Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Role of the voltage window on the capacity retention of P2-Na2/3[Fe1/2Mn1/2]O2 cathode material for rechargeable sodium-ion batteries

Role of the voltage window on the capacity retention of P2-Na2/3[Fe1/2Mn1/2]O2 cathode material... ARTICLE https://doi.org/10.1038/s42004-022-00628-0 OPEN Role of the voltage window on the capacity retention of P2-Na [Fe Mn ]O cathode 2/3 1/2 1/2 2 material for rechargeable sodium-ion batteries 1,5,6 2 1 1 1,3 Maider Zarrabeitia , Francesco Nobili , Oier Lakuntza , Javier Carrasco , Teófilo Rojo , 1,4 1,7 Montse Casas-Cabanas & Miguel Ángel Muñoz-Márquez P2-Na [Fe Mn ]O layered oxide is a promising high energy density cathode material 2/3 1/2 1/2 2 for sodium-ion batteries. However, one of its drawbacks is the poor long-term stability in the operating voltage window of 1.5–4.25 V vs Na /Na that prevents its commercialization. In this work, additional light is shed on the origin of capacity fading, which has been analyzed using a combination of experimental techniques and theoretical methods. Electrochemical impedance spectroscopy has been performed on P2-Na [Fe Mn ]O half-cells operat- 2/3 1/2 1/2 2 ing in two different working voltage windows, one allowing and one preventing the high voltage phase transition occurring in P2-Na [Fe Mn ]O above 4.0 V vs Na /Na; so as 2/3 1/2 1/2 2 to unveil the transport properties at different states of charge and correlate them with the existing phases in P2-Na [Fe Mn ]O . Supporting X-ray photoelectron spectroscopy 2/3 1/2 1/2 2 experiments to elucidate the surface properties along with theoretical calculations have concluded that the formed electrode-electrolyte interphase is very thin and stable, mainly composed by inorganic species, and reveal that the structural phase transition at high voltage from P2- to “Z”/OP4-oxygen stacking is associated with a drastic increased in the bulk electronic resistance of P2-Na [Fe Mn ]O electrodes which is one of the causes of the 2/3 1/2 1/2 2 observed capacity fading. Centre for Cooperative Research on Alternative Energies (CIC energiGUNE), Basque Research and Technology Alliance (BRTA), Alava Technology Park, Albert Einstein 48, 01510 Vitoria-Gasteiz, Spain. School of Science and Technology—Chemistry Division, University of Camerino, Via Madonna delle Carceri, 62032 Camerino, Italy. Departamento de Química Inorgánica, Universidad del País Vasco UPV/EHU, P.O. Box 664, 48080 Leioa, Spain. IKERBASQUE, Basque Foundation for Science, María Díaz de Haro 3, 48013 Bilbao, Spain. Present address: Helmholtz Institute Ulm (HIU), 6 7 Helmholtzstrasse 11, 89081 Ulm, Germany. Present address: Karlsruhe Institute of Technology (KIT), P.O. Box 3640, 76021 Karlsruhe, Germany. Present address: School of Science and Technology—Chemistry Division, University of Camerino, Via Madonna delle Carceri, 62032 Camerino, Italy. email: miguel.munoz@unicam.it COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 1 1234567890():,; ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 + + eversible extraction/insertion of Na into host structures (Na p.f.unit = 0.8), the P2-phase coexists with a new distorted was already demonstrated in the 1980s with layered P’2-phase (S.G: Cmcm) . In addition, the described structural 1,2 Roxides . Nowadays, pushed by the increasing need for mechanism of P2-Na [Fe Mn ]O is not affected by the elec- 2/3 1/2 1/2 2 16,20–22 more efficient low-cost energy storage devices, sodium-ion bat- trolyte used . On the other hand, ex-situ and in-situ XRD teries (SIBs) are becoming an alternative for large-scale applica- show that all the structural transformations are reversible , 3–5 tions and light electromobility . This is nested in the fact that although a slight broadening of the peaks and the gradual intensity sodium precursors are evenly distributed in the Earth’s crust and decrease in the second cycle suggest that the crystallinity of the 6–8 21 are cheaper and more abundant than lithium ones . Due to material is reduced upon electrochemical cycling .Indeed,ex- their similar chemical properties, many lithium-based analog situ electronic microscopy analysis —scanning electronic micro- electrode materials have been proposed as cathodes for SIBs . scopy (SEM) and transmission electronic microscopy (TEM) Therefore, a wide variety of sodium-based cathode materials have — clearly shows the exfoliation of the P2-type layered oxides after been studied, including polyanionic materials (phosphates, pyr- cycling, which has been attributed as one of the causes of the poor 25,26 ophosphates, and mixed polyanions), organic compounds, Prus- capacity retention of P2-Na [Fe Mn ]O .The structural 2/3 1/2 1/2 2 9–14 sian Blue analogs (PBAs), and layered oxides . Each one of transformations are known to be detrimental to the reversible these materials has advantages and limitations, for example, capacity —mainly due to the huge volume changes involved— , polyanionic materials exhibit very good capacity retention due to although the exact mechanism and effect on the physical properties their stable 3D framework, in contrast, their specific capacity is of such impoverishment are not understood yet. 9,10 typically lower than many layered oxides ; organic compounds Several strategies have been developed in order to improve its show massive capacity fading due to their dissolution in the long-term stability. One of the most successful approaches is to organic carbonate-based electrolyte ; PBAs usually deliver lower partially substitute the TM by an electrochemically active or capacities at low operating voltage than layered oxides . Layered inactive element/s (such as Ni, Co, Cu, Mg, Al, Ti, K, Li, and so oxides with general formula Na TMO (TM = transition metal/s on) giving rise to ternary or quaternary compounds which, x 2 such as Co, Mn, Fe, Ni, Ti, V, etc., as well as alkali metals namely despite typically displaying lower capacity values, exhibit better Li, K, and Mg) can deliver high specific capacity, but their cycle life capacity retention and in some cases higher operating 13–15 27–34 should be even greater enhanced . Among all layered oxides, P2- voltage . The origin of such enhanced electrochemical per- Na [Fe Mn ]O is one of the most promising cathode materials formance is still unclear and has been attributed to different 2/3 1/2 1/2 2 in terms of cost-efficiency and energy density .Itismadefrom factors, such as improvement of the structural stability, reduction + + Earth abundant elements and delivers a high reversible capacity of of the volume change between Na extracted and Na inserted −1 190 mAh g when it is cycled in the voltage range of 1.5–4.3 V vs states, increase of the sodium interlayer distance, buffering of the Na /Na using metallic sodium as the counter electrode. Moreover, Jahn-Teller induced distortion on Mn(III), and or controlling the +32,34–38 in a full-cell configuration, using hard carbon (HC) as an anode, P2- distribution of Na . Interestingly, the (slightly) doped P2- −1 Na [Fe Mn ]O delivers a reversible capacity of 185 mAh g layered oxides that exhibit improved cycling stability, do not 2/3 1/2 1/2 2 with an average cell voltage of 2.75 V ; which is comparable to the show a high voltage phase transition or the volume changes 19,30,33,34,39,40 prototype cells developed by Faradion using a quaternary layered between P2 and “Z”/OP4 significantly reduced . This oxide (Na Ni Mn Mg Ti O ) as cathode and HC as observation suggested that the high voltage phase formation a (1−x−y−z) x y z 2 anode . Albeit the good energy density values obtained from P2- should be avoided. Consequently, an alternative approach to Na [Fe Mn ]O layered oxide, capacity retention is still one of improve the capacity retention has been studied -not only for P2- 2/3 1/2 1/2 2 its major weaknesses and it has been related to phase transitions Na [Fe Mn ]O but also for other P2-layered oxides— which 2/3 1/2 1/2 2 involving different stacking sequences when the sodium con- is the reduction of the operating voltage window, from >4.3 to centration changes while inducing large volume changes and exfo- 4.0 V vs Na /Na, avoiding the phase transition that has been liation of the layered oxide at particle surface . observed upon cycling at high voltage, although at the cost of a 21,23,24,41,42 The structural evolution of P2-Na [Fe Mn ]O (P6 / lower capacity as well . The exact impact of reducing 2/3 1/2 1/2 2 3 mmc) occurs via a solid-solution mechanism in the voltage the operating voltage window on the electrochemical perfor- + + range of 2.0–4.0 V vs Na /Na (from 0.67 to 0.36 Na per for- mance of layered oxides has been attributed to the improvement 15,20 mula (p.f.) unit) . The repulsion between [TMO ]slabs, of the structural stability upon electrochemical cycling, however, induced by Na extraction from the interlayer space, leads to a further studies should be carried out to clearly discern the role of decrease of a parameter while c parameter increases. Above the operating voltage window on the electrochemical, physical, + + + 4.0 V (Na p.f.unit = 0.36) and up to 4.1 V vs Na /Na (Na and structural properties of P2-layered oxides. p.f.unit = 0.26) a second phase appears showing a biphasic In this work, P2-Na [Fe Mn ]O cathode material is studied 2/3 1/2 1/2 2 region. Above 4.1 V, the newly formed phase propagates as by means of electrochemical impedance spectroscopy (EIS) in two solid-solution until the end of the Na extraction process operating voltage windows. On the one hand, from 1.5 to 4.25 V + + (4.1–4.3 V vs Na /Na; Na p.f.unit = 0.19). The exact struc- (P2-NFMO-LV), where the P2- to “Z”/OP4-type phase transition ture of this second phase at high voltage is still controversial will occur. On the other hand, from 2.0 to 4.0 V (P2-NFMO-SV), due to the fact that it is formed by layer gliding and therefore avoiding the above-mentioned phase transition. These studies are structural disorder increases. Indeed, this phase could not be completed with compositional studies of the electrode-electrolyte clearly indexed by powder X-ray diffraction (XRD) measure- interface by means of X-ray photoelectron spectroscopy (XPS) and ments and it is mainly referred as “Z” phase, OP4-type struc- with density functional theory (DFT) simulations of the electronic ture (scape group (S.G): P-6m2or P6 )orP–Ointergrowth structure. The results show that the phase transition occurring at 16,20–24 phase , which consists of alternating layers with P- and high voltage has a profound impact on the electronic and ionic O-interlayer sites that result from the gliding of half [TMO ] transport properties of P2-Na [Fe Mn ]O electrode, and such 2 2/3 1/2 1/2 2 16 + layers as determined by synchrotron XRD .During Na changes directly impact in capacity retention. The obtained results insertion, the reverse reaction mechanism is observed: starting in this investigation can be extrapolated to other P2-layered oxides from “Z”/OP4 solid-solution region, followed first by a biphasic with different TM compositions and will help clarify the role of the + + region until 3.10 V vs Na /Na (Na p.f.unit = 0.44) and later operating voltage window and high voltage phase formation in the by a P2 solid-solution region. Finally, below 2.0 V vs Na /Na electrochemical properties of the P2-layered oxides. 2 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE Fig. 1 Nyquist plot of P2-NFMO-LV electrode at OCV (2.43 V vs Na /Na). The impedance data from a 100 kHz to 5 mHz and b 100 kHz to 1 Hz. The frequency regions are highlighted (HF-red, MF-green, and LF-blue). Results and discussion model, the poor electronic conductivity of the P2-layered oxides P2-Na [Fe Mn ]O cycled in the 1.5–4.25 V vs Na /Na has been considered as an extra resistance and capacitance labeled 2/3 1/2 1/2 2 voltage window: P2-NFMO-LV electrode. For the first voltage as bulk electronic resistance (R ) and capacitance arising from elec window studied, the Nyquist plot of the impedance dispersion charge accumulation (C ). elec recorded at an open circuit voltage (OCV—2.43 V vs Na /Na) is Figures 2 and 3 reveal that the main change in the Nyquist shown in Fig. 1a. Three semicircles are observed at different plots is observed in the LF semicircle (blue region of Fig. 2a, c), frequencies: low-frequency (LF) below 10 Hz, medium-frequency indicating a variation of the R (blue square of Fig. 3) upon Na elec (MF) in the 5 kHz–10 Hz range and, high-frequency (HF) above extraction/insertion which can be related to modifications in the 5 kHz. The latter one is better observed once the impedance dis- crystalline structure. Similar changes in the EIS behavior at LFs persion is enlarged (see Fig. 1b). The HF semicircle corresponds have been observed in other layered oxide cathodes and + 53,54,57–60 to the Na migration resistance through the electrode-electrolyte anodes . interphase (EEI) and is labeled as R . The oxidation of the After comparing the first with the second cycle, it is found that EEI electrolyte and subsequent formation of the EEI is not expected at R (blue square of Fig. 3) continuously increases although elec OCV, but a similar EEI is chemically formed before cycling, as several oscillations take place depending on the voltage and Na confirmed by XPS and in agreement with previous EEI studies of content, which are related to the structural evolution of 43–46 SIB electrodes . The high reactivity upon air/moisture expo- P2-Na [Fe Mn ]O . In the P2 solid-solution region (blue 2/3 1/2 1/2 2 47 + sure of the layered oxide , the dehydrofluorination reaction of region of Fig. 3), R decreases upon Na extraction while elec polivinylidene fluoride (PVdF), which takes place during elec- increases upon Na insertion after raising the minimum R elec trode preparation as confirmed by solid-state nuclear magnetic values due to the OP4 effect. Except for the first Na extraction in 43 + resonance , and the use of metallic sodium as counter electrode the 3.00–3.33 V vs Na /Na range, probably due to the fact that 48–50 3+ 4+ contribute to form a passivation surface layer composed by the electron transfer is easier in mixed valence states (Fe /Fe + 4+ 3+ carbonate species originated by electrolyte decomposition reac- couple —3.5 V vs Na /Na) than when only Mn and Fe are tions. The MF semicircle corresponds to charge-transfer resis- present on the P2-Na [Fe Mn ]O . This means that in the 2/3 1/2 1/2 2 tance (R ) and accumulation of charge in the interfacial double P2 solid-solution region, the P2-NFMO-LV electrode exhibits a CT layer (C ), while the LF region process can be correlated to the reversible behavior in terms of electronic conductivity, becoming DL bulk electronic conductivity, as already described for other a better electronic conductor during Na extraction. Never- 51–54 layered oxides electrode materials among others . Addition- theless, the R trend is interrupted when the P2-phase is elec ally, in the very LF region (40–5 mHz) a sloping line at ~45° with transformed into “Z ”/OP4-type structure, above 4.0 V vs Na /Na respect to the real axis (Z´) can be observed which corresponds to during Na extraction (violet region of Fig. 3), and when the + + the Na solid-state diffusion. distorted P’2 phase is observed, below 2.0 V vs Na /Na during Na The EIS spectrum of the P2-NFMO-LV electrode were insertion (yellow region of Fig. 3). These increments of R at elec collected every 45 mV during the first two cycles in the voltage 4.0 and 2.0 V vs Na /Na progressively reduce the bulk electronic range of 1.5–4.25 V vs Na /Na and the Nyquist plots shown in conductivity of P2-NFMO-LV electrode upon electrochemical Fig. 2 correspond to some relevant potential values of the first Na cycling, as indicated by the overall increase of R , where the elec extraction (Fig. 2a, b) and insertion (Fig. 2c, d) processes. The increment of R at the second cycle is even higher than in the elec overall trend of the resistance values obtained from the fit to the first one. This massive R increase suggests that the large elec equivalent circuit detailed in the experimental section of all volume change occurring during these phase transitions —mainly impedance spectra (133 in total) measured during the first two from P2 to “Z”/OP4— result in electrical isolation (loss of cycles is shown in Fig. 3. contact) among the active material particles, as well as the The described equivalent circuit (more details in the Methods conductive carbon, and current collector. section) is a modification of the surface model proposed by In the MF semicircle of the Nyquist plot, R (green triangle of CT 55,56 Aurbach and co-workers . This model, developed for graphite Fig. 3) constantly increases overall upon electrochemical cycling. electrodes, assumes that the active material has good electronic This is expected since the phase transitions upon electrochemical conductivity, but it is known that layered oxides are poor cycling induce the irreversible formation of grain boundaries, a electronic conductors. Therefore, in the used equivalent circuit mosaic texture, exfoliation of the layer or/and increase of interfacial COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 3 ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 Fig. 2 Nyquist plots of P2-NFMO-LV electrode upon the first cycle. In the top panels, during first Na extraction: at 2.53 V (red point), 3.46 V (green hexagon), 3.69 V (blue square), 4.02 V (black rhombus), 4.15 V (navy triangle), and 4.25 V (pink pentagon) in the frequency range of a 100 kHz to 5 mHz and b 100 kHz to 1 Hz/50 mHz which is a zoom-in of the very low real impedance component. In the bottom panels, during first Na insertion: at 4.25 V (pink pentagon), 3.45 V (red hexagon), 2.64 V (blue point), 2.42 V (black square), 2.19 V (green triangle), 2.07 V (brown rhombus), and 1.50 V (orange square) in the frequency range of c 100 kHz to 5 mHz and d 100 kHz to 1 Hz/50 mHz which is a zoom-in of the very low real impedance component. The LF region is highlighted in blue. microstrains, which was attributed as one of the causes of the from PVdF in the F 1s region (red line) (see also the concentration capacity fading of P2-layered oxides, including P2-Na [Fe Mn ] of the mentioned species in Supplementary Fig. S2). The EEI 2/3 1/2 1/2 25,26,61 O . formation at OCV occurs by the reduction of solvents and salt In parallel, R (red points of Fig. 3) constantly increases upon decomposition, as also observed in other Na-based electrodes and EEI + + first Na extraction, while above 3.00 V vs Na /Na it remains attributed to the reductive nature of layered oxides and instability of 19,44–46,48 constant. This behavior suggests that the formed EEI is overall metallic sodium . Second, at further electrochemical cycling, stable during the electrochemical cycling. Nonetheless, there is a the formed EEI is composed of C-O-C species, such as polyethylene + 64 small drop in the second Na insertion, more previously in the P oxide (PEO, (-CH -CH -O-) ) originated from direct polymeriza- 2 2 n + 65 ′2 region (below 2.1 V vs Na /Na). There are two main factors tion of ethylene carbonate (EC) and diethyl carbonate (DEC) ,as that can cause the lower R . On the one hand, some electrical well as NaCO R(R= alkyl group/s) and Na CO from EC and DEC EEI 3 2 3 45,66–68 contact problem —note that there is a glitch at 2.4 V, while R reduction as shown in C 1s and O 1s spectra . Indeed, the CT also suffers the drop. On the other hand, some slight outermost surface region is mainly composed of Na CO as 2 3 modifications to the EEI. However, the stability of local regions confirmed by the Na 1s spectra (Supplementary Fig. S3). Besides of the interphase, i.e., the outermost surface region, cannot be the mentioned carbonaceous/oxygenated species, the F 1s spectrum corroborated by means of EIS. reveals NaF formation due to the PVdF dehydrofluorination 43 46,69 The formation and stability of the outermost EEI of P2-Na reaction ,aswellasNaPF decomposition reactions . 2/ 6 [Fe Mn ]O electrodes cycled at different states of charge On the other hand, the first evidence of the EEI stability can be 3 1/2 1/2 2 (SOC - as indicated in the galvanostatic profile of Supplementary observed at a glance, note that the most significant difference is Fig. S1) has been measured by means of XPS. Figure 4 shows the found between the XPS spectra of the pristine and OCV electrodes, C 1s, O 1s, and F 1s photoemission lines while the binding while the XPS spectra from cycled electrodes are rather similar to the energies of the observed species are collected in Supplementary ones from OCV electrode, confirming that the surface composition Table S1. of the electrodes does not undergo major changes upon electro- First, the formation of the EEI is confirmed at OCV by the chemical cycling (see also Supplementary Fig. S2, illustration of the intensity decrease of electrode component photoelectron peaks: C65 concentration percentage of the electrode components). Indeed, the component in the C 1s region (black line) ,P2-Na [Fe Mn ]O presence of stable components corresponding to the pristine 2/3 1/2 1/2 2 component in the O 1s region (cyan line) and -CF component electrode (C65 (C1s), P2-Na [Fe Mn ]O (O 1s), and -CF 2 2/3 1/2 1/2 2 2 4 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE electrode becomes a worse electronic conductor. Such variation in the electronic properties is nested in the structural changes, namely the increase of the c parameter, owing to the electrostatic 15,20 repulsion between oxygen layers ; that ultimately results in an increase of the bulk electronic conductivity, as already observed 53,54 for the Li-based LiCoO layered oxide . In addition, the intrinsic electronic structure of the involved phases has been assessed using DFT calculations, as detailed in the experimental section, to better understand the origin of this observation. The density of states (DOS) for P2-Na [Fe Mn 2/3 1/2 1/ O ] and O2-Na [Fe Mn ]O model structures are shown in 2 2 1/5 1/2 1/2 2 Supplementary Fig. S5a, b, respectively. The O2-type structure should be considered as a proxy of the more complex “Z”/OP4- type structure, which is the most plausible phase formed at high 16,20–23 voltage as has been experimentally reported . Using an actual OP4 structural model involves a very high computational cost because large supercells and a prohibitive number of combinatorial Na /vacancy and Fe/Mn site distributions would need to be considered, which is beyond the scope of this work. In fact, a simple O2-based structural model is representative of the main possible orderings and local environments of the transition metals under the presence of Na /vacancy octahedral sites within OP4-type structures. Such octahedral environments are the main local structural difference with respect to the prismatic environ- ments found in the P2-phase. For the P2-type structure, the computed DOS reveals a valence band maximum populated with Mn states and the presence of a band gap of 0.5–0.8 eV between the valence and conduction bands (Supplementary Fig. S5a). In the case of the O2-type structure, the onset of the conduction band slightly increases by less than 0.3 eV with respect the P2 structure, and two discrete Fe spin-up states appear within the band gap yielding a pseudo-half-metallic character to the system (Supplementary Fig. S5b). Such differences between the electronic structures of P2- and O2-type model structures are not sufficiently large to suggest a noticeable qualitative difference in the intrinsic electronic conductivity between P2- and OP4-type Fig. 3 Fitted resistance values for the first two electrochemical cycles of phases. This is in stark contrast with the measured R values elec the P2-NFMO-LV electrode. a Resistance trend of R (red point), R that reveal the more insulating character of the Na extracted EEI CT (green triangle), R (blue square) and b α factor of the CPE associated cathode electrode with respect to the more sodiated one. At this elec with C (blue inverted triangle). The structural evolution highlighted as point, several factors should be considered upon the interpreta- elec colored regions and labeled upon electrochemical cycling is included in the tion of the DFT results. Indeed, the calculated band gaps 15, 20 figure for guidance . correspond to the active material while the EIS measurements correspond to the behavior of the whole electrode where the active material properties are deconvoluted. Therefore, the group of PVdF (F 1s)) is in agreement with the constant R values EEI eventual changes in particle size, interface and/or exfoliation, measured by EIS. The peak shape and intensity slightly vary during staking faults, etc. appearing upon cycling are not considered by Na insertion, for example, the NaCO R and NaF concentration the DFT calculations; certainly, these factors influence the bulk increases, which is due to theirpartial dissolutioninthe electronic conductivity of the material. organic carbonate-based electrode during Na extraction, and In order to observe experimentally these possible changes that 43,67 might be the responsible of the R drop observed below 2.1 V . EEI can affect the bulk electronic conductivity of the electrode, the This has been previously reported for similar layered oxides (note measured α value of C has been analyzed. α is the exponential elec that the Na CO R (x= 1) and NaF are highly soluble in 2−x 3 x factor of the electronic constant phase element (CPE) and for an 44,45 organic carbonate-based electrolytes) .Itcan be concludedthat, ideal surface, α is equal to 1. Figure 3b summarizes the α values of although electrolyte oxidation/reduction is not expected in the P2- the first two electrochemical cycles, and it shows an overall decrease NFMO-LV electrode since the operating voltage window is inside of α during the electrochemical cycling (variations are observed in the electrochemical stability window of the organic carbonate-based the phase transition regions). The α factor strongly depends on the electrolyte , a thin EEI is formed on the active material and 51 surface homogeneity, roughness and degree of polycrystallinity ,so conductive carbon already at OCV SOC, whose thickness and its continuous decrease is suggesting that the surface of the chemical composition slightly vary at different SOC. Indeed, the thin electrode is becoming more heterogeneous and/or the crystallinity is EEI cannot be detected by Fourier-transform Infrared spectroscopy being loss upon electrochemical cycling. Since these aspects cannot (FTIR), as shown in Supplementary Fig. S4, where the FTIR spectra be considered by the performed DFT calculations, and even if the of the cycled electrodes are identical. P2- and O2-phases can be representative of the OP4-phase, this As a result, the overall increase of R and R in the second CT elec increase of the heterogeneity, roughness, and/or polycrystallinity of cycle with respect to the first cycle indicates that the Na insertion the active material might be the origin of the discrepancy between into the electroactive material is hindered and the P2-NFMO-LV the DFT and EIS results. COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 5 ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 Fig. 4 XPS spectra of P2-NFMO-LV electrodes. C 1s, O 1s, and F 1s photoemission lines of P2-NFMO-LV electrodes before cycling (pristine) and at + + + different SOC (first Na extraction, first Na insertion, and second Na extraction) during electrochemical cycling as highlighted in Fig. S1. capacity retention than the P2-NFMO-LV electrode. Therefore, in order to study the effect of the transport properties, and mainly the R , in the smallest operating voltage window, where the elec phase transition at high voltage is avoided (P2 solid-solution region, taken over the distorted P′2 phase at around 2.0 V), the same EIS methodology applied to the P2-NFMO-LV electrode has been carried out between 2.0 and 4.0 V (P2-NFMO-SV electrode). This approach helps to corroborate the effect of the high volume change of the “Z”/OP4-type structure in the trans- port properties and its role in the R increase. elec The Nyquist plot of the P2-NFMO-SV electrode at OCV (2.68 V vs Na /Na, Supplementary Fig. S7) shows a very similar lineshape to the one from the P2-NFMO-LV electrode (Fig. 1). In fact, three semicircles at the same frequency regions are observed: HF region above 5 kHz (red region), MF region in the 5 kHz–10 Hz range (green region), and LF region below 10 Hz (blue region) with a sloping line at the lowest frequencies. The impedance spectra of the P2-NFMO-SV electrode during + + first Na insertion and Na extraction at different SOC (see Nyquist plots of Supplementary Fig. S8) show that it follows the same behavior as the P2-NFMO-LV electrode. This means that Fig. 5 Fitted resistance values for the first two electrochemical cycles of the material becomes a better electronic conductor (R elec the P2-NFMO-SV electrode. R (red points), R (green triangle), and EEI CT + + decrease) during the Na extraction process, while, upon Na R (blue square). The structural evolution highlighted as colored regions elec insertion, the active material exhibits the opposite trend, and labeled upon electrochemical cycling is included in the figure for becoming a worse electronic conductor (see Fig. 5). Interestingly, 15, 20 guidance . R is more stable than in the P2-NFMO-LV electrode during elec the first two cycles, which most probably is due to the blocking of the “Z”/OP4-type structure formation. However, at ~2.0 V vs Na P2-Na [Fe Mn ]O cycled in the 2.0–4.0 V vs Na /Na 2/3 1/2 1/2 2 voltage window: P2-NFMO-SV electrode. One of the approa- /Na an increment of the LF semicircle size (Supplementary ches to enhance the capacity retention of layered oxides is by Fig. S8c, pink curve) and the R value (a yellow region in Fig. 5) elec controlling the operating voltage window, obtaining the best is observed, owing to the fact that, at this voltage, the phase results in the 2.0–4.0 V vs Na /Na window, as mentioned transition from ordered P2 to distorted P′2 occurs. The R is elec 22,24,29,59,71 above . Supplementary Fig. S6 shows that, after 45 reversible within this voltage range, indicating that the “Z”/OP4- cycles, the P2-NFMO-LV electrode delivers the same capacity as type structure formation is the main responsible for the resistance the P2-NFMO-SV electrode, although the initial capacity is increase/loos of the electronic conductivity. −1 45 mAh g higher. The capacity retention of both electrodes is R (green triangles of Fig. 5) increment is also more stable in CT illustrated in Supplementary Fig. S6b, and it can be clearly the P2-NFMO-SV electrode. The slight increase during the observed that the P2-NFMO-SV electrode exhibits higher electrochemical cycling, likewise for the P2-NFMO-LV electrode, 6 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE Fig. 6 Nyquist plots of P2-NFMO-LV and P2-NFMO-SV electrodes upon electrochemical cycling. P2-NFMO-LV at a 4.25 V and b 1.50 V and P2-NFMO- + st nd rd th th SV electrode at c 4.0 V and d 2.0 V (vs Na /Na) of the 1 (black square), 2 (red point), 3 (blue triangle), 4 (green inverted triangle), and 20 cycle (magenta hexagon). suggests that by avoiding the structural change at high voltage, electrodes (P2-NFMO-LV and P2-NFMO-SV), the overall resis- the formation of grain boundaries, mosaic texture, exfoliation, tance increases upon cycling. Nevertheless, the increment rate of and/or interfacial microstrains is reduced. the overall resistance in the P2-NFMO-SV electrode is much Finally, the R (red points of Fig. 5) trend is the same as for lower than for the P2-NFMO-LV electrode. EEI the P2-NFMO-LV electrode. The R drop observed at 2.1 V vs R of P2-NFMO-LV shows a remarkable increase in the first EEI elec + + Na /Na during the second Na extraction might be correlated four cycles. Meanwhile, the total resistance of the P2-NFMO-SV with slight changes in the EEI surface, also observed in the second electrode is one order of magnitude smaller in the fourth cycle Na insertion on the P2-NFMO-LV electrode. Even so, after even if the phase transition at low voltage (P2 to P′2) is not further cycling, the R stabilizes at a constant value. avoided. Furthermore, in the twentieth cycle, the total resistance EEI Henceforth, avoiding the “Z”/OP4-type structure formation by of the P2-NFMO-SV electrode is still lower than that of the P2- reducing the operating voltage window, which is one of the NFMO-LV electrode in the fourth cycle. strategies (note that by chemical tunning also the high voltage Therefore, although it is considered that the structural evolution 30,31,33,34,39,40 transition can be avoided) , the electrode displays a of P2-Na [Fe Mn ]O is reversible, the XRD data acquired 2/3 1/2 1/2 2 high bulk electronic conductivity with stable R values. The after more than 75 cycles of the P2-NFMO-LV electrode clearly elec electronic conductivity loos due to the “Z”/OP4-type structure show that the full-width at half-maximum (FWHM) is drastically formation results due to the huge volume change that isolates the increasing upon electrochemical cycling (Supplementary Fig. S9). active particles from the conducting carbon and/or current The P2-Na [Fe Mn ]O (002) reflection of the cycled electrode 2/3 1/2 1/2 2 collector, while triggering strong internal stress. In turn, the until 4.25 V (>75 cycles) exhibits a much larger FWHM value than electronic conductivity is lost and/or reduced , which ultimately the pristine one (0.23° vs 0.09°). These phase transitions induce a influences the capacity fading. [TMO ]-gliding, which for the P2- to OP4-phase transition, the interlayer distance is reduced from 5.65 to 5.30–5.05 Å, while in the structural change at 2.0 V (P2 to P′2) the difference is 0.10 Å lower Further cycling of P2-NFMO-LV and P2-NFMO-SV electrodes. for P′2-type . As it was mentioned, these phase transitions lead to The impedance spectra have also been recorded after 1, 2, 3, 4, an unavoidable volume change and a significant increase of the and 20 cycles for both electrodes (Fig. 6). Although the fits interfacial microstrain, staking faults formation, and/or exfoliation become more complicated after long cycling because of the large of thelayersasconfirmed by previous reports on the grounds of semicircle developed at LF (highlighted in blue in Fig. 6) which 25,26,61 SEM/TEM experiments . These structural modifications are overlaps with the contribution from other processes at higher the responsibility of the decrease of electronic conductivity and frequencies (i.e., R and R ), it can be observed that, in both CT EEI COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 7 ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 greatly contribute to the reported capacity loss. On the other hand, GF/D) as separator and 1 M NaPF in EC:DEC in a 1:1 wt% (Acros Organics) as electrolyte. The experiments were carried out in a Maccor Series 4000 battery tester although in the first cycles the phase transition at low voltage (from in two operating voltage windows: (i)1.5–4.25 V (P2-NFMO-LV) and (ii) P2 to P′2) is reversible, it is clearly observed that upon further cycle + −1 2.0–4.0 V (P2-NFMO-SV) vs Na /Na, at 0.05C C-rate (1C = 260 mA g , which (see Fig. 6c, d), the overall resistance of the electrode also increases + theoretically corresponds to the exchange of one Na ). EIS experiments were (lower than in P2-NFMO-LV) that at long-term cycling will be an performed in three electrode Swagelok-T-type cells. Metallic sodium disks were employed as counter and reference electrodes, and the same electrolyte as for the issue, also contributing in the capacity fading. Hence, avoiding the galvanostatic experiments was used (1 M NaPF in EC:DEC in a 1:1 wt%). The phase evolution, for example by using a reduced operating voltage experiments were carried out at room temperature in the same two operating window (2.0–4.0 V vs Na /Na), better capacity retention is voltage windows used for the galvanostatic tests: (i) 1.5–4.25 V (P2-NFMO-LV) obtained, although lower initial capacity values are delivered and (ii) 2.0–4.0 V (P2-NFMO-SV) vs Na /Na in a VMP3 potentiostat (Bio-Logic) through potentiostatic intermittent titration technique (PITT) where the potential (Supplementary Fig. S6), due to the transport properties maintain of the electrode was controlled with a 45 mV step scan. A sinusoidal perturbation stable. of 5 mV was applied in the frequency range of 100 kHz–5 mHz and before EIS data were taken a 4 h of equilibrium condition was set at a constant potential. Impe- dance dispersion data (133 for MFMO-LV and 76 for NFMO-SV impedance Conclusions spectra in total) were fitted by Boukamp’s Equivalent Circuit software . This EIS experiments show that the responsible, in terms of transport model is based on the following parameters: (i)Na resistance across the elec- properties, of the poor capacity retention of P2-Na [Fe Mn ] 2/3 1/2 1/2 trolyte (R ), (ii) resistance and capacitance of the EEI (R and C ), (iii) R sol EEI EEI CT O layered oxide cathode material is the formation of the “Z”/ 2 and C ,(iv) R and C ,(v) Warburg diffusion (Z ) related to the solid-state DL elec elec w OP4 phase that induces the loss of the bulk electronic con- diffusion of Na , and finally (vi) intercalation capacity (C ) due to the charge 51,55,56 accumulation (see top of Supplementary Fig. S12) . The resistance and ductivity of the electrode. DFT calculations rule out the possibility capacitance of each process are connected in parallel. Additionally, in order to take that such observation is due to significant changes in the intrinsic into account any deviation from an ideal material such as surface inhomogeneity, electronic structure of the bulk materials during cycling. Since the roughness or degree of polycrystallinity, the C elements and Z have been replaced XPS studies show that the EEI is rather stable and the DFT results by CPE. In order to illustrate the goodness of the fits, some fitted EIS spectra are shown as examples in Supplementary Fig. S12a (P2-NFMO-LV electrode) and suggest that no significant changes should be expected for dif- S12b (P2-NFMO-SV electrode), while the obtained values from the fit are collected ferent orderings of this structure, such bulk electron conductivity in Supplementary Tables S2 and S3, respectively. loss is most probably due to the large volume changes occurring in the active material when cycling above 4.0 V vs Na /Na where Characterization of EEI. The EEI was studied, by means of XPS using a Phoibos a P2- to “Z”/OP4-type phase transition occurs. These volume 150 spectrometer (Specs GmbH), at several SOC (cycled at 0.05C): pristine, OCV, changes lead to interfacial microstrain, staking faults formation, st + nd + 1 Na extraction and insertion and 2 Na extraction states as highlighted in the and/or exfoliation of the layers that, besides reducing the bulk galvanostatic profile of Supplementary Fig. S1. The EEI composition and stability are followed by analyzing the C 1s, O 1s, F 1s, and Na 1s photoemission lines. Since electron conductivity, lead to the progressive isolation of the the Auger peaks and photoelectron peaks were overlapped in some cases, two active particle with respect to the conducting carbon and current different X-ray sources were used: a non-monochromatic Mg K source (hν = collector. When the high cut-off voltage is set to avoid the 1253.6 eV), which was applied for C 1s and O 1s photoelectron peaks and a non- structural change, the electronic resistance is maintained since the monochromatic Al K source (hν = 1486.6 eV) for all photoemission lines (C 1s, F 1s, and Na 1s). The electrodes were stopped at the selected SOC, rinsed with DEC, [TMO ]-layers are not glided, hence, the P2-Na [Fe Mn ]O 2 2/3 1/2 1/2 2 and dried before being inserted into the XPS vacuum chamber by means of an electrode exhibits much better structural reversibility, as well as argon-filled transfer system, never exposing the electrodes to air. High-resolution reversible resistance values and more stable transport properties scans at low potential were acquired at 100 W, 20 eV pass energy, and 0.1 eV and in turn better capacity retention. Such results might be energy step. Charging effects were compensated by flood gun at 10 μA and 1.5 V in order to correct slight variations of the binding energy . Calibration of the binding extrapolated to other P2-layered oxides which, despite exhibiting energy was performed using the C 1s graphitic signal as a reference at 284.4 eV. good theoretical capacity when cycled up to 4.3 V vs Na /Na, The recorded spectra were fitted by CasaXPS software using a nonlinear Shirley- they should not be used as a high voltage cathode electrodes in type background and a Voigt profile (70% Gaussian and 30% Lorentzian) . The order to keep good capacity retention and excellent cycle life for composition of the EEI and subsurface region has also been evaluated by means of the SIB. FTIR using an Agilent Technologies Cary 630 benchtop system placed inside an inert atmosphere glove box. The electrodes were rinsed in DEC prior to FTIR experiments to remove electrolyte salt traces. Methods Synthesis of the active cathode material. P2-Na [Fe Mn ]O was synthe- 2/3 1/2 1/2 2 Theoretical methods. DFT calculations have been performed within the Vienna sized by the solid-state method. First, stoichiometric amounts of Na CO ·H O 2 3 2 75,76 (99.5%, Sigma Aldrich), Fe O (99%, Alfa Aesar), and Mn O (98%, Alfa Aesar) ab initio simulation package , employing the PBE functional. In order to 2 3 2 3 describe the localized nature of Fe and Mn 3d states, the DFT + U scheme of were mixed in a mortar for 2 h. The mixed powder was compressed in pellets and heated up to 900 °C for 12 h under air atmosphere followed by liquid nitrogen Dudarev et al. has been applied, with U = 4.0 eV for Fe and Mn atoms as 16 78,79 quenching . The obtained sample was transferred and stored in an argon-filled suggested in the literature . All calculations were spin-polarized, starting from a glove box (MBraun, H O and O < 1 ppm) in order to avoid any contact with the high-spin configuration. PBE-based projector augmented wave potentials were 2 2 atmosphere. used to replace core electrons, whereas we treated explicitly the Na (3s), Fe (3p, 3d, 4s), Mn (3p, 3d, 4s), and O (2s, 2p) electrons as valence electrons and their wave- functions were expanded in plane-waves with cut-off energy of 600 eV. The irre- Structural and morphological characterization. The structural characterization ducible Brillouin zone was sampled using a Monkhorst–Pack grid with of the synthesized active cathode material was performed by powder XRD using a 7 × 7 × 1 k-point sampling per (1 × 1) unit cell. Ground state energies for every Bruker Advance D8 instrument with copper radiation (Cu Kα λ = 1.5406 Å, 1,2 configuration were computed allowing the lattice parameters, cell shape, and 1.5444 Å). The powder XRD pattern was refined by the Le Bail method and shows −1 atomic positions to relax with a residual force threshold of 0.02 eV Å . It is known a pure P2-type structure with …AABBAA… staking sequence (Supplementary that in binary or ternary transition metal P2-compounds, the presence of Mn Fig. S10). The morphology was analyzed using SEM (Quanta 200 FEG-FEI model) + 42,69,82 atoms can suppress long-range Na /vacancy orderings . For the P2-struc- operated at 30 kV. The SEM images show that most particles crystalize as hex- ture, the Na /vacancy ordering used was previously found for a range of other P2- agonal platelets of 5–10 µm of diameter (Supplementary Fig. S11). + 83 84 phases with 2/3 Na content: P2-Na CoO , P2-Na Fe Mn O , and P2- 2/3 2 2/3 2/3 1/3 2 Na Mn Ni O . This ordering consists of a zig-zag pattern of Na1 and 2/3 2/3 1/3 2 Electrochemical characterization. Electrodes were prepared by mixing 80% of Na2 sites arranged in triangular units, involving a supercell with in-plane lattice P2-Na [Fe Mn ]O active material, 10% carbon Super C65 (Timcal C-Nergy- vectors a = 3a + 2b and b = 2a + 3b, containing 24 formula units with 16 2/3 1/2 1/2 2 super super TM), and 10% PVdF (Solef® Arkema Group) dissolved in N-methyl-2-pyrrolidone Na, 12 Fe (III), 12 Mn (III) and 8 Mn (IV) and 48 O atoms. On the other hand, (NMP—Sigma Aldrich). The slurry was cast on battery-grade aluminum foil and three different Na /vacancy orderings were considered for the O2-structure, one dried under vacuum overnight at 120 °C. 11 mm electrodes were pressed at 5 tons with Na atoms in rows, the other in a hexagonal ordering, and the last one mixing for 1 min before assembling cells in an argon-filled glove box (MBraun, H O and both of them. In this case, we have used a supercell containing 36 formula units O < 1 ppm). The galvanostatic experiments were carried out in CR2032 type coin- with 8 Na, 8 Fe (III), 10 Fe (IV), 18 Mn (IV), and 72 O atoms. All of them are in a cells, using P2-Na [Fe Mn ]O electrodes as a working electrode and metallic high spin configuration, with a total magnetic moment of 100.0 and 134.0 for the 2/3 1/2 1/2 2 sodium disk (99.8% Acros Organics) as a counter electrode, glass fiber (Whatman P2- and O2- structures, respectively. The Na /vacancy ordering assumed in this 8 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE work should be considered as a low-energy, reasonable approximation, since from 20. Singh, G., López Del Amo, J. M., Galceran, M., Pérez-Villar, S. & Rojo, T. a computational viewpoint, the modeling of such arrangements would require large Structural evolution during sodium deintercalation/intercalation in Na [Fe 2/3 1/ supercells which are unaffordable at the DFT level. For the Fe/Mn ordering in both Mn ]O . J. Mater. Chem. A 3, 6954–6961 (2015). 2 1/2 2 oxides layers, we have thoroughly sampled a range of local structures within the 21. Mortemard De Boisse, B., Carlier, D., Guignard, M., Bourgeois, L. & Delmas, considered supercell. In particular, for each investigated Na /vacancy orderings, 50 C. P2-Na Mn Fe O phase used as positive electrode in Na batteries: x 1/2 1/2 2 different configurations were randomly generated and computed their DFT energy, Structural changes induced by the electrochemical (De)intercalation process. in order to choose the most stable ones. The magnetic moment was used to Inorg. Chem. 53, 11197–11205 (2014). distinguish the oxidation state of Mn and Fe. Finally, the DOS was calculated, to 22. Pang, W. K. et al. Interplay between electrochemistry and phase evolution of obtain the bandgap of the structures. In order to accurately determine Fermi the P2-type Na (Fe Mn )O cathode for use in sodium-ion batteries. Chem. x 1/2 1/2 2 energies, single-point DOS calculations were performed with twice the k-point Mater. 27, 3150–3158 (2015). density 14 × 14 × 2 per (1 × 1) unit cell as compared to geometrical optimization 23. Bai, Y. et al. Enhanced sodium ion storage behavior of P2-Type Na Fe 2/3 1/ calculations. Mn O synthesized via a chelating agent assisted route. ACS Appl. Mater. 2 1/2 2 Interfaces 8, 2857–2865 (2016). 24. Talaie, E., Duffort, V., Smith, H. L., Fultz, B. & Nazar, L. F. Structure of the Data availability high voltage phase of layered P2-Na [Mn Fe ]O and the positive effect 2/3−z 1/2 1/2 2 The optimized atomic coordinates of the P2- and O2-phases are included in the of Ni substitution on its stability. Energy Environ. Sci. 8, 2512–2523 (2015). Supplementary Data 1 and 2 files, respectively. Restrictions apply to the availability of the 25. Alvarado, J. et al. Improvement of the cathode electrolyte interphase on P2- datasets generated during and/or analyzed during the current study, and so are not Na Ni Mn O by atomic layer deposition. ACS Appl. Mater. Interfaces 9, 2/3 1/3 2/3 2 publicly available. Data are however available from the authors upon reasonable request 26518–26530 (2017). and with permission of CIC energiGUNE and HIU Batteries. 26. Liu, Y. et al. Layered P2-Na [Ni Mn ]O as high-voltage cathode for 2/3 1/3 2/3 2 sodium-ion batteries: the capacity decay mechanism and Al O surface 2 3 Code availability modification. Nano Energy 27,27–34 (2016). Computer codes used during the current study are available from Dr. O. Lakuntza or 27. Wang, S., Sun, C., Wang, N. & Zhang, Q. Ni- and/or Mn-based layered Dr. J. Carrasco upon reasonable request and with permission of CIC energiGUNE. transition metal oxides as cathode materials for sodium ion batteries: Status, challenges and countermeasures. J. Mater. Chem. A 7, 10138–10158 (2019). 28. Li, Z. Y. et al. Unveiling the role of Co in improving the high-rate capability Received: 9 July 2021; Accepted: 12 January 2022; and cycling performance of layered Na Mn Ni Co O cathode 0.7 0.7 0.3−x x 2 materials for sodium-ion batteries. ACS Appl. Mater. Interfaces 8, 15439–15448 (2016). 29. Talaie, E., Kim, S. Y., Chen, N. & Nazar, L. F. Structural evolution and redox processes involved in the electrochemical cycling of P2- Na [Mn Fe Cu ]O . Chem. Mater. 29, 6684–6697 (2017). 0.67 0.66 0.20 0.14 2 References 30. Tapia-Ruiz, N. et al. High voltage structural evolution and enhanced Na-ion 1. Delmas, C., Branconnier, J.-J., Fouassier, C. & Hagenmuller, P. diffusion in P2-Na Ni Mg Mn O (0 ≤ x ≤ 0.2) cathodes from 2/3 1/3−x x 2/3 2 Electrochemical intercalation of sodium in Na CoO bronzes. Solid State Ion. x 2 diffraction, electrochemical and ab initio studies. Energy Environ. Sci. 11, 4, 165–169 (1981). 1470–1479 (2018). 2. Maazaz, A., Delmas, C. & Hagenmuller, P. A study of the NaxTiO2 system by 31. Kumakura, S., Tahara, Y., Sato, S., Kubota, K. & Komaba, S. P′2-Na 2/ electrochemical deintercalation. J. Incl. Phenom. 1,45–51 (1983). Mn Me O (Me = Mg, Ti, Co, Ni, Cu, and Zn): correlation between 3 0.9 0.1 2 3. Bauer, A. et al. The scale-up and commercialization of nonaqueous Na-ion orthorhombic distortion and electrochemical property. Chem. Mater. 29, battery technologies. Adv. Energy Mater. 8,1–13 (2018). 8958–8962 (2017). 4. Li, Y. et al. Advanced sodium-ion batteries using superior low cost pyrolyzed 32. Liu, Z. et al. Ultralow volume change of P2-type layered oxide cathode for Na- anthracite anode: towards practical applications. Energy Storage Mater. 5, ion batteries with controlled phase transition by regulating distribution of Na 191–197 (2016). +. Angew. Chem. Int. Ed. 60, 20960–20969 (2021). 5. Goikolea, E. et al. Na-ion batteries—approaching old and new challenges. Adv. 33. Liu, Y. et al. Mitigation of Jahn-Teller distortion and Na+/vacancy ordering in Energy Mater. 10,1–21 (2020). a distorted manganese oxide cathode material by Li substitution. Chem. Sci. 6. Muñoz-Márquez, M. Á. et al. Na-ion batteries for large scale applications: a 12, 1062–1067 (2021). review on anode materials and solid electrolyte interphase formation. Adv. 34. Wang, C. et al. Tuning local chemistry of P2 layered-oxide cathode for high Energy Mater. 7, 1700463 (2017). energy and long cycles of sodium-ion battery. Nat. Commun. 12, 2256 (2021). 7. Larcher, D. & Tarascon, J. M. Towards greener and more sustainable batteries 35. Li, Y. et al. Air-stable copper-based P2-Na Cu Fe Mn O as a new 7/9 2/9 1/9 2/3 2 for electrical energy storage. Nat. Chem. 7,19–29 (2015). positive electrode material for sodium-ion batteries. Adv. Sci. 2,1–7 (2015). 8. Vaalma, C., Buchholz, D., Weil, M. & Passerini, S. A cost and resource analysis 36. Park, J. K., Park, G. G., Kwak, H. H., Hong, S. T. & Lee, J. W. Enhanced rate of sodium-ion batteries. Nat. Rev. Mater. 3, 18013 (2018). capability and cycle performance of titanium-substituted P2-Type 9. Zhao, L. N., Zhang, T., Zhao, H. L. & Hou, Y. L. Polyanion-type electrode Na Fe Mn O as a cathode for sodium-ion batteries. ACS Omega 3, 0.67 0.5 0.5 2 materials for advanced sodium-ion batteries. Mater. Today Nano 10, 100072 361–368 (2017). (2020). 37. Singh, G. et al. High voltage Mg-doped Na Ni Mg Mn O (x = 0.05, 0.67 0.3−x x 0.7 2 10. Jin, T. et al. Polyanion-type cathode materials for sodium-ion batteries. Chem. 0.1) Na-ion cathodes with enhanced stability and rate capability. Chem. Mater. Soc. Rev. 49, 2342–2377 (2020). 28, 5087–5094 (2016). 11. Rajagopalan, R., Tang, Y., Jia, C., Ji, X. & Wang, H. Understanding the sodium 38. Zarrabeitia, M. et al. Unraveling the role of Ti in the stability of positive storage mechanisms of organic electrodes in sodium ion batteries: Issues and layered oxide electrodes for rechargeable Na-ion batteries. J. Mater. Chem. A solutions. Energy Environ. Sci. 13, 1568–1592 (2020). 7, 14169–14179 (2019). 12. Hurlbutt, K., Wheeler, S., Capone, I. & Pasta, M. Prussian blue analogs as 39. Han, M. H. et al. High-performance P2-phase Na Mn Fe Ti O cathode 2/3 0.8 0.1 0.1 2 battery materials. Joule 2, 1950–1960 (2018). material for ambient-temperature sodium-ion batteries. Chem. Mater. 28, 13. Liu, Q. et al. Sodium transition metal oxides: the preferred cathode 106–116 (2016). choice for future sodium-ion batteries? Energy Environ. Sci. 14, 158–179 40. Wang, L. et al. Copper-substituted Na Ni Cu Mn O cathode 0.67 0.3−x x 0.7 2 (2021). materials for sodium-ion batteries with suppressed P2-O2 phase transition. J. 14. Gonzalo, E., Zarrabeitia, M., Drewett, N. E., López del Amo, J. M. & Rojo, T. Mater. Chem. A 5, 8752–8761 (2017). Sodium manganese-rich layered oxides: Potential candidates as positive 41. Xu, J., Chou, S. L., Wang, J. L., Liu, H. K. & Dou, S. X. Layered P2- electrode for sodium-ion batteries. Energy Storage Mater. 34, 682–707 (2021). Na Fe Mn O cathode material for rechargeable sodium-ion batteries. 0.66 0.5 0.5 2 15. Liu, Z. et al. Recent progress of P2-type layered transition-metal oxide ChemElectroChem 1, 371–374 (2014). cathodes for sodium-ion batteries. Chemistry 26, 7747–7766 (2020). 42. Liu, L. et al. High-performance P2-type Na (Mn Fe Co )O cathode 2/3 1/2 1/4 1/4 2 16. Yabuuchi, N. et al. P2-type Na [Fe Mn ]O made from earth-abundant x 1/2 1/2 2 material with superior rate capability for Na-ion batteries. Adv. Energy Mater. elements for rechargeable Na batteries. Nat. Mater. 11, 512–517 (2012). 5,1–5 (2015). 17. Yabuuchi, N., Kubota, K., Dahbi, M. & Komaba, S. Research development on 43. Muñoz-Márquez, M. A. et al. Composition and evolution of the solid- sodium-ion batteries. Chem. Rev. 114, 11636–11682 (2014). electrolyte interphase in Na Ti O electrodes for Na-Ion batteries: XPS 2 3 7 18. Sayers, R., Barker, J. & Heap, R. WO2015177544 (A1) (2014). and auger parameter analysis. ACS Appl. Mater. Interfaces 7,7801–7808 19. Kubota, K., Kumakura, S., Yoda, Y., Kuroki, K. & Komaba, S. Electrochemistry (2015). and solid-state chemistry of NaMeO (Me = 3d transition metals). Adv. 44. Doubaji, S. et al. Passivation layer and cathodic redox reactions in sodium-ion Energy Mater. 8,1–30 (2018). batteries probed by HAXPES. ChemSusChem 9,97–108 (2016). COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 9 ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 45. Zarrabeitia, M. et al. Toward stable electrode/electrolyte interface of P2- 71. Yoshida, J. et al. New P2-Na Mn Ni Co O layered oxide as electrode 0.70 0.60 0.30 0.10 2 layered oxide for rechargeable Na-ion batteries. ACS Appl. Mater. Interfaces material for Na-ion batteries. J. Electrochem. Soc. 161, A1987–A1991 (2014). 11, 28885–28893 (2019). 72. Boukamp, B. A. A Nonlinear Least Squares Fit procedure for analysis of 46. Zarrabeitia, M., Casas-Cabanas, M. & Muñoz-Márquez, M. Á. Understanding immittance data of electrochemical systems. Solid State Ion. 20,31–44 (1986). the electrode—electrolyte interphase of high voltage positive electrode 73. Wagner, C. D., Riggs, W. M., Davis, L. E. & Moulder, J. F. Handbook of X-ray Na Co (PO ) P O for rechargeable sodium-ion batteries. Electrochim. Acta Photoelectron Spectroscopy. (Perkin-Elmer Corporation. Physical Electronics 4 3 4 2 2 7 372, 137846 (2021). Division, 1979). 47. Sathiya, M., Hemalatha, K., Ramesha, K., Tarascon, J.-M. & Prakash, A. S. 74. Walton, J., Wincott, P., Fairley, N. & Carrick, A. Peak Fitting with CasaXPS: A Synthesis, structure, and electrochemical properties of the layared sodium Casa Pocket Book (2010). insertion cathode material: NaNi Mn Co O . Chem. Mater. 24, 75. Kresse, G. & Hafner, J. Ab initio molecular dynamics for liquid metals. Phys. 1/3 1/3 1/3 2 1846–1853 (2012). Rev. B 47, 558–561 (1993). 48. Zarrabeitia, M., Muñoz-Márquez, M., Nobili, F., Rojo, T. & Casas-Cabanas, M. 76. Kresse, G. & Furthmüller, J. Efficiency of ab-initio total energy calculations for Influence of using metallic na on the interfacial and transport properties of metals and semiconductors using a plane-wave basis set. Comput. Mater. Sci. Na-ion batteries. Batteries 3, 16 (2017). 6,15–50 (1996). 49. Conder, J. & Villevieille, C. How reliable is the Na metal as a counter electrode 77. Dudarev, S. & Botton, G. Electron-energy-loss spectra and the structural in Na-ion half cells? Chem. Commun. 55, 1275–1278 (2019). stability of nickel oxide: An LSDA + U study. Phys. Rev. B 57, 1505–1509 50. Pfeifer, K. et al. Can metallic sodium electrodes affect the electrochemistry of (1998). sodium‐ion batteries? Reactivity Issues and Perspectives. ChemSusChem 12, 78. Lee, D. H., Xu, J. & Meng, Y. S. An advanced cathode for Na-ion batteries with 3312–3319 (2019). high rate and excellent structural stability. Phys. Chem. Chem. Phys. 15, 51. Barsoukov, E. & Macdonald, J. R. Impedance Spectroscopy Theory Experiment 3304–3312 (2013). and Applications. (John Wiley & Sons, Inc, 2005). 79. Hautier, G., Ong, S. P., Jain, A., Moore, C. J. & Ceder, G. Accuracy of density 52. Zhuang, S. S., Qiu, Q. C. & Xu, X. Y. Diagnosis of Electrochemical Impedance functional theory in predicting formation energies of ternary oxides from Spectroscopy in Lithium-Ion Batteries. Lithium Ion Batteries-New Development. binary oxides and its implication on phase stability. Phys. Rev. B 85, 155208 Ilias Belharouak, IntechOpen, https://doi.org/10.5772/26749 (2012). (2012). 53. Nobili, F., Tossici, R., Marassi, R., Croce, F. & Scrosati, B. An AC impedance 80. Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave spectroscopic study of Li CoO at different temperatures. J. Phys. Chem. B method. Phys. Rev. B 59, 1758–1775 (1999). x 2 106, 3909–3915 (2002). 81. Monkhorst, H. & Pack, J. D. Special points for Brillouin-zone integrations. 54. Nobili, F., Dsoke, S., Minicucci, M., Croce, F. & Marassi, R. Correlation of Ac- Phys. Rev. B 13, 5188–5192 (1976). impedance and in situ X-ray spectra of LiCoO . J. Phys. Chem. B 110, 82. Wang, X., Tamaru, M., Okubo, M. & Yamada, A. Electrode properties of P2- 11310–11313 (2006). Na Mn Co O as cathode materials for sodium-ion batteries. J. Phys. 2/3 y 1−y 2 55. Levi, M. D. & Aurbach, D. Simultaneous measurements and modeling of the Chem. C 117, 15545–15551 (2013). electrochemical impedance and the cyclic voltammetric characteristics of 83. Platova, T. A., Mukhamedshin, I. R., Alloul, H., Dooglav, A. V. & Collin, G. graphite electrodes doped with lithium. J. Phys. Chem. B 101,4630–4640 (1997). Nuclear quadrupole resonance and X-ray investigation of the structure of 56. Aurbach, D. et al. Common electroanalytical behavior of li intercalation Na CoO . Phys. Rev. B 80, 224106 (2009). 2/3 2 processes into graphite and transition metal oxides. J. Electrochem. Soc. 145, 84. Katcho, N. A. et al. Origins of bistability and Na ion mobility difference in P2- 3024–3034 (1998). and O3-Na Fe Mn O cathode polymorphs. Adv. Energy Mater. 7,1–9 2/3 2/3 1/3 2 57. Croce, F. et al. An electrochemical impedance spectroscopic study of the transport (2017). properties of LiNi Co O . Electrochem. Commun. 1, 605–608 (1999). 0.75 0.25 2 58. Nobili, F. et al. Sol-gel synthesis and electrochemical characterization of Mg-/ Acknowledgements Zr-doped LiCoO cathodes for Li-ion batteries. J. Power Sources 197, 276–284 M.Z. thanks the Government of the Basque Country for Ph.D. funding through a Pre- (2012). doctoral fellowship and her stage at the University of Camerino by “EGONLABUR” 59. Hasa, I., Passerini, S. & Hassoun, J. Toward high energy density cathode Fellowship. B. Acebedo and M. Jauregui are acknowledged for their technical support materials for sodium-ion batteries: investigating the beneficial effect of with material synthesis and powder XRD measurements. O.L. thanks J.X Lian for his aluminum doping on the P2-type structure. J. Mater. Chem. A 5, 4467–4477 insight into generating the DOS graphs. Financial support from the Basque Government (2017). (Elkartek20 CIC energiGUNE) and from the Ministerio de Economía y Competitividad of 60. Zarrabeitia, M., Nobili, F., Muñoz-Márquez, M. Á., Rojo, T. & Casas-Cabanas, the Spanish Government (ENE2013-44330-R) is also acknowledged. M. Direct observation of electronic conductivity transitions and solid electrolyte interphase stability of Na Ti O electrodes for Na-ion batteries. J. 2 3 7 Power Sources 330,78–83 (2016). Author contributions 61. Xu, G. L. et al. Insights into the structural effects of layered cathode materials for M.Z.: Investigation, methodology, impedance, electrochemical and XPS data analysis, high voltage sodium-ion batteries. Energy Environ. Sci. 10, 1677–1693 (2017). visualization, funding acquisition, writing—original draft & editing. F.N.: EIS data 62. Blyth, R. I. R. et al. XPS studies of graphite electrode materials for lithium ion supervision, EIS methodology. O.L.: DFT calculations. J.C.: DFT data supervision, writing batteries. Appl. Surf. Sci. 167,99–106 (2000). —original draft. T.R.: Project administration, writing—review, funding acquisition. 63. Beamson, G. & Briggs, D. High resolution XPS of organic polymers: the M.C.C.: Project administration, supervision, writing—review draft. M.A.M.M.: Con- scienta ESCA300 database. J. Chem. Educ. 70, A25 (1993). ceptualization, methodology, investigation, formal analysis, funding acquisition, project 64. Andersson, A. M., Henningson, A., Siegbahn, H., Jansson, U. & Edström, K. administration, supervision, writing—review and editing. Electrochemically lithiated graphite characterised by photoelectron spectroscopy. J. Power Sources 119–121, 522–527 (2003). 65. Vogdanis, L. & Heitz, W. Carbon dioxide as a refrigerant. J. Frankl. Inst. 196, Competing interests 713 (1923). The authors declare no competing interests. 66. Schroder, K. W., Celio, H., Webb, L. J. & Stevenson, K. J. Examining solid electrolyte interphase formation on crystalline silicon electrodes: influence of Additional information electrochemical preparation and ambient exposure conditions. J. Phys. Chem. Supplementary information The online version contains supplementary material C 116, 19737–19747 (2012). available at https://doi.org/10.1038/s42004-022-00628-0. 67. Rezvani, S. J. et al. SEI dynamics in metal oxide conversion electrodes of Li-ion batteries. J. Phys. Chem. C 121, 26379–26388 (2017). Correspondence and requests for materials should be addressed to Miguel Ángel 68. Rezvani, S. J. et al. Does alumina coating alter the solid permeable interphase Muñoz-Márquez. dynamics in LiMn O cathodes? J. Phys. Chem. C 124, 26670–26677 (2020). 2 4 69. Dahbi, M. et al. Effect of hexafluorophosphate and fluoroethylene carbonate Peer review information Communications Chemistry thanks the anonymous reviewers on electrochemical performance and the surface layer of hard carbon for for their contribution to the peer review of this work. sodium-ion batteries. ChemElectroChem 3, 1856–1867 (2016). 70. Bhide, A., Hofmann, J., Katharina Dürr, A., Janek, J. & Adelhelm, P. Reprints and permission information is available at http://www.nature.com/reprints Electrochemical stability of non-aqueous electrolytes for sodium-ion batteries and their compatibility with Na CoO . Phys. Chem. Chem. Phys. 16, 0.7 2 Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in 1987–1998 (2014). published maps and institutional affiliations. 10 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/ licenses/by/4.0/. © The Author(s) 2022 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 11 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Communications Chemistry Springer Journals

Role of the voltage window on the capacity retention of P2-Na2/3[Fe1/2Mn1/2]O2 cathode material for rechargeable sodium-ion batteries

Loading next page...
 
/lp/springer-journals/role-of-the-voltage-window-on-the-capacity-retention-of-p2-na2-3-fe1-Ot9SNpmNCo

References (94)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2022
eISSN
2399-3669
DOI
10.1038/s42004-022-00628-0
Publisher site
See Article on Publisher Site

Abstract

ARTICLE https://doi.org/10.1038/s42004-022-00628-0 OPEN Role of the voltage window on the capacity retention of P2-Na [Fe Mn ]O cathode 2/3 1/2 1/2 2 material for rechargeable sodium-ion batteries 1,5,6 2 1 1 1,3 Maider Zarrabeitia , Francesco Nobili , Oier Lakuntza , Javier Carrasco , Teófilo Rojo , 1,4 1,7 Montse Casas-Cabanas & Miguel Ángel Muñoz-Márquez P2-Na [Fe Mn ]O layered oxide is a promising high energy density cathode material 2/3 1/2 1/2 2 for sodium-ion batteries. However, one of its drawbacks is the poor long-term stability in the operating voltage window of 1.5–4.25 V vs Na /Na that prevents its commercialization. In this work, additional light is shed on the origin of capacity fading, which has been analyzed using a combination of experimental techniques and theoretical methods. Electrochemical impedance spectroscopy has been performed on P2-Na [Fe Mn ]O half-cells operat- 2/3 1/2 1/2 2 ing in two different working voltage windows, one allowing and one preventing the high voltage phase transition occurring in P2-Na [Fe Mn ]O above 4.0 V vs Na /Na; so as 2/3 1/2 1/2 2 to unveil the transport properties at different states of charge and correlate them with the existing phases in P2-Na [Fe Mn ]O . Supporting X-ray photoelectron spectroscopy 2/3 1/2 1/2 2 experiments to elucidate the surface properties along with theoretical calculations have concluded that the formed electrode-electrolyte interphase is very thin and stable, mainly composed by inorganic species, and reveal that the structural phase transition at high voltage from P2- to “Z”/OP4-oxygen stacking is associated with a drastic increased in the bulk electronic resistance of P2-Na [Fe Mn ]O electrodes which is one of the causes of the 2/3 1/2 1/2 2 observed capacity fading. Centre for Cooperative Research on Alternative Energies (CIC energiGUNE), Basque Research and Technology Alliance (BRTA), Alava Technology Park, Albert Einstein 48, 01510 Vitoria-Gasteiz, Spain. School of Science and Technology—Chemistry Division, University of Camerino, Via Madonna delle Carceri, 62032 Camerino, Italy. Departamento de Química Inorgánica, Universidad del País Vasco UPV/EHU, P.O. Box 664, 48080 Leioa, Spain. IKERBASQUE, Basque Foundation for Science, María Díaz de Haro 3, 48013 Bilbao, Spain. Present address: Helmholtz Institute Ulm (HIU), 6 7 Helmholtzstrasse 11, 89081 Ulm, Germany. Present address: Karlsruhe Institute of Technology (KIT), P.O. Box 3640, 76021 Karlsruhe, Germany. Present address: School of Science and Technology—Chemistry Division, University of Camerino, Via Madonna delle Carceri, 62032 Camerino, Italy. email: miguel.munoz@unicam.it COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 1 1234567890():,; ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 + + eversible extraction/insertion of Na into host structures (Na p.f.unit = 0.8), the P2-phase coexists with a new distorted was already demonstrated in the 1980s with layered P’2-phase (S.G: Cmcm) . In addition, the described structural 1,2 Roxides . Nowadays, pushed by the increasing need for mechanism of P2-Na [Fe Mn ]O is not affected by the elec- 2/3 1/2 1/2 2 16,20–22 more efficient low-cost energy storage devices, sodium-ion bat- trolyte used . On the other hand, ex-situ and in-situ XRD teries (SIBs) are becoming an alternative for large-scale applica- show that all the structural transformations are reversible , 3–5 tions and light electromobility . This is nested in the fact that although a slight broadening of the peaks and the gradual intensity sodium precursors are evenly distributed in the Earth’s crust and decrease in the second cycle suggest that the crystallinity of the 6–8 21 are cheaper and more abundant than lithium ones . Due to material is reduced upon electrochemical cycling .Indeed,ex- their similar chemical properties, many lithium-based analog situ electronic microscopy analysis —scanning electronic micro- electrode materials have been proposed as cathodes for SIBs . scopy (SEM) and transmission electronic microscopy (TEM) Therefore, a wide variety of sodium-based cathode materials have — clearly shows the exfoliation of the P2-type layered oxides after been studied, including polyanionic materials (phosphates, pyr- cycling, which has been attributed as one of the causes of the poor 25,26 ophosphates, and mixed polyanions), organic compounds, Prus- capacity retention of P2-Na [Fe Mn ]O .The structural 2/3 1/2 1/2 2 9–14 sian Blue analogs (PBAs), and layered oxides . Each one of transformations are known to be detrimental to the reversible these materials has advantages and limitations, for example, capacity —mainly due to the huge volume changes involved— , polyanionic materials exhibit very good capacity retention due to although the exact mechanism and effect on the physical properties their stable 3D framework, in contrast, their specific capacity is of such impoverishment are not understood yet. 9,10 typically lower than many layered oxides ; organic compounds Several strategies have been developed in order to improve its show massive capacity fading due to their dissolution in the long-term stability. One of the most successful approaches is to organic carbonate-based electrolyte ; PBAs usually deliver lower partially substitute the TM by an electrochemically active or capacities at low operating voltage than layered oxides . Layered inactive element/s (such as Ni, Co, Cu, Mg, Al, Ti, K, Li, and so oxides with general formula Na TMO (TM = transition metal/s on) giving rise to ternary or quaternary compounds which, x 2 such as Co, Mn, Fe, Ni, Ti, V, etc., as well as alkali metals namely despite typically displaying lower capacity values, exhibit better Li, K, and Mg) can deliver high specific capacity, but their cycle life capacity retention and in some cases higher operating 13–15 27–34 should be even greater enhanced . Among all layered oxides, P2- voltage . The origin of such enhanced electrochemical per- Na [Fe Mn ]O is one of the most promising cathode materials formance is still unclear and has been attributed to different 2/3 1/2 1/2 2 in terms of cost-efficiency and energy density .Itismadefrom factors, such as improvement of the structural stability, reduction + + Earth abundant elements and delivers a high reversible capacity of of the volume change between Na extracted and Na inserted −1 190 mAh g when it is cycled in the voltage range of 1.5–4.3 V vs states, increase of the sodium interlayer distance, buffering of the Na /Na using metallic sodium as the counter electrode. Moreover, Jahn-Teller induced distortion on Mn(III), and or controlling the +32,34–38 in a full-cell configuration, using hard carbon (HC) as an anode, P2- distribution of Na . Interestingly, the (slightly) doped P2- −1 Na [Fe Mn ]O delivers a reversible capacity of 185 mAh g layered oxides that exhibit improved cycling stability, do not 2/3 1/2 1/2 2 with an average cell voltage of 2.75 V ; which is comparable to the show a high voltage phase transition or the volume changes 19,30,33,34,39,40 prototype cells developed by Faradion using a quaternary layered between P2 and “Z”/OP4 significantly reduced . This oxide (Na Ni Mn Mg Ti O ) as cathode and HC as observation suggested that the high voltage phase formation a (1−x−y−z) x y z 2 anode . Albeit the good energy density values obtained from P2- should be avoided. Consequently, an alternative approach to Na [Fe Mn ]O layered oxide, capacity retention is still one of improve the capacity retention has been studied -not only for P2- 2/3 1/2 1/2 2 its major weaknesses and it has been related to phase transitions Na [Fe Mn ]O but also for other P2-layered oxides— which 2/3 1/2 1/2 2 involving different stacking sequences when the sodium con- is the reduction of the operating voltage window, from >4.3 to centration changes while inducing large volume changes and exfo- 4.0 V vs Na /Na, avoiding the phase transition that has been liation of the layered oxide at particle surface . observed upon cycling at high voltage, although at the cost of a 21,23,24,41,42 The structural evolution of P2-Na [Fe Mn ]O (P6 / lower capacity as well . The exact impact of reducing 2/3 1/2 1/2 2 3 mmc) occurs via a solid-solution mechanism in the voltage the operating voltage window on the electrochemical perfor- + + range of 2.0–4.0 V vs Na /Na (from 0.67 to 0.36 Na per for- mance of layered oxides has been attributed to the improvement 15,20 mula (p.f.) unit) . The repulsion between [TMO ]slabs, of the structural stability upon electrochemical cycling, however, induced by Na extraction from the interlayer space, leads to a further studies should be carried out to clearly discern the role of decrease of a parameter while c parameter increases. Above the operating voltage window on the electrochemical, physical, + + + 4.0 V (Na p.f.unit = 0.36) and up to 4.1 V vs Na /Na (Na and structural properties of P2-layered oxides. p.f.unit = 0.26) a second phase appears showing a biphasic In this work, P2-Na [Fe Mn ]O cathode material is studied 2/3 1/2 1/2 2 region. Above 4.1 V, the newly formed phase propagates as by means of electrochemical impedance spectroscopy (EIS) in two solid-solution until the end of the Na extraction process operating voltage windows. On the one hand, from 1.5 to 4.25 V + + (4.1–4.3 V vs Na /Na; Na p.f.unit = 0.19). The exact struc- (P2-NFMO-LV), where the P2- to “Z”/OP4-type phase transition ture of this second phase at high voltage is still controversial will occur. On the other hand, from 2.0 to 4.0 V (P2-NFMO-SV), due to the fact that it is formed by layer gliding and therefore avoiding the above-mentioned phase transition. These studies are structural disorder increases. Indeed, this phase could not be completed with compositional studies of the electrode-electrolyte clearly indexed by powder X-ray diffraction (XRD) measure- interface by means of X-ray photoelectron spectroscopy (XPS) and ments and it is mainly referred as “Z” phase, OP4-type struc- with density functional theory (DFT) simulations of the electronic ture (scape group (S.G): P-6m2or P6 )orP–Ointergrowth structure. The results show that the phase transition occurring at 16,20–24 phase , which consists of alternating layers with P- and high voltage has a profound impact on the electronic and ionic O-interlayer sites that result from the gliding of half [TMO ] transport properties of P2-Na [Fe Mn ]O electrode, and such 2 2/3 1/2 1/2 2 16 + layers as determined by synchrotron XRD .During Na changes directly impact in capacity retention. The obtained results insertion, the reverse reaction mechanism is observed: starting in this investigation can be extrapolated to other P2-layered oxides from “Z”/OP4 solid-solution region, followed first by a biphasic with different TM compositions and will help clarify the role of the + + region until 3.10 V vs Na /Na (Na p.f.unit = 0.44) and later operating voltage window and high voltage phase formation in the by a P2 solid-solution region. Finally, below 2.0 V vs Na /Na electrochemical properties of the P2-layered oxides. 2 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE Fig. 1 Nyquist plot of P2-NFMO-LV electrode at OCV (2.43 V vs Na /Na). The impedance data from a 100 kHz to 5 mHz and b 100 kHz to 1 Hz. The frequency regions are highlighted (HF-red, MF-green, and LF-blue). Results and discussion model, the poor electronic conductivity of the P2-layered oxides P2-Na [Fe Mn ]O cycled in the 1.5–4.25 V vs Na /Na has been considered as an extra resistance and capacitance labeled 2/3 1/2 1/2 2 voltage window: P2-NFMO-LV electrode. For the first voltage as bulk electronic resistance (R ) and capacitance arising from elec window studied, the Nyquist plot of the impedance dispersion charge accumulation (C ). elec recorded at an open circuit voltage (OCV—2.43 V vs Na /Na) is Figures 2 and 3 reveal that the main change in the Nyquist shown in Fig. 1a. Three semicircles are observed at different plots is observed in the LF semicircle (blue region of Fig. 2a, c), frequencies: low-frequency (LF) below 10 Hz, medium-frequency indicating a variation of the R (blue square of Fig. 3) upon Na elec (MF) in the 5 kHz–10 Hz range and, high-frequency (HF) above extraction/insertion which can be related to modifications in the 5 kHz. The latter one is better observed once the impedance dis- crystalline structure. Similar changes in the EIS behavior at LFs persion is enlarged (see Fig. 1b). The HF semicircle corresponds have been observed in other layered oxide cathodes and + 53,54,57–60 to the Na migration resistance through the electrode-electrolyte anodes . interphase (EEI) and is labeled as R . The oxidation of the After comparing the first with the second cycle, it is found that EEI electrolyte and subsequent formation of the EEI is not expected at R (blue square of Fig. 3) continuously increases although elec OCV, but a similar EEI is chemically formed before cycling, as several oscillations take place depending on the voltage and Na confirmed by XPS and in agreement with previous EEI studies of content, which are related to the structural evolution of 43–46 SIB electrodes . The high reactivity upon air/moisture expo- P2-Na [Fe Mn ]O . In the P2 solid-solution region (blue 2/3 1/2 1/2 2 47 + sure of the layered oxide , the dehydrofluorination reaction of region of Fig. 3), R decreases upon Na extraction while elec polivinylidene fluoride (PVdF), which takes place during elec- increases upon Na insertion after raising the minimum R elec trode preparation as confirmed by solid-state nuclear magnetic values due to the OP4 effect. Except for the first Na extraction in 43 + resonance , and the use of metallic sodium as counter electrode the 3.00–3.33 V vs Na /Na range, probably due to the fact that 48–50 3+ 4+ contribute to form a passivation surface layer composed by the electron transfer is easier in mixed valence states (Fe /Fe + 4+ 3+ carbonate species originated by electrolyte decomposition reac- couple —3.5 V vs Na /Na) than when only Mn and Fe are tions. The MF semicircle corresponds to charge-transfer resis- present on the P2-Na [Fe Mn ]O . This means that in the 2/3 1/2 1/2 2 tance (R ) and accumulation of charge in the interfacial double P2 solid-solution region, the P2-NFMO-LV electrode exhibits a CT layer (C ), while the LF region process can be correlated to the reversible behavior in terms of electronic conductivity, becoming DL bulk electronic conductivity, as already described for other a better electronic conductor during Na extraction. Never- 51–54 layered oxides electrode materials among others . Addition- theless, the R trend is interrupted when the P2-phase is elec ally, in the very LF region (40–5 mHz) a sloping line at ~45° with transformed into “Z ”/OP4-type structure, above 4.0 V vs Na /Na respect to the real axis (Z´) can be observed which corresponds to during Na extraction (violet region of Fig. 3), and when the + + the Na solid-state diffusion. distorted P’2 phase is observed, below 2.0 V vs Na /Na during Na The EIS spectrum of the P2-NFMO-LV electrode were insertion (yellow region of Fig. 3). These increments of R at elec collected every 45 mV during the first two cycles in the voltage 4.0 and 2.0 V vs Na /Na progressively reduce the bulk electronic range of 1.5–4.25 V vs Na /Na and the Nyquist plots shown in conductivity of P2-NFMO-LV electrode upon electrochemical Fig. 2 correspond to some relevant potential values of the first Na cycling, as indicated by the overall increase of R , where the elec extraction (Fig. 2a, b) and insertion (Fig. 2c, d) processes. The increment of R at the second cycle is even higher than in the elec overall trend of the resistance values obtained from the fit to the first one. This massive R increase suggests that the large elec equivalent circuit detailed in the experimental section of all volume change occurring during these phase transitions —mainly impedance spectra (133 in total) measured during the first two from P2 to “Z”/OP4— result in electrical isolation (loss of cycles is shown in Fig. 3. contact) among the active material particles, as well as the The described equivalent circuit (more details in the Methods conductive carbon, and current collector. section) is a modification of the surface model proposed by In the MF semicircle of the Nyquist plot, R (green triangle of CT 55,56 Aurbach and co-workers . This model, developed for graphite Fig. 3) constantly increases overall upon electrochemical cycling. electrodes, assumes that the active material has good electronic This is expected since the phase transitions upon electrochemical conductivity, but it is known that layered oxides are poor cycling induce the irreversible formation of grain boundaries, a electronic conductors. Therefore, in the used equivalent circuit mosaic texture, exfoliation of the layer or/and increase of interfacial COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 3 ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 Fig. 2 Nyquist plots of P2-NFMO-LV electrode upon the first cycle. In the top panels, during first Na extraction: at 2.53 V (red point), 3.46 V (green hexagon), 3.69 V (blue square), 4.02 V (black rhombus), 4.15 V (navy triangle), and 4.25 V (pink pentagon) in the frequency range of a 100 kHz to 5 mHz and b 100 kHz to 1 Hz/50 mHz which is a zoom-in of the very low real impedance component. In the bottom panels, during first Na insertion: at 4.25 V (pink pentagon), 3.45 V (red hexagon), 2.64 V (blue point), 2.42 V (black square), 2.19 V (green triangle), 2.07 V (brown rhombus), and 1.50 V (orange square) in the frequency range of c 100 kHz to 5 mHz and d 100 kHz to 1 Hz/50 mHz which is a zoom-in of the very low real impedance component. The LF region is highlighted in blue. microstrains, which was attributed as one of the causes of the from PVdF in the F 1s region (red line) (see also the concentration capacity fading of P2-layered oxides, including P2-Na [Fe Mn ] of the mentioned species in Supplementary Fig. S2). The EEI 2/3 1/2 1/2 25,26,61 O . formation at OCV occurs by the reduction of solvents and salt In parallel, R (red points of Fig. 3) constantly increases upon decomposition, as also observed in other Na-based electrodes and EEI + + first Na extraction, while above 3.00 V vs Na /Na it remains attributed to the reductive nature of layered oxides and instability of 19,44–46,48 constant. This behavior suggests that the formed EEI is overall metallic sodium . Second, at further electrochemical cycling, stable during the electrochemical cycling. Nonetheless, there is a the formed EEI is composed of C-O-C species, such as polyethylene + 64 small drop in the second Na insertion, more previously in the P oxide (PEO, (-CH -CH -O-) ) originated from direct polymeriza- 2 2 n + 65 ′2 region (below 2.1 V vs Na /Na). There are two main factors tion of ethylene carbonate (EC) and diethyl carbonate (DEC) ,as that can cause the lower R . On the one hand, some electrical well as NaCO R(R= alkyl group/s) and Na CO from EC and DEC EEI 3 2 3 45,66–68 contact problem —note that there is a glitch at 2.4 V, while R reduction as shown in C 1s and O 1s spectra . Indeed, the CT also suffers the drop. On the other hand, some slight outermost surface region is mainly composed of Na CO as 2 3 modifications to the EEI. However, the stability of local regions confirmed by the Na 1s spectra (Supplementary Fig. S3). Besides of the interphase, i.e., the outermost surface region, cannot be the mentioned carbonaceous/oxygenated species, the F 1s spectrum corroborated by means of EIS. reveals NaF formation due to the PVdF dehydrofluorination 43 46,69 The formation and stability of the outermost EEI of P2-Na reaction ,aswellasNaPF decomposition reactions . 2/ 6 [Fe Mn ]O electrodes cycled at different states of charge On the other hand, the first evidence of the EEI stability can be 3 1/2 1/2 2 (SOC - as indicated in the galvanostatic profile of Supplementary observed at a glance, note that the most significant difference is Fig. S1) has been measured by means of XPS. Figure 4 shows the found between the XPS spectra of the pristine and OCV electrodes, C 1s, O 1s, and F 1s photoemission lines while the binding while the XPS spectra from cycled electrodes are rather similar to the energies of the observed species are collected in Supplementary ones from OCV electrode, confirming that the surface composition Table S1. of the electrodes does not undergo major changes upon electro- First, the formation of the EEI is confirmed at OCV by the chemical cycling (see also Supplementary Fig. S2, illustration of the intensity decrease of electrode component photoelectron peaks: C65 concentration percentage of the electrode components). Indeed, the component in the C 1s region (black line) ,P2-Na [Fe Mn ]O presence of stable components corresponding to the pristine 2/3 1/2 1/2 2 component in the O 1s region (cyan line) and -CF component electrode (C65 (C1s), P2-Na [Fe Mn ]O (O 1s), and -CF 2 2/3 1/2 1/2 2 2 4 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE electrode becomes a worse electronic conductor. Such variation in the electronic properties is nested in the structural changes, namely the increase of the c parameter, owing to the electrostatic 15,20 repulsion between oxygen layers ; that ultimately results in an increase of the bulk electronic conductivity, as already observed 53,54 for the Li-based LiCoO layered oxide . In addition, the intrinsic electronic structure of the involved phases has been assessed using DFT calculations, as detailed in the experimental section, to better understand the origin of this observation. The density of states (DOS) for P2-Na [Fe Mn 2/3 1/2 1/ O ] and O2-Na [Fe Mn ]O model structures are shown in 2 2 1/5 1/2 1/2 2 Supplementary Fig. S5a, b, respectively. The O2-type structure should be considered as a proxy of the more complex “Z”/OP4- type structure, which is the most plausible phase formed at high 16,20–23 voltage as has been experimentally reported . Using an actual OP4 structural model involves a very high computational cost because large supercells and a prohibitive number of combinatorial Na /vacancy and Fe/Mn site distributions would need to be considered, which is beyond the scope of this work. In fact, a simple O2-based structural model is representative of the main possible orderings and local environments of the transition metals under the presence of Na /vacancy octahedral sites within OP4-type structures. Such octahedral environments are the main local structural difference with respect to the prismatic environ- ments found in the P2-phase. For the P2-type structure, the computed DOS reveals a valence band maximum populated with Mn states and the presence of a band gap of 0.5–0.8 eV between the valence and conduction bands (Supplementary Fig. S5a). In the case of the O2-type structure, the onset of the conduction band slightly increases by less than 0.3 eV with respect the P2 structure, and two discrete Fe spin-up states appear within the band gap yielding a pseudo-half-metallic character to the system (Supplementary Fig. S5b). Such differences between the electronic structures of P2- and O2-type model structures are not sufficiently large to suggest a noticeable qualitative difference in the intrinsic electronic conductivity between P2- and OP4-type Fig. 3 Fitted resistance values for the first two electrochemical cycles of phases. This is in stark contrast with the measured R values elec the P2-NFMO-LV electrode. a Resistance trend of R (red point), R that reveal the more insulating character of the Na extracted EEI CT (green triangle), R (blue square) and b α factor of the CPE associated cathode electrode with respect to the more sodiated one. At this elec with C (blue inverted triangle). The structural evolution highlighted as point, several factors should be considered upon the interpreta- elec colored regions and labeled upon electrochemical cycling is included in the tion of the DFT results. Indeed, the calculated band gaps 15, 20 figure for guidance . correspond to the active material while the EIS measurements correspond to the behavior of the whole electrode where the active material properties are deconvoluted. Therefore, the group of PVdF (F 1s)) is in agreement with the constant R values EEI eventual changes in particle size, interface and/or exfoliation, measured by EIS. The peak shape and intensity slightly vary during staking faults, etc. appearing upon cycling are not considered by Na insertion, for example, the NaCO R and NaF concentration the DFT calculations; certainly, these factors influence the bulk increases, which is due to theirpartial dissolutioninthe electronic conductivity of the material. organic carbonate-based electrode during Na extraction, and In order to observe experimentally these possible changes that 43,67 might be the responsible of the R drop observed below 2.1 V . EEI can affect the bulk electronic conductivity of the electrode, the This has been previously reported for similar layered oxides (note measured α value of C has been analyzed. α is the exponential elec that the Na CO R (x= 1) and NaF are highly soluble in 2−x 3 x factor of the electronic constant phase element (CPE) and for an 44,45 organic carbonate-based electrolytes) .Itcan be concludedthat, ideal surface, α is equal to 1. Figure 3b summarizes the α values of although electrolyte oxidation/reduction is not expected in the P2- the first two electrochemical cycles, and it shows an overall decrease NFMO-LV electrode since the operating voltage window is inside of α during the electrochemical cycling (variations are observed in the electrochemical stability window of the organic carbonate-based the phase transition regions). The α factor strongly depends on the electrolyte , a thin EEI is formed on the active material and 51 surface homogeneity, roughness and degree of polycrystallinity ,so conductive carbon already at OCV SOC, whose thickness and its continuous decrease is suggesting that the surface of the chemical composition slightly vary at different SOC. Indeed, the thin electrode is becoming more heterogeneous and/or the crystallinity is EEI cannot be detected by Fourier-transform Infrared spectroscopy being loss upon electrochemical cycling. Since these aspects cannot (FTIR), as shown in Supplementary Fig. S4, where the FTIR spectra be considered by the performed DFT calculations, and even if the of the cycled electrodes are identical. P2- and O2-phases can be representative of the OP4-phase, this As a result, the overall increase of R and R in the second CT elec increase of the heterogeneity, roughness, and/or polycrystallinity of cycle with respect to the first cycle indicates that the Na insertion the active material might be the origin of the discrepancy between into the electroactive material is hindered and the P2-NFMO-LV the DFT and EIS results. COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 5 ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 Fig. 4 XPS spectra of P2-NFMO-LV electrodes. C 1s, O 1s, and F 1s photoemission lines of P2-NFMO-LV electrodes before cycling (pristine) and at + + + different SOC (first Na extraction, first Na insertion, and second Na extraction) during electrochemical cycling as highlighted in Fig. S1. capacity retention than the P2-NFMO-LV electrode. Therefore, in order to study the effect of the transport properties, and mainly the R , in the smallest operating voltage window, where the elec phase transition at high voltage is avoided (P2 solid-solution region, taken over the distorted P′2 phase at around 2.0 V), the same EIS methodology applied to the P2-NFMO-LV electrode has been carried out between 2.0 and 4.0 V (P2-NFMO-SV electrode). This approach helps to corroborate the effect of the high volume change of the “Z”/OP4-type structure in the trans- port properties and its role in the R increase. elec The Nyquist plot of the P2-NFMO-SV electrode at OCV (2.68 V vs Na /Na, Supplementary Fig. S7) shows a very similar lineshape to the one from the P2-NFMO-LV electrode (Fig. 1). In fact, three semicircles at the same frequency regions are observed: HF region above 5 kHz (red region), MF region in the 5 kHz–10 Hz range (green region), and LF region below 10 Hz (blue region) with a sloping line at the lowest frequencies. The impedance spectra of the P2-NFMO-SV electrode during + + first Na insertion and Na extraction at different SOC (see Nyquist plots of Supplementary Fig. S8) show that it follows the same behavior as the P2-NFMO-LV electrode. This means that Fig. 5 Fitted resistance values for the first two electrochemical cycles of the material becomes a better electronic conductor (R elec the P2-NFMO-SV electrode. R (red points), R (green triangle), and EEI CT + + decrease) during the Na extraction process, while, upon Na R (blue square). The structural evolution highlighted as colored regions elec insertion, the active material exhibits the opposite trend, and labeled upon electrochemical cycling is included in the figure for becoming a worse electronic conductor (see Fig. 5). Interestingly, 15, 20 guidance . R is more stable than in the P2-NFMO-LV electrode during elec the first two cycles, which most probably is due to the blocking of the “Z”/OP4-type structure formation. However, at ~2.0 V vs Na P2-Na [Fe Mn ]O cycled in the 2.0–4.0 V vs Na /Na 2/3 1/2 1/2 2 voltage window: P2-NFMO-SV electrode. One of the approa- /Na an increment of the LF semicircle size (Supplementary ches to enhance the capacity retention of layered oxides is by Fig. S8c, pink curve) and the R value (a yellow region in Fig. 5) elec controlling the operating voltage window, obtaining the best is observed, owing to the fact that, at this voltage, the phase results in the 2.0–4.0 V vs Na /Na window, as mentioned transition from ordered P2 to distorted P′2 occurs. The R is elec 22,24,29,59,71 above . Supplementary Fig. S6 shows that, after 45 reversible within this voltage range, indicating that the “Z”/OP4- cycles, the P2-NFMO-LV electrode delivers the same capacity as type structure formation is the main responsible for the resistance the P2-NFMO-SV electrode, although the initial capacity is increase/loos of the electronic conductivity. −1 45 mAh g higher. The capacity retention of both electrodes is R (green triangles of Fig. 5) increment is also more stable in CT illustrated in Supplementary Fig. S6b, and it can be clearly the P2-NFMO-SV electrode. The slight increase during the observed that the P2-NFMO-SV electrode exhibits higher electrochemical cycling, likewise for the P2-NFMO-LV electrode, 6 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE Fig. 6 Nyquist plots of P2-NFMO-LV and P2-NFMO-SV electrodes upon electrochemical cycling. P2-NFMO-LV at a 4.25 V and b 1.50 V and P2-NFMO- + st nd rd th th SV electrode at c 4.0 V and d 2.0 V (vs Na /Na) of the 1 (black square), 2 (red point), 3 (blue triangle), 4 (green inverted triangle), and 20 cycle (magenta hexagon). suggests that by avoiding the structural change at high voltage, electrodes (P2-NFMO-LV and P2-NFMO-SV), the overall resis- the formation of grain boundaries, mosaic texture, exfoliation, tance increases upon cycling. Nevertheless, the increment rate of and/or interfacial microstrains is reduced. the overall resistance in the P2-NFMO-SV electrode is much Finally, the R (red points of Fig. 5) trend is the same as for lower than for the P2-NFMO-LV electrode. EEI the P2-NFMO-LV electrode. The R drop observed at 2.1 V vs R of P2-NFMO-LV shows a remarkable increase in the first EEI elec + + Na /Na during the second Na extraction might be correlated four cycles. Meanwhile, the total resistance of the P2-NFMO-SV with slight changes in the EEI surface, also observed in the second electrode is one order of magnitude smaller in the fourth cycle Na insertion on the P2-NFMO-LV electrode. Even so, after even if the phase transition at low voltage (P2 to P′2) is not further cycling, the R stabilizes at a constant value. avoided. Furthermore, in the twentieth cycle, the total resistance EEI Henceforth, avoiding the “Z”/OP4-type structure formation by of the P2-NFMO-SV electrode is still lower than that of the P2- reducing the operating voltage window, which is one of the NFMO-LV electrode in the fourth cycle. strategies (note that by chemical tunning also the high voltage Therefore, although it is considered that the structural evolution 30,31,33,34,39,40 transition can be avoided) , the electrode displays a of P2-Na [Fe Mn ]O is reversible, the XRD data acquired 2/3 1/2 1/2 2 high bulk electronic conductivity with stable R values. The after more than 75 cycles of the P2-NFMO-LV electrode clearly elec electronic conductivity loos due to the “Z”/OP4-type structure show that the full-width at half-maximum (FWHM) is drastically formation results due to the huge volume change that isolates the increasing upon electrochemical cycling (Supplementary Fig. S9). active particles from the conducting carbon and/or current The P2-Na [Fe Mn ]O (002) reflection of the cycled electrode 2/3 1/2 1/2 2 collector, while triggering strong internal stress. In turn, the until 4.25 V (>75 cycles) exhibits a much larger FWHM value than electronic conductivity is lost and/or reduced , which ultimately the pristine one (0.23° vs 0.09°). These phase transitions induce a influences the capacity fading. [TMO ]-gliding, which for the P2- to OP4-phase transition, the interlayer distance is reduced from 5.65 to 5.30–5.05 Å, while in the structural change at 2.0 V (P2 to P′2) the difference is 0.10 Å lower Further cycling of P2-NFMO-LV and P2-NFMO-SV electrodes. for P′2-type . As it was mentioned, these phase transitions lead to The impedance spectra have also been recorded after 1, 2, 3, 4, an unavoidable volume change and a significant increase of the and 20 cycles for both electrodes (Fig. 6). Although the fits interfacial microstrain, staking faults formation, and/or exfoliation become more complicated after long cycling because of the large of thelayersasconfirmed by previous reports on the grounds of semicircle developed at LF (highlighted in blue in Fig. 6) which 25,26,61 SEM/TEM experiments . These structural modifications are overlaps with the contribution from other processes at higher the responsibility of the decrease of electronic conductivity and frequencies (i.e., R and R ), it can be observed that, in both CT EEI COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 7 ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 greatly contribute to the reported capacity loss. On the other hand, GF/D) as separator and 1 M NaPF in EC:DEC in a 1:1 wt% (Acros Organics) as electrolyte. The experiments were carried out in a Maccor Series 4000 battery tester although in the first cycles the phase transition at low voltage (from in two operating voltage windows: (i)1.5–4.25 V (P2-NFMO-LV) and (ii) P2 to P′2) is reversible, it is clearly observed that upon further cycle + −1 2.0–4.0 V (P2-NFMO-SV) vs Na /Na, at 0.05C C-rate (1C = 260 mA g , which (see Fig. 6c, d), the overall resistance of the electrode also increases + theoretically corresponds to the exchange of one Na ). EIS experiments were (lower than in P2-NFMO-LV) that at long-term cycling will be an performed in three electrode Swagelok-T-type cells. Metallic sodium disks were employed as counter and reference electrodes, and the same electrolyte as for the issue, also contributing in the capacity fading. Hence, avoiding the galvanostatic experiments was used (1 M NaPF in EC:DEC in a 1:1 wt%). The phase evolution, for example by using a reduced operating voltage experiments were carried out at room temperature in the same two operating window (2.0–4.0 V vs Na /Na), better capacity retention is voltage windows used for the galvanostatic tests: (i) 1.5–4.25 V (P2-NFMO-LV) obtained, although lower initial capacity values are delivered and (ii) 2.0–4.0 V (P2-NFMO-SV) vs Na /Na in a VMP3 potentiostat (Bio-Logic) through potentiostatic intermittent titration technique (PITT) where the potential (Supplementary Fig. S6), due to the transport properties maintain of the electrode was controlled with a 45 mV step scan. A sinusoidal perturbation stable. of 5 mV was applied in the frequency range of 100 kHz–5 mHz and before EIS data were taken a 4 h of equilibrium condition was set at a constant potential. Impe- dance dispersion data (133 for MFMO-LV and 76 for NFMO-SV impedance Conclusions spectra in total) were fitted by Boukamp’s Equivalent Circuit software . This EIS experiments show that the responsible, in terms of transport model is based on the following parameters: (i)Na resistance across the elec- properties, of the poor capacity retention of P2-Na [Fe Mn ] 2/3 1/2 1/2 trolyte (R ), (ii) resistance and capacitance of the EEI (R and C ), (iii) R sol EEI EEI CT O layered oxide cathode material is the formation of the “Z”/ 2 and C ,(iv) R and C ,(v) Warburg diffusion (Z ) related to the solid-state DL elec elec w OP4 phase that induces the loss of the bulk electronic con- diffusion of Na , and finally (vi) intercalation capacity (C ) due to the charge 51,55,56 accumulation (see top of Supplementary Fig. S12) . The resistance and ductivity of the electrode. DFT calculations rule out the possibility capacitance of each process are connected in parallel. Additionally, in order to take that such observation is due to significant changes in the intrinsic into account any deviation from an ideal material such as surface inhomogeneity, electronic structure of the bulk materials during cycling. Since the roughness or degree of polycrystallinity, the C elements and Z have been replaced XPS studies show that the EEI is rather stable and the DFT results by CPE. In order to illustrate the goodness of the fits, some fitted EIS spectra are shown as examples in Supplementary Fig. S12a (P2-NFMO-LV electrode) and suggest that no significant changes should be expected for dif- S12b (P2-NFMO-SV electrode), while the obtained values from the fit are collected ferent orderings of this structure, such bulk electron conductivity in Supplementary Tables S2 and S3, respectively. loss is most probably due to the large volume changes occurring in the active material when cycling above 4.0 V vs Na /Na where Characterization of EEI. The EEI was studied, by means of XPS using a Phoibos a P2- to “Z”/OP4-type phase transition occurs. These volume 150 spectrometer (Specs GmbH), at several SOC (cycled at 0.05C): pristine, OCV, changes lead to interfacial microstrain, staking faults formation, st + nd + 1 Na extraction and insertion and 2 Na extraction states as highlighted in the and/or exfoliation of the layers that, besides reducing the bulk galvanostatic profile of Supplementary Fig. S1. The EEI composition and stability are followed by analyzing the C 1s, O 1s, F 1s, and Na 1s photoemission lines. Since electron conductivity, lead to the progressive isolation of the the Auger peaks and photoelectron peaks were overlapped in some cases, two active particle with respect to the conducting carbon and current different X-ray sources were used: a non-monochromatic Mg K source (hν = collector. When the high cut-off voltage is set to avoid the 1253.6 eV), which was applied for C 1s and O 1s photoelectron peaks and a non- structural change, the electronic resistance is maintained since the monochromatic Al K source (hν = 1486.6 eV) for all photoemission lines (C 1s, F 1s, and Na 1s). The electrodes were stopped at the selected SOC, rinsed with DEC, [TMO ]-layers are not glided, hence, the P2-Na [Fe Mn ]O 2 2/3 1/2 1/2 2 and dried before being inserted into the XPS vacuum chamber by means of an electrode exhibits much better structural reversibility, as well as argon-filled transfer system, never exposing the electrodes to air. High-resolution reversible resistance values and more stable transport properties scans at low potential were acquired at 100 W, 20 eV pass energy, and 0.1 eV and in turn better capacity retention. Such results might be energy step. Charging effects were compensated by flood gun at 10 μA and 1.5 V in order to correct slight variations of the binding energy . Calibration of the binding extrapolated to other P2-layered oxides which, despite exhibiting energy was performed using the C 1s graphitic signal as a reference at 284.4 eV. good theoretical capacity when cycled up to 4.3 V vs Na /Na, The recorded spectra were fitted by CasaXPS software using a nonlinear Shirley- they should not be used as a high voltage cathode electrodes in type background and a Voigt profile (70% Gaussian and 30% Lorentzian) . The order to keep good capacity retention and excellent cycle life for composition of the EEI and subsurface region has also been evaluated by means of the SIB. FTIR using an Agilent Technologies Cary 630 benchtop system placed inside an inert atmosphere glove box. The electrodes were rinsed in DEC prior to FTIR experiments to remove electrolyte salt traces. Methods Synthesis of the active cathode material. P2-Na [Fe Mn ]O was synthe- 2/3 1/2 1/2 2 Theoretical methods. DFT calculations have been performed within the Vienna sized by the solid-state method. First, stoichiometric amounts of Na CO ·H O 2 3 2 75,76 (99.5%, Sigma Aldrich), Fe O (99%, Alfa Aesar), and Mn O (98%, Alfa Aesar) ab initio simulation package , employing the PBE functional. In order to 2 3 2 3 describe the localized nature of Fe and Mn 3d states, the DFT + U scheme of were mixed in a mortar for 2 h. The mixed powder was compressed in pellets and heated up to 900 °C for 12 h under air atmosphere followed by liquid nitrogen Dudarev et al. has been applied, with U = 4.0 eV for Fe and Mn atoms as 16 78,79 quenching . The obtained sample was transferred and stored in an argon-filled suggested in the literature . All calculations were spin-polarized, starting from a glove box (MBraun, H O and O < 1 ppm) in order to avoid any contact with the high-spin configuration. PBE-based projector augmented wave potentials were 2 2 atmosphere. used to replace core electrons, whereas we treated explicitly the Na (3s), Fe (3p, 3d, 4s), Mn (3p, 3d, 4s), and O (2s, 2p) electrons as valence electrons and their wave- functions were expanded in plane-waves with cut-off energy of 600 eV. The irre- Structural and morphological characterization. The structural characterization ducible Brillouin zone was sampled using a Monkhorst–Pack grid with of the synthesized active cathode material was performed by powder XRD using a 7 × 7 × 1 k-point sampling per (1 × 1) unit cell. Ground state energies for every Bruker Advance D8 instrument with copper radiation (Cu Kα λ = 1.5406 Å, 1,2 configuration were computed allowing the lattice parameters, cell shape, and 1.5444 Å). The powder XRD pattern was refined by the Le Bail method and shows −1 atomic positions to relax with a residual force threshold of 0.02 eV Å . It is known a pure P2-type structure with …AABBAA… staking sequence (Supplementary that in binary or ternary transition metal P2-compounds, the presence of Mn Fig. S10). The morphology was analyzed using SEM (Quanta 200 FEG-FEI model) + 42,69,82 atoms can suppress long-range Na /vacancy orderings . For the P2-struc- operated at 30 kV. The SEM images show that most particles crystalize as hex- ture, the Na /vacancy ordering used was previously found for a range of other P2- agonal platelets of 5–10 µm of diameter (Supplementary Fig. S11). + 83 84 phases with 2/3 Na content: P2-Na CoO , P2-Na Fe Mn O , and P2- 2/3 2 2/3 2/3 1/3 2 Na Mn Ni O . This ordering consists of a zig-zag pattern of Na1 and 2/3 2/3 1/3 2 Electrochemical characterization. Electrodes were prepared by mixing 80% of Na2 sites arranged in triangular units, involving a supercell with in-plane lattice P2-Na [Fe Mn ]O active material, 10% carbon Super C65 (Timcal C-Nergy- vectors a = 3a + 2b and b = 2a + 3b, containing 24 formula units with 16 2/3 1/2 1/2 2 super super TM), and 10% PVdF (Solef® Arkema Group) dissolved in N-methyl-2-pyrrolidone Na, 12 Fe (III), 12 Mn (III) and 8 Mn (IV) and 48 O atoms. On the other hand, (NMP—Sigma Aldrich). The slurry was cast on battery-grade aluminum foil and three different Na /vacancy orderings were considered for the O2-structure, one dried under vacuum overnight at 120 °C. 11 mm electrodes were pressed at 5 tons with Na atoms in rows, the other in a hexagonal ordering, and the last one mixing for 1 min before assembling cells in an argon-filled glove box (MBraun, H O and both of them. In this case, we have used a supercell containing 36 formula units O < 1 ppm). The galvanostatic experiments were carried out in CR2032 type coin- with 8 Na, 8 Fe (III), 10 Fe (IV), 18 Mn (IV), and 72 O atoms. All of them are in a cells, using P2-Na [Fe Mn ]O electrodes as a working electrode and metallic high spin configuration, with a total magnetic moment of 100.0 and 134.0 for the 2/3 1/2 1/2 2 sodium disk (99.8% Acros Organics) as a counter electrode, glass fiber (Whatman P2- and O2- structures, respectively. The Na /vacancy ordering assumed in this 8 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE work should be considered as a low-energy, reasonable approximation, since from 20. Singh, G., López Del Amo, J. M., Galceran, M., Pérez-Villar, S. & Rojo, T. a computational viewpoint, the modeling of such arrangements would require large Structural evolution during sodium deintercalation/intercalation in Na [Fe 2/3 1/ supercells which are unaffordable at the DFT level. For the Fe/Mn ordering in both Mn ]O . J. Mater. Chem. A 3, 6954–6961 (2015). 2 1/2 2 oxides layers, we have thoroughly sampled a range of local structures within the 21. Mortemard De Boisse, B., Carlier, D., Guignard, M., Bourgeois, L. & Delmas, considered supercell. In particular, for each investigated Na /vacancy orderings, 50 C. P2-Na Mn Fe O phase used as positive electrode in Na batteries: x 1/2 1/2 2 different configurations were randomly generated and computed their DFT energy, Structural changes induced by the electrochemical (De)intercalation process. in order to choose the most stable ones. The magnetic moment was used to Inorg. Chem. 53, 11197–11205 (2014). distinguish the oxidation state of Mn and Fe. Finally, the DOS was calculated, to 22. Pang, W. K. et al. Interplay between electrochemistry and phase evolution of obtain the bandgap of the structures. In order to accurately determine Fermi the P2-type Na (Fe Mn )O cathode for use in sodium-ion batteries. Chem. x 1/2 1/2 2 energies, single-point DOS calculations were performed with twice the k-point Mater. 27, 3150–3158 (2015). density 14 × 14 × 2 per (1 × 1) unit cell as compared to geometrical optimization 23. Bai, Y. et al. Enhanced sodium ion storage behavior of P2-Type Na Fe 2/3 1/ calculations. Mn O synthesized via a chelating agent assisted route. ACS Appl. Mater. 2 1/2 2 Interfaces 8, 2857–2865 (2016). 24. Talaie, E., Duffort, V., Smith, H. L., Fultz, B. & Nazar, L. F. Structure of the Data availability high voltage phase of layered P2-Na [Mn Fe ]O and the positive effect 2/3−z 1/2 1/2 2 The optimized atomic coordinates of the P2- and O2-phases are included in the of Ni substitution on its stability. Energy Environ. Sci. 8, 2512–2523 (2015). Supplementary Data 1 and 2 files, respectively. Restrictions apply to the availability of the 25. Alvarado, J. et al. Improvement of the cathode electrolyte interphase on P2- datasets generated during and/or analyzed during the current study, and so are not Na Ni Mn O by atomic layer deposition. ACS Appl. Mater. Interfaces 9, 2/3 1/3 2/3 2 publicly available. Data are however available from the authors upon reasonable request 26518–26530 (2017). and with permission of CIC energiGUNE and HIU Batteries. 26. Liu, Y. et al. Layered P2-Na [Ni Mn ]O as high-voltage cathode for 2/3 1/3 2/3 2 sodium-ion batteries: the capacity decay mechanism and Al O surface 2 3 Code availability modification. Nano Energy 27,27–34 (2016). Computer codes used during the current study are available from Dr. O. Lakuntza or 27. Wang, S., Sun, C., Wang, N. & Zhang, Q. Ni- and/or Mn-based layered Dr. J. Carrasco upon reasonable request and with permission of CIC energiGUNE. transition metal oxides as cathode materials for sodium ion batteries: Status, challenges and countermeasures. J. Mater. Chem. A 7, 10138–10158 (2019). 28. Li, Z. Y. et al. Unveiling the role of Co in improving the high-rate capability Received: 9 July 2021; Accepted: 12 January 2022; and cycling performance of layered Na Mn Ni Co O cathode 0.7 0.7 0.3−x x 2 materials for sodium-ion batteries. ACS Appl. Mater. Interfaces 8, 15439–15448 (2016). 29. Talaie, E., Kim, S. Y., Chen, N. & Nazar, L. F. Structural evolution and redox processes involved in the electrochemical cycling of P2- Na [Mn Fe Cu ]O . Chem. Mater. 29, 6684–6697 (2017). 0.67 0.66 0.20 0.14 2 References 30. Tapia-Ruiz, N. et al. High voltage structural evolution and enhanced Na-ion 1. Delmas, C., Branconnier, J.-J., Fouassier, C. & Hagenmuller, P. diffusion in P2-Na Ni Mg Mn O (0 ≤ x ≤ 0.2) cathodes from 2/3 1/3−x x 2/3 2 Electrochemical intercalation of sodium in Na CoO bronzes. Solid State Ion. x 2 diffraction, electrochemical and ab initio studies. Energy Environ. Sci. 11, 4, 165–169 (1981). 1470–1479 (2018). 2. Maazaz, A., Delmas, C. & Hagenmuller, P. A study of the NaxTiO2 system by 31. Kumakura, S., Tahara, Y., Sato, S., Kubota, K. & Komaba, S. P′2-Na 2/ electrochemical deintercalation. J. Incl. Phenom. 1,45–51 (1983). Mn Me O (Me = Mg, Ti, Co, Ni, Cu, and Zn): correlation between 3 0.9 0.1 2 3. Bauer, A. et al. The scale-up and commercialization of nonaqueous Na-ion orthorhombic distortion and electrochemical property. Chem. Mater. 29, battery technologies. Adv. Energy Mater. 8,1–13 (2018). 8958–8962 (2017). 4. Li, Y. et al. Advanced sodium-ion batteries using superior low cost pyrolyzed 32. Liu, Z. et al. Ultralow volume change of P2-type layered oxide cathode for Na- anthracite anode: towards practical applications. Energy Storage Mater. 5, ion batteries with controlled phase transition by regulating distribution of Na 191–197 (2016). +. Angew. Chem. Int. Ed. 60, 20960–20969 (2021). 5. Goikolea, E. et al. Na-ion batteries—approaching old and new challenges. Adv. 33. Liu, Y. et al. Mitigation of Jahn-Teller distortion and Na+/vacancy ordering in Energy Mater. 10,1–21 (2020). a distorted manganese oxide cathode material by Li substitution. Chem. Sci. 6. Muñoz-Márquez, M. Á. et al. Na-ion batteries for large scale applications: a 12, 1062–1067 (2021). review on anode materials and solid electrolyte interphase formation. Adv. 34. Wang, C. et al. Tuning local chemistry of P2 layered-oxide cathode for high Energy Mater. 7, 1700463 (2017). energy and long cycles of sodium-ion battery. Nat. Commun. 12, 2256 (2021). 7. Larcher, D. & Tarascon, J. M. Towards greener and more sustainable batteries 35. Li, Y. et al. Air-stable copper-based P2-Na Cu Fe Mn O as a new 7/9 2/9 1/9 2/3 2 for electrical energy storage. Nat. Chem. 7,19–29 (2015). positive electrode material for sodium-ion batteries. Adv. Sci. 2,1–7 (2015). 8. Vaalma, C., Buchholz, D., Weil, M. & Passerini, S. A cost and resource analysis 36. Park, J. K., Park, G. G., Kwak, H. H., Hong, S. T. & Lee, J. W. Enhanced rate of sodium-ion batteries. Nat. Rev. Mater. 3, 18013 (2018). capability and cycle performance of titanium-substituted P2-Type 9. Zhao, L. N., Zhang, T., Zhao, H. L. & Hou, Y. L. Polyanion-type electrode Na Fe Mn O as a cathode for sodium-ion batteries. ACS Omega 3, 0.67 0.5 0.5 2 materials for advanced sodium-ion batteries. Mater. Today Nano 10, 100072 361–368 (2017). (2020). 37. Singh, G. et al. High voltage Mg-doped Na Ni Mg Mn O (x = 0.05, 0.67 0.3−x x 0.7 2 10. Jin, T. et al. Polyanion-type cathode materials for sodium-ion batteries. Chem. 0.1) Na-ion cathodes with enhanced stability and rate capability. Chem. Mater. Soc. Rev. 49, 2342–2377 (2020). 28, 5087–5094 (2016). 11. Rajagopalan, R., Tang, Y., Jia, C., Ji, X. & Wang, H. Understanding the sodium 38. Zarrabeitia, M. et al. Unraveling the role of Ti in the stability of positive storage mechanisms of organic electrodes in sodium ion batteries: Issues and layered oxide electrodes for rechargeable Na-ion batteries. J. Mater. Chem. A solutions. Energy Environ. Sci. 13, 1568–1592 (2020). 7, 14169–14179 (2019). 12. Hurlbutt, K., Wheeler, S., Capone, I. & Pasta, M. Prussian blue analogs as 39. Han, M. H. et al. High-performance P2-phase Na Mn Fe Ti O cathode 2/3 0.8 0.1 0.1 2 battery materials. Joule 2, 1950–1960 (2018). material for ambient-temperature sodium-ion batteries. Chem. Mater. 28, 13. Liu, Q. et al. Sodium transition metal oxides: the preferred cathode 106–116 (2016). choice for future sodium-ion batteries? Energy Environ. Sci. 14, 158–179 40. Wang, L. et al. Copper-substituted Na Ni Cu Mn O cathode 0.67 0.3−x x 0.7 2 (2021). materials for sodium-ion batteries with suppressed P2-O2 phase transition. J. 14. Gonzalo, E., Zarrabeitia, M., Drewett, N. E., López del Amo, J. M. & Rojo, T. Mater. Chem. A 5, 8752–8761 (2017). Sodium manganese-rich layered oxides: Potential candidates as positive 41. Xu, J., Chou, S. L., Wang, J. L., Liu, H. K. & Dou, S. X. Layered P2- electrode for sodium-ion batteries. Energy Storage Mater. 34, 682–707 (2021). Na Fe Mn O cathode material for rechargeable sodium-ion batteries. 0.66 0.5 0.5 2 15. Liu, Z. et al. Recent progress of P2-type layered transition-metal oxide ChemElectroChem 1, 371–374 (2014). cathodes for sodium-ion batteries. Chemistry 26, 7747–7766 (2020). 42. Liu, L. et al. High-performance P2-type Na (Mn Fe Co )O cathode 2/3 1/2 1/4 1/4 2 16. Yabuuchi, N. et al. P2-type Na [Fe Mn ]O made from earth-abundant x 1/2 1/2 2 material with superior rate capability for Na-ion batteries. Adv. Energy Mater. elements for rechargeable Na batteries. Nat. Mater. 11, 512–517 (2012). 5,1–5 (2015). 17. Yabuuchi, N., Kubota, K., Dahbi, M. & Komaba, S. Research development on 43. Muñoz-Márquez, M. A. et al. Composition and evolution of the solid- sodium-ion batteries. Chem. Rev. 114, 11636–11682 (2014). electrolyte interphase in Na Ti O electrodes for Na-Ion batteries: XPS 2 3 7 18. Sayers, R., Barker, J. & Heap, R. WO2015177544 (A1) (2014). and auger parameter analysis. ACS Appl. Mater. Interfaces 7,7801–7808 19. Kubota, K., Kumakura, S., Yoda, Y., Kuroki, K. & Komaba, S. Electrochemistry (2015). and solid-state chemistry of NaMeO (Me = 3d transition metals). Adv. 44. Doubaji, S. et al. Passivation layer and cathodic redox reactions in sodium-ion Energy Mater. 8,1–30 (2018). batteries probed by HAXPES. ChemSusChem 9,97–108 (2016). COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 9 ARTICLE COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 45. Zarrabeitia, M. et al. Toward stable electrode/electrolyte interface of P2- 71. Yoshida, J. et al. New P2-Na Mn Ni Co O layered oxide as electrode 0.70 0.60 0.30 0.10 2 layered oxide for rechargeable Na-ion batteries. ACS Appl. Mater. Interfaces material for Na-ion batteries. J. Electrochem. Soc. 161, A1987–A1991 (2014). 11, 28885–28893 (2019). 72. Boukamp, B. A. A Nonlinear Least Squares Fit procedure for analysis of 46. Zarrabeitia, M., Casas-Cabanas, M. & Muñoz-Márquez, M. Á. Understanding immittance data of electrochemical systems. Solid State Ion. 20,31–44 (1986). the electrode—electrolyte interphase of high voltage positive electrode 73. Wagner, C. D., Riggs, W. M., Davis, L. E. & Moulder, J. F. Handbook of X-ray Na Co (PO ) P O for rechargeable sodium-ion batteries. Electrochim. Acta Photoelectron Spectroscopy. (Perkin-Elmer Corporation. Physical Electronics 4 3 4 2 2 7 372, 137846 (2021). Division, 1979). 47. Sathiya, M., Hemalatha, K., Ramesha, K., Tarascon, J.-M. & Prakash, A. S. 74. Walton, J., Wincott, P., Fairley, N. & Carrick, A. Peak Fitting with CasaXPS: A Synthesis, structure, and electrochemical properties of the layared sodium Casa Pocket Book (2010). insertion cathode material: NaNi Mn Co O . Chem. Mater. 24, 75. Kresse, G. & Hafner, J. Ab initio molecular dynamics for liquid metals. Phys. 1/3 1/3 1/3 2 1846–1853 (2012). Rev. B 47, 558–561 (1993). 48. Zarrabeitia, M., Muñoz-Márquez, M., Nobili, F., Rojo, T. & Casas-Cabanas, M. 76. Kresse, G. & Furthmüller, J. Efficiency of ab-initio total energy calculations for Influence of using metallic na on the interfacial and transport properties of metals and semiconductors using a plane-wave basis set. Comput. Mater. Sci. Na-ion batteries. Batteries 3, 16 (2017). 6,15–50 (1996). 49. Conder, J. & Villevieille, C. How reliable is the Na metal as a counter electrode 77. Dudarev, S. & Botton, G. Electron-energy-loss spectra and the structural in Na-ion half cells? Chem. Commun. 55, 1275–1278 (2019). stability of nickel oxide: An LSDA + U study. Phys. Rev. B 57, 1505–1509 50. Pfeifer, K. et al. Can metallic sodium electrodes affect the electrochemistry of (1998). sodium‐ion batteries? Reactivity Issues and Perspectives. ChemSusChem 12, 78. Lee, D. H., Xu, J. & Meng, Y. S. An advanced cathode for Na-ion batteries with 3312–3319 (2019). high rate and excellent structural stability. Phys. Chem. Chem. Phys. 15, 51. Barsoukov, E. & Macdonald, J. R. Impedance Spectroscopy Theory Experiment 3304–3312 (2013). and Applications. (John Wiley & Sons, Inc, 2005). 79. Hautier, G., Ong, S. P., Jain, A., Moore, C. J. & Ceder, G. Accuracy of density 52. Zhuang, S. S., Qiu, Q. C. & Xu, X. Y. Diagnosis of Electrochemical Impedance functional theory in predicting formation energies of ternary oxides from Spectroscopy in Lithium-Ion Batteries. Lithium Ion Batteries-New Development. binary oxides and its implication on phase stability. Phys. Rev. B 85, 155208 Ilias Belharouak, IntechOpen, https://doi.org/10.5772/26749 (2012). (2012). 53. Nobili, F., Tossici, R., Marassi, R., Croce, F. & Scrosati, B. An AC impedance 80. Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave spectroscopic study of Li CoO at different temperatures. J. Phys. Chem. B method. Phys. Rev. B 59, 1758–1775 (1999). x 2 106, 3909–3915 (2002). 81. Monkhorst, H. & Pack, J. D. Special points for Brillouin-zone integrations. 54. Nobili, F., Dsoke, S., Minicucci, M., Croce, F. & Marassi, R. Correlation of Ac- Phys. Rev. B 13, 5188–5192 (1976). impedance and in situ X-ray spectra of LiCoO . J. Phys. Chem. B 110, 82. Wang, X., Tamaru, M., Okubo, M. & Yamada, A. Electrode properties of P2- 11310–11313 (2006). Na Mn Co O as cathode materials for sodium-ion batteries. J. Phys. 2/3 y 1−y 2 55. Levi, M. D. & Aurbach, D. Simultaneous measurements and modeling of the Chem. C 117, 15545–15551 (2013). electrochemical impedance and the cyclic voltammetric characteristics of 83. Platova, T. A., Mukhamedshin, I. R., Alloul, H., Dooglav, A. V. & Collin, G. graphite electrodes doped with lithium. J. Phys. Chem. B 101,4630–4640 (1997). Nuclear quadrupole resonance and X-ray investigation of the structure of 56. Aurbach, D. et al. Common electroanalytical behavior of li intercalation Na CoO . Phys. Rev. B 80, 224106 (2009). 2/3 2 processes into graphite and transition metal oxides. J. Electrochem. Soc. 145, 84. Katcho, N. A. et al. Origins of bistability and Na ion mobility difference in P2- 3024–3034 (1998). and O3-Na Fe Mn O cathode polymorphs. Adv. Energy Mater. 7,1–9 2/3 2/3 1/3 2 57. Croce, F. et al. An electrochemical impedance spectroscopic study of the transport (2017). properties of LiNi Co O . Electrochem. Commun. 1, 605–608 (1999). 0.75 0.25 2 58. Nobili, F. et al. Sol-gel synthesis and electrochemical characterization of Mg-/ Acknowledgements Zr-doped LiCoO cathodes for Li-ion batteries. J. Power Sources 197, 276–284 M.Z. thanks the Government of the Basque Country for Ph.D. funding through a Pre- (2012). doctoral fellowship and her stage at the University of Camerino by “EGONLABUR” 59. Hasa, I., Passerini, S. & Hassoun, J. Toward high energy density cathode Fellowship. B. Acebedo and M. Jauregui are acknowledged for their technical support materials for sodium-ion batteries: investigating the beneficial effect of with material synthesis and powder XRD measurements. O.L. thanks J.X Lian for his aluminum doping on the P2-type structure. J. Mater. Chem. A 5, 4467–4477 insight into generating the DOS graphs. Financial support from the Basque Government (2017). (Elkartek20 CIC energiGUNE) and from the Ministerio de Economía y Competitividad of 60. Zarrabeitia, M., Nobili, F., Muñoz-Márquez, M. Á., Rojo, T. & Casas-Cabanas, the Spanish Government (ENE2013-44330-R) is also acknowledged. M. Direct observation of electronic conductivity transitions and solid electrolyte interphase stability of Na Ti O electrodes for Na-ion batteries. J. 2 3 7 Power Sources 330,78–83 (2016). Author contributions 61. Xu, G. L. et al. Insights into the structural effects of layered cathode materials for M.Z.: Investigation, methodology, impedance, electrochemical and XPS data analysis, high voltage sodium-ion batteries. Energy Environ. Sci. 10, 1677–1693 (2017). visualization, funding acquisition, writing—original draft & editing. F.N.: EIS data 62. Blyth, R. I. R. et al. XPS studies of graphite electrode materials for lithium ion supervision, EIS methodology. O.L.: DFT calculations. J.C.: DFT data supervision, writing batteries. Appl. Surf. Sci. 167,99–106 (2000). —original draft. T.R.: Project administration, writing—review, funding acquisition. 63. Beamson, G. & Briggs, D. High resolution XPS of organic polymers: the M.C.C.: Project administration, supervision, writing—review draft. M.A.M.M.: Con- scienta ESCA300 database. J. Chem. Educ. 70, A25 (1993). ceptualization, methodology, investigation, formal analysis, funding acquisition, project 64. Andersson, A. M., Henningson, A., Siegbahn, H., Jansson, U. & Edström, K. administration, supervision, writing—review and editing. Electrochemically lithiated graphite characterised by photoelectron spectroscopy. J. Power Sources 119–121, 522–527 (2003). 65. Vogdanis, L. & Heitz, W. Carbon dioxide as a refrigerant. J. Frankl. Inst. 196, Competing interests 713 (1923). The authors declare no competing interests. 66. Schroder, K. W., Celio, H., Webb, L. J. & Stevenson, K. J. Examining solid electrolyte interphase formation on crystalline silicon electrodes: influence of Additional information electrochemical preparation and ambient exposure conditions. J. Phys. Chem. Supplementary information The online version contains supplementary material C 116, 19737–19747 (2012). available at https://doi.org/10.1038/s42004-022-00628-0. 67. Rezvani, S. J. et al. SEI dynamics in metal oxide conversion electrodes of Li-ion batteries. J. Phys. Chem. C 121, 26379–26388 (2017). Correspondence and requests for materials should be addressed to Miguel Ángel 68. Rezvani, S. J. et al. Does alumina coating alter the solid permeable interphase Muñoz-Márquez. dynamics in LiMn O cathodes? J. Phys. Chem. C 124, 26670–26677 (2020). 2 4 69. Dahbi, M. et al. Effect of hexafluorophosphate and fluoroethylene carbonate Peer review information Communications Chemistry thanks the anonymous reviewers on electrochemical performance and the surface layer of hard carbon for for their contribution to the peer review of this work. sodium-ion batteries. ChemElectroChem 3, 1856–1867 (2016). 70. Bhide, A., Hofmann, J., Katharina Dürr, A., Janek, J. & Adelhelm, P. Reprints and permission information is available at http://www.nature.com/reprints Electrochemical stability of non-aqueous electrolytes for sodium-ion batteries and their compatibility with Na CoO . Phys. Chem. Chem. Phys. 16, 0.7 2 Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in 1987–1998 (2014). published maps and institutional affiliations. 10 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem COMMUNICATIONS CHEMISTRY | https://doi.org/10.1038/s42004-022-00628-0 ARTICLE Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/ licenses/by/4.0/. © The Author(s) 2022 COMMUNICATIONS CHEMISTRY | (2022) 5:11 | https://doi.org/10.1038/s42004-022-00628-0 | www.nature.com/commschem 11

Journal

Communications ChemistrySpringer Journals

Published: Feb 1, 2022

There are no references for this article.