, J. Mater. Chem. A 2013, 1, 4194.

[248] J. Liu, D. Chang, P. Whitfield, Y. Janssen, X. Yu, Y. Zhou, J. Bai, J. Ko, K.-W. Nam, L. Wu, Y. Zhu, M. Feygenson, G. Amatucci, A. Van der Ven, X.-Q. Yang, P. Khalifah, Chem. Mater. 2014, 26, 3295.

[249] N. Yabuuchi, M. Kajiyama, J. Iwatate, H. Nishikawa, S. Hitomi, R. Okuyama, R. Usui, Y. Yamada, S. Komaba, Nat. Mater. 2012, 11, 512.

[250] J. Zhao, J. Xu, D. H. Lee, N. Dimov, Y. S. Meng, S. Okada, J. Power Sources 2014, 264, 235.

[251] P. Barpanda, G. Oyama, S. Nishimura, S.-C. Chung, A. Yamada, Nat. Commun. 2014, 5, 4358.

[252] P. Han, X. Han, J. Yao, Z. Liu, X. Cao, G. Cui, Electrochem. Commun. 2015, 61, 84.

[253] Z. Zhu, F. Cheng, Z. Hu, Z. Niu, J. Chen, J. Power Sources 2015, 293, 626.

[254] I. Hasa, X. Dou, D. Buchholz, Y. Shao-Horn, J. Hassoun, S. Passerini, B. Scrosati, J. Power Sources 2016, 310, 26.

[255] W. Ren, Z. Zhu, Q. An, L. Mai, Small 2017, 1604181, 1.[256] B. Zhang, C. M. Ghimbeu, C. Laberty, C. Vix-Guterl, J. M. Tarascon,

Adv. Energy Mater. 2016, 6, 1.[257] M. Keller, C. Vaalma, D. Buchholz, S. Passerini, ChemElectroChem

2016, 3, 1030.[258] W. Luo, C. Bommier, Z. Jian, X. Li, R. Carter, S. Vail, Y. Lu, J. J. Lee,

X. Ji, ACS Appl. Mater. Interfaces 2015, 7, 2626.[259] F. Shen, H. Zhu, W. Luo, J. Wan, L. Zhou, J. Dai, B. Zhao, X. Han,

K. Fu, L. Hu, ACS Appl. Mater. Interfaces 2015, 7, 23291.[260] H. Wang, Y. Xiao, C. Sun, C. Lai, X. Ai, RSC Adv. 2015, 5, 106519.[261] J. Hassoun, F. Bonaccorso, M. Agostini, M. Angelucci, M. G. Betti,

R. Cingolani, M. Gemmi, C. Mariani, S. Panero, V. Pellegrini, B. Scrosati, Nano Lett. 2014, 14, 1.

[262] Y. Li, S. Xu, X. Wu, J. Yu, Y. Wang, Y.-S. Hu, H. Li, L. Chen, X. Huang, J. Mater. Chem. A 2015, 3, 71.

[263] R. Dugas, B. Zhang, P. Rozier, J. M. Tarascon, J. Electrochem. Soc. 2016, 163, A867.

[264] B. Zhang, R. Dugas, G. Rousse, P. Rozier, A. M. Abakumov, J.-M. Tarascon, Nat. Commun. 2016, 7, 10308.

[265] J. Ni, S. Fu, C. Wu, Y. Zhao, J. Maier, Y. Yu, L. Li, Adv. Energy Mater. 2016, 6, 2.

[266] N. Wang, Z. Bai, Y. Qian, J. Yang, Adv. Mater. 2016, 28, 4126.[267] H. Li, L. Peng, Y. Zhu, D. Chen, X. Zhang, G. Yu, W. Yang, Y. Lu,

D. Evans, J.-J. Lee, Energy Environ. Sci. 2016, 9, 3399.[268] Y. Sun, L. Zhao, H. Pan, X. Lu, L. Gu, Y.-S. Hu, H. Li, M. Armand,

Y. Ikuhara, L. Chen, X. Huang, Nat. Commun. 2013, 4, 1870.[269] K. Zaghib, M. Simoneau, M. Armand, M. Gauthier, J. Power

Sources 1999, 81–82, 300.[270] Y. Wang, X. Yu, S. Xu, J. Bai, R. Xiao, Y.-S. Hu, H. Li, X.-Q. Yang,

L. Chen, X. Huang, Nat. Commun. 2013, 4, 1.[271] A. Rudola, K. Saravanan, S. Devaraj, H. Gong, P. Balaya, A. Kuhn,

J. Liu, K. Gotoh, K. Fujiwara, Chem. Commun. 2013, 49, 7451.[272] H. Xiong, M. D. Slater, M. Balasubramanian, C. S. Johnson,

T. Rajh, J. Phys. Chem. Lett. 2011, 2, 2560.[273] J. Wang, J. Polleux, J. Lim, B. Dunn, J. Phys. Chem. C 2007, 2, 14925.[274] N. Wang, Z. Bai, Y. Qian, J. Yang, Adv. Mater. 2016, 28, 4126.[275] V. Raju, J. Rains, C. Gates, W. Luo, X. Wang, W. F. Stickle,

G. D. Stucky, X. Ji, Nano Lett. 2014, 14, 4119.[276] A. Moretti, M. Secchiaroli, D. Buchholz, G. Giuli, R. Marassi,

S. Passerini, J. Electrochem. Soc. 2015, 162, A2723.[277] J. W. Kim, V. Augustyn, B. Dunn, Adv. Energy Mater. 2012, 2, 141.[278] S. Hariharan, K. Saravanan, P. Balaya, Electrochem. Commun.

2013, 31, 5.[279] X. Xia, Q. Hao, W. Lei, W. Wang, H. Wang, X. Wang, J. Mater.

Chem. 2012, 22, 8314.[280] H. Ye, Y. Wang, F. Zhao, W. Huang, N. Han, J. Zhou, M. Zeng,

Y. Li, J. Mater. Chem. A 2016, 4, 1754.[281] S. Komaba, T. Mikumo, A. Ogata, Electrochem. Commun. 2008,

10, 1276.[282] P. R. Kumar, Y. H. Jung, K. K. Bharathi, C. H. Lim, D. K. Kim,

Electrochim. Acta 2014, 146, 503.[283] S. M. Oh, S.-T. T. Myung, C. S. Yoon, J. Lu, J. Hassoun, B. Scrosati,

K. Amine, Y.-K. K. Sun, Nano Lett. 2014, 14, 1620.[284] X. Xie, K. Kretschmer, J. Zhang, B. Sun, D. Su, G. Wang, Nano

Energy 2015, 13, 208.[285] I. Hasa, S. Passerini, J. Hassoun, RSC Adv. 2015, 5, 48928.[286] W. Li, S. L. Chou, J. Z. Wang, J. H. Kim, H. K. Liu, S. X. Dou, Adv.

Mater. 2014, 26, 4037.





[287] P. R. Abel, M. G. Fields, A. Heller, C. B. Mullins, ACS Appl. Mater. Interfaces 2014, 6, 15860.

[288] Y. Zhao, A. Manthiram, Chem. Mater. 2015, 27, 3096.[289] M. Walter, S. Doswald, M. V. Kovalenko, J. R. Harding,

R. R. Mitchell, M. S. Whittingham, Y. Shao-Horn, H. Zhou, L. Hu, C. P. Grey, K. W. Chapman, P. J. Chupas, X.-Q. Yang, C. P. Grey, J. Mater. Chem. A 2016, 4, 7053.

[290] D. Y. W. Yu, P. V Prikhodchenko, C. W. Mason, S. K. Batabyal, J. Gun, S. Sladkevich, A. G. Medvedev, O. Lev, Nat. Commun. 2013, 4, 2922.

[291] W. Sun, X. Rui, D. Yang, Z. Sun, B. Li, W. Zhang, Y. Zong, S. Madhavi, S. Dou, Q. Yan, ACS Nano 2015, 9, 11371.

[292] Y. C. Lu, C. Ma, J. Alvarado, N. Dimov, Y. S. Meng, S. Okada, J. Mater. Chem. A 2015, 3, 16971.

[293] Z. Hu, Z. Zhu, F. Cheng, K. Zhang, J. Wang, C. Chen, J. Chen, P. Johansson, M. R. Palacin, Y. Huang, T. Duong, Energy Environ. Sci. 2015, 8, 1309.

[294] Y. Liu, N. Zhang, H. Kang, M. Shang, L. Jiao, J. Chen, Chem. - Eur. J. 2015, 21, 11878.

[295] P. R. Kumar, Y. H. Jung, D. K. Kim, E. Levi, E. Sominski, A. Gedanken, D. Gonbeau, G. Yushin, Y. Shi, H. Zhang, C. S. Lai, L. J. Li, RSC Adv. 2015, 5, 79845.

[296] C. Wu, Y. Jiang, P. Kopold, P. A. van Aken, J. Maier, Y. Yu, Adv. Mater. 2016, 28, 7276.

[297] B. Qu, C. Ma, G. Ji, C. Xu, J. Xu, Y. S. Meng, T. Wang, J. Y. Lee, Adv. Mater. 2014, 26, 3854.

[298] S. Peng, X. Han, L. Li, Z. Zhu, F. Cheng, M. Srinivansan, S. Adams, S. Ramakrishna, Small 2016, 12, 1359.

[299] C. D. Wessells, S. V. Peddada, R. A. Huggins, Y. Cui, Nano Lett. 2011, 11, 5421.

[300] C. D. Wessells, R. A. Huggins, Y. Cui, Nat. Commun. 2011, 2, 550.[301] M. Pasta, C. D. Wessells, R. A. Huggins, Y. Cui, Nat. Commun.

2012, 3, 1149.[302] D. J. Kim, Y. H. Jung, K. K. Bharathi, S. H. Je, D. K. Kim, A. Coskun,

J. W. Choi, Adv. Energy Mater. 2014, 4, 1400133.[303] R. Y. Wang, C. D. Wessells, R. A. Huggins, Y. Cui, Nano Lett. 2013,

13, 5748.[304] P. R. Kumar, Y. H. Jung, C. H. Lim, D. K. Kim, J. Mater. Chem.

A 2015, 3, 6271.[305] C. Deng, S. Zhang, Z. Dong, Y. Shang, Nano Energy 2014, 4, 49.[306] C. Liu, Z. G. Neale, G. Cao, Mater. Today 2016, 19, 109.[307] Y. Noguchi, E. Kobayashi, L. S. Plashnitsa, S. Okada, J. Yamaki,

Electrochim. Acta 2013, 101, 59.[308] E. Kobayashi, L. S. Plashnitsa, T. Doi, S. Okada, J. Yamaki, Electro-

chem. Commun. 2010, 12, 894.[309] S. Guo, H. Yu, P. Liu, Y. Ren, T. Zhang, M. Chen, M. Ishida,

H. Zhou, Energy Environ. Sci. 2015, 8, 1237.[310] M. Palanisamy, H. W. Kim, S. Heo, E. Lee, Y. Kim, ACS Appl. Mater.

Interfaces 2017, 9, 10618.[311] L. S. Plashnitsa, E. Kobayashi, Y. Noguchi, S. Okada, J. Yamaki,

J. Electrochem. Soc. 2010, 157, A536.[312] S. Li, Y. Dong, L. Xu, X. Xu, L. He, L. Mai, Adv. Mater. 2014,

26, 3545.[313] P. Senguttuvan, G. Rousse, M. E. Arroyo Y De Dompablo,

H. Vezin, J. M. Tarascon, M. R. Palacín, J. Am. Chem. Soc. 2013, 135, 3897.

[314] S. Komaba, T. Itabashi, T. Ohtsuka, H. Groult, N. Kumagai, B. Kaplan, H. Yashiro, J. Electrochem. Soc. 2005, 152, A937.

[315] M. K. Aydinol, A. F. Kohan, G. Ceder, K. Cho, J. Joannopoulos, Phys. Rev. B, 1997, 56, 1354.

[316] Y. Wang, R. Xiao, Y.-S. Hu, M. Avdeev, L. Chen, Nat. Commun. 2015, 6, 6954.

[317] P.-F. Wang, Y. You, Y.-X. Yin, Y.-S. Wang, L.-J. Wan, L. Gu, Y.-G. Guo, Angew. Chem., Int. Ed. 2016, 55, 7445.

[318] C. Ma, J. Alvarado, J. Xu, R. J. Clément, M. Kodur, W. Tong, C.P. Grey, Y. S. Meng, J. Am. Chem. Soc. 2017, 139, 4835.

[319] Y. Cao, L. Xiao, W. Wang, D. Choi, Z. Nie, J. Yu, L. V. Saraf, Z. Yang, J. Liu, Adv. Mater. 2011, 23, 3155.

[320] Y. Zhong, X. Xia, J. Zhan, X. Wang, J. Tu, J. Xiao, Z. Nie, L. V. Saraf, Z. Yang, J. Liu, X. J. Huang, J. Mater. Chem. A 2016, 4, 11207.

[321] D. Kim, E. Lee, M. Slater, W. Lu, S. Rood, C. S. Johnson, Electro-chem. Commun. 2012, 18, 66.

[322] L. Mu, S. Xu, Y. Li, Y.-S. Hu, H. Li, L. Chen, X. Huang, Adv. Mater. 2015, 27, 6928.

[323] S. Xu, Y. Wang, L. Ben, Y. Lyu, N. Song, Z. Yang, Y. Li, L. Mu, H.-T. Yang, L. Gu, Y.-S. Hu, H. Li, Z.-H. Cheng, L. Chen, X. Huang, Adv. Energy Mater. 2015, 5, 1501156.

[324] S.-M. Oh, S.-T. Myung, J.-Y. Hwang, B. Scrosati, K. Amine, Y.-K. Sun, Chem. Mater. 2014, 26, 6165.

[325] J.-K. Kim, Y. J. Lim, H. Kim, G.-B. Cho, Y. Kim, Energy Environ. Sci. 2015, 8, 3589.

[326] Z. Yuan, L. Si, X. Zhu, J. Mater. Chem. A 2015, 3, 23090.[327] A. Ponrouch, R. Dedryvère, D. Monti, A. E. Demet,

J. M. Ateba Mba, L. Croguennec, C. Masquelier, P. Johansson, M. R. Palacín, Energy Environ. Sci. 2013, 6, 2361.

[328] M. Nose, H. Nakayama, K. Nobuhara, H. Yamaguchi, S. Nakanishi, H. Iba, J. Power Sources 2013, 234, 175.

[329] D. Yang, J. Xu, X.-Z. Liao, Y.-S. He, H. Liu, Z.-F. Ma, Chem. Commun 2014, 50, 13377.

[330] J. Song, L. Wang, Y. Lu, J. Liu, B. Guo, P. Xiao, J.-J. Lee, X.-Q. Yang, G. Henkelman, J. B. Goodenough, J. Am. Chem. Soc. 2015, 137, 2658.

[331] L. Wang, J. Song, R. Qiao, L. Andrew Wray, M. A. Hossain, Y.-D. Chuang, W. Yang, Y. Lu, D. Evans, J.-J. Lee, S. Vail, X. Zhao, M. Nishijima, S. Kakimoto, J. B. Goodenough, J. Am. Chem. Soc. 2015, 137, 2548.

[332] K. Chihara, N. Chujo, A. Kitajou, S. Okada, Electrochim. Acta 2013, 110, 240.

[333] C. Ding, T. Nohira, R. Hagiwara, A. Fukunaga, S. Sakai, K. Nitta, Electrochim. Acta 2015, 176, 344.

[334] C.-Y. Yu, J.-S. Park, H.-G. Jung, K.-Y. Chung, D. Aurbach, Y.-K. Sun, S.-T. Myung, S. Komaba, H. Li, Z. W. Fu, X. Q. Yang, Energy Environ. Sci. 2015, 8, 2019.

[335] K. Zhang, Z. Hu, X. Liu, Z. Tao, J. Chen, Adv. Mater. 2015, 27, 3305.

[336] S. Peng, X. Han, L. Li, Z. Zhu, F. Cheng, M. Srinivansan, S. Adams, S. Ramakrishna, Small 2016, 12, 1359.

[337] H. Wang, Y. Liu, M. Li, H. Huang, H. M. Xu, R. J. Hong, H. Shen, Optoelectron. Adv. Mater., Rapid Commun. 2010, 4, 1166.

[338] L. Chen, W. Li, Y. Wang, C. Wang, Y. Xia, RSC Adv. 2014, 4, 25369.

[339] H. Banda, D. Damien, K. Nagarajan, M. Hariharan, M. M. Shaijumon, J. Mater. Chem. A 2015, 3, 10453.

[340] G. L. Xu, Z. Chen, G. M. Zhong, Y. Liu, Y. Yang, T. Ma, Y. Ren, X. Zuo, X. H. Wu, X. Zhang, K. Amine, Nano Lett. 2016, 16, 3955.

[341] M. Fan, Y. Chen, Y. Xie, T. Yang, X. Shen, N. Xu, H. Yu, C. Yan, Adv. Funct. Mater. 2016, 26, 5019.

[342] S. Hartung, N. Bucher, V. S. Nair, C. Y. Ling, Y. Wang, H. E. Hoster, M. Srinivasan, ChemPhysChem 2014, 15, 2121.

[343] S. Guo, P. Liu, Y. Sun, K. Zhu, J. Yi, M. Chen, M. Ishida, H. Zhou, Angew. Chem., Int. Ed. 2015, 54, 11701.

[344] S. Zhang, Y. Liu, N. Zhang, K. Zhao, J. Yang, S. He, J. Power Sources 2016, 329, 1.

[345] Y. Zhang, H. Zhao, Y. Du, S. Jin, Y. Zhou, Y. He, S. He, F. Zhao, H. Hou, D. Chen, L. Chen, L. Chen, J. Mater. Chem. A 2016, 4, 7155.

[346] D. Wang, X. Bie, Q. Fu, D. Dixon, N. Bramnik, Y. Hu, F. Fauth, Y. Wei, H. Ehrenberg, G. Chen, F. Du, Nat. Commun. 2017, 8, 1.


Readiness Level of Sodium‐Ion Battery Technology: A ...mse2.kaist.ac.kr/~ncrl/pub/2018_Readiness...

Documents

Transcript of Readiness Level of Sodium‐Ion Battery Technology: A ...mse2.kaist.ac.kr/~ncrl/pub/2018_Readiness...