Na-Ion Battery Anodes: Materials And Electrochemistry
Articlepubs.acs.org/accountsNa-Ion Battery Anodes: Materials and ElectrochemistryWei Luo,†,‡ Fei Shen,† Clement Bommier,§ Hongli Zhu,† Xiulei Ji,*,§ and Liangbing Hu*,††Department of Materials Science and Engineering and ‡Department of Mechanical Engineering, University of Maryland, CollegePark, Maryland 20742, United States§Department of Chemistry, Oregon State University, Corvallis, Oregon 97331, United StatesCONSPECTUS: The intermittent nature of renewable energy sources, such as solar and wind, callsfor sustainable electrical energy storage (EES) technologies for stationary applications. Li will besimply too rare for Li-ion batteries (LIBs) to be used for large-scale storage purposes. In contrast,Na-ion batteries (NIBs) are highly promising to meet the demand of grid-level storage because Na istruly earth abundant and ubiquitous around the globe. Furthermore, NIBs share a similar rockingchair operation mechanism with LIBs, which potentially provides high reversibility and long cyclinglife. It would be most efficient to transfer knowledge learned on LIBs during the last three decadesto the development of NIBs. Following this logic, rapid progress has been made in NIB cathodematerials, where layered metal oxides and polyanionic compounds exhibit encouraging results. Onthe anode side, pure graphite as the standard anode for LIBs can only form NaC64 in NIBs if solventco-intercalation does not occur due to the unfavorable thermodynamics. In fact, it was the utilization of a carbon anode in LIBsthat enabled the commercial successes. Anodes of metal-ion batteries determine key characteristics, such as safety and cycling life;thus, it is indispensable to identify suitable anode materials for NIBs.In this Account, we review recent development on anode materials for NIBs. Due to the limited space, we will mainly discusscarbon-based and alloy-based anodes and highlight progress made in our groups in this field. We first present what is knownabout the failure mechanism of graphite anode in NIBs. We then go on to discuss studies on hard carbon anodes, alloy-typeanodes, and organic anodes. Especially, the multiple functions of natural cellulose that is used as a low-cost carbon precursor formass production and as a soft substrate for tin anodes are highlighted. The strategies of minimizing the surface area of carbonanodes for improving the first-cycle Coulombic efficiency are also outlined, where graphene oxide was employed as dehydrationagent and 2,2,6,6-tetramethylpiperidine-1-oxyl (TEMPO) was used to unzip wood fiber. Furthermore, surface modification byatomic layer deposition technology is introduced, where we discover that a thin layer of Al2O3 can function to encapsulate Snnanoparticles, leading to a much enhanced cycling performance. We also highlight recent work about the phosphorene/grapheneanode, which outperformed other anodes in terms of capacity. The aromatic organic anode is also studied as anode with veryhigh initial sodiation capacity. Furthermore, electrochemical intercalation of Na ions into reduced graphene oxide is applied forfabricating transparent conductors, demonstrating the great feasibility of Na ion intercalation for optical applications.1. INTRODUCTIONOver the past three decades, Li-ion batteries (LIBs) haveachieved tremendous success as power sources for portableelectronic devices and electric vehicles (EVs). However,technological improvement of LIBs cannot address the rarityof Li resources, which may lead to a risk that EVs powered byLIBs will no longer be affordable with their exhaustive usage.1Thus, it is critical to develop alternative battery technologiesbeyond LIBs based on earth abundant elements. One highlypromising candidate is Na, which is adjacent to Li on the periodictable. It shares many similar alkali metal chemistries with Li and isvery abundant and widely distributed.1 3 The concept of Na-ionbatteries (NIBs) is not new: they were investigated together withLIBs back in the 1980s;4 however, by the early 1990s, theresearch community quickly lost interest in NIBs due to thelower energy density of NIBs and the advance of LIBs.Recently, ambient-temperature NIBs have raised muchattention again for grid-level applications considering thesustainability advantages of NIBs. Significant progress has beenmade for NIB cathodes by adapting the knowledge learned onLIBs.5,6 As for anode materials, graphite, the commercial anode 2016 American Chemical Societyfor LIBs, does not function in NIBs due to its extremely lowcapacity. This can be alleviated with solvent co-intercalation, butsuch a method brings about its own set of challenges.7 Therefore,numerous attempts have been made to find suitable anodes forNIBs.Although there have been several review articles related to NIBanodes,8 10 this Account aims to summarize recent progressesmade in carbon based and alloy based anodes for NIBs, especiallyfrom our research groups. We will highlight high performancehard carbon anodes and new insights on the Na ion storagemechanism in hard carbon. We will then discuss alloying typeanodes, such as tin (Sn) and phosphorus (P), which demonstratesome of the highest reported capacities but have their ownunique set of challenges and solutions. Lastly, metal oxides,sulfides, and organic based anodes will be reviewed, albeit brieflydue to the limited space.Received: October 27, 2015Published: January 19, 2016231DOI: 10.1021/acs.accounts.5b00482Acc. Chem. Res. 2016, 49, 231 240
ArticleAccounts of Chemical ResearchFigure 1. (a) Potential profiles of graphite electrodes in LIBs and NIBs.26 Reproduced by permission of The Electrochemical Society. (b) Cointercalation of Na ion and diglyme into graphite in NIBs. Reproduced from ref 7. Copyright 2014 WILEY. (c) Superior cycling performance ofexpanded graphite anode in NIBs. Reprinted by permission from Macmillan Publishers Ltd: Nat. Commun. (ref 19), copyright 2014. (d) Na ionintercalation into RGO enables a transparent conductor. Reproduced from ref 25. Copyright 2015 American Chemical Society.2. CARBON ANODESCarbon based anodes are leading candidates for LIBs due to theirlow potential, high capacity, abundance, and low cost.11 For thesame reasons, carbon anodes are also among the most promisingchoices for NIBs.solvents causes a high level of volume change, inherently limitingits practical application.Another approach to utilize graphitic carbons is to expandtheir interlayer distance.18 20 For example, Wang’s groupreported an expanded graphite anode with an interlayer distanceof 0.43 nm.19 The as-obtained expanded graphite shows a highreversible capacity of 300 mA·h/g at 20 mA/g and a stableperformance for 2000 cycles (Figure 1c). By comparing variousreduction conditions, Singh demonstrated that increasing orderand decreasing interlayer spacing of reduced graphene oxide(RGO) lead to a poorer performance.20 Recently, Mitlin and coworkers produced graphene-like materials from peat moss, whichexhibited enlarged intergraphene spacing (0.388 nm) andpromising Na ion storage properties.21 Ji and co-workers alsodiscovered that a narrower interlayer spacing of graphitizablecarbon leaded to a lower capacity.22On the other hand, Na ion or Li ion intercalation can be usedto tune the properties of two-dimensional (2D) materials.23,24For example, Wang et al. reported that Li ion intercalation caneffectively tune the structure and properties of MoS2. Theydemonstrated that lithiated MoS2 exhibited an enhancedhydrogen evolution reaction activity. Inspired by this, for thefirst time, we have successfully applied the electrochemicalintercalation of Na ions to build a transparent electrode.25 Asshown in Figure 1d, printed RGO films become much moretransparent after intercalation of Na ions (from 36% to 79%).Meanwhile, the sheet resistance shows a 270 times decrease(from 83 000 to 311 Ω/sq), which is attributed to the2.1. Graphitic CarbonLi ions are readily inserted into graphite with a finalstoichiometry of LiC6, which is equivalent to a capacity of 372mA·h/g. However, only a small amount of Na atoms can beintercalated into graphite (Figure 1a).12,13 This limited capacitycan be explained from a thermodynamic perspective, which isassociated with Na plating on the carbon surface before formingthe graphite intercalation compounds (GICs).14 16 For example,Grande et al. performed a density functional theory (DFT) studyand calculated the binding energy between Li, Na, K, andgraphene sheets, which showed that NaC6 was the onlyintercalation compound that was not energetically favorable.14Furthermore, solvation energy plays a significant role in thefeasibility of intercalation, though this has yet to becomprehensively studied.Recently, Adelhelm et al. proposed a co-intercalation approachto form a Na solvent graphite ternary GIC anode for NIBs.7With use of a diglyme-based electrolyte, graphite exhibited areversible capacity of 100 mA·h/g with a potential plateau at 0.6V at 0.1 C (Figure 1b). Such an unexpected capacity is due to aco-intercalation of diglyme solvated Na ions into graphite.Recently, Kang’s group further developed an ether-basedelectrolyte for graphite anodes.17 However, co-intercalation of232DOI: 10.1021/acs.accounts.5b00482Acc. Chem. Res. 2016, 49, 231 240
ArticleAccounts of Chemical ResearchFigure 2. (a, b) Cellulose nanofiber derived CNFs as a long-life NIB anode. Reproduced from ref 31 with permission from The Royal Society ofChemistry. (c, d) Specific capacity as a function of DFT pore volume and BET surface area. Reprinted from ref 34, with permission from Elseiver. (e)Mechanism of treating cellulose by TEMPO. Adapted from ref 40. Copyright 2015 American Chemical Society. (f) A three-tiered mechanism for Na ionstorage in hard carbon anode. Reproduced from ref 42. Copyright 2015 American Chemical Society.(1D) hard carbon anodes.32,33 Note that it is still elusive whetherthe stable cycling is directly linked to the 1D morphology or theshorter ion diffusion distance and enhanced stress tolerance.Ji’s group then studied the correlation between the opennanoporosity and the specific capacity of hard carbon anodes,where Bommier et al. found that increased surface area via CO2activation led to lower reversible capacity (Figure 2c,d).34Furthermore, porous carbon anodes typically exhibit very lowfirst-cycle Coulombic efficiency (FCCE),35,36 which is a seriousissue. Noticeably, the low capacity and poor FCCE is caused bythe more prominent formation of a solid electrolyte interphase(SEI) layer on the large surface area.37,38enhancement of carrier density in RGO and better contactbetween RGO layers by Na ion intercalation.2.2. Hard CarbonHard carbon, also known as non-graphitizable carbon, cannot begraphitized by thermal treatment.11 In 2000, Stevens and Dahndemonstrated that glucose-derived hard carbon exhibits adesodiation capacity of 300 mA·h/g.26,27 Inspired by theirpioneering work, there have been many reports on hard carbonanodes.28 30 Luo et al. studied the impact of morphology oncycling performance of hard carbon, where carbon nanofibersderived from cellulose exhibited a stable capacity of 176 mA·h/gat 200 mA/g over 600 cycles (Figure 2a,b).31 Similarperformance is also demonstrated in other one-dimensional233DOI: 10.1021/acs.accounts.5b00482Acc. Chem. Res. 2016, 49, 231 240
ArticleAccounts of Chemical ResearchFigure 3. (a) The structural evolution of Sn anode upon sodiation. Reproduced from ref 45. Copyright 2012 American Chemical Society. (b)Depositing Sn nanoparticles onto a 3D Ni-coated tobacco mosaic virus (TMV) as anodes for NIBs. Reproduced from ref 46. Copyright 2013 AmericanChemical Society. (c) Optimizing ternary Sn/Ge/Sb thin film alloys as anodes for NIBs. Reproduced from ref 49. Copyright 2014 American ChemicalSociety.mechanisms. Certainly, further studies are demanded to fullyunderstand the mechanisms of Na storage in hard carbon,particularly for the plateau region.The results related to surface area motivated us to minimizethe surface area of hard carbon anodes. Luo et al. employed GOas a 2D dehydration agent to prevent foaming duringcaramelization of sucrose, which extends the burnoff durationof sucrose caramel over a wider temperature range. 39Accordingly, the specific surface area of the resulting carbonreduces from 137.2 to 5.4 m2/g, and the FCCE is improved from74% to 83%. To increase FCCE, Hu’s group discovered thatpretreating cellulose fiber with 2,2,6,6-tetramethylpiperidine-1oxyl (TEMPO) before thermal carbonization can reduce surfacearea from 586 to 126 m2/g.40 This is made possible becauseTEMPO can unzip the cellulose fibers by oxidizing hydroxylgroups to carboxyl groups, thus loosening the hydrogen bond(Figure 2e). After the paper-making process, the flat structure ofribbon-like TEMPO-treated fiber results in a much denser paperand leads to the lower surface area after carbonization. FCCE isgreatly enhanced from 28% to 72%, and a stable cyclingperformance of 200 mA·h/g at 100 mA/g for 200 cycles isobtained.It has been widely believed that Na ion storage in hard carbonfollows the sequential intercalation into turbostratic nanodomains (TNs) and pore filling into the voids between TNs.26This model is known as the “card-house” model supported byexperimental evidence; however, some recent experimentalresults showed discrepancies. For example, Cao et al. proposedthat Na ion intercalation into the TNs of hollow carbonnanowires corresponds to the potential plateau at lowpotentials.30 Moreover, ex situ XRD by both Komaba et al. andMitlin et al. showed a reversible dilation and contraction of theTNs in the low voltage plateau.21,29,41 Recently, Bommier et al.suggested that the storage mechanism may be three tiered, wherethe sloping capacity was assigned to defect sites, supported by exsitu total neutron scattering/associated pair distribution function(PDF) studies (Figure 2f).42 From galvanostatic intermittenttitration technique (GITT), the authors observed a substantialincrease in diffusivity at voltages close to Na metal plating, whichbreaks the plateau region down to two possible storage3. ALLOY ANODESAlloy-type anodes are attractive for their high capacities in NIBs.However, the large volume change of alloy anodes uponelectrochemical cycling may cause electrode pulverization, lossof contact with the current collector, and consequent capacityfading.43 To tackle this problem, several strategies such as usingsmart substrates, fabricating nanostructures, and using specialbinder have been developed.3.1. TinBy forming Na15Sn4, Sn exhibits a high theoretical capacity of 847mA·h/g.44 In 2012, Huang et al. discovered that Sn initially formsthe NaxSn (x 0.5) phase, which converts to its final Na15Sn4structure.45 With coupling with in situ transmission electronmicroscopy (TEM) technology, they proved that NaxSn (x 0.5) and Na15Sn4 correspond to 60% and 420% volumetricexpansion (Figure 3a). To overcome the impact of volumechange, Wang et al. coated Sn onto three-dimensional (3D)current collectors with nickel nanofibers, which are later coveredby carbon and greatly improved cycling (Figure 3b).46Additionally, it has been suggested that optimizing binder orcomponents plays a critical role in Sn based anodes.47 49 Forexample, Sn50Ge25Sb25 exhibits high capacity (833 mA·h/g at 85mA/g), excellent rate capability (381 mA·h/g at 8500 mA/g),and stable cycling performance (662 mA·h/g after 50 cycles at 85mA/g) (Figure 3c).49For improving cycling performance of Sn anodes, Hu’s grouphas contributed two new strategies. The first involves ahierarchical wood fiber substrate, which yields a stable cyclingperformance over 400 cycles.50 With performing experiment andcontinuum chemomechanical modeling, we discovered that thesoft texture of wood fiber can effectively buffer the mechanicalstresses of Sn anode upon alloying/dealloying. In addition, theporous nature of the substrate functions as an electrolyte234DOI: 10.1021/acs.accounts.5b00482Acc. Chem. Res. 2016, 49, 231 240
ArticleAccounts of Chemical ResearchFigure 4. Wood fiber substrates are applied for Sn anode: (a) hierarchical structure of wood fiber; (b) structural wrinkling of wood fiber effectivelyreleases sodiation generated stresses; (c, d) chemomechanical modeling of the hoop stresses in Sn@wood fiber anode at the fully sodiated (c) anddesodiated (d) state. Reproduced from ref 50. Copyright 2013 American Chemical Society.Figure 5. ALD technology is employed for improving the cycling performance of Sn anodes: (a, b) comparison of bare Sn and ALD-Al2O3 coated Snnanoparticle on carbon nanofiber; (c) in situ TEM images of the first three cycles; (d) FEM modeling. Reproduced from ref 51. Copyright 2014American Chemical Society.Al2O3 coating remarkably boosts the cycling performance.51With in situ TEM, we unveiled the dynamic mechanicalprotection of the ALD-Al2O3 coating by coherently deformingreservoir that enables dual ion transportation through thesubstrate (Figure 4). Aside from substrate selection, surfacemodification of Sn anode with an atomic layer deposition (ALD)235DOI: 10.1021/acs.accounts.5b00482Acc. Chem. Res. 2016, 49, 231 240
ArticleAccounts of Chemical ResearchFigure 6. (a) Potential profiles of Sb/C composite. Reproduced from 52 with permission of The Royal Society of Chemistry. (b, c) Cycling performanceand TEM image of Sb nanofiber. Reproduced from ref 56. Copyright 2013 American Chemical Society. (d) Rate capability of Sb electrode with vaporground carbon fibers as the conductive additive and carboxymethyl cellulose binder. Adapted from ref 57. Copyright 2012 American Chemical Society.(e) Cycling performance and CE of Sb/CNT composites. Reproduced from ref 58 with permission of The Royal Society of Chemistry. (f) Onedimensional peapod-like Sb@C sub-micrometer structures. Reproduced from ref 60 with permission from the Royal Society of Chemistry.Figure 7. (a c) Ball-milling red P/carbon composite as an anode for NIBs. Reproduced with permission from ref 62. Copyright 2013 WILEY. (d, e)Phosphorene graphene anodes. Reprinted by permission from Macmillan Publishers Ltd: Nat. Nanotechnol. (ref 66), copyright 2015.236DOI: 10.1021/acs.accounts.5b00482Acc. Chem. Res. 2016, 49, 231 240
ArticleAccounts of Chemical ResearchFigure 8. (a) First-principle calculations illustrate the Na diffusion path along the [010] direction of TiO2 from Na1 to Na10 sites. Reprinted bypermission from Macmillan Publishers Ltd: Nat. Commun. (ref 75), copyright 2015. (b) Planar microscale battery for in situ AFM measurements onMoS2 electrode. Reproduced from ref 81. Copyright 2015 American Chemical Society. (c) Morphology of MoS2 nanoplates embedded in carbonnanofibers and (d) corresponding potential profiles. Reproduced from ref 82. Copyright 2014 WILEY. (e) Electrochemical reactions between Na ionsand PTCDA. Reproduced from ref 87. Copyright 2014 WILEY.3.3. Phosphoruswith Sn nanoparticle under the huge volume changes uponalloying/dealloying. Chemomechanical simulations clearlyshowed that bare Sn nanoparticles become disconnected fromthe underlying substrate upon charging. By contrast, the ALDAl2O3 coating acts as ion-conductive nanoglue and robustlyanchors the Sn nanoparticle anode to the substrate, therebyeffectively enhancing the cyclability (Figure 5).P exhibits a theoretical capacity of 2600 mA·h/g by formingNa3P. Lee et al. and Yang et al. reported P anodes in NIBs at thesame time.62,63 They discovered that mechanical ball-milling ofred P and carbon black at an optimized P/C ratio (7:3) leads toan amorphous P/C composite with great Na storage properties,including extremely high capacity ( 2000 mA·h/g), excellentrate capability, and stable cycling performance (Figure 7a c).Later, carbon nanotubes (CNTs)64 and graphene65 were alsointroduced into P electrodes, which showed comparableperformance to amorphous P/C composites. Most recently,Cui et al. demonstrated a black P/graphene nanostructure, wherethey fabricated a sandwich-structure comprising a fewphosphorene layers a
Na-Ion Battery Anodes: Materials and Electrochemistry Wei Luo,†,‡ Fei Shen,† Clement Bommier,§ Hongli Zhu,† Xiulei Ji,*,§ and Liangbing Hu*,† †Department of Materials Science and Engineering and ‡Department of Mechanical Engineering, University of Maryland, College Park, Maryland 20742, United States §Department of Chemistry, Oregon State University, Corvallis, Oregon 97331 .
polyatomic ions: a. amm onium ion b. sulfate ion c. sulfite ion d. carbonate ion e. nitrate ion f. permanganate ion g. hypochlorite ion h. phosphate ion i. cyanide ion j. hydroxide ion 9.2 Naming and Writing Formulas for Ionic Compounds A. _ Ionic Compounds 1. What are Binary Ionic Compounds? 2.
ION 7550 / ION 7650 User Guide The ION meter in an Enterprise Energy Management System Chapter 1 - Introduction Page 11 The ION meter in an Enterprise Energy Management System You can use ION 7550 and ION 7650 meters as standalone devices, but their extensive capabilities are full y realized when used with ION software as part of an
Table 3 Additional Ion kits used with the Ion 16S Metagenomics Kit Description Cat. no.[1] Quantity Ion Plus Fragment Library Kit 4471252 1 kit Ion Xpress Barcode Adapters 1-16 Kit[2] 4471250 1 kit Ion PGM Hi‑Q OT2 Kit A27739 1 kit Ion PGM Enrichment Beads 4478525 1 kit Ion PGM Hi‑Q Sequencing Kit (use with Ion PGM .
Ni-H. 2 (76 cells in series) Li-Ion (30 cells in series) BCDU Li-Ion . Adapter Plate . Data Cable . BIU . BCDU . Ni-H. 2. Battery A Battery B . Ni-H. 2. Cells Ni-H. 2. Cells . BCDU: Battery Charge / Discharge Unit . BIU: Battery Interface Unit . BSCCM: Battery Signal Conditioning and Control Module . Battery . BSCCM . BSCCM - Main Power Path .
Ni-H 2 (76 81 Ah cells in series) Li-Ion (30 134 Ah cells in series) BCDU Li-Ion Adapter Plate Data Cable BIU BCDU Ni-H 2 Battery A Battery B Ni-H 2 Cells Ni-H 2 Cells BCDU: Battery Charge / Discharge Unit BIU: Battery Interface Unit BSCCM: Battery Signal Conditioning and Control Module Battery BSCCM BSCCM -Main Power Path Commands & Data .
into carbonaceous materials and transition metal oxides used as anodes and cathodes, respectively. Graphite’s intrinsic layered structure has made this material a workhorse in the battery community for Li-ion battery anodes. In the past few decades, research on the layered properties of graphitic anode
User Management: System administrators can create/edit/delete users and define . ION 7550 1,2,3,4 YES YES YES ION 7600 1,2,3,4 YES YES YES Schneider Electric ION 7650 1,2,3,4 YES YES YES ION 7700 1,2,3,4 YES YES YES ION 7750 1,2,3,4 YES YES YES ION 8300 1,2,3,4 YES YES YES ION 8400 1,2,3,4 YES YES YES ION 8500 1,2,3,4 YES YES YES .
is produced to comply with ASTM C167-64, the "Standard Test Method for Thickness and Density of Blanket- or Batt-Type Thermal Insulating Material". Wool fiberglass insulation production lines usually consist of the following processes: (1) preparation of molten glass, (2) formation of fibers into a wool fiberglass mat, (3) curing the binder-coated fiberglass mat, (4) cooling the mat, and (5 .
