# Fileset

[89_Kawaguchi_Commun Mater 2025_Amorphous oxide for rt RMBs.pdf](https://mdr.nims.go.jp/filesets/06e7daac-4f2f-4b6c-9eb9-3c8cbe45773a/download)

## Creator

[Tomoya Kawaguchi](https://orcid.org/0000-0002-7600-4847), Hikari Sakurai, Shusuke Fukui, [Xiatong Ye](https://orcid.org/0000-0002-5932-5964), [Hongyi Li](https://orcid.org/0000-0003-1890-3825), [Toshihiko Mandai](https://orcid.org/0000-0002-2403-7794), [Norihiko L. Okamoto](https://orcid.org/0000-0003-0199-7271), [Tetsu Ichitsubo](https://orcid.org/0000-0002-1127-3034)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Amorphous oxide cathode enabling room-temperature rechargeable magnesium batteries](https://mdr.nims.go.jp/datasets/e5880490-c67b-4732-be87-8b75f0c7354e)

## Fulltext

Amorphous oxide cathode enabling room-temperature rechargeable magnesium batteriescommunicationsmaterials ArticleA Nature Portfolio journalhttps://doi.org/10.1038/s43246-025-00921-0Amorphous oxide cathode enabling room-temperature rechargeable magnesiumbatteriesCheck for updatesTomoya Kawaguchi 1,4 , Hikari Sakurai1,2, Shusuke Fukui1,2, Xiatong Ye 1,2, Hongyi Li 1,Toshihiko Mandai 3, Norihiko L. Okamoto 1 & Tetsu Ichitsubo 1Rechargeable magnesium batteries (RMBs) have faced challenges in utilizing oxide cathodes due tothe inherently sluggish Mg diffusion and poor compatibility with electrolytes, despite the high redoxpotential. Herein, we present a prototype RMB that is operational at room temperature, consisting of ananoparticulate amorphous oxide cathode, fluorinated alkoxyborate (Mg[B(HFIP)4]2) as theelectrolyte, and a Mgmetal anode. The amorphous MgxTi1/9Mo2/9O cathode contains a considerablefree volume formed by ion exchange between monovalent and divalent cations, facilitating Mgdiffusion and eventually realizing reversible Mg insertion/extraction at room temperature. Thereasonable compatibility of the present cathode with the electrolyte enables full cell operation with aMgmetal anode, and various analyses have demonstrated that Mg intercalation is responsible for thebattery performance. The discharging capacity is ~150mAh g−1, and 70mAh g−1 is maintained after200 cycles. These findings demonstrate the feasibility of RMBs with oxide cathodes that areoperational at room temperature.Rechargeable magnesium batteries (RMBs) are potential alternatives forhigh-energy-density chemical energy storage due to the abundance ofmagnesium in the Earth’s crust1; the development of RMBs would mitigatethe soledependenceon scarce lithium,which is essential for current lithium-ion batteries (LIBs). The high energy density of RMBs is due to the avail-ability of magnesium metal anodes, which exhibit a specific capacity of2205mAh g−1 and a relatively low redox potential of −2.38 V vs. SHE.Additionally, these anodes have a favorable plating morphology in selectedelectrolytes and conditions1–3, unlike Li and Na anodes, which tend to formproblematic dendrites4. This makes RMBs a safe and viable choice forapplications such as stationary energy storage systems. Despite their pro-mise, however, the development of practical RMBs has faced obstacles sincethe RMB prototype in 2000, which consisted of a sulfide cathode and aGrignard-based electrolyte5, primarily due to the lack of available cathodematerials6,7.Oxide materials8–14 are generally beneficial for achieving high energydensities due to their inherently high redox potential (2–3 V vs. Mg2+/Mg),which fully leverages the low potential of magnesium metal anodes com-pared with other chalcogenides15–20, such as sulfides and selenides. Fur-thermore, the advent of halide-free electrolytes21–26 that are capable ofrelatively reversible Mg deposition/stripping opens up the possibility ofutilizing oxide electrodes. Among them, electrolytes with weakly coordi-nating anions24 such asMg[B(HFIP)4]225 andMg[Al(HFIP)4]226 (B(HFIP)4:tetrakis(hexafluoro-iso-propoxy)borate, Al(HFIP)4: tetrakis(hexafluoro-iso-propoxy)aluminate) are among the most promising due to their highoxidative resistance in comparison with that of amine-based/amine-addedelectrolytes21–23 and conventional Grignard-based electrolytes of dichlorocomplexes and phenyl complexes7. Nevertheless, it is still challenging tobuild a full cell prototype based on an oxide cathode27 due to the inherentlysluggish Mg diffusion in oxide cathodes28, unfavorable rocksalt transfor-mation accompanied byMg insertion8,11,12, and poor electrode compatibilitywith the electrolyte29. Indeed, the low diffusivity of Mg drastically narrowsthe choice of oxide cathodes, such as α-MnO2 and α-V2O56, that areoperational at room temperature. α-MnO2 is a promising cathodematerial11, yielding a capacity of 110mAh g−1 due to substantial Mgintercalation29 in a half-cell with an electrolyte consisting of Mg[TFSA]2 intriglyme (TFSA: bis(trifluoromethanesulfonyl)amide; triglyme (G3): trie-thylene glycol dimethyl ether). The reason why the system is defined as ahalf-cell is that the electrolyte containing Mg[TFSA]2 induces a largeoverpotential (>2 V) for stripping the Mg metal anode, which poses chal-lenges in full cell operation. A simple solution for this issue is to replace thesalt in Mg[TFSA]2 with Mg[B(HFIP)4]2, which is capable of reversible Mg1Institute for Materials Research, Tohoku University, Sendai, Japan. 2Graduate School of Engineering, Tohoku University, Sendai, Japan. 3Research Center forEnergy and Environmental Materials, National Institute forMaterials Science (NIMS), Tsukuba, Japan. 4Present address: Argonne National Laboratory, Lemont, IL,USA. e-mail: tkawaguchi@anl.gov; tichi@imr.tohoku.ac.jpCommunications Materials |           (2025) 6:203 11234567890():,;1234567890():,;http://crossmark.crossref.org/dialog/?doi=10.1038/s43246-025-00921-0&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s43246-025-00921-0&domain=pdfhttp://crossmark.crossref.org/dialog/?doi=10.1038/s43246-025-00921-0&domain=pdfhttp://orcid.org/0000-0002-7600-4847http://orcid.org/0000-0002-7600-4847http://orcid.org/0000-0002-7600-4847http://orcid.org/0000-0002-7600-4847http://orcid.org/0000-0002-7600-4847http://orcid.org/0000-0002-5932-5964http://orcid.org/0000-0002-5932-5964http://orcid.org/0000-0002-5932-5964http://orcid.org/0000-0002-5932-5964http://orcid.org/0000-0002-5932-5964http://orcid.org/0000-0003-1890-3825http://orcid.org/0000-0003-1890-3825http://orcid.org/0000-0003-1890-3825http://orcid.org/0000-0003-1890-3825http://orcid.org/0000-0003-1890-3825http://orcid.org/0000-0002-2403-7794http://orcid.org/0000-0002-2403-7794http://orcid.org/0000-0002-2403-7794http://orcid.org/0000-0002-2403-7794http://orcid.org/0000-0002-2403-7794http://orcid.org/0000-0003-0199-7271http://orcid.org/0000-0003-0199-7271http://orcid.org/0000-0003-0199-7271http://orcid.org/0000-0003-0199-7271http://orcid.org/0000-0003-0199-7271http://orcid.org/0000-0002-1127-3034http://orcid.org/0000-0002-1127-3034http://orcid.org/0000-0002-1127-3034http://orcid.org/0000-0002-1127-3034http://orcid.org/0000-0002-1127-3034mailto:tkawaguchi@anl.govmailto:tichi@imr.tohoku.ac.jpwww.nature.com/commsmatstripping/deposition. However, the drastic change in the solvation structureof Mg cations due to the use of Mg[B(HFIP)4]2 salt results in the decom-position of α-MnO2 during discharge, which is accompanied by significantside reactions with the electrolyte, inhibiting the construction of operationalRMB full cells29.In the present study, we develop a oxide cathode material,Mg0.27Li0.09Ti0.11Mo0.22O (MLTMO), which is a composite of majoramorphous and minor rocksalt phases (Fig. 1a). The Mg-containingamorphous oxide is prepared via amorphization of the Li-containing oxidesfollowed by ion exchange between Li and Mg; this process spontaneouslyintroduces a considerable free volume into the resultant amorphous phaseas remnants of Li cation vacancies due to the valence state discrepancybetween monovalent Li and divalent Mg. This amorphous structure with alarge free volume enables facile Mg insertion and extraction. Additionally,the significant difference between the amorphous and rocksalt structuressuppresses detrimental rocksalt transformation, in contrast to spineloxides8, whose oxygen sublattice is similar to that of rocksalt oxide. Weselected Mg[B(HFIP)4]2-G3 electrolyte24,25 due to its high oxidative resis-tance, reversible Mg stripping/plating with relatively low overpotential,compatibility with aluminum cell components, and potential for com-mercial availability30. We demonstrate that this material is capable of Mgextraction/insertion at room temperature and fairly stable inMg[B(HFIP)4]2-G3 electrolyte and evaluate its full cell feasibility at roomtemperature in an orthodox split cell, which consists of a composite cathodeelectrode, a Mg metal anode, a separator, and a relatively lean electrolyte(Fig. 1b).Thedischarging capacity remains at 70mAh g−1 formore than200cycles, and rigorous chemical analysis via inductively coupled plasmaoptical emission spectroscopy (ICP‒OES) confirms thatMg intercalation isthe primary contributor to the electrochemically observed capacity duringcycling.Cathode preparation and electrochemical propertiesThe cathode material MLTMO was prepared by a modified solutioncombustion synthesis (SCS)31 followed by ion exchange between Li andMgfrom the ion-exchange precursor Li2Ti1/3Mo2/3O313, which is a pseudo-binary oxide of Li2MoO3 and Li2TiO3. The ion-exchange precursorLi2Ti1/3Mo2/3O3 has a Li-rich layered rocksalt structure (Fig. 2a and Sup-plementary Fig. S1), whose crystalline size was estimated to be 3.65(2) nmfrom the 104 diffraction peak according to the Scherrer equation. In con-trast, the secondary particles (white regions in the high-angle annular darkfield scanning transmission electron microscopy (HAADF-STEM) image)were well dispersed and ranged from 7 to 10 nm in diameter, as determinedfrom HAADF-STEM images (Fig. 2b). The dark gray interstitial spacecorresponds to amorphous carbon, into which the oxide particles wereembedded (Supplementary Fig. S2). Fuel-rich combustion31 followed bycalcination at 600 °C in a reductive environment of 10% H2-Ar yieldedamorphous carbon covering the target oxide, which presumably decreasedthe oxide particle size by suppressing excessive growth at high temperature(Fig. 2c), in contrast to particles obtained by the conventional Pechinimethod13. Energy-dispersiveX-ray spectroscopy (EDS) images revealed thatthe composition was virtually homogeneous, while the low fluorescenceyield of Ti and O relative to that of Mo obscured the particle outlines in thecorresponding EDS images. Constituent Li was exchanged with Mg byimmersing the ion-exchange precursor Li2Ti1/3Mo2/3O3 inMg[TFSA]2-G3electrolyte at 90 °C for 24 h (Supplementary Fig. S3), which yielded thecomposition Mg0.269(8)Li0.085(2)Ti0.111Mo0.215(6)O, according to ICP‒OES.X-ray powder diffraction (XRPD, Fig. 2a) indicated thatMLTMOconsistedof major amorphous andminor crystalline phases with a rocksalt structure,where the ball-milling of the ion-exchange precursor Li2Ti1/3Mo2/3O3before ion exchange induced amorphization (Supplementary Fig. S1).The cell performance was evaluated using a three-electrode split celloperated at room temperature. The cell consisted of a composite electrodewithMLTMOas the cathode,Mgmetal foil as the anode, a glassfiberfilter asthe separator, and 0.53MMg[B(HFIP)4]2 in G3 as the electrolyte. The cellwas also equipped with a reference electrode of Li metal immersed in 1MLiTFSA in G3, which was separated from the main bath by a ceramic filter.Charge/discharge tests were conducted at a constant current density of10mA g−1 with cutoff potentials of 1.8 and 4.1 V vs. Li RE (Fig. 3 andSupplementary Fig. S4).The discharge capacity was highest at ~100mAh g−1 at the ~6th cycleand remained at ~75mAh g−1 after 200 cycles. The present cell, therefore,showed much better cyclability than previously reported oxide cathodesoperated at 60 °C13 and 90 °C12. ICP‒OES analysis of the material after the5th cycle demonstrated that reversible Mg insertion/extraction was pre-dominantly responsible for the electrochemically observed capacity, inwhich the changes in Mg and Li composition were equivalent to electro-chemical capacities of 69.3 and 8.2mAh g−1, respectively (SupplementaryTable S1). This contrasts with the case of an α-MnO2 cathode discharged ina similarweakly coordinating electrolyte, inwhich side reactions rather thanMg insertion contributed the most to the electrochemically observed dis-charging capacity29. The present cell was equipped with a Li referenceelectrode separated from the main bath by a ceramic filter due to the highpotential stability of Li metal. Nevertheless, the small amount of Li in themain bath derived from the reference electrode hardly affected the electrodeproperties. Indeed, the amount of Li in the electrolyte, derived from thereference electrode, was confirmed to be negligibly small (Li/Mg <10−3) byFig. 1 | Advantages of utilizing amorphous oxide cathodes and cell configurationof present rechargeable magnesium batteries (RMBs). a Features of amorphousoxide cathodes. External perturbations cause amorphization of Li-rich layeredoxides, and subsequent ion exchange yields a Mg-containing oxide cathode. Thevalence-state discrepancy betweenmonovalent Li and divalentMg forms a large freevolume. The free volume in the amorphous oxide cathode ensures a percolative Mgmigration path and reversible Mg intercalation. The significant difference instructure between the amorphous oxide and rocksalt oxide suppresses the phasetransformation into the electrochemically inactive rocksalt structure in the Mg-richcomposition. b Cell configuration of present RMBs, in which aMgmetal anode andits compatible electrolyte, Mg[B(HFIP)4]2 in triglyme, is used with an amorphouscathode.https://doi.org/10.1038/s43246-025-00921-0 ArticleCommunications Materials |           (2025) 6:203 2www.nature.com/commsmatthe ICP-OESanalysis in a similar system13. Thiswas further corroboratedbythe virtually identical cell performance obtained in the same cell config-uration with a Mg reference electrode (Fig. 3a and Supplementary Fig. S5),while a Mg reference tends to be more sensitive to external influences andless stable in potential (Supplementary Fig. S6). The small changes in Licomposition at the discharged state of the 5th and 109th cycle observed viaICP‒OES (Supplementary Table S1) also support these observations.The charging capacity always exceeded the discharging capacity, whichwas attributed to oxidative decomposition of the electrolyte because excesscharging capacity was hardly observed in the high-concentration electrolyteMg[TFSA]2-G3 even at an elevated temperature of 60 °C13. The excesscharging capacitywas conspicuous in thefirst 20 cycles,whereas it decreasedin the later cycles, reflecting the corresponding Coulombic efficiency (CE)values: 45–70% in the first 20 cycles and more than 70% in the later cycles.This cycledependenceofCE implies that this parasitic charging capacity canbe attributed to the formation of a cathode electrolyte interphase. Forinstance, the amount of boron on the electrode, derived only from theelectrolyte, considerably increased after long-term cycling, indicating thatcontinuous electrolyte decomposition during charging/discharging deliv-ered the parasitic capacities (Supplementary Table S1).The rate capability of discharging was evaluated by changing thecurrent density from 5 to 500mA g−1, where the cell was charged at aFig. 2 | Characterization of the oxide cathodematerials and synthesis mechanism. a X-ray dif-fraction profiles of (black broken line) Li2Ti1/3Mo2/3O3 and (blue solid line) ion-exchangedMg0.27Li0.09Ti0.11Mo0.22O. h and c in the diffractionindexes represent hexagonal and cubic lattices,respectively. The asterisk indicates the halo peak ofthe amorphous phase. The simulated XRD profilesof Li2MoO3 and MgO are also presented at thebottom of the panel as references to the hexagonallayered rocksalt structure and cubic rocksalt struc-ture, respectively. b STEM-HAADF and EDS ima-ges of crystalline Li2Ti1/3Mo2/3O3. The scale barrepresents 20 nm. c Schematic of the modifiedsolution combustion synthesis.Fig. 3 | Electrochemical performance with respect to reversible capacity, dis-charge rate, and long-term cyclability. a Potential profiles ofMg0.27Li0.09Ti0.11Mo0.22O at room temperature with a current density of 10 mA g−1in the potential window of 4.1 V for charging and 1.8 V vs. Li RE for discharging. Theinset shows the rate capability for discharging to 1.7 V vs. Li RE, in which the cell wascharged to 4.2 V vs. Li RE at a common current density of 10 mA g−1 before eachdischarging test. b Cell voltage profiles obtained from EWE− ECE in (a), where thecounter electrode was pure Mg metal. The vertical axis was, therefore, convertedfrom the electrode potential into net cell voltage. cCapacity retention correspondingto (a). Filled and open circles indicate the discharging and charging capacity,respectively. Green diamonds represent the Coulombic efficiency.https://doi.org/10.1038/s43246-025-00921-0 ArticleCommunications Materials |           (2025) 6:203 3www.nature.com/commsmatconstant current density of 10mA g−1 before discharging at various currentdensities. Only discharging was evaluated because the charging capacitywould include the contribution from the oxidative decomposition of theelectrolyte. The upper limit of the electrochemical window is estimated to be~3.5 V vs. Mg2+/Mg on the Pt electrode in the present electrolyte system24.Namely, the electrolyte decomposition is minimal when discharging below3.5 V vs.Mg2+/Mg, while chargingwould be accompanied by the electrolytedecomposition, because the electrode potential reaches close the upper limitof the electrochemicalwindow.Note that theupper limit changesdependingon the electrode and tends to be lower in the composite electrode relative tothe Pt electrode due to the large surface area of the composite electrode andthe catalytic activity of the cathode materials32. The slowest discharge of5mA g−1 yielded the largest discharging capacity of ~150mAh g−1. Thedischarging capacity remained at ~70mAh g−1 even at the fastest dischargerate of 500mA g−1, implying the high rate capability of the present cathodematerial and cell configuration.The scan rate dependence of the cyclic voltammetry (CV)9,33,34 results(Supplementary Fig. S7) indicated that the diffusion-limited redox reaction,rather than the electrochemical capacitor reaction, was predominant atrelatively slow charging/discharging rates of up to 50mA g−1, which isapproximately equivalent to 0.5 mV s−1 in CV,while fast discharge involveda considerable contribution from the electrochemical capacitor reaction asseen in the CV result at 5.0 mV s−1. The relatively good rate capability can,therefore, be attributed to two factors: Mg-ion diffusion and pseudocapa-citive contributions33. The small particle size of the present MLTMO mat-ches the diffusion length of Mg at room temperature, thereby enablingbattery operation under ambient conditions. Additionally, reducing particlesize increases the specific surface area, which in turn enhances the pseu-docapacitive contributions, particularly during high-rate operation.The differential capacity curve (dQ/dE) of the charge/discharge test(Supplementary Fig. S8) was in good agreement with the CV profile at thelowest sweeping rate. The differential capacity increased monotonicallyduring charging and decreased during discharging, without showing asignificant peak. This profile suggests that charge compensation of Mo, asdiscussed in the X-ray absorption spectroscopy section below, occurs over abroad potential range in the operating cathode material. The overpotentialof the Mg metal anode was within 0.2 and 0.45 V during stripping (dis-charge) and deposition (charge), respectively (Supplementary Fig. S9),which is in good agreement with the previous studies24,25,30 on the presentMg[B(HFIP)4]2-G3 electrolyte. Amine-additive21 or amine-based22 elec-trolytes with Mg[TFSA]2 salt represent an alternative electrolyte option,offering broader availability and reasonably low overpotential. However,enhancing the oxidative resistance of amine-based compounds remains achallenge22, particularly for high-voltage oxide cathodes.Notably, the present cell configuration is equivalent to what is com-monly referred to as a full cell, which consists of a fully operational cathode,anode, and electrolyte. Indeed, the charge/discharge performancemeasuredin the two-electrode coin cell shows good agreement with that obtainedfrom the three-electrode split cells (Supplementary Fig. S10). Nevertheless,in the present study, the working electrode potential was controlled in thecharge/discharge tests with the split cell to meticulously evaluate the cellbehavior without being disturbed by the unstable potential of theMgmetalanode and electrolyte decomposition.We, therefore, defined the present celloperation as a reference-controlled full cell (RC full cell). Note that the RCfull cell is distinct from the conventional half-cells used in previous RMBstudies, in which either only the cathode or anode is fully operational, suchas in the case of oxide cathodes withMg[TFSA]2-G3 electrolyte12,13. In suchcells, the passivated Mg metal anode hampers substantial cell dischargebecause of the significant overpotential of Mg stripping. Thus, the presentstudywith theRC full cell contrastswith studies based on conventional half-cells. The present results imply that strictly defined full cell operation isfeasible by adjusting the galvanostat setting to control the cutoff voltagebased on the net cell voltage (i.e., EWE− ECE) rather than on the electrodepotential of the cathode (EWE) measured via the reference electrode. Here,EWE and ECE represent the electrode potentials of the cathode (workingelectrode) and anode (counter electrode), respectively.Charge compensation mechanismX-ray absorption spectroscopy (XAS) was used to clarify the redoxmechanism and cyclic evolution of the charge and discharge states. X-rayabsorption near edge structure (XANES) demonstrated that the chargeduring charging and discharging was predominantly compensated bychanges in the valence state of Mo between 6+ and lower valence states,while the effect of Tiwasminimal, based on the same composition of Ti-Mooxides13. The superimposed spectra exhibited isosbestic points (indicated byarrows inFig. 4a), implying thatMLTMOcharged anddischargedvia a two-phase reaction. Indeed, STEM-EDS analysis revealed two distinct types ofamorphousmaterials, i.e., Mg-rich andMg-poor amorphousmaterials, in aprevious study on coarseMLTMO operated at 60 °C13. Figure 4b shows themagnifiedpre-edgepeaks indicatedbyanasterisk inFig. 4a, representing theaverage valence states of Mo in different states of charge and discharge.The intensity of the pre-edge peak represents the average valence stateof Mo based on the reference materials Mo(IV)O2 and Mo(VI)O3; astronger pre-edge peak corresponds to a higher valence state. The spectracorresponding to the 10th cycle (green solid and dashed lines in Fig. 4b)showed the largest change in pre-edge peak intensity with charging anddischarging, in good agreement with the almost largest capacity observed inthis cycle (Fig. 3c). The intermediate charging/discharging states andFig. 4 | Charge compensation mechanism duringcharging and discharging. aMo K-edge XANES.Arrows indicated the isosbestic points observed inthe various states of charge/discharge in this energyregion. The asterisk represents a pre-edge peak.bMagnified spectra of the pre-edge peak indicatedin (a) by the asterisk.https://doi.org/10.1038/s43246-025-00921-0 ArticleCommunications Materials |           (2025) 6:203 4www.nature.com/commsmatcompletely charged/discharged states were examined in the 3rd cycle, wherethe discharging capacity increased with increasing number of cycles. Thecharged state in the 3rd cycle (pink solid line) was virtually identical to thatin the 10th cycle, indicating that repeated cycling minimally affected thecharged state. The intermediate charging state (pink broken line) fellsomewhat between the charged (pink solid line) and discharged states (bluesolid line), indicating that electrochemical charging continuously oxidizedthe constituent Mo cations even though the electrolyte was also oxidativelydecomposed in the high potential region. In contrast to charging, dischar-ging reduced the valence state of Mo up to only ~70% of the electro-chemically observed capacity, while further discharging did not reduceMo,judging from the very similar spectra corresponding to a discharging state of~70% (blue dashed line in Fig. 4b) and the discharged state (blue solid line).This implies that side reactions of the composite electrode or electrolytepartially consumed the electrochemically observed charge at the end ofdischarging.Nevertheless, the compatibility ofMLTMOwith the electrolytewasmuchbetter than that ofα-MnO2, as ICP‒OES indicated substantialMginsertion. Indeed, side reactions during discharging hinder Mg insertioninto the α-MnO2 cathode material due to the inherent incompatibilitybetween bare α-MnO2 and the electrolyte29. Further cycling improved theelectrochemical reduction of Mo, as observed in the most reduced state atthe 10th cycle. This behavior contrastswith those of the charged states in the3rd (red solid line) and 10th (green broken line) cycles, where the valencestate of Mo was similar, independent of the number of cycles. Thus, withincreasing number of cycles, the active range of Mo valence states expandstoward the reduced state rather than the oxidized state after charging. ThisextendedMovalence changewithcycling canbe attributed to the increase inthe amount of active material accessible throughMg intercalation from theelectrolyte, as revealed by STEM-EDS13.Free volume in the amorphous structureFree volume is estimated via an amorphous structure simulated by mole-cular dynamics and ab initio calculations. In the crystal system, cationvacancies are siteswhere cations are absent, and cations can possibly occupyvia an intercalation or elemental step of diffusion. Likewise, in the presentstudy, we define free volume in amorphous oxide as a counterpart of thecation vacancy in the crystal structure, since no vacancies can be defined inthe amorphous structure. Namely, the free volume is the region whereatoms, especially cations in the present study, can possibly occupy35,36. Notethat the free volumedefined in thepresent study isnotnecessarily equivalentto that generally used for studies of amorphous materials37. The simulatedamorphous structurewas prepared via amelt-quenchmethod starting fromthe Li2Ti1/3Mo2/3O3 (Li0.67Ti0.11Mo0.22O) crystal of 1000 atoms, followed byreplacing Li atoms with Mg. Half of the original Li atoms were randomlysubstituted with Mg, and the remaining half were removed, in accordancewith the ion-exchange strategy used in the corresponding experiment. Theresulting amorphous structure was then fully relaxed using ab initio densityfunctional theory (DFT) calculations (Fig. 5a).The free volume ratio, f V , was calculated by f V ¼ 1� ΣiniVi=Vcell,whereni is the number of atomsof element, i,Vi is theVoronoi volume, andVcell is the volume for the simulation cell of the amorphous structure.Vi forMg, Ti, andMowere determined from the simulated amorphous structure,whileVO from crystallineMgOwas used to evaluate the free volume, due toits well-defined coordination environment. f V was estimated to be 9.7%,which is equivalent to 52%of theVMg.Namely, the chemical formula can benominally described as Mg0.33□0.17Ti0.11Mo0.22O when explicitly describ-ing the amount of free volume, □, in the structure. Therefore, approxi-mately half of the free volume introduced by the ion exchange is secured inthe amorphous structure according to the theoretical prediction. The dis-tribution of the free volume is further visualized based on the bond valencesum(BVS) in the simulated amorphous structure (Fig. 5b).Thegreenregionin Fig. 5b represents an isosurface defined by vBVS � videal�� �� < 0:4, wherevBVS is a calculatedBVS at eachmeshpoint, and videal is a nominal valence ofdivalent Mg2+. The widespread and well-dispersed distribution of the greenregions indicates that the free volume is uniformly distributed throughoutthe amorphous matrix, without forming large voids or aggregated clusters.Such free volume would contribute as diffusion path of Mg in the amor-phous structure.Outlook for full cell realizationThe present demonstration of RC full cell operation at room temperaturewith MLTMO implies the following general requirements for oxide cath-odes for RMBs: an atomic structure with considerable vacancy/free volume,a small particle size, a redox potential within the electrochemical window ofthe electrolyte, and compatible cell components. Interactions betweenneighboring Mg cations significantly affect the energy barrier in cathodematerials38; in addition to strong Coulombic interactions between divalentMg cations and surroundingOanions,which lead to a relatively high energybarrier for Mg migration. The former factor is easily overlooked since suchinteractions are virtually negligible in LIB cathodes39. However, for instance,a previous study38 showed that it was extremely challenging to substantiallyextract/insert Mg from/into Mg-rich disordered rocksalt oxides whileintroducing cation vacancies to facilitate Mg migration and eventuallyactivate the cathode material12,38. Specifically, substantial Mg migrationrequires considerable vacancies/free volume in the initial structure. Thiscould also explain why previously proposed oxide cathodes that areoperational at room temperature, such as MnO2, V2O5, and TiO2, containlittle Mg6. Consequently, candidate materials with adequate space toaccommodate Mg are limited to specific structures.In contrast to conventional material synthesis, the present studyemployed ion exchange between monovalent alkaline and divalent Mgcations, which simultaneously introduced vacancies/free volume (Fig. 1b).Ideally, the amount of vacancies/free volume is the same as that ofexchanged Mg cations, given that the host structure maintains space formovable cations during ion exchange. This vacancy introduction, realizedby balancing the electric charge between monovalent and divalent ions, isFig. 5 | Free volume region that divalent Mg canoccupy. a Atomic structure of amorphousMg0.33Ti0.11Mo0.22O obtained via moleculardynamics and ab initio calculations. Mo–O andTi–O coordination polyhedra are shown in purpleand light blue, respectively. Orange spheres repre-sent Mg atoms. Green regions indicate free volumeregions within the structure where bond valencesum (BVS) values are between 1.6 and 2.4, suggest-ing favorable sites for Mg occupancy. b Samestructure as in a with coordination polyhedraremoved to highlight the spatial distribution of Mgatoms and BVS-based free volume.https://doi.org/10.1038/s43246-025-00921-0 ArticleCommunications Materials |           (2025) 6:203 5www.nature.com/commsmatlikely critical for facilitating Mg migration in oxide cathodes by mitigatingthe interactions betweenMg cations. If the cations completely exchange, therelevant chemical reaction can be described as AMxOy+ 0.5Mg2+→Mg0.5□0.5MxOy+A+, whereA is an alkaline element,M is ametal element,and□ is a vacancy/free volume in the atomic structure. The present studyemployed Li as the alkaline metal, A, because Li-containing compoundshave beenwell studied for use in LIBs. However,A is not necessarily limitedto Li; Na andKwould also be candidates for this procedure. The cations canbe exchanged chemically and electrochemically; moreover, the chemicalprocess of simply immersing the Li-containing compound in excess Mgelectrolyte should be milder for oxide cathodes because this process cir-cumvents the formation of higher valence states of redox-active transitionmetals, which are frequently unstable. The present study also implies thatMg-containing cathodes such asMLTMOcan ultimately lead to anode-freecells with no Mg metal in the anode, even though a demonstration of thisidea is beyond the scope of this work. This approach would be beneficial forthe industrialmanufacturing of RMBs becauseMgmetal ismuch safer thanalkaline metals but is prone to oxidization in ambient air.The residual alkaline cations in the cathode materials may enhanceelectrochemical properties by facilitatingMg diffusion in the materials12,40,41or suppressing adverse phase transitions42. This diffusion-enhancementeffect was elaborated in the Chevrel compound with Li40 and furtherextended to oxides incorporating various alkaline cations of Li, Na, and K41,where the extent of enhancementdependson the specific cation species.Thesmall amount of residual Li in the present MLTMO would affect the elec-trode performances, while the detailed analysis is beyond the scope of thiswork. These effects can potentially be leveraged in the aforementioned ion-exchange approach to obtain Mg-containing oxide cathodes.This concept has been intensively explored in so-called “dual-saltbatteries”43,44 and also referred to as “hybrid batteries” in later studies45,46.The dual-salt/hybrid batteries employ electrolytes containing both mono-and multi-valent cations (e.g., Li+ and Mg2+), where the primary reactionstypically involve Li intercalation at cathode, and Mg plating/stripping atanode. However, co-intercalation of both Li and Mg into the cathode hasalso been explored as a strategy to increase the overall energy density of thedual-salt/hybrid battery systems44,45. The charge/discharge mechanism ofthe present MLTMO partially aligns with the dual-salt/hybrid batteryconcept, particularly in terms of the co-intercalation of Mg2+ and Li+.However, the present system exhibits a dominant contribution from Mgintercalation ( ~ 90%) and does not include any Li salt in the initial elec-trolyte. These features contrast from the original dual-salt/hybrid batterydesign, although categorical distinction between dual-salt/hybrid batteriesand RMBs employing the cathode containing monovalent cations remainsinherently ambiguous.It is alsowell known that particle size is an important factor influencingthe activity of cathodematerials, as it should be comparable to the diffusionlength. The diffusion length, ld, is approximated by ld �ffiffiffiffiffiDtp, where D isthe chemical diffusion coefficient and t is the time; cathode materials forRMBs typically have an ld in the range of one nanometer at roomtemperature27. The particle size of coarse MLTMO in the previous study13was tens of nm (Supplementary Fig. S11c), as heat treatment inevitablyaccompanies particle growth. Consequently, the course cathode materialcould deliver less than 30mAh g−1 at room temperature (SupplementaryFig. S11a). In contrast to the previous study, we suppressed the particlegrowth during heat treatment by embedding the nucleus in the carbonmatrix in themodified solution combustion synthesis (Fig. 2c). This yieldedthe cathode materials with a primary particle size of 3.65(2) nm, which wascomparable to the typical diffusion length of Mg at room temperature. Thesmaller particle size increased the particle regions for the substantial Mgextraction/insertion, and therefore, delivered the capacity of up to150mAh g−1 at room temperature.The substantial electrochemical window of the electrolyte, specificallythe oxidative decomposition potential, tends to become narrower oncathode composite electrodes because of the large surface area. A well-defined archetypal electrode, such as Pt foil, is typically used to evaluate theelectrochemical window using linear sweep voltammetry (LSV) in electro-lyte studies. The potential where the current drastically increases is definedas the upper limit of the electrochemical window. However, even a smalloxidative current within the defined electrochemical window preceding thedrastic current increase effectively limits the maximum potential for char-ging a composite electrode with a large surface area. The different types ofthe materials used for these purposes, namely, Pt for LSV and transitionmetal oxides for the cell, may also affect the catalytic activity of oxidativeelectrolyte decomposition and, ultimately, the practical electrochemicalwindow. For instance, an excessive charging capacity resulted in amoderateCE of ~70%. This value should be increased to ~100% for practical RMBs inthe future by developing more electrochemically stable electrolytes and/ormodifying the surface of the cathode materials to suppress the catalyticactivity of electrolyte decomposition.The last but critical challenge of developing operational RMBs iscompatibility between the cell components.Mg[Al(HFIP)4]2, a derivative ofthe compound Mg[B(HFIP)4]2 used in the present study, is corrosive to Almetal at high potential26,29,47 and is typically used as the cathode currentcollector inRMBs aswell as LIBs.We confirmed that the present electrolyte,i.e.,Mg[B(HFIP)4]2 inG3,was not corrosive to the present cell components,including the Al current collector. The most critical compatibility is thatbetween the electrolyteand the cathodematerial. For instance, bareα-MnO2used as a cathode material is not compatible with hexafluoroisopropoxide-derived electrolytes during discharging29. This is because the solvationstructure of the solvate-separated ion pair (SSIP) is highly reactive tounstable surface oxygen on the α-MnO2 cathode, whereas the contact ionpair (CIP) in the Mg[TFSA]2-glyme electrolyte can form a stable interfacethat suppresses the continuous reaction between α-MnO2 and the electro-lyte. Such incompatibility results in a considerable discrepancy between theelectrochemically observed discharging capacity and that calculated fromthe composition; Mg insertion into the cathode material is minimal despitethe substantial discharging capacity. In contrast to conventional oxidecathodes, MLTMO exhibited reasonable compatibility with theMg[B(HFIP)4]2-G3 electrolyte, judging from the substantial Mg insertion/extraction demonstrated by chemical composition analysis via ICP‒OES.For instance, the electrochemical discharging capacity of approximately100mAh g−1 at the 5th cycle (Fig. 3c) was in good agreement with thecapacity of 77mAh g−1 estimated from the composition change (Supple-mentary Table S1). This consistency implies the high stability of Ti-Mooxides in the presence of the Mg[B(HFIP)4]2-G3 electrolyte, which may beattributed to the inherent high chemical resistance of Ti and Mo oxides ingeneral.Capacity retention is also a crucial factor in realizing practical full cells.Reported degradation mechanisms of cathode materials can be categorizedas follows: (1) conversion or semi-topotactic reactions8,11,48,49, (2) electrolytedecomposition12,48,50, (3) surface reactions51,52, and (4) decomposition ordissolution53. Conversion and semi-topotactic reactions do not necessarilydegrade capacity retention, but cyclability is compromised when thesereactions are not fully reversible. A representative example is the semi-topotactic transition from spinel to rocksalt induced by Mg insertion. Thereverse transition from rocksalt back to spinel is often incomplete due tosluggish Mg diffusion within the rocksalt structure, resulting in poor cyclelife10. This issue can be mitigated by limiting the Mg composition range tomaintain structural integrity during charge/discharge, as demonstrated indefect spinel cathodes9. The present MLTMO material would circumventthis issue by incorporating elements such asMoandTi, which are less proneto forming rocksalt structures, and by leveraging the large structural dif-ference between the amorphous and rocksalt phases.Electrolyte decomposition, particularly at high voltages, also leads topoor cyclability, either through the formation of insulating decompositionproducts on the cathode surface12 or due to a limited electrochemical win-dow that restricts full cathode utilization. When the electrochemical win-dow is fixed, such limitations stem from the intrinsic redox potential of thecathode material and the overpotential during Mg extraction, which islinked to intercalation kinetics48,50. Therefore, the redox potential must behttps://doi.org/10.1038/s43246-025-00921-0 ArticleCommunications Materials |           (2025) 6:203 6www.nature.com/commsmatcarefully tuned by selecting appropriate redox-active elements, and thestructure should be designed to enable fast Mg-extraction kinetics.Surface reactions generally result in the formation ofMg-impermeablelayers51,52, which are often tied to the chemical compatibility between thecathode and the electrolyte. Decomposition and dissolution are observed in,for example, sulfide cathodes53, which dissolve into the electrolyte duringcycling. However, such degradation may also occur in oxide cathodes,particularly when halide-based electrolytes are used. As described above,capacity retention is influenced by multiple interrelated factors, such asredox potential, Mg intercalation kinetics, phase transition behavior, andelectrolyte compatibility. At the current stage of RMB development,improving cyclability requires addressing these challenges in an integratedmanner during material design, possibly with the aid of computationalmaterials science, rather than relying on a single specific strategy.The present MLTMO is further compared with the typical oxide andpolyanion cathode materials previously reported, considering not only theelectrode performance but also the compatibility between cell componentsand rigor of the cathode characterization (Supplementary Table S2). Thefollowing conditions are considered such that each study implies the fea-sibility of the full cell operation at room temperature, and is comparable tothe present study.Webelieve that essential characterizations todemonstratereversibleMg insertion/extraction into/from the cathodes are bulk-sensitiveanalyses of the chemical composition, crystal structure, and chemical state.This is because excessive side reactions, such as electrolyte decomposition29and proton insertion54, can falsely give seemingly good electrochemicalperformances and lead tomisinterpretation that reversibleMg insertion andextraction are occurring. We also evaluate the electrolyte used in terms ofcompatibility with the Mg metal anode. Six materials out of the 32 listedmaterials fulfill the three conditions of cathode characterizations, electrolytecompatibility, and operation at room temperature. It is further narroweddown to two materials when ruling out the use of halide-containing elec-trolytes of all phenyl complex and dichloro complex, which are corrosive tothe oxide materials and whose electrochemical window is limited by halideions. The screened materials are α-MnO248 and γ-MnO2-basedMg0.15MnO221, both of which show the highest capacity of ~270mAh g−1and the average potential/voltage of ~1.0 and ~2.4 V for α-MnO2 andMg0.15MnO2, respectively. The present MLTMO cathode material deliversthe energy density of ~255Wh kg−1 (=1.7 × 150), which is comparable to α-MnO2 (270Wh kg−1 (=1.0 × 270)), but less thanMg0.15MnO2 (648Wh kg−1(=2.4 × 270)). However, the MLTMO shows the relatively good capacityretention of 80% after 200 cycles (=80/100, Fig. 3c), while α-MnO2 andMg0.15MnO2 show 19% (=30/160) at 30 cycles and 70% (=190/270) at 25cycles, respectively.The feasibility of using the present cell configuration as an RMBoperational at room temperature was further examined by using a minimalsetup of a 2032 coin cell connected to a commercial blue LED (Fig. 6). Thecurrent‒voltage curve (I‒V curve) of the LED (Supplementary Fig. S12)indicated that a substantial current flowed above ~2.5 V with blue lightemission. Therefore, cells consisting of conventional sulfide-based cathodes,which have a potential of 1–1.5 V, hardly drive the blue LED. Additionally,even cells with oxide cathodes do not necessarily deliver a voltage higherthan 2.5 V when combined with electrolytes that passivate the Mg metalanode. This is the case for, for example, Mg[TFSA]2-based electrolytes,whereas Mg[TFSA]2-based electrolytes are widely used for cathode char-acterization due to their high stability with respect to oxide cathodes. Asingle 2032 coin cell charged to 3.5 V after eight charging/discharging cyclescontinuously activated the blue LED for over 7min. This finding demon-strates that the present RMB prototype delivers a relatively high voltage dueto effective use of the oxide cathode material.The present study demonstrated the feasibility of RMBs consisting ofan oxide cathode and Mg[B(HFIP)4]2-G3 as the electrolyte that areoperational at room temperature. An MLTMO cathode material wasprepared by amodified SCS followed by ion exchange between Li andMg,which simultaneously introduces a considerable free volume into theresulting amorphous structure. The small particle size and substantial freevolume enable reversibleMg extraction/insertion at room temperature. Afull cell consisting of the composite cathode and a Mg metal anodeseparated by a glass fiber separator and containing a relatively leanMg[B(HFIP)4]2-G3 electrolyte exhibited the largest capacity of150mAh g−1 at a current density of 5 mA g−1 and good capacity retentionof 70mAh g−1 over 200 cycles. Chemical composition analysis via ICP‒OES confirmed that the electrochemically observed capacity was pre-dominantly derived from Mg insertion/extraction. Notably, the currentresults imply the feasibility of anode-free RMBs since the MLTMOcathode containsMg, unlike previous cathodematerials, such as α-MnO2and α-V2O5. Furthermore, the present study reveals that ion-exchangedamorphous oxides are viable cathodematerials for RMBs, whichwould beadvantageous for Mg diffusion and compatibility with the electrolyte.Importantly, the RC full cell demonstrated in the present study providesopportunities to investigate the charging/discharging mechanism in awhole cell, in which the crosstalk and compatibility of multiple materialsand cell components affect cell performance, ultimately leading to therealization of practical RMBs.MethodsSynthesis. Mg0.27Li0.09Ti0.11Mo0.22O (MLTMO) was produced by amodified solution combustion synthesis (SCS) followed by ion exchangebetween Li and Mg from the ion exchange precursor Li2Ti1/3Mo2/3O3(LTMO). Stoichiometric quantities of (NH4)6Mo7O24-4H2O (Wako,purity 99%), titanium tetraisopropoxide ([(CH3)2CHO]4Ti, Wako,purity 95%) and a 5% excess of LiNO3 (Wako) were mixed with propy-lene glycol C3H8O2 (Wako, purity 98%), citric acid anhydrous (Wako,purity 98%), and ammonium nitrate (Wako, purity 98%) to prepare aprecursor gel. The molar ratio of [metal element]:[propylene gly-col]:[citric acid]:[ammonium nitrate] was set to 1:2:2:4.5. The molyb-denum salt, lithium salt, propylene glycol, and ammonium nitrate weredissolved in deionized water to create a Mo‒Li solution. A citric acidsolution (CA solution) was prepared by dissolving citric acid in dehy-drated ethanol predried with 3 Å molecular sieves overnight. Titaniumtetraisopropoxide was mixed with dehydrated ethanol to create a Tisolution. For batch synthesis of 100 mmol LTMO, these three solutionswere prepared with ~250 ml of solvent each.The CA solution was added to the Ti solution and mixed thoroughlyfor ~10min. Then, the Mo–Li solution was carefully dripped into the Tisolution under rapid stirring so that the transparency of the mixed solutionwas maintained. The mixed solution was heated at 60 °C overnight. Thetemperature was then changed to 65 °C, and the mixture was heated for afew days until the solution changed to a light orange viscous paste.Fig. 6 | Blue LED emission test using a single prototype coin cell consisting of aMg0.27Li0.09Ti0.11Mo0.22O cathode, a Mg metal anode, and Mg[B(HFIP)4]2 in G3 asthe electrolyte.https://doi.org/10.1038/s43246-025-00921-0 ArticleCommunications Materials |           (2025) 6:203 7www.nature.com/commsmatThe paste was placed on an alumina boat and calcined at 600 °C for10 h with 10% H2–Ar gas flow. The paste spontaneously combusted atapproximately 300 °C during the temperature increase in the furnace. Theobtained puff-like black product was transferred to an Ar-filled glove boxwith minimal exposure to moist air and hand-milled using an agate mortarand pestle. The powder was further ball-milled at 500 rpm for 200min intotal with 0.5mm-ϕ zirconia balls in dimethyl carbonate in an Ar atmo-sphere. The pristine LTMO powder contained 24 wt% residual amorphouscarbon derived from propylene glycol and citric acid, as determined byinfrared absorption spectroscopy after combustion.Ion exchange and electrochemical tests were performed with compo-site electrodes fabricated by mixing the synthesized LTMO with carbonblack (SuperC65, Timcal) andpolyvinylidene difluoride (PVDF,Kureha) ataweight ratio of 8:1:3whilemaintaining an inert Ar atmosphere in amixingcontainer. The obtained slurry was uniformly applied onto aluminum foiland vacuum dried at 120 °C for 12 h. The cropped electrode, typically16mm ϕ, was immersed in an electrolyte consisting ofMg(TFSA)2 (Wako)and triglyme at a molar ratio of 1:2.22 for 24 h at 90 °C to promote ionexchange between Li and Mg. The ion-exchanged electrodes were washedwith ultradry acetonitrile (Wako, >99.8%), dried, and used for electro-chemical evaluation and characterization. Ion exchange and subsequentwashing were performed in an Ar-filled glove box. The triglyme, whichwasobtained from Tokyo Chemical Industry, was predried with 3 Å molecularsieves.Electrochemical testing. Three-electrode split cells (SB9, EC Frontier),were assembled and operated for electrochemical testing in Ar-filledglove boxes. Al foil covered the cell casing on the cathode side to preventcontact between the stainless steel cell body and the electrolyte. Polishedpure magnesium foil (Nilaco or Rikazai, 60–100 μm thick, >99.9%) wasused as the counter electrode. The reference electrode was lithium foil(Honjo Metal) immersed in 1M LiTFSA in triglyme as the electrolyte,separated from the main bath by a ceramic filter. The working electrodepotential was converted to V vs. Mg2+/Mg by using the experimentallydetermined potential difference of 1.0 V between Li+/Li andMg2+/Mg inthe system. A glass fiber filter (GF/F, Whatman) was used as a separator.The electrolyte was 0.53M Mg[B(HFIP)4]2 in triglyme. The electrolytewas pretreated before cell evaluation by applying constant potentials of1.3 and 4 V vs. Li+/Li for 48 h at each potential to eliminate any redox-active impurities. For electrolyte pretreatment, Pt foil and Mg ribbonwere used as working and counter electrodes, and the electrolyte wasstirred.Two-electrode 2032 coin cells (Hosen) were assembled using the samecomponents as the three-electrode split cells except for the absence of areference electrode. The cathode side of the cell housing was coated withaluminum.Electrochemical measurements were conducted using VMP3, VSP-300, and VSP apparatuses (Biologic) at room temperature. Galvanostaticcharging/discharging tests were conducted at a current density of 10mA g−1within a potential window of 1.8–4.1 V versus the Li RE. To evaluate thedischarge rate capability, the cell was charged at a common current densityof 10mA g−1 to 4.2 V versus the Li RE before being discharged at variouscurrent densities to 1.7 V versus the Li RE.Sample characterization. After electrochemical testing, all the electrodesamples were washed with ultradry acetonitrile for further character-ization. X-ray powder diffraction (XRPD) was performed with a Smar-tLab apparatus (Rigaku) equipped with a molybdenum X-ray source anda 1DD/teX detector (Rigaku) inDebye–Scherrer geometry. The electrodesamples were encapsulated within Lindeman glass capillaries (500 μmdiameter) in Ar-filled glove boxes to prevent air exposure. The zirconiumKβ filter was placed upstream of the sample to suppress the Mo Kαfluorescence from the Mo-containing samples, which is excited by thehigh-energy continuous X-ray from the X-ray tube, in addition to theoriginal purpose ofMoKβ elimination. The backgroundwas numericallysubtracted from the XRPD profiles using the XRPD profile of a blankcapillary.The cation composition was determined by inductively coupledplasma‒optical emission spectroscopy (ICP‒OES). The carbon content inthe pristine sample was evaluated via infrared absorption spectroscopy aftercombustion.Transmission electron microscopy (TEM), scanning transmissionelectron microscopy (STEM), and EDS were performed utilizing aspherical-aberration-corrected JEM-ARM200F microscope (JEOL) oper-ating at 200 kV. The samples were prepared for STEM observation bysuspending themon copper-mesh grids coatedwith thin perforated carbon.X-ray absorption spectroscopy (XAS) was performed at the BL11S2beamline of AichiSR, Japan. Washed electrode samples were mixed withboron nitride, compressed into pellets 3 mm in diameter, and sealed withpolyimide film tape to avoid exposure to air during measurement. The MoK-edge absorption spectra were measured in transmission mode.Amorphous model and valence-based analysis. To generate anamorphous Mg1/2Ti1/3Mo2/3O3 structure, a 5 × 5 × 5 supercell of arocksalt-type Li2Ti1/3Mo2/3O3 configuration was constructed, in whichLi, Ti, and Mo cations were randomly distributed over the cation sub-lattice. The 1000-atom system was subjected to classical moleculardynamics simulations using LAMMPS55 with the CHGNet56 machine-learning potential.The structure was melted at 3000 K using the Langevin thermostat for10 ps in the NVE ensemble with a damping parameter of 0.02 ps, followedby a linear quench to 300 Kover 20 ps. The time stepwas set to 1 fs.After thequench, energy minimization and isotropic cell relaxation were performedto remove residual stress. Final equilibration was conducted at 300 K for10 ps using the NVT ensemble.To achieve the target composition, 50% of the Li atoms were replacedwith Mg and the other 50% were with vacancy. The resulting structure wasrelaxed by density functional theory (DFT) calculations using VASP57–60,employing the PBE functional and the projector augmented-wave (PAW)method61,62. Ionic relaxation was performed with a force convergencethreshold of 0.005 eV/Åand an energy convergence criterion of 5 × 10−4 eV.The cell shape and volume were allowed to change, and Γ-point samplingwas used.Bond valence sum (BVS) analysis was conducted on the amorphousMg1/2Ti1/3Mo2/3O3 structure to identify energetically favorable ionsites63. A uniform three-dimensional 200 × 200 × 200 grid was generatedwithin the unit cell (a,b,c = 20.38 Å) with amesh spacing of ~0.1 Å alongeach axis. Grid points within 1.7 Å of any existing atomwere excluded toavoid unphysical overlaps. At each remaining grid point, the BVS wascalculated using the bond valence expression:vBVS ¼ Σi exp R0 � di� �=b� �, where di denotes the distance to the ithoxygen atom within 6.0 Å, R0 = 1.693 Å is the bond valence parameterfor the Mg–O ion pair, and b = 0.37 Å is the universal constant com-monly applied in bond valence modeling64. The expected valence for Mgwas set to 2.0, and the absolute valence mismatch vBVS � 2:0�� �� wasevaluated at each grid point.Data availabilityThe data that support the findings of this study are available from thecorresponding authors upon reasonable request.Received: 5 April 2025; Accepted: 8 August 2025;References1. Liang, Y., Dong, H., Aurbach, D. & Yao, Y. Current status and futuredirections of multivalent metal-ion batteries. Nat. Energy 5, 646–656(2020).2. Matsui, M. Study on electrochemically deposited Mg metal. J. PowerSources 196, 7048–7055 (2011).https://doi.org/10.1038/s43246-025-00921-0 ArticleCommunications Materials |           (2025) 6:203 8www.nature.com/commsmat3. Leuppert, L. et al. Electrolyte-dependent deposition morphology onmagnesium metal utilizing MeMgCl, Mg[B(hfip)4]2 andMg(HMDS)2–2AlCl3 electrolytes. RSC Appl. Interfaces 1, 1142–1155(2024).4. Liu,W., Liu, P. &Mitlin,D.Reviewof emergingconcepts inSEI analysisand artificial SEImembranes for lithium, sodium, andpotassiummetalbattery anodes. Adv. Energy Mater. 10, 2002297 (2020).5. Aurbach, D. et al. Prototype systems for rechargeable magnesiumbatteries. Nature 407, 724–727 (2000).6. Li, Z., Häcker, J., Fichtner, M. & Zhao‐Karger, Z. Cathode materialsand chemistries for magnesium batteries: challenges andopportunities. Adv. Energy Mater. 13, 2300682 (2023).7. Blázquez, J. A. et al. A practical perspective on the potential ofrechargeable Mg batteries. Energy Environ. Sci. 16, 1964–1981(2023).8. Okamoto, S. et al. Intercalation and push‐out process with spinel‐to‐rocksalt transition on Mg insertion into spinel oxides in magnesiumbatteries. Adv. Sci. 2, 1500072 (2015).9. Shimokawa, K. et al. Structure design of long‐life spinel‐oxidecathodematerials for magnesium rechargeable batteries.Adv.Mater.33, 2007539 (2021).10. Shimokawa, K. et al. Zinc-based spinel cathode materials formagnesium rechargeable batteries: toward the reversible spinel-rocksalt transition. J. Mater. Chem. A 7, 12225–12235 (2019).11. Hatakeyama, T. et al. Accelerated kinetics revealing metastablepathways of magnesiation-induced transformations in MnO2polymorphs. Chem. Mater. 33, 6983–6996 (2021).12. Kawaguchi, T. et al. Securing cation vacancies to enable reversibleMg insertion/extraction in rocksalt oxides. J. Mater. Chem. A 12,9088–9101 (2024).13. Kawaguchi, T., Nemoto,N., Sakurai, H.,Okamoto,N. L. & Ichitsubo, T.Utilization of delithiation-induced amorphous oxide as a cathode forrechargeable magnesium batteries. Chem. Mater. 36, 4877–4887(2024).14. Koketsu, T. et al. Reversiblemagnesium and aluminium ions insertionin cation-deficient anatase TiO2. Nat. Mater. 16, 1142–1148 (2017).15. Sun, X. et al. A high capacity thiospinel cathode for Mg batteries.Energy Environ. Sci. 9, 2273–2277 (2016).16. Blanc, L. et al. Toward the development of a high-voltageMg cathodeusing a chromium sulfide host. ACS Mater. Lett. 3, 1213–1220(2021).17. Mao, M. et al. Tuning anionic chemistry to improve kinetics of Mgintercalation. Chem. Mater. 31, 3183–3191 (2019).18. Canepa, P. et al. Odyssey of multivalent cathode materials: openquestions and future challenges.Chem. Rev. 117, 4287–4341 (2017).19. Li, Z. et al. Multi-electron reactions enabled by anion-based redoxchemistry for high-energymultivalent rechargeable batteries.Angew.Chem. Int. Ed. 59, 11483–11490 (2020).20. Dey, S. et al. Exploring cation–anion redox processes in one-dimensional linear chain vanadium tetrasulfide rechargeablemagnesium ion cathodes. J. Am. Chem. Soc. 142, 19588–19601(2020).21. Hou, S. et al. Solvation sheath reorganization enables divalent metalbatteries with fast interfacial charge transfer kinetics. Science 374,172–178 (2021).22. Wang, F. et al. Solvent molecule design enables excellent chargetransfer kinetics for a magnesium metal anode. ACS Energy Lett. 8,780–789 (2023).23. Wang,M. et al. Synergybetween the coordination and trace ionizationof co-solvents enables reversiblemagnesiumelectroplating/strippingbehavior. Energy Environ. Sci. 17, 630–641 (2024).24. Mandai, T., Youn, Y. & Tateyama, Y. Remarkable electrochemical andion-transport characteristics of magnesium-fluorinatedalkoxyaluminate–diglyme electrolytes for magnesium batteries.Mater. Adv. 2, 6283–6296 (2021).25. Zhao-Karger, Z., Gil Bardaji, M. E., Fuhr, O. & Fichtner, M. A new classof non-corrosive, highly efficient electrolytes for rechargeablemagnesium batteries. J. Mater. Chem. A 5, 10815–10820 (2017).26. Herb, J. T., Nist-Lund, C. A. & Arnold, C. B. A fluorinatedalkoxyaluminate electrolyte formagnesium-ionbatteries.ACSEnergyLett. 1, 1227–1232 (2016).27. Kobayashi, H. et al. Ultraporous, ultrasmall MgMn2O4 spinel cathodefor a room-temperature magnesium rechargeable battery. ACS Nano17, 3135–3142 (2023).28. Rong, Z. et al. Materials design rules for multivalent ion mobility inintercalation structures. Chem. Mater. 27, 6016–6021 (2015).29. Ye, X. et al. Examining electrolyte compatibility on polymorphicMnO2cathodes for room-temperature rechargeable magnesium batteries.ACS Appl. Mater. Interfaces 14, 56685–56696 (2022).30. Mandai, T. Critical issues of fluorinated alkoxyborate-basedelectrolytes in magnesium battery applications. ACS Appl. Mater.Interfaces 12, 39135–39144 (2020).31. Varma, A., Mukasyan, A. S., Rogachev, A. S. & Manukyan, K. V.Solution combustion synthesis of nanoscale materials. Chem. Rev.116, 14493–14586 (2016).32. Han, J., Yagi, S., Takeuchi, H., Nakayama,M. & Ichitsubo, T. Catalyticmechanismof spinel oxides for oxidative electrolyte decomposition inMg rechargeable batteries. J.Mater. Chem. A 9, 26401–26409 (2021).33. Wang, J., Polleux, J., Lim, J. & Dunn, B. Pseudocapacitivecontributions to electrochemical energy storage in TiO2 (anatase)nanoparticles. J. Phys. Chem. C 111, 14925–14931 (2007).34. Wang, L. et al. Solvent-controlled charge storage mechanisms ofspinel oxideelectrodes inMgorganohaloaluminateelectrolytes.NanoLett. 18, 763–772 (2018).35. Doweidar, H. Volume of ionic sites in silicate glasses. J. Phys.Condens. Matter 19, 156102 (2007).36. Nhan, N. T., Hung, P. K., Nghiep, D. M. & Kim, H. S. Moleculardynamics investigation on microstructure and void in amorphousSiO2.Mater. Trans. 49, 1212–1218 (2008).37. Turnbull, D. & Cohen, M. H. Free‐volume model of the amorphousphase: glass transition. J. Chem. Phys. 34, 120–125 (1961).38. Robert, H. et al. Mg-rich disordered rocksalt oxide cathodes for Mg-ion batteries. J. Mater. Chem. A 12, 27303–27310 (2024).39. Lee, J. et al. Unlocking the potential of cation-disordered oxides forrechargeable lithium batteries. Science 343, 519–522 (2014).40. Li, H. et al. Fast diffusion of multivalent ions facilitated by concertedinteractions in dual‐ion battery systems. Adv. Energy Mater. 8,1801475 (2018).41. Qi, Y. et al. Alkali cation additives assisting magnesium cationintercalation in hollandite-typemanganesedioxide cathodes. J. Phys.Chem. C 127, 21271–21278 (2023).42. Roy, A. et al. Improving rechargeable magnesium batteries throughdual cation co-intercalation strategy. Nat. Commun. 15, 492 (2024).43. Yagi, S. et al. A concept of dual-salt polyvalent-metal storage battery.J. Mater. Chem. A 2, 1144–1149 (2014).44. Ichitsubo, T. et al. Toward “rocking-chair type” Mg–Li dual-saltbatteries. J. Mater. Chem. A 3, 10188–10194 (2015).45. Gao, T. et al. Hybrid Mg2+/Li+ battery with long cycle life and high ratecapability. Adv. Energy Mater. 5, 1401507 (2015).46. Zhang, J. et al. Current design strategies for rechargeablemagnesium-based batteries. ACS Nano 15, 15594–15624 (2021).47. Mandai, T. et al. Toward improved anodic stability of ether-basedelectrolytes for rechargeable magnesium batteries. J. Phys. Chem. C127, 10419–10433 (2023).48. Iimura, R. et al. Ultrasmall α ‐MnO2 with low aspect ratio: applicationsto electrochemical multivalent‐ion intercalation hosts and aerobicoxidation catalysts. Small 21, 2411493 (2025).49. Ling, C., Zhang, R., Arthur, T. S. & Mizuno, F. How general is theconversion reaction in Mg battery cathode: a case study of themagnesiation of α-MnO2. Chem. Mater. 27, 5799–5807 (2015).https://doi.org/10.1038/s43246-025-00921-0 ArticleCommunications Materials |           (2025) 6:203 9www.nature.com/commsmat50. Sun, X., Bonnick, P. & Nazar, L. F. Layered TiS2 positive electrode forMg batteries. ACS Energy Lett. 1, 297–301 (2016).51. Arthur, T. S. et al. Understanding the electrochemical mechanism ofK-αMnO2 for magnesium battery cathodes. ACS Appl. Mater.Interfaces 6, 7004–7008 (2014).52. Zhang, R. & Ling, C. Unveil the chemistry of olivine FePO4 asmagnesium battery cathode. ACS Appl. Mater. Interfaces 8,18018–18026 (2016).53. Shimokawa, K. et al. Electrochemically synthesized liquid-sulfur/sulfide composite materials for high-rate magnesium batterycathodes. J. Mater. Chem. A 9, 16585–16593 (2021).54. Verrelli, R. et al. On the strange case of divalent ions intercalation inV2O5. J. Power Sources 407, 162–172 (2018).55. Thompson, A. P. et al. LAMMPS—a flexible simulation tool forparticle-based materials modeling at the atomic, meso, andcontinuum scales. Comput. Phys. Commun. 271, 108171 (2022).56. Deng, B. et al. CHGNet as a pretrained universal neural networkpotential for charge-informedatomisticmodelling.Nat.Mach. Intell.5,1031–1041 (2023).57. Kresse, G. & Hafner, J. Ab initio molecular dynamics for liquid metals.Phys. Rev. B 47, 558–561 (1993).58. Kresse, G. &Hafner, J. Ab initio molecular-dynamics simulation of theliquid-metal–amorphous-semiconductor transition in germanium.Phys. Rev. B 49, 14251–14269 (1994).59. Kresse, G. & Furthmüller, J. Efficiency of ab-initio total energycalculations formetals and semiconductors using a plane-wavebasisset. Comput. Mater. Sci. 6, 15–50 (1996).60. Kresse, G. & Furthmüller, J. Efficient iterative schemes for ab initiototal-energy calculations using a plane-wave basis set. Phys. Rev. B54, 11169–11186 (1996).61. Kresse, G. & Hafner, J. Norm-conserving and ultrasoftpseudopotentials for first-row and transition elements. J. Phys.Condens. Matter 6, 8245–8257 (1994).62. Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to theprojector augmented-wave method. Phys. Rev. B 59, 1758–1775(1999).63. Mori, K. et al. Visualization of conduction pathways in lithiumsuperionic conductors: Li2S-P2S5 glasses and Li7P3S11glass–ceramic. Chem. Phys. Lett. 584, 113–118 (2013).64. Brown, I. D. & Altermatt, D. Bond-valence parameters obtained fromasystematic analysis of the Inorganic Crystal Structure Database.ActaCrystallogr. B 41, 244–247 (1985).65. Momma, K. & Izumi, F. VESTA: a three-dimensional visualizationsystem for electronic and structural analysis. J. Appl. Crystallogr. 41,653–658 (2008).AcknowledgementsThe authors thank Dr. Hiroshi Tanimura for fruitful discussions and technicalsupport. The work of T.K. was supported by JSPS KAKENHI (GrantNumbers 21H01646, 22KK0068, and 24K01188). This work was partiallysupported by the GteX Program of Japan (Grant Number JPMJGX23S1).The work of TI was partially supported by JSPS KAKENHI (Grant Number23H05452). The XAS measurements were performed with the approval ofAichSR (Proposal No. 2024D4004). The ICP analyses were supported by F.Sakamoto, K. Nakayama, and Y. Kabasawa at the Analytical ResearchCorefor Advanced Materials, IMR, Tohoku University. The atomic structure wasdrawn using Vesta65.Author contributionsT.K. and T.I. designed the experiments. T.K. and H.S. synthesized the oxidematerials. T.K.,H.S. andS.F.performed theelectrochemicalmeasurements.T.M.synthesized theelectrolyte. T.K.andH.S. characterized thesamplesviaXRPD, XAS, and SEM/EDS. N.L.O. performed the TEM/STEM/EDSobservations. T.K., H.S., X.Y. and H.L. developed the operational full cells.T.K. wrote the manuscript, and all the authors discussed the results andapproved the final version of the manuscript.Competing interestsThe authors declare no competing interests.Additional informationSupplementary information The online version containssupplementary material available athttps://doi.org/10.1038/s43246-025-00921-0.Correspondence and requests for materials should be addressed toTomoya Kawaguchi or Tetsu Ichitsubo.Peer review informationCommunicationsMaterials thanksQinyouAn andthe other, anonymous, reviewer(s) for their contribution to the peer review ofthis work.Reprints and permissions information is available athttp://www.nature.com/reprintsPublisher’s note Springer Nature remains neutral with regard tojurisdictional claims in published maps and institutional affiliations.Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing,adaptation, distribution and reproduction in anymedium or format, as longas you give appropriate credit to the original author(s) and the source,provide a link to the Creative Commons licence, and indicate if changeswere made. The images or other third party material in this article areincluded in the article's Creative Commons licence, unless indicatedotherwise in a credit line to the material. If material is not included in thearticle's Creative Commons licence and your intended use is not permittedby statutory regulation or exceeds the permitted use, you will need toobtain permission directly from the copyright holder. To view a copy of thislicence, visit http://creativecommons.org/licenses/by/4.0/.© The Author(s) 2025https://doi.org/10.1038/s43246-025-00921-0 ArticleCommunications Materials |           (2025) 6:203 10https://doi.org/10.1038/s43246-025-00921-0http://www.nature.com/reprintshttp://creativecommons.org/licenses/by/4.0/www.nature.com/commsmat Amorphous oxide cathode enabling room-temperature rechargeable magnesium batteries Cathode preparation and electrochemical properties Charge compensation mechanism Free volume in the amorphous structure Outlook for full cell realization Methods Synthesis Electrochemical testing Sample characterization Amorphous model and valence-based analysis Data availability References Acknowledgements Author contributions Competing interests Additional information