# Fileset

[Advanced Energy Materials - 2023 - Igarashi - New Template Synthesis of Anomalously Large Capacity Hard Carbon for Na‐ and.pdf](https://mdr.nims.go.jp/filesets/b4fc1aa3-2c9e-4257-bcf7-4b0f72e3067f/download)

## Creator

Daisuke Igarashi, Yoko Tanaka, [Kei Kubota](https://orcid.org/0000-0001-8941-3650), Ryoichi Tatara, Hayato Maejima, Tomooki Hosaka, Shinichi Komaba

## Rights

[Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/)

## Other metadata

[New Template Synthesis of Anomalously Large Capacity Hard Carbon for Na‐ and K‐Ion Batteries](https://mdr.nims.go.jp/datasets/51ce52b9-3c9e-4bb6-8795-e471a2a83458)

## Fulltext

New Template Synthesis of Anomalously Large Capacity Hard Carbon for Na‐ and K‐Ion BatteriesRESEARCH ARTICLEwww.advenergymat.deNew Template Synthesis of Anomalously Large CapacityHard Carbon for Na- and K-Ion BatteriesDaisuke Igarashi, Yoko Tanaka, Kei Kubota, Ryoichi Tatara, Hayato Maejima,Tomooki Hosaka, and Shinichi Komaba*Hard carbon (HC) is a promising negative-electrode material for Na-ionbatteries. HC electrochemically stores Na+ ions, resulting in anon-stoichiometric chemical composition depending on their nanoscalestructure, including the carbon framework, and interstitial pores. Therefore,optimizing these structures for Na storage by altering the synthesisconditions can enhance the capacity of Na-ion batteries. In this study, HCsusing MgO, ZnO, and CaCO3 as nanopore templates are systematicallyinvestigated, and the ZnO template is found to be particularly effective. Byoptimizing the concentration of ZnO embedded in the carbon matrix, utilizinga blend of zinc gluconate, and zinc acetate as starting materials, the optimalZnO-template HC demonstrates a reversible capacity of 464 mAh g−1(corresponding to NaC4.8) with high initial coulombic efficiency of 91.7% andlow average potential of 0.18 V versus Na+/Na. Thus, a Na-ion battery full cellconsisting of Na5/6Ni1/3Fe1/6Mn1/6Ti1/3O2 and the optimized ZnO-templateHC demonstrates a remarkable energy density of 312 Wh kg−1, comparable tothat of a Li-ion battery with LiFePO4 and graphite. Moreover, theZnO-template HC in a K half-cell also displays a significant capacity of381 mAh g−1, that is, KC5.8 where the alkali content is higher than stage-1graphite intercalation compounds, LiC6 and KC8.1. IntroductionNa-ion batteries (NIBs) and K-ion batteries (KIBs) are emergingas next-generation batteries that eschew the use of minor andD. Igarashi, Y. Tanaka, K. Kubota, R. Tatara, H. Maejima, T. Hosaka,S. KomabaDepartment of Applied ChemistryTokyo University of Science1-3 Kagurazaka, Shinjuku-ku, Tokyo 162-8601, JapanE-mail: komaba@rs.tus.ac.jpK. KubotaResearch Center for Energy and Environmental Materials (GREEN)National Institute for Materials Science (NIMS)1-1 Namiki, Tsukuba, Ibaraki 305-0044, JapanThe ORCID identification number(s) for the author(s) of this articlecan be found under https://doi.org/10.1002/aenm.202302647© 2023 The Authors. Advanced Energy Materials published byWiley-VCH GmbH. This is an open access article under the terms of theCreative Commons Attribution-NonCommercial-NoDerivs License,which permits use and distribution in any medium, provided the originalwork is properly cited, the use is non-commercial and no modificationsor adaptations are made.DOI: 10.1002/aenm.202302647expensive elements such as Li, Co, andCu. In recent years, these systems haveattracted considerable attention as sus-tainable energy conversion technologiesand future alternatives to Li-ion batter-ies (LIBs).[1,2] To develop practical NIBs,large-capacity electrode materials mustbe developed for achieving a perfor-mance comparable to that of LIBs. Thisis due to the inherent challenges of NIBs,such as relatively small capacity of alkali-containing positive electrode attributedto the three times heavier atomic weightof Na than Li and the lower voltagecaused by the Na+/Na potential being0.2–0.3 V higher than that of Li+/Li.[1]Hard carbon (HC) is essentially non-graphitizable and emerges as one of themost promising negative electrode ma-terials for NIBs.[3] Unlike LIBs, graphitecannot be used as a negative elec-trode in NIBs because of the thermo-dynamic instability of Na-graphite in-tercalation compounds (Na-GICs).[4] HCis a type of low-crystallinity carbonconsisting of two different nanosizeddomain structures. One is a randomly oriented pseudo-graphiticdomain composed of several stacked graphene sheets as a tur-bostratic structure with a variety of defects, and the other is in-ternal micropores existing between pseudo-graphitic domainsor graphene sheets as a nanosized interlayer space. The struc-ture of HC varies and depends significantly on the raw mate-rial and heat treatment temperature. Generally, HCs preparedat higher temperatures have higher crystallinity in their pseudo-graphitic domains, larger pores, and fewer defects such as car-bon defects and hetero-elements.[5–7] A unique feature of HC asa rechargeable battery material is its ability to store alkali met-als such as Li and Na through the electrochemical formationof a non-stoichiometric composition[8,9] depending on the car-bon structure. This is in contrast to graphite, where the max-imum alkali metal storage capacity is determined by the stoi-chiometry of graphite intercalation compounds (GIC), such asLiC6 and KC8.[10,11] The non-stoichiometric process of alkali stor-age in HCs has the potential to exceed the capacity of graphite,even in Li and K cells. Since the 1990s, HC has been utilized asa negative electrode in the earliest LIBs,[12] and its application inKIBs continues to be a subject of ongoing research.[2]Adv. Energy Mater. 2023, 13, 2302647 2302647 (1 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbHhttp://crossmark.crossref.org/dialog/?doi=10.1002%2Faenm.202302647&domain=pdf&date_stamp=2023-11-09www.advancedsciencenews.com www.advenergymat.deGiven that the structural characteristics of carbon materials areassociated with their alkali-metal storage properties, new synthe-sis methods have been devised to control the structure of HCsand maximize their capacity as battery materials. For example,hetero-elements doping,[13,14] closing open pores on activated car-bon, [15,16] bulk etching,[17] and template methods[15] have beenreported after the simple strategy of changing the raw materialsand heat treatment temperatures.[6] Although the detailed mech-anism of Na insertion into HC is still under debate, recent stud-ies, including various characterizations, and theoretical calcula-tions of Na-inserted HC, have suggested the following sequenceof three steps by sweeping from higher to lower potentials: 1)Na adsorption at various defect sites, 2) Na intercalation into theinterlayers of pseudo-graphitic domains, and 3) the formation ofquasi-metallic Na clusters within internal pores. In the final step,Na cluster formation in nanopores by applying a lower potentialhas been established by multiple groups through computationalstudies,[18–20] and experimental characterizations, including X-ray scattering methods,[3,21–24] solid state Na-NMR for chemicalstate analysis,[24–26] and other analyses, such as spectroscopy[27]and dilatometry.[7]Based on these findings, the most intuitive approach for en-hancing HC capacity is to design a pore structure optimized forNa storage. In line with this strategy, our group reported thesynthesis of a HC with a reversible capacity of 478 mAh g−1[15]using the MgO template method. In this process, MgO formedby the pyrolysis of Mg gluconate (Mg Glu) serves as a templatefor the formation of closed pores in the HC.[15] Although MgOis a well-known template material for controlling the structureof carbon,[28] other inorganic compounds such as zeolites,[29]silica,[30] ZnO,[31] and CaCO3[32] have also been used to preparemesoporous carbons.In this study, we propose a new template carbon derived fromgluconate salts of Zn and Ca, hereafter denoted as “Zn Glu” and“Ca Glu,” respectively, to shed light on the potential of new tem-plate inorganics to synthesize HCs for NIBs and KIBs. The diva-lent nature of these three gluconate salts[33] serves as the startingmaterial for synthesizing large-capacity HC materials, employ-ing the same approach as that for the MgO-template carbon. Fur-thermore, we investigated the electrochemical properties of thetemplate carbons as sodium and potassium insertion materials,considering that HCs with high Na storage capabilities often ex-hibit superior K storage properties, as per previous studies.2. Results and Discussion2.1. Synthesis and Characterization of PrecursorsFigure 1a shows a schematic illustration of the HC synthesis pro-cess employed in this study, which is based on the concept of tem-plate synthesis methods from our recent study.[15] The chemicalstructures of the gluconate salts and starting materials are shownin Figure S1a (Supporting Information), and their powder mor-phologies observed by scanning electron microscopy (SEM) areshown in Figure S1b–d (Supporting Information). Gluconatesare widely used as precursors for template carbons.[28] During thepreheating process at 600 °C, the starting material was pyrolyzedand transformed into a composite of precursor carbon and inor-ganic particles, which served as the nanopore templates. In thesubsequent acid leaching phase, excess inorganic particles on thecarbon surface were eliminated, leaving open pores in their place.Through the final post-heating process in an inert atmosphere,the precursor carbon becomes HC, developing graphitic domainsand closed nanopores.[34] Template oxides embedded in the car-bon matrix of the precursor undergo reduction via a carbother-mal reaction, forming elemental metals that are volatilized andremoved at high temperatures.[35,36] The detailed sample prepara-tion method is described in Supporting Information. Hereafter,the sample names of precursor carbons, washed precursor car-bon via acid-leaching, and HCs as final products are referred toas “PRE-X,” “wPRE-X,” and “HC-X,” respectively, where X repre-sents the metal element in the starting material.SEM images of both PRE-X and wPRE-X (X = Mg, Zn, orCa) are shown in Figure S2 (Supporting Information). Notably,the morphology of PRE-X did not replicate that of the respectivestarting materials. Figure 1b shows a comparison of the X-raydiffraction (XRD) patterns of PRE-X and wPRE-X. The diffrac-tion peaks of MgO, ZnO, CaO, and CaCO3 are identified in thePRE-X pattern. In the wPRE-X pattern, however, the peak inten-sities of these inorganic materials noticeably diminished, and thebroad peaks originating from low-crystalline carbon located at 2𝜃= 22 ° and 43 ° became relatively pronounced after acid leaching.This suggests that a significant proportion of the inorganic parti-cles that served as templates was removed by acid leaching, whichis consistent with the SEM-energy dispersive X-ray spectroscopy(EDS) analysis shown in Figure S3 (Supporting Information).As reported in our earlier study on MgO-template carbon,the MgO in (w)PRE-Mg consists of small nanoparticles.[15] Thiswas evident from the broad MgO peaks in the XRD patterns,and the MgO nanoparticles were distinctly visible in the dark-field transmission electron microscopy (TEM) image of PRE-Mg(Figure 1c).By contrast, sharper ZnO-derived diffraction peaks are ob-served in the pattern of PRE-Zn, indicating the presence of highlycrystalline ZnO with a larger particle size. However, the wPRE-Zn pattern shows broad ZnO peaks, suggesting that highly crys-talline ZnO is primarily located on the carbon surface and canbe removed by acid leaching. In fact, the SEM-EDS mapping inFigure S3 (Supporting Information) and the high-magnificationSEM images of the PRE-Zn and wPRE-Zn samples in FigureS4 (Supporting Information) show white ZnO particles on PRE-Zn, whereas no such particles are visible in the SEM image ofwPRE-Zn. Figure S5 (Supporting Information) presents a de-tailed analysis of the ZnO-derived diffraction peaks of PRE-Znand wPRE-Zn. The 100, 002, and 101 diffraction peaks within2𝜃 = 30–40° in the patterns of PRE-Zn can be deconvoluted intosharp and broad components, whereas those of wPRE-Zn are fit-ted only with broad components. The ZnO crystallite size cal-culated through Scherrer’s equation with the full width at halfmaximum (FWHM) of each 002 diffraction are 39.3 nm for thesharp component of PRE-Zn and 3.4 and 4.1 nm for the broadcomponent of PRE-Zn and wPRE-Zn, respectively. The presenceof ZnO nanoparticles in (w)PRE-Zn, as predicted from the XRDanalysis, aligns with the dark-field TEM image of PRE-Zn shownin Figure 1d.In the case of PRE-Ca, crystalline phases of both CaO andCaCO3 were detected using XRD. Based on the literature, CaCO3is preferentially formed, and CaO is produced by the thermalAdv. Energy Mater. 2023, 13, 2302647 2302647 (2 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewww.advancedsciencenews.com www.advenergymat.deFigure 1. a) Illustration of template synthesis method. b) XRD patterns of PRE-X and wPRE-X, where X = Mg (left), Zn (middle), or Ca (right). Dark-fieldTEM images of c) PRE-Mg, and d) PRE-Zn, in which MgO and ZnO are indicated as bright spots. e) Bright-field TEM images of PRE-Ca.decomposition of CaCO3 under high-temperature conditions.[37]According to the TEM image in Figure 1e, the particle size of CaOor CaCO3 in PRE-Ca is several tens of nanometers, which is con-siderably larger than that of MgO in PRE-Mg and ZnO in PRE-Znand cannot serve as a “template” for the nanopores of HC.As shown in the thermogravimetry-differential thermal analy-sis (TG-DTA) curves in Figure S6 (Supporting Information), thepyrolysis behaviors of Mg Glu, Zn Glu, and Ca Glu are differentin terms of their exothermic and endothermic properties, as wellas weight loss temperatures. Although the pyrolysis process oforganic compounds is complex, making the decomposition pro-cess of each starting material difficult to discern in detail, thesedifferences in pyrolysis behavior would affect the distribution andparticle (crystallite) size of the inorganic particles, depending onthe type of divalent metal involved.2.2. Characterization of HCsSEM images of the HCs synthesized from the three different glu-conates are shown in Figure 2a. The morphology of each HC-X retained features of the corresponding wPRE-X (Figure S3,Supporting Information). The XRD patterns of HC-Mg, HC-Zn,and HC-Ca are shown in Figure 2b. All samples representedbroad peaks typical for low-crystalline carbons at 2𝜃 = 21–23°and 42–44°, which can be assigned to 002 and 100 diffractions,respectively, of pseudo-graphitic domains in HCs. The calcu-lated average interlayer distances, d002, are 0.3767, 0.3766, and0.3564 nm for HC-Mg, HC-Zn, and HC-Ca, respectively. Nopeaks other than those of HC were observed in the XRD pat-terns, indicating that most of the templates were removed dur-ing the post-heat treatment process. The removal of the tem-plates was also confirmed by STEM-EDS, as shown in FigureS7 (Supporting Information). Mg and Zn were barely detectablein HC-Mg and HC-Zn, while a small amount of Ca was de-tected in HC-Ca. Elemental mapping revealed the coexistence ofdense Ca and O spots in the carbon matrix. Because the boilingpoint of metallic Ca is >1400 °C, it is presumed that it did notvolatilize during the post-heat treatment and remained as CaO orCaCO3.The pore structures of the HCs were characterized usingsmall-angle X-ray scattering (SAXS), as shown in Figure 2c. TheAdv. Energy Mater. 2023, 13, 2302647 2302647 (3 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewww.advancedsciencenews.com www.advenergymat.deFigure 2. a) SEM images of HC-Mg (left), HC-Zn (middle), and HC-Ca (right). b) XRD patterns of HCs. c) SAXS patterns of HCs and graphite. d) HRTEMimages of HC-Mg (left), HC-Zn (middle), and HC-Ca (right).SAXS pattern of natural graphite is also displayed for compari-son, as graphite does not contain pores. Notably, shoulder scat-tering peaks were clearly observed in the SAXS patterns of HC-Mg and HC-Zn. By pattern fitting and assumptions,[38] we calcu-lated average pore sizes of 1.25 and 1.44 nm for the HC-Mg andHC-Zn samples, respectively. By contrast, the pattern for HC-Cashows a shoulder peak with a very low intensity, which indicatesthat HC-Ca has a smaller or negligible number of closed poresthan HC-Mg and HC-Zn. Based on our previous observations,the nanopore size of HCs obtained by simple carbonization oforganic precursors at 1300–1500 °C is 1.1–1.2 nm,[6,39,40] there-fore, HC-Mg and HC-Zn possess relatively larger sized nanoporestructure. From these results, it can be concluded that the MgOand ZnO nanoparticles contained in PRE-Mg and PRE-Zn act astemplates, enabling the formation of nanopores in HC-Mg andHC-Zn.Adv. Energy Mater. 2023, 13, 2302647 2302647 (4 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewww.advancedsciencenews.com www.advenergymat.deThe pore structures of the HCs were further character-ized by microstructural observations using high-resolution TEM(HRTEM), as shown in Figure 2d. The HRTEM images of PRE-Mg and PRE-Zn show dense microstructures on the scale of a fewnanometers, whereas larger voids (5–10 nm in diameter) were ob-served in HC-Ca. Although the CaO or CaCO3 particles formed inPRE-Ca may lead to the formation of these large voids, such largespaces are considered unsuitable for the electrochemical forma-tion of Na clusters.[23,24]N2-sorption measurement of the HCs were conducted, and theobserved isotherms of the HCs are displayed in Figure S8 (Sup-porting Information). A typical type-I isotherm was obtained forHC-Mg, whereas the isotherm for HC-Zn appeared to be type-II, but with a very small amount of N2 adsorption. According tothe IUPAC Technical Report, HC-Mg should have open microp-ores on its surface, while HC-Zn is a nonporous material or hasonly macropores.[41] The specific surface areas calculated basedon Brunauer–Emmett–Teller (BET) theory (denoted as “SBET”)were 582 and 85 m2 g−1 for HC-Mg and HC-Zn, respectively. Theisotherm of HC-Ca shows a hysteresis loop from ambient pres-sure to the middle range of the relative pressure, which impliesthe presence of mesopores on the surface of HC-Ca. AlthoughBET analysis is not entirely suitable for such materials, the ten-tative SBET of HC-Ca, calculated from the adsorption curve, was574 m2 g−1. SBET and the distribution of open pores calculatedfrom gas sorption such as N2 and CO2 are often used to evaluatethe “surface” structure of carbons, expecting that such insightinto “surface” could be associated with passivation of electrodesurface of solid electrolyte interphase (SEI) formation by cover-ing with electrolytic decomposition products on the particles.[3]Considering that the solvent molecules and anions present in theelectrolyte are larger than the N2 or CO2 molecules, the surfacearea determined by gas sorption does not necessarily representthe contact area between the liquid electrolyte and the electrodeactive material.[16,42] Therefore, we performed cyclic voltammetry(CV) on the HC electrodes to evaluate the contact areas in the po-tential range of 2.2–2.8 V versus Na+/Na. This specific range wasselected because of the absence of faradaic electrochemical reac-tions, while allowing for the formation of electric double layers atthe electrode/electrolyte interface. The cyclic voltammograms ob-tained are shown in Figure S9 (Supporting Information). All thevoltammograms show rectangular curves, which are typical char-acteristics of capacitors. The calculated average electric-doublelayer capacities in this potential range are 0.15 mAh g−1 for HC-Mg, 0.12 mAh g−1 for HC-Zn, and 0.97 mAh g−1 for HC-Ca(corresponding to 0.90, 0.72, and 5.8 F g−1, respectively). Thesedouble-layer capacitance values are assumed to be linearly pro-portional to the contact area between the HC particles and theelectrolyte solution. Therefore, we believe that the irreversible ca-pacity of SEI formation on the HC electrodes in the Na cell isrelated to the capacitance, as described below.2.3. Electrochemical Performance of HCs as NIB NegativeElectrodeCharge–discharge tests (corresponding to the insertion and ex-traction of Na, respectively, at the HC electrode) were performedto evaluate the performance of the HC electrodes for NIBs.Figure 3a shows the initial charge–discharge curves of the HCelectrodes in the Na half-cells. The HC composite electrodesshow high reversible capacity of > 200 mAh g−1 with a voltagevariation curve totally resembling those of other HC reportedpreviously.[3] Thus, the redox activity of the template-synthesizedHC electrodes is attributed to reversible sodium insertion.Among the three samples, HC-Zn exhibited a superior bat-tery performance, demonstrating an initial reversible capacityof 418 mAh g−1 and initial efficiency of 90%. HC-Mg also ex-hibits a high initial capacity and efficiency of 383 mAh g−1 and85%, respectively; however, HC-Ca delivers a lower capacity of258 mAh g−1 and an efficiency of 56% in the first cycle. Thedifference in reversible capacity can be explained by the charac-teristics of the closed pores evaluated by SAXS, that is, HC-Znand HC-Mg show superior sodium storage performance com-pared to HC-Ca owing to their carbon structures with larger-sizedclosed pores and a higher proportion of closed pores. This isthe first study to demonstrate that Zn-Glu-derived carbonaceousmaterials have remarkable properties as negative electrodes forrechargeable NIBs. This is notable because these materials havebeen previously studied in the context of other energy materi-als such as electrodes for double-layer capacitors,[43] and oxygen-reduction reaction catalysts.[44]The capacities corresponding to the higher-potential slope re-gion (>0.15 V) were 80, 89, and 81 mAh g−1, and those ofthe lower-potential plateau region (<0.15 V) were 303, 329, and177 mAh g−1 for HC-Mg, HC-Zn, and HC-Ca, respectively. Thecapacities and potential variations of the slope region were almostsimilar for each sample, ≈80 mAh g−1, whereas the capacities ofthe plateau region differed among the three samples. This resultaligns well with the material design; the pore structure designenhanced the formation of Na clusters and extended the plateaucapacity at lower potentials. Other studies have also suggestedthat the nanopore structure of HC has a significant effect on thesodiation capacity in the plateau region.[7,23,45] The variation inthe initial Coulombic efficiency of the HCs, related to irreversibleelectrolyte decomposition and SEI formation at the initial cycle,aligns more closely with the capacitances obtained from the CVdiscussed above rather than with the BET surface areas calculatedfrom N2 sorption, as shown previously. The smooth particle sur-face of Zn-HC observed by SEM (Figure 2a) likely contributed toreducing the contact area with the electrolyte, leading to a lowerirreversible capacity and higher initial efficiency.Figure 3b shows the cycle performance and correspondingcoulombic efficiency of HC-Zn. Except for the first cycle, whereelectrolyte decomposition and SEI formation occurred in theearly stages of sodiation, stable reversible cycling, and high ef-ficiencies > 99% were obtained second cycle onward. These re-sults demonstrate that the large reversible capacity does not in-clude any plating/stripping of Na metal, which typically resultsin very low Coulombic efficiency in common electrolytes[46] butoriginates from reversible Na insertion into HC.To confirm that the large capacity of HC-Zn originated fromNa cluster formation in the nanopores, characterizations of so-diated HC electrodes were performed. Figure 3c shows the exsitu SAXS patterns of the pristine, fully sodiated, and fully des-odiated HC electrodes. The intensity of the shoulder peak ob-served in the pristine sample decreased with sodiation and thenincreased again with desodiation, reverting to an intensity thatAdv. Energy Mater. 2023, 13, 2302647 2302647 (5 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewww.advancedsciencenews.com www.advenergymat.deFigure 3. a) Initial charge–discharge curves of HCs. b) Cycle performance of HC-Zn. c) ex situ SAXS patterns of HC-Zn at different sodiation states.d) ex situ WAXS patterns of HC-Zn at different sodiation states.closely aligned with the pristine state. The decreased peak in-tensity observed in the sodiated sample can be attributed tothe reduced electron density contrast between the carbon ma-trix and pores, resulting from the insertion of Na atoms fillingthe nanopores.[3,21,22] The filling of the Na cluster is more clearlydemonstrated in the ex situ wide-angle X-ray scattering (WAXS)patterns shown in Figure 3d, alongside the simulated scatteringpattern of the Na metal crystal. The broad peaks located ≈q =1.5, 3, and 5.2 Å−1 in the pattern of pristine sample correspondto the 002, 100, and 110 diffractions of graphite-like structure inHC. Along with peaks corresponding to the host carbon struc-ture, additional scattering peaks were observed for the sodiatedHC, which can be attributed to metallic Na crystals. These SAXSand WAXS results are consistent with the WAXS data reportedby Morikawa et al.[23] and the X-ray total scattering observed byStratford et al.,[24] confirming that the large capacity of HC-Znis indeed due to the formation of sodium clusters within thenanopores. The completely reversible change in the WAXS andSAXS patterns during sodium insertion and extraction provesthat the electrochemical formation and removal of Na clusterswithin these nanopores is a highly reversible process. BecauseHC-Zn showed superior performance in the Na cells, we furtheroptimized HC-Zn by diluting and enriching the amount of ZnOin the carbon matrix.2.4. Dilution and Enrichment of ZnO as Nanopore TemplateThe size and distribution of the ZnO templates are expected tobe critical factors for enhancing Na storage capacity. Thus, we in-vestigated the effect of template dilution and enrichment on thecapacity of the ZnO-template HCs by introducing glucose (Glc)as an additional carbon source and zinc acetate (Zn(OAc)2) as anadditional template source for Zn Glu. According to the litera-ture, only ZnO remains after the pyrolysis of Zn(OAc)2 even inan inert gas atmosphere,[47] and the C-atoms in acetate ions areremoved as CO2 or acetone.[48] Therefore, the amount of ZnO inthe preheated precursors derived from the mixtures of Zn Gluand Zn(OAc)2 was reasonably expected to be greater than that inPRE-Zn.The starting materials were prepared by homogeneously mix-ing Glc, Zn, Glu, and Zn(OAc)2 via freeze-drying (see Support-ing Information for details). The samples are designated as PRE-Zn[x-y-z], wPRE- Zn[x-y-z], or HC-Zn[x-y-z] where the mixingmolar ratio of Glc, Zn Glu, and Zn(OAc)2 is expressed as x :y :z,which is appended after “Zn. ” The freeze-drying process was em-ployed for pure Zn Glu for comparison; thus, the sample prepa-ration conditions of HC-Zn[0-100-0] differed slightly from thoseof HC-Zn, even though both originated from pure Zn Glu.The XRD patterns of the unwashed and washed precursors,as well as the HCs, are shown in Figure S10 (Supporting Infor-mation). The synthesis process of the HCs appeared to be con-sistent as there were no significant differences in the patterns,except for variations in the intensity and breadth of the ZnOdiffraction peaks. The crystallite sizes of ZnO in PRE-Zn[x-y-z]and wPRE-Zn[x-y-z], calculated in the same manner as shown inFigure S5 (Supporting Information), are summarized in Table S1(Supporting Information). The crystallite sizes of ZnO calculatedfrom Scherrer’s equation with broad diffraction peaks derivedfrom ZnO nanoparticles, which are available as templates fornanopores, showed no substantial variation with different mix-ing ratios of raw materials. Even with an increasing amount oftemplate, the Zn residue was hardly detected in the final product,HC, because ZnO was reduced, and the resultant Zn metal wasvolatilized during the post-heat treatment, as evidenced by theSTEM-EDS mapping of HC[0-75-25] in Figure S11 (SupportingAdv. Energy Mater. 2023, 13, 2302647 2302647 (6 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewww.advancedsciencenews.com www.advenergymat.deFigure 4. a) Initial charge–discharge curves of HC-Zn[x-y-z] in Na half-cell. b) Cycle performance of HC-Zn[0-75-25] in Na half-cell. c) Relationshipbetween mixing ratio of the raw materials and the specific capacity of corresponding HCs. d) SAXS patterns of HC-Zn[x-y-z]. e) Charge–discharge curvesof Na5/6Ni1/3Fe1/6Mn1/6Ti1/3O2//HC-Zn[0-75-25] full cell. f) Comparison of gravimetric energy densities among NIB in this study, NIB reported in 2011in ref. [3] and LIB consisting of LiFePO4 and graphite.Information). Although a sample derived from a mixture of Glc,Zn Glu, and Zn(OAc)2 in the ratio of 100:0:0 contains no Zn atall, the sample is formally denoted as “HC-Zn[100-0-0]” for sim-plicity. The XRD patterns of PRE-Zn[100-0-0] and HC-Zn[100-0-0], shown in Figure S12 (Supporting Information), confirm theabsence of crystalline ZnO in the sample.Figure 4a shows the initial charge and discharge curves of HCssynthesized from a freeze-dried mixture with different ratios ofGlc, Zn Glu, and Zn(OAc)2. In Figure 4a, the dilution of the tem-plate by the addition of Glc to Zn Glu resulted in a decrease in thecapacity. On the contrary, enrichment of the template by adding25% Zn(OAc)2 effectively enhanced a reversible capacity up toAdv. Energy Mater. 2023, 13, 2302647 2302647 (7 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewww.advancedsciencenews.com www.advenergymat.de464 mAh g−1 (corresponding to NaC4.8 formation by assuming noirreversible Na insertion) and high initial Coulombic efficiencyof 91.7% for HC-Zn[0-75-25]. Although the sodiation potential ofHC-Zn[0-75-25] is close to the Na plating potential, it was compa-rable to that of HC-Zn and the previously reported MgO templateHC, as shown in Figure S13 (Supporting Information).Figure 4b reveals that HC-Zn[0-75-25] electrode exhibits goodlong-term cycling stability over 200 cycles, and its capacity re-tention after 200 cycles was 93% of the initial capacity. The cy-cle performance was not compromised, even with a 48 mAh g−1increase in capacity compared to that of HC-Zn (418 mAh g−1),as shown in Figure 3. Although several HCs with enhancedcapacities of ≈500 mAh g−1 have been reported in recentyears,[13,14,16,17,49] there are no other reports of HCs that possessall of sufficient cycle stability, high initial efficiency (>90%), andminimal potential hysteresis leading to high energy efficiency,aside from MgO-template HCs.[15] Since a high initial Coulom-bic efficiency is an essential requirement for practical full-celloperation,[50] the template synthesis method is believed to be veryeffective in increasing the energy density of NIB full cells.In Figure 4c, the total initial capacity, capacity of the plateauregion (<0.15 V), and that of the slope region (>0.15 V) of theZnO-template HCs are plotted against the composition of theraw materials. The slope capacity of the HCs was nearly con-stant and independent of the composition of the raw materials,and the variation in the total capacity was primarily dominatedby that of the plateau region. In general, the plateau capacitiesof HCs strongly depend on their closed-pore structure. There-fore, a high reversible capacity could be achieved by varying theamount and distribution of the ZnO template. Actually, we foundthat HC-Zn[0-75-25], with the minor addition of Zn(OAc)2 to ZnGlu, demonstrated the maximum capacity, whereas HC-Zn[0-50-50], with an increased Zn(OAc)2 quantity of 50%, presented only341 mAh g−1 as seen in Figure 4c. This result indicates that thereis an optimum amount of ZnO as a template for nanopores. Ourprevious work showed that a MgO-template HC derived from a50:50 mixture of Mg Glu, and Glc exhibited a maximum capac-ity of 478 mAh g−1. The difference in optimum values for theMgO and ZnO template amounts can be attributed to differencesin the crystallite size, crystallinity, and grain size distribution ofMgO and ZnO embedded in the carbon precursors, as discussedabove.The relationship between the mixing ratio of the raw materi-als and the Na storage performance can be elucidated from thestructural characteristics of the HCs. The pore structures of theHCs were characterized using SAXS, as shown in Figure 4d. Theshoulder peak profiles varied significantly depending on the mix-ing ratio of the raw materials. In particular, HC-Zn[0-50-50] pos-sessed an unusual pattern with a notably larger peak than theother samples in the region q = 0.02–0.03 Å−1. According to theXRD results in Figure S10 (Supporting Information), the broaddiffraction peaks of ZnO were barely noticeable in the patterns ofPRE-Zn [0-50-50] and wPRE-Zn[0-50-50], suggesting that the for-mation of fine ZnO template nanoparticles was limited in thesesamples. Excessive addition of Zn(OAc)2 likely led to the over-growth of ZnO crystals, thereby reducing the fraction of fine ZnOnanoparticles suitable as templates for nanopores.The HCs derived from the mixtures containing Glc also ex-hibited SAXS patterns that could not be satisfactorily fitted to asingle peak. Our analysis requires the assumption of two shoul-der peaks to fit the patterns of HC-Zn[75-25-0], HC-Zn, [50-50-0],and HC-Zn[25-75-0] (see Figure S14a—c, Supporting Informa-tion). However, for Glc-free HC-Zn[0-100-0] and HC-Zn[0-75-25],a single peak was observed to fit the experimental data (FigureS14d,e, Supporting Information). The structural parameters ofHC-Zn[x-y-z], such as the interlayer distance d002, crystallite sizeof the pseudo-graphitic domain Lc, stacking number calculatedfrom the XRD data, and pore size DSAXS are summarized inTable S2 (Supporting Information). According to our experienceand recent studies on HC, it is still challenging to determine astraightforward relationship between these parameters and Nastorage capacity because of the difficulty in distinguishing be-tween extended interlayers in the pseudo-graphitic domains andnanopores along with a highly disordered framework, and thusthese parameters roughly correlate with each other.[24] However,as a general trend, the large-capacity HCs in this study have thefollowing features: 1) they possess a large d002 and small Lc, andtherefore a small number of stacking layers n; 2) their size dis-tribution, as observed by SAXS, is highly uniform, and the aver-age pore size is relatively large. This trend was also supported bythe focused ion beam processed (FIB)-STEM observations of HC-Zn[25-75-0] and HC-Zn[0-75-25], as shown in Figure S15 (Sup-porting Information). The high-magnification STEM image ofHC-Zn[25-75-0] confirms a heterogeneous structure with a void-like structure of ≈5 nm, whereas the HC-Zn[0-75-25] sample hasa dense structure with high uniformity. The 5 nm scale pore-like structure observed in the STEM image of HC-Zn[25-75-0]is consistent with the pore diameter of 4.34 nm calculated fromthe low-q peak in the SAXS pattern of HC-Zn[25-75-0] (see TableS2, Supporting Information). Therefore, we conclude that opti-mizing the mixing ratio of the raw materials and achieving auniform distribution of nanopores with a suitable size for Naclustering maximizes the capacity of the ZnO-template carbons.This consideration is in agreement with the recent discussion byStratford et al., describing the pore size distribution is more im-portant than the average pore size in controlling the Na storagecapacity.[24]A Na-ion full cell consisting of HC-Zn[0-75-25], whichexhibits a superior reversible capacity as a negative elec-trode material, and a Na-containing layered transition oxideNa5/6Ni1/3Fe1/6Mn1/6Ti1/3O2 as a positive electrode material[51]were fabricated with 1 mol dm−3 Na(PF6)0.8(FSA)0.2/EC:PC (1:1v/v) solution with the addition of 0.2 vol% vinylene carbonate.Although the reversible capacity of HC-Zn[0-75-25] is slightlysmaller than that of HC-Mg, [50-50-0] the optimized MgO-template HC in ref., [15] the simulated discharge curves of thefull cells are almost identical as shown in Figure S16 (SupportingInformation), partly because the former exhibits a better initialcoulombic efficiency that the latter. The charge–discharge curvesof the NIB are shown in Figure 4e, and their capacity retentionover 100 cycles is shown in Figure S17a (Supporting Informa-tion). The specific capacity of 113 mAh g−1 based on positive elec-trode mass (corresponding to 373 mAh g−1 based on negativeelectrode mass) was demonstrated at the initial cycle, and satis-factory cycle performance was achieved over the subsequent 100cycles. The discharge rate capability of the NIB full cell is shownin Figure S17b (Supporting Information). The capacity and cellpolarization are hardly changed and deteriorated by increasingAdv. Energy Mater. 2023, 13, 2302647 2302647 (8 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewww.advancedsciencenews.com www.advenergymat.deFigure 5. a) Initial charge–discharge curves of HC-Zn[x-y-z] in K half-cell. b) Cycle performance of HC-Zn[0-75-25] in K half-cell. c) ex situ SAXS patternsof HC-Zn in different potassiation states. d) Charge–discharge curves of K2Mn[Fe(CN)6]//HC-Zn[0-75-25] full cell.current rates from 0.1 to 1 C. Even at a high rate of 10 C, the cellretained ≈76% of its capacity compared to that at 0.1 C, demon-strating the robust rate capability of the full cell.The discharge curves and gravimetric energy densities of theNIB cells fabricated in this study, those from a 2011 study,[3]and those of a conventional LIB consisting of LiFePO4 (LFP)and graphite were experimentally compared under compara-ble test conditions, as shown in Figure 4f. The energy den-sity of NIB fabricated in this study reached 312 Wh kg−1.This value is >1.5 times larger than that of the O3-type lay-ered NaNi0.5Mn0.5O2//Kureha’s HC Na-ion cell reported in 2011,which was one of the earliest demonstrations of NIBs free fromNa-metal electrodes, and it is comparable to the energy densityof widely commercialized LFP-type LIB. Although there remainchallenges, such as the longer cycle life of a full cell and Ni-freehigh-capacity positive electrode, we succeeded in developing alarge-capacity and high-initial coulombic efficiency HC-Zn[0-75-25] material, which enables NIB to have a comparable or higherenergy density than LIBs.2.5. Application to KIB Negative ElectrodesThe ZnO-template HCs were further applied as negative elec-trodes in KIBs. The charge and discharge curves (correspond-ing to potassium insertion and extraction, respectively) of the Khalf-cells are shown in Figure 5a. Because the potential variationand reversibility were similar to those of potassium insertion intoHC reported previously,[6,39,40] the redox activity was caused byreversible electrochemical potassium insertion. For these ZnO-template HCs, the specific capacity of the K cells followed a trendconsistent with that of the Na cells. For instance, HC-Zn[0-75-25],which exhibits the largest capacity in a Na half-cell, also demon-strated a large capacity of 381 mAh g−1 in a K half-cell. This re-versible capacity corresponds to KC5.8 formation by assuming allcharges passing through the electrode are consumed for K inser-tion. The K concentration was higher than that of stage-1 KC8 andthe Li concentration of stage-1 LiC6. The trend of high Na storagecapacity in HCs correlating with high K storage capacity is com-monly observed in conventional HCs derived from cellulose,[39]sucrose,[6] and phenolic resins.[40] The high reversibility of HC-Zn[0-75-25] over 50 cycles (Figure 5b) can be attributed to thereversible K insertion into the carbon structure rather than theplating/stripping of K metal.Given that several samples delivered large reversible capaci-ties, far exceeding the theoretical capacity of graphite in K cells(279 mAh g−1 for KC8 chemical composition),[52] it is reasonableto conclude that K storage sites can accommodate K atoms at sig-nificantly higher concentrations than the KC8 in-plane arrange-ment typically found in K-GICs. A recent study suggested the pos-sibility of the formation of quasi-metallic K clusters in the closedpores of HC materials.[53,54] It has been suggested that the size ofthe closed pores suitable for alkali metal clustering depends onthe alkali metal species, and Na clustering requires a larger poresize than that of Li.[26] In fact, the HCs developed in this study hadsmaller capacities in the Li half-cells than in the Na and K half-cells, as shown in Figure S18 (Supporting Information). To un-derstand the mechanism of K-insertion, ex situ SAXS measure-ments were performed. Figure 5c displays the ex situ SAXS pat-terns of pristine, potassiated, and depotassiated HC-Zn (see theinitial charge–discharge curves in Figure S19, Supporting Infor-mation). The identical SAXS patterns of the pristine and depotas-siated samples indicate that K insertion is completely reversible.Contrastingly, the pattern of the potassiated sample shows aAdv. Energy Mater. 2023, 13, 2302647 2302647 (9 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewww.advancedsciencenews.com www.advenergymat.dedecrease in the scattering intensity over a wide q range. Simi-lar to the pattern of the sodiated sample (Figure 3c), the decreasein the shoulder peak intensity indicates a reduction in the elec-tron density contrast between the nanopores and carbon matrix.Therefore, we anticipate the formation of K clusters in HC in thelow-potential region based on the correlation between the Na andK-storage capacities in various HCs[6,39,40] as well as the notable K-storage capacity of the ZnO-template HC, which clearly exceedsthe stoichiometric KC8 composition. A comparison of the energydensities of hypothetical KIB full cells with graphite or HC-Zn[0-75-25] as negative electrodes is shown in Figure S20 (SupportingInformation). Because the average depotassiation potentials ofgraphite and HC-Zn[0-75-25] are 0.37 and 0.41 V versus K+/K,respectively, HC-Zn[0-75-25] has a disadvantageous working po-tential. However, the energy density of the hypothetical full cellper weight of the negative electrode is much higher for HC-Zn[0-75-25] than for graphite owing to its anomalously large capacity.Given that ZnO-template HCs were found to deliver a largercapacity as negative electrodes for KIBs, a KIB full cell consist-ing of K2Mn[Fe(CN)6] and HC-Zn[0-75-25] with a 1 mol dm−3K(PF6)0.8(FSA)0.2/EC:PC solution[55] + 0.2 vol% vinylene carbon-ate was fabricated with an optimal mass ratio of 2.0:1 for the pos-itive and negative electrodes. The charge–discharge curves andcycle performance are shown in Figure 5d and Figure S21 (Sup-porting Information), respectively. The KIB showed an averagedischarge voltage of 3.3 V and a substantial specific capacity of107 mAh (g of K2Mn[Fe(CN)6])−1, without any pretreatment ofthe electrolyte and electrodes, such as pre-cycling. These resultsconfirm that a large-capacity HC is compatible with KIBs whencombined with proper positive electrodes and is a promising al-ternative to graphite for KIB-negative electrodes.[56] We believethat a new series of templated porous carbon materials has po-tential as active materials for next-generation batteries, such asNIB and KIB, and will possibly be enhanced by rational designdepending on the battery and redox system for future energy de-vices.3. ConclusionA novel ZnO-template HC featuring large capacity, high initialCoulombic efficiency, and low working potential for negativeelectrodes suitable for Na- and KIBs was developed. Systematiccharacterization and electrochemical evaluation of the precursorsand HCs derived from Mg Glu, Zn Glu, and Ca Glu suggestedthat the ZnO-template carbon is a promising NIB negative elec-trode, comparable to the MgO-template HC. A ZnO-template HCderived from a mixture of Zn Glu and Zn(OAc)2 in an appropri-ate mixing ratio exhibited a maximum capacity of 464 mAh g−1.Based on comprehensive experiments on the structure and elec-trochemical properties of various samples derived from Glc–ZnGlu–Zn(OAc)2 mixtures with different mixing ratios, the largecapacity of the optimized ZnO-template HC was attributed to itsfavorable pore size and uniform pore distribution for Na cluster-ing. An NIB full cell with a ZnO-template HC provides an energydensity of over 300 Wh kg−1, which is comparable to that of LFP-based LIBs, overcoming the challenges associated with the heavyatomic weight of Na and the high standard electrode potentialof Na+/Na. Furthermore, an unprecedented large K-storage ca-pacity was also found for the ZnO-template HCs, and a KIB fullcell with a ZnO-template HC without any precycling treatmentwas successfully demonstrated. This finding proves that HCs arepromising candidates for KIB negative electrodes as an alterna-tive to graphite. We hope that this study will trigger improve-ments in the energy densities of NIBs and KIBs to a new levelbeyond that of LIBs.Supporting InformationSupporting Information is available from the Wiley Online Library or fromthe author.AcknowledgementsThis study was partially funded by the MEXT Program, Data Creation, andUtilization Type Materials Research and Development Project (grant no.JPMXP 1122712807) and JST-CREST (grant no. JPMJCR21O6), the NEDOIntensive Support Program for Young Promising Researchers (grant no.JPNP20004), and JSPS KAKENHI (Grant No. JP20H02849, JP21K14724,JP21K20561, JP22K14772, and JP23K13829). The authors are grateful toDr. Toshinari Ichihashi and Prof. Yasushi Idemoto at the Tokyo Universityof Science for the TEM measurements and to Haruka Yoshimo, KazushiMagara, Nobuhito Noro, Tatsuo Matsuyama, Yuki Fujii, and Yuki Hoshi fortheir experimental support. D.I. thanks the JST for establishing universityfellowships for the Creation of Science and Technology Innovation (grantno. JPMJFS2144), and the academic crowdfunding platform “academist”and its supporters. R.T. thanks the TEPCO Memorial Foundation researchgrant (basic research), the ECSJ Kanto branch research grant, and the Taka-hashi Industrial and Economic Research Foundation.Conflict of InterestThe authors declare no conflict of interest.Data Availability StatementThe datasets in this study are available from the corresponding author onreasonable request.Keywordsanode materials, hard carbon, K-ion batteries, Na-ion batteries,nanopores, template synthesisReceived: August 12, 2023Revised: September 11, 2023Published online: November 9, 2023[1] N. Yabuuchi, K. Kubota, M. Dahbi, S. Komaba, Chem. Rev. 2014, 114,11636.[2] T. Hosaka, K. Kubota, A. S. Hameed, S. Komaba, Chem. Rev. 2020,120, 6358.[3] S. Komaba, W. Murata, T. Ishikawa, N. Yabuuchi, T. Ozeki, T.Nakayama, A. Ogata, K. Gotoh, K. Fujiwara, Adv. Funct. Mater. 2011,21, 3859.[4] Y. Kondo, T. Fukutsuka, K. Miyazaki, Y. Miyahara, T. Abe, J. Elec-trochem. Soc. 2019, 166, A5323.Adv. Energy Mater. 2023, 13, 2302647 2302647 (10 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensewww.advancedsciencenews.com www.advenergymat.de[5] H. Au, H. Alptekin, A. C. S. Jensen, E. Olsson, C. A. O’keefe, T. Smith,M. Crespo-Ribadeneyra, T. F. Headen, C. P. Grey, Q. Cai, A. J. Drew,M.-M. Titirici, Energy Environ. Sci. 2020, 13, 3469.[6] K. Kubota, S. Shimadzu, N. Yabuuchi, S. Tominaka, S. Shiraishi, M.Abreu-Sepulveda, A. Manivannan, K. Gotoh, M. Fukunishi, M. Dahbi,S. Komaba, Chem. Mater. 2020, 32, 2961.[7] H. Alptekin, H. Au, A. C. Jensen, E. Olsson, M. Goktas, T. F. Headen,P. Adelhelm, Q. Cai, A. J. Drew, M.-M. Titirici, ACS Appl. Energy Mater.2020, 3, 9918.[8] K. Tatsumi, J. Conard, M. Nakahara, S. Menu, P. Lauginie, Y. Sawada,Z. Ogumi, J. Power Sources 1999, 81–82, 397.[9] Z. Jian, Z. Xing, C. Bommier, Z. Li, X. Ji, Adv. Energy Mater. 2016, 6,1501874.[10] J. R. Dahn, Phys. Rev. B 1991, 44, 9170.[11] H. Onuma, K. Kubota, S. Muratsubaki, W. Ota, M. Shishkin, H. Sato,K. Yamashita, S. Yasuno, S. Komaba, J. Mater. Chem. A 2021, 9, 11187.[12] Y. Nishi, Chem. Rec. 2001, 1, 406.[13] J. Chen, T. Hu, Z. Zou, Q. Zeng, Y. Jiang, C. Tang, C. Tang, W. Li, C.Fang, W. Sun, L. Zeng, C. M. Li, Carbon 2022, 186, 193.[14] Z. Lu, J. Wang, W. Feng, X. Yin, X. Feng, S. Zhao, C. Li, R. Wang, Q.-A.Huang, Y. Zhao, Adv. Mater. 2023, 35, 2211461.[15] A. Kamiyama, K. Kubota, D. Igarashi, Y. Youn, Y. Tateyama, H. Ando,K. Gotoh, S. Komaba, Angew. Chem., Int. Ed. 2021, 60, 5114.[16] Q. Li, X. Liu, Y. Tao, J. Huang, J. Zhang, C. Yang, Y. Zhang, S. Zhang,Y. Jia, Q. Lin, Y. Xiang, J. Cheng, W. Lv, F. Kang, Y. Yang, Q.-H. Yang,Natl. Sci. Rev. 2022, 9, nwac084.[17] X. Yin, Z. Lu, J. Wang, X. Feng, S. Roy, X. Liu, Y. Yang, Y. Zhao, J. Zhang,Adv. Mater. 2022, 34, 2109282.[18] V. L. Deringer, C. Merlet, Y. Hu, T. H. Lee, J. A. Kattirtzi, O. Pecher, G.Csányi, S. R. Elliott, C. P. Grey, Chem. Commun. 2018, 54, 5988.[19] Y. Youn, B. Gao, A. Kamiyama, K. Kubota, S. Komaba, Y. Tateyama,NPJ Comput. Mater. 2021, 7, 48.[20] J. Li, C. Peng, J. Li, J. Wang, H. Zhang, Energy Fuels 2022, 36, 5937.[21] D. A. Stevens, J. R. Dahn, J. Electrochem. Soc. 2000, 147, 4428.[22] D. A. Stevens, J. R. Dahn, J. Electrochem. Soc. 2001, 148, A803.[23] Y. Morikawa, S.-I. Nishimura, R.-I. Hashimoto, M. Ohnuma, A.Yamada, Adv. Energy Mater. 2020, 10, 1903176.[24] J. M. Stratford, A. K. Kleppe, D. S. Keeble, P. A. Chater, S. S. Meysami,C. J. Wright, J. Barker, M.-M. Titirici, P. K. Allan, C. P. Grey, J. Am.Chem. Soc. 2021, 143, 14274.[25] J. M. Stratford, P. K. Allan, O. Pecher, P. A. Chater, C. P. Grey, Chem.Commun. 2016, 52, 12430.[26] R. Morita, K. Gotoh, M. Fukunishi, K. Kubota, S. Komaba, N.Nishimura, T. Yumura, K. Deguchi, S. Ohki, T. Shimizu, H. Ishida,J. Mater. Chem. A 2016, 4, 13183.[27] J. S. Weaving, A. Lim, J. Millichamp, T. P. Neville, D. Ledwoch, E.Kendrick, P. F. Mcmillan, P. R. Shearing, C. A. Howard, D. J. L. Brett,ACS Appl. Energy Mater. 2020, 3, 7474.[28] T. Morishita, T. Tsumura, M. Toyoda, J. Przepiórski, A. W. Morawski,H. Konno, M. Inagaki, Carbon 2010, 48, 2690.[29] T. Kyotani, Z. Ma, A. Tomita, Carbon 2003, 41, 1451.[30] A. A. Zakhidov, R. H. Baughman, Z. Iqbal, C. Cui, I. Khayrullin, S. O.Dantas, J. Marti, V. G. Ralchenko, Science 1998, 282, 897.[31] B. Yan, J. Zheng, F. Wang, L. Zhao, Q. Zhang, W. Xu, S. He, Mater.Des. 2021, 201, 109518.[32] C. Zhao, W. Wang, Z. Yu, H. Zhang, A. Wang, Y. Yang, J. Mater. Chem.2010, 20, 976.[33] Y. Maeno, K. Hagino, T. Ishiguro, Found. Chem. 2021, 23, 201.[34] E. R. Buiel, A. E. George, J. R. Dahn, Carbon 1999, 37, 1399.[35] B.-S. Kim, J.-M. Yoo, J.-T. Park, J.-C. Lee, Mater. Trans. 2006, 47, 2421.[36] P. Strubel, S. Thieme, T. Biemelt, A. Helmer, M. Oschatz, J. Brückner,H. Althues, S. Kaskel, Adv. Funct. Mater. 2015, 25, 287.[37] F. J. W. J. Labuschagné, W. W. Focke, J. Mater. Sci. 2003, 38, 1249.[38] K. Omote, Y. Ito, S. Kawamura, Appl. Phys. Lett. 2003, 82, 544.[39] H. Yamamoto, S. Muratsubaki, K. Kubota, M. Fukunishi, H.Watanabe, J. Kim, S. Komaba, J. Mater. Chem. A 2018, 6, 16844.[40] A. Kamiyama, K. Kubota, T. Nakano, S. Fujimura, S. Shiraishi, H.Tsukada, S. Komaba, ACS Appl. Energy Mater. 2020, 3, 135.[41] M. Thommes, K. Kaneko, A. V. Neimark, J. P. Olivier, F. Rodriguez-Reinoso, J. Rouquerol, K. S. W. Sing, Pure Appl. Chem. 2015, 87,1051.[42] J. Tan, Z. Li, M. Ye, J. Shen, ACS Appl. Mater. Interfaces 2022, 37259.[43] G. Duan, J. Xiao, L. Chen, C. Zhang, S. Jian, S. He, F. Wang, J. EnergyStorage 2023, 67, 107559.[44] L.-J. He, J.-Q. Zhong, Y.-M. Liu, Z.-J. Zhou, W.-H. Yang, ChemistrySelect2020, 5, 12759.[45] Z. V. Bobyleva, O. A. Drozhzhin, A. M. Alekseeva, K. A. Dosaev, G. S.Peters, G. P. Lakienko, T. I. Perfilyeva, N. A. Sobolev, K. I. Maslakov,S. V. Savilov, A. M. Abakumov, E. V. Antipov, ACS Appl. Energy Mater.2023, 6, 181.[46] H. Yang, J. Hwang, Y. Tonouchi, K. Matsumoto, R. Hagiwara, J. Mater.Chem. A 2021, 9, 3637.[47] N. Horzum, M. E. Hilal, T. Isik, New J. Chem. 2018, 42, 11831.[48] A. Moezzi, A. Mcdonagh, A. Dowd, M. Cortie, Inorg. Chem. 2013, 52,95.[49] S. Li, H. Yuan, C. Ye, Y. Wang, L. Wang, K. Ni, Y. Zhu, J. Mater. Chem.A 2023, 11, 9816.[50] J. A. S. Oh, G. Deysher, P. Ridley, Y.-T. Chen, D. Cheng, A. Cronk, S.-Y.Ham, D. H. S. Tan, J. Jang, L. H. B. Nguyen, Y. S. Meng, Adv. EnergyMater. 2023, 13, 2300776.[51] S. Komaba, Chem. Lett. 2020, 49, 1507.[52] Z. Jian, W. Luo, X. Ji, J. Am. Chem. Soc. 2015, 137, 11566.[53] Q. Li, J. Zhang, L. Zhong, F. Geng, Y. Tao, C. Geng, S. Li, B. Hu, Q.-H.Yang, Adv. Energy Mater. 2022, 12, 2201734.[54] A. Vasileiadis, Y. Li, Y. Lu, Y.-S. Hu, M. Wagemaker, ACS Appl. EnergyMater. 2023, 6, 127.[55] T. Hosaka, T. Matsuyama, K. Kubota, S. Yasuno, S. Komaba, ACS Appl.Mater. Interfaces 2020, 12, 34873.[56] X. Bie, K. Kubota, T. Hosaka, K. Chihara, S. Komaba, J. Mater. Chem.A 2017, 5, 4325.Adv. Energy Mater. 2023, 13, 2302647 2302647 (11 of 11) © 2023 The Authors. Advanced Energy Materials published by Wiley-VCH GmbH 16146840, 2023, 47, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/aenm.202302647 by National Institute For, Wiley Online Library on [06/01/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License