Structural Evolution and Redox Processes Involved in the

Article pubs.acs.org/cm

Structural Evolution and Redox Processes Involved in the Electrochemical Cycling of P2−Na0.67[Mn0.66Fe0.20Cu0.14]O2 Elahe Talaie,† Se Young Kim,† Ning Chen,‡ and Linda F. Nazar*,† †

Department of Chemistry, Waterloo Institute for Nanotechnology, University of Waterloo, 200 University Avenue West, Waterloo, Ontario N2L 3G1, Canada ‡ Canadian Light Source, 44 Innovation Boulevard, Saskatoon, Saskatchewan S7N 2V3, Canada S Supporting Information *

ABSTRACT: We report the synthesis and structural evolution of P2−Na0.67[Mn0.66Fe0.20Cu0.14]O2 during charge and discharge as a positive electrode for Na-ion batteries. Operando X-ray diffraction analysis revealed the existence of two phase transitions in the voltage window between 1.5−4.3 V. Ex situ pair distribution function analysis was used to characterize the local structure of the high-voltage phase which is disordered along the c-axis and is derived by the migration of a fraction of iron ions into the interlayer space. Operando Xray absorption spectroscopy was employed to monitor the local structural evolution of the transition metals and shows the existence of Mn3+/4+ redox below ∼3.4 V, followed by the redox activity of Cu and Fe ions at higher voltages. However, no change was observed in the K-edge X-ray absorption near-edge spectrum of any of the transition metals at voltages above 4.1 V, where the growth of the high-voltage phase is initiated. The combination of these results implies the reversible contribution of oxide ion redox to the capacity, which is coincident with metal migration. The above processes result in voltage fading over cycling, similar to that exhibited by Li2MnO3-based positive electrode materials in Li-ion batteries.

■

voltage.12−14 Partial substitution of Fe atoms by Mn, Ni, Co, and Ti enhances the structural stability, capacity retention, and rate capability.12,14−22 Sodium metal oxides with Cu2+/3+ redox activity have been explored over the past few years as potential positive electrode materials for Na-ion batteries and are interesting because of the cost and safety advantages of Cu compared to Ni and Co. The O3-NaCuO2 phase was reported by Takahashi et al.23 to deliver a specific capacity of more than 140 mAh·g−1 in the voltage range of 0.75−3.0 V, but with poor capacity retention. P2− Na0.68[Cu0.34Mn0.66]O2, P2−Na0.68[Cu0.34Mn0.50Ti0.16]O2, and O3−Na0.9[Cu1/4Fe1/4Mn1/4Ti1/4]O2 materials exhibited limited cycling behavior in seminal studies on these materials.24,25 Later, P2−Na2/3[Cu1/3Mn2/3]O2 was reported to deliver 67 mAh·g−1 capacity on the first discharge cycled in the voltage range of 3.0−4.2 V at a rate of C/4 (where a 1C rate is defined as charge/discharge in 1 h) and survived 1000 cycles with 61% capacity retention.26 In the same study, X-ray absorption spectroscopy (XAS) revealed a charge compensation mechanism based on the Cu2+/3+ redox couple, whereas manganese ions were found to be electrochemically inactive. In a different study, Cu2+/3+ and Fe3+/4+ ions were found to be responsible for

INTRODUCTION The development of sustainable energy generation and storage technologies is garnering increasing attention due to the growing demand for energy that parallels environmental concerns regarding fossil fuel dependency. Li-ion batteries are widely used in portable electronic devices and their application is rapidly expanding to vehicle electrification owing to their high energy densities. Na-ion battery (NIB) technology is considered as a promising alternative for LIBs for large-scale applications.1−3 This is especially true for grid storage where cost is an important criterion. Resource availability of lithium and hence its cost may change dramatically in the future, whereas sodium will always be a highly abundant element. Layered sodium metal oxide host materials composed of lowcost and nontoxic transition metals such as Mn and Fe have been the focus of many studies and are considered promising candidates for positive electrodes for Na-ion batteries.4−9 The demonstration of Fe3+/4+ redox activity in layered sodium metal oxides,10,11 which is not observed in their lithium counterparts, attracted a great deal of attention from the research community because of the possible sustainability and high-voltage advantages of such materials. However, subsequent studies revealed their structural instability during cycling. Migration of a fraction of transition-metal ions from the octahedral site of MO2 layers (M: transition metal) into the tetrahedral site of the interlayer space was shown to occur in NaxMO2 (x ≤ 1) at high © 2017 American Chemical Society

Received: March 23, 2017 Revised: June 25, 2017 Published: July 5, 2017 6684

DOI: 10.1021/acs.chemmater.7b01146 Chem. Mater. 2017, 29, 6684−6697

Article

Chemistry of Materials

Figure 1. (a) XRD patterns of pristine Na0.67[Mn0.65Fe0.2Cu0.15]O2 (blue), pristine Na0.67[Mn0.66Fe0.2Cu0.14]O2 (red), and aged Na0.67[Mn0.66Fe0.2Cu0.14]O2 in air for 3 months (black). Symbol (↓) marks the CuO impurity peak; (b) schematic presentation of P2−NaxMO2; (c−e) SEM images of pristine Na0.67[Mn0.66Fe0.2Cu0.14]O2 (c) and the sample aged in air for 3 weeks (d) and 3 months (e).

the redox activity of O3−Na0.9[Cu0.22Fe0.30Mn0.48]O2.27 Hu et al.28 studied P2−Na7/9[Cu2/9Fe1/9Mn2/3]O2 and reported good capacity retention (87% after 150 cycles at a 1C cycling rate), persistence of the P2 structure over the voltage window of 2.5− 4.2 V, and air stability for this material. During the preparation of this paper, it came to our attention that P2 Na2/3Ni1/3‑xCuxMn2/3O2 has also been investigated and is reported to retain a P2/OP4 structure when cycled to high voltages.29 Understanding the impact of copper substitution on the stability of NaxMO2 materials requires in-depth characterization to probe phase transitions during cycling, the reactivity with ambient atmosphere, and charge compensation mechanisms. Clear insight into these issues is necessary to optimize the composition to obtain suitable stability with minimal compromise of the energy density. Herein, the structural evolution of P2−Na0.67[Mn0.66Fe0.20Cu0.14]O2 during the first charge and discharge is studied by operando X-ray diffraction (XRD) and ex situ pair distribution function (PDF) analyses to examine the effect of phase transitions on the electrochemical performance of the material. The results are compared with those of our previously reported study of Na0.67[Mn0.65Fe0.20Ni0.15]O2,14 effectively providing a systematic comparison of the effect of Ni and Cu-substitution into P2-type sodium manganese iron oxides. The redox processes during the first charge/discharge of Na0.67[Mn0.66Fe0.20Cu0.14]O2 is also probed by operando X-ray absorption spectroscopy (XAS). This technique has been widely employed to investigate the redox behavior in layered lithium metal oxides,30−32 but very few such studies have been reported for layered sodium metal oxides.33,34 Our results demonstrate no evidence of transitionmetal redox participation in Nax[Mn0.66Fe0.20Cu0.14]O2 at high potential, and suggest the contribution of oxide ion redox to partially account for the electrochemical capacity. Over this voltage range, the material converts from the initial P2 phase to a new phase characterized by transition metal migration into

the interlayer space. Electrochemical measurements show that avoiding the high-voltage transition enhances the capacity retention and voltage stability.

■

EXPERIMENTAL SECTION

P2−Na0.67[Mn0.66Fe0.20Cu0.14]O2 and P2−Na0.67[Mn0.65Fe0.20Ni0.15]O2 were synthesized by solid-state methods. A mixture of stoichiometric amounts of Na2CO3 (EMD Millipore, ≥ 99.5%), Mn2O3 (SigmaAldrich, 99%), Fe2O3 (Sigma-Aldrich, ≥ 99%) and CuO (J. T. Baker, 99.5%) or NiO (Sigma-Aldrich,99.8%) was ball-milled at 250 rpm for 1 h and then pelletized. The pellets were heated at 700 °C for 4 h, followed by heating at 900 °C for 6 h in air. They were cooled to 600 °C under an argon flow for 4 h and then to room temperature prior to being transferred to an Ar-filled glovebox in a sealed tube. To chemically oxidize Na0.67[Mn0.66Fe0.20Cu0.14]O2, the material was mixed with an overstoichiometric amount of NO2BF4 in acetonitrile in an Ar-filled glovebox. After 2 days, the resulting powder was washed with acetonitrile and dried at room temperature under vacuum. X-ray powder diffraction (XRPD) data were collected on a PANalytical Empyrean instrument using Cu Kα radiation equipped with a PIXcel bidimensional detector with a Ni Kβ filter. For Rietveld refinement, XRD data was collected on a sample that was loaded in a 0.3 mm glass capillary in an Ar-filled glovebox in order to protect the sample from exposure to air. Rietveld refinement was performed within the FullProf35 software suite. Operando XRD experiments were performed using a homemade hermetically sealed cell. During the acquisition, the cell was cycled at a rate of C/20 with a pattern collection time of ≈20 min (Δx ≈ 0.02 in NaxMO2). Lattice parameter evolution during charge/discharge was determined using the Le Bail method.36 Using the same diffractometer, the reciprocal space data for pair distribution function (PDF) analysis was obtained in transmission mode using Ag Kα radiation, a Rh Kβ filter, and a NaI scintillation point detector. The sample was loaded in a 1 mm glass capillary in an Ar-filled glovebox. The total collection time was 48 h. The data was converted to PDF using PDFGetX337 software, and real space data was fitted using PDFgui38 software. The damping factor associated with this instrumental configuration was calibrated by refining the structure of a silicon sample. 6685


Article

Chemistry of Materials The SEM images were recorded using a Zeiss Ultra Plus scanning electron microscope (FE-SEM) at 15 kV in secondary electron mode. The metal cation molar ratios were confirmed by energy dispersive Xray spectroscopy (EDS) (EDAX) and inductively coupled plasma atomic emission spectroscopy (ICP-AES; Prodigy high dispersion ICP, Teledyne Leeman Laboratories). The electrochemical performance of the material as a positive electrode was investigated in 2325 coin cells using Na metal (SigmaAldrich, ACS reagent) as the negative electrode and 1 M NaClO4 (Alfa Aesar, ≥ 98%) in propylene carbonate (BASF, 99.98%) with 2 vol % of 4-fluoro-1, 3-dioxolan-2-one (FEC) (Sigma-Aldrich, 99%) as the electrolyte. To prepare the positive electrodes, the active material was mixed with 10 wt % carbon black (TIMCAL) and 10 wt % polyvinylidene fluoride (PVDF) (Aldrich average Mw ≈ 534 000) suspended in N-methyl-2-pyrrolidinone (NMP) (Sigma-Aldrich, 99.5%) and cast on aluminum foil with a typical loading of ≈7 mg/ cm2. The electrodes were separated by two glass fiber separators (Merck Millipore). The electrode preparation procedure and cell assembly were performed in an Ar-filled glovebox. Galvanostatic cycling experiments were performed at room temperature with a current density of 13 mA·g−1 (corresponding to a C/20 rate), unless stated otherwise. For operando XRD experiments, the electrode preparation and cell assembly were similar to those described above, but the electrode was deposited on a thin glassy carbon substrate to allow the penetration of X-rays. For operando XAS experiments, modified coin cells were used (Figure S1). An 8 μm aluminized Kapton (Sheldahl) window was placed on the positive case of a 2325 coin cell for the penetration of the X-ray beam. Cells were continuously charged up to 4.3 V vs Na/ Na+ at a rate of C/20 and discharged to 1.5 V at a rate of C/10 in galvanostatic mode during data collection. Each cell was relaxed at open circuit voltage (OCV) for an hour at the end of charge. Mn, Fe, Cu K-edge XANES spectra were collected at the Hard X-ray MicroAnalysis (HXMA) beamline (06ID-1) at the Canadian Light Source (CLS). The CLS storage ring operated at 2.9 GeV and 200−250 mA in the injection mode during the experiment. The beamline was equipped with Si (111) and Si (220) double crystal monochromators and Rh -and Pt-coated focusing mirrors. Reduction of higher-order harmonics was achieved by detuning the second monochromator crystal to 50% of the peak intensity. Data collection was performed in the fluorescence mode using a 32 element Ge detector. The scan stepsizes were 10 eV/step, 0.25 eV/step, and 0.1 Å−1/step for pre-edge, XANES, and EXAFS regions, respectively. The data acquisition time was set to 1 s/step for the pre-edge and XANES regions and was progressively increased to 5 s/step for the EXAFS region. The total collection time for each pattern was 40 min, equivalent to Δx ≈ 0.03 and 0.07 in NaxMO2 during charge and discharge, respectively. The energy calibration for each spectrum was performed using the first inflection point of the reference channel spectrum, which was simultaneously collected from the metal foil. The reference spectra were aligned for comparison of the ensembled data. The raw XAS data were corrected for pre-edge and postedge background and normalized to unit step height. The X-ray absorption near edge structure (XANES) data were analyzed using the Athena software package.39

■

Figure 2. Rietveld refinement of X-ray powder diffraction data of Na0.67[Mn0.66Fe0.2Cu0.14]O2. The observed data is shown in red markers, the calculated pattern is shown in black, the difference curve is shown in blue, and the Bragg reflections are shown in green. The unit cell parameters and atomic parameters are presented in the figure inset.

with the (100) reflection of the P2 phase and hence not be visible. The appearance of CuO or another unidentified impurity was reported before in the synthesis of coppersubstituted P2 compositions.24,26,40 These observations suggest limited solubility of Cu2+ in the P2 layered structure, probably due to its larger ionic radius (0.73 Å) compared to that of other transition-metal cations Mn4+ (0.53 Å), Mn3+ (high spin, 0.65 Å), Fe3+ (high spin, 0.65 Å), and Ni2+ (0.69 Å).41 A single phase material was reproducibly achieved when 10 wt % less copper oxide was used to prepare the precursors, resulting in the Na0.67[Mn0.66Fe0.20Cu0.14]O2 composition (Figure 1a, red). Figure 2 shows the results of Rietveld refinement of the X-ray powder diffraction data of Na0.67[Mn0.66Fe0.20Cu0.14]O2. The analysis was performed on an air-protected sample, loaded into a glass capillary in an Ar-filled glovebox and sealed from exposure to air. The lattice parameters (a = 2.9280 Å and c = 11.1800 Å) are slightly larger than those of Na0.67[Mn0.65Fe0.20Ni0.15]O2 (a = 2.9207 Å and c = 11.1598 Å)14 in agreement with the larger size of Cu2+ compared to Ni2+. We note that hkl-dependent peak broadening was observed in the XRD pattern of Na0.67[Mn0.66Fe0.20Cu0.14]O2, similar to previously reported NaxMO2 materials:4,42 namely, (10l) Bragg peaks are broader than (00l) reflections. Application of a microstrain correction43 improved the peak profile modeling. Figure 1c shows an SEM image of as-prepared Na0.67[Mn0.66Fe0.20Cu0.14]O2. The crystals have a plate-like morphology, with submicron thickness and several micron diameter basal dimensions. No impurity phase is distinguishable by SEM based on particle morphology. However, the SEM image of a sample exposed to air for 3 weeks (Figure 1d) shows particles with a different morphology (marked by a yellow arrow). The SEM SEM image of the sample aged in air for three months (Figure 1e) clearly demonstrates a new phase with ribbon-like morphology. EDS analysis showed the constituent elements to be C, O, and Na, suggesting the formation of Na2CO3. The XRD pattern of the aged sample (Figure 1a, black) shows major new peaks in addition to the P2 phase. The growth of sodium carbonate ribbons formed by reaction of the material with CO2 in air, and a similar evolution of the XRD pattern (due to the intercalation of water molecules) were previously observed for an aged sample of its parent composition, P2−Na2/3[Mn1/2Fe1/2]O2.42 Recently, it was reported that no air reactivity was observed for P2−

RESULTS AND DISCUSSION

Structure and Air Stability. The target composition Na0.67[Mn0.65Fe0.20Cu0.15]O2 was synthesized by a solid-state method. The XRD pattern of the material (Figure 1a, blue) confirms its P2 structure (space group P63/mmc). This structure (Figure 1b) can be described as a stack of edgesharing MO6 layers accommodating two different prismatic sodium sitesNae and Nafwhich share edges and faces, respectively, with the MO6 octahedra. The XRD pattern of the material reveals the presence of a minor impurity phase (marked by ↓), which is likely CuO. The small peak at ≈38.7° is in accord with the (111) CuO reflection, whereas the second most intense reflection of CuO at ≈ 35.5° would be overlapped 6686


Article


Figure 3. (a) Color map illustration of the evolution of the XRD pattern of Nax[Mn0.66Fe0.2Cu0.14]O2 recorded over the first discharge during galvanostatic cycling at a rate of C/20, along with the voltage profile of the cell vs time. The hkl reflections corresponding to the P2 phase are marked on the figure; (b) XRD patterns of the pristine Nax[Mn0.66Fe0.2Cu0.14]O2 electrode (black), the electrode charged to 4.3 V (Z-phase, red), and the electrode discharged to 1.5 V (P′2 phase, blue).

of P2−Na7/9[Cu2/9Fe1/9Mn2/3]O2 must have another origin. Subsequent investigation, of Na0.67[Mn0.66Fe0.20Cu0.14]O2 in our work were performed on air-protected samples to ensure that no side-reactions occurred. Operando X-ray Diffraction Analysis (XRD). The phase transitions of Nax[Mn0.66Fe0.20Cu0.14]O2 during galvanostatic charge and discharge were examined by operando XRD analysis. Figure 3a presents a color map of the evolution of the Nax[Mn0.66Fe0.20Cu0.14]O2 diffraction pattern over the first discharge between 4.3 to 1.5 V at a cycling rate of C/20. Recent studies44,45 showed the dependency of phase transitions in layered NaxMO2 on the cycling rate. Operando XRD analysis showed that the phase transitions induced by insertion and extraction of sodium ions into/from Nax[Mn0.66Fe0.20Cu0.14]O2 ar e sim i l ar t o t hos e in Na x [Mn 0 . 5 Fe 0 . 5 ] O 2 and Nax[Mn0.65Fe0.20Ni0.15]O2.8,14 The initial P2 structure is preserved over a wide range of composition range but converts to a low-crystallinity phase (referred to as the Z-phase) at high voltage and a P′2 phase at low voltage. The P2 stability voltage window in Nax[Mn0.66Fe0.20Cu0.14]O2, 2.1−4.1 V (Figure 4a), is slightly wider compared to its parent composition Nax[Mn0.5Fe0.5]O2, 2.1−4.0 V. The inplane lattice parameter of the P2 structure (ahex), corresponding to the first-neighbor M−M distance, increases on insertion of sodium ions (Figure 4c, blue). This is agreement with the reduction of transition-metal ions. On the other hand, the interlayer distance contracts (Figure 4b, blue) in part due to the increased screening effect of the Na+ ions that weakens the electrostatic repulsion between adjacent oxygen layers. The low voltage P′2 phase (space group: Cmcm) can be described as the orthorhombic distortion of the hexagonal P2 phase (space group: P63/mmc) which is favored by the

Figure 4. (a) Phase evolution of Nax[Mn0.66Fe0.2Cu0.14]O2 vs sodium content over the first cycle. The asterisk * shows the cycling starting point; (b) evolution of the average interlayer distance; and (c) in-plane lattice parameter of Nax[Mn0.66Fe0.2Cu0.14]O2 as a function of sodium content during the first discharge.

Na7/9[Cu2/9Fe1/9Mn2/3]O2 up to 2 months,28 but our findings are not consistent with this. Our results imply that incorporation of copper into P2−Na2/3[Mn1/2Fe1/2]O2 could mitigate its reactivity, but does not suppress it, and the stability 6687


Article


Figure 5. (a) Fit of the PDF curve of Z-Na0.1[Mn0.66Fe0.2Cu0.14]O2 in the interatomic distance range of 1.5−5.5 Å. Rw = 0.24; (b) comparison of the experimental PDF data of Z-Nax[Mn0.66Fe0.20Cu0.14]O2 with P2 and Z- Nax[Mn0.5Fe0.5]O2. The data for the Nax[Mn0.5Fe0.5]O2 are adapted from ref 14. (c) Representation of interatomic distances d1, d2, d3, d4, and d5, corresponding to the peaks in PDF in the pristine P2 structure, O2 stacking scheme of the Z phase, and MO2 layer. The purple and red spheres represent the transition metals and oxygen atoms, respectively, and yellow and green polyhedra represent two different NaO6 polyhedra. The bilayer model is described in more detail in Figure S3.

content compared to the onset of the transition in Na x [Mn 0.5 Fe 0.5 ]O 2 and Na x [Mn 0.65 Fe 0.20 Ni 0.15 ]O 2 (x ≈ 0.81).14 All of these three compositions are expected to have a similar concentration of Jahn−Teller active Mn3+ ions at a specific x (in NaxMO2). The fact that the P2−P′2 transition starts at a lower sodium concentration in the copper-substituted composition is attributed to the role of Jahn−Teller active Cu2+ ions (d9 electronic state).47 It was reported that no distortion of the hexagonal cell was observed when P2− Na2/3[Mn1/2Fe1/4Co1/4]O219 was discharged to 1.5 V. That extended domain of stability was attributed to the reduction of Co3+ to Co2+ during discharge instead of reduction of Mn4+ to Jahn−Teller active Mn3+ ions. An undistorted hexagonal phase can be described in an orthorhombic cell with the following relationship between lattice parameters; aortho = ahex, bortho = ahex + 2bhex (e.g., aortho = bortho/√3) and cortho = chex. Therefore, the divergence of bortho/aortho from √3 implies a greater distortion from the hexagonal cell. Increased orthorhombic distortion is observed in P′2-Nax[Mn0.66Fe0.20Cu0.14]O2 as more sodium ions are inserted into the structure (Figure 4c, green). Transition to the Z-phase at high voltage initiates at 4.1 V, x ≈ 0.30 in Nax[Mn0.66Fe0.2Cu0.14]O2 at the first charge and proceeds through a two-phase mechanism until 4.2 V and x ≈ 0.21, where only the Z-phase exists (Figure 4a). This accounts for an electrochemical capacity of Δx ≈ 0.46, based on the starting composition Na0.67[Mn0.66Fe0.2Cu0.14]O2. Figure 3 shows that the (100)P2 and (110)P2 reflections are maintained in the XRD pattern of the charged material, Na0.1[Mn0.66Fe0.2Cu0.14]O2, proving that the structural coherence within MO2 layers is maintained (Figure 3b, red). Nevertheless, the substantial broadening of (00l)P2 and (10l)P2 peaks indicates the loss of long-range order along the c-axis. The transition from the P2 structure to this low crystalline phase was also reported for iron-containing layered oxide compositions Nax[Mn1/2Fe1/2]O2,4,8,14 Nax[Mn0.65Fe0.20Ni0.15]O2,14 and Nax[Mn1/2Fe1/4Co1/4]O2.19 In our previous study,14 a model was proposed to describe the Z-phase. We concluded

Figure 6. Ga l v a no st at ic c ha r g e / d is c ha r ge p ro fi les of Na0.67[Mn0.66Fe0.2Cu0.14]O2 at the cycling rate of C/20 (a,c) and the discharge profile evolution (b,d), for cells cycled within the voltage range of 1.5−4.3 V (a,b) and 2.1−4.1 V (c,d); (e) specific energy vs the cycle number of Na0.67[Mn0.66Fe0.2Cu0.14]O2 cycled within the voltage window of 1.5−4.3 V at a rate of C/20 (black circles) and cycled within the voltage range of 2.1−4.1 V at different rates (colored circles).

cooperative effect of Jahn−Teller active ions.8,46 It is evident (Figure 3a) that the (10l) reflections of the P2 phase split into two in the P′2 domain, whereas (00l) peaks are preserved. The P2−P′2 phase transition in Nax[Mn0.66Fe0.20Cu0.14]O2 starts at x ≈ 0.72 and 2.1 V (Figure 4a), which is at a lower sodium 6688


Article


Figure 7. (a) Voltage profile of a Nax[Mn0.66Fe0.20Cu0.14]O2 electrode (vs Na metal) from which the X-ray absorption spectra at the Mn K-edge were collected. The marks on the graph show the points at which data collection was started; (b−d), normalized XANES spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 at the Mn K-edge during the first charge (b), first discharge (c), and the comparison of the initial state and the end of discharge (d).

compared to Z-Nax[Mn0.66Fe0.20Cu0.14]O2 prepared in the operando cell by charging to 4.3 V. The PDF curve of Na0.1[Mn0.66Fe0.20Cu0.14]O2 was fit (Figure 5a) using a modification of the “bilayer model” previously proposed for the parent composition, Z-Na0.1[Mn0.5Fe0.5]O2.14 The bilayer model comprises a stack of bilayers along the c-axis that are randomly shifted in the ab-plane. Each bilayer has an O2-stacking scheme of two MO2 layers and the interlayer tetrahedral sites are partially occupied by transition metals (Figure 5c, center). Sodium ions are neglected for simplicity, owing to their low concentration and scattering factor. A comparison of the experimental PDF data of ZNa0.1[Mn0.66Fe0.20Cu0.14]O2 with Z-Na0.1[Mn0.5Fe0.5]O2 and the pristine phase, P2−Na0.67[Mn0.5Fe0.5]O2, is shown in Figure 5b in the interatomic distance range of 1−5.5 Å, (the data for Nax[Mn0.5Fe0.5]O2 (x = 0.67, 0.1) are adapted from ref 14). For all three compositions, the peaks located at ∼2 Å and ∼4.5 Å correspond to the first M−O (d1) and third M−O (d4) correlation lengths (Figure 5c), respectively. The second M−O correlation length gives rise to a small peak at ≈ 3.6 Å for the starting composition P2−Na0.67[Mn0.5Fe0.5]O2 (Figure 5b, blue markers) with low intensity (the number of third oxygen neighbors is twice as large as the number of second oxygen neighbors for each M atom). This peak is screened by the peak at 3.5 Å in the oxidized samples. The peaks located at ∼3 Å and ∼5 Å are ascribed to the first M−M (d2) and second M−M (d5) correlation lengths, respectively. Generally, a peak attributed to an M−M correlation is stronger than M−O at the same distance and with the same coordination number, because of the higher X-ray scattering factor of transition metals

that some transition metals (mainly iron) migrate into the interlayer space at high voltage. This explains the trend observed in the lattice parameter evolution in Figure 4b. Migration of transition metals into interlayer tetrahedral sites upon charge was also proposed in O3−NaxFeO213 and O3− NaxCrO248 on the basis of ex situ XRD and XAS analyses. This structural rearrangement is irreversible in O3 oxides, and results in severe capacity fading. The Z-phase does convert back to the P2 phase during discharge, as demonstrated by XRD analysis.14 However ,the slow conversion still affects capacity retention and voltage stability (see below). Pair Distribution Function (PDF) Analysis of the ZPhase. Pair distribution function (PDF) analysis using laboratory X-ray data was employed to investigate the poorly crystalline Z-Nax[Mn0.66Fe0.20Cu0.14]O2 phase obtained at high voltage. The PDF represents the probability of finding any pair of atoms at a specific interatomic distance, regardless of the crystallinity of the structure. Because of the sensitivity of the PDF technique to amorphous phases, a chemically oxidized sample was prepared for this study to avoid complications that could arise from the addition of carbon and binder in an electrochemically oxidized electrode sample. The oxidized sample exhibits an interlayer distance of ≈5.1 Å. On this basis, its composition is estimated to be Na0.1[Mn0.66Fe0.20Cu0.14]O2 by interpolation of the data in Figure 4b. Compared to electrochemical oxidation (which results in a higher interlayer d-spacing of 5.2 Å in the material), chemical oxidation extracts a slightly more sodium. The XRD pattern (Figure S2) also demonstrates more peak broadening (higher disorder), in agreement with a higher sodium vacancy 6689


Article


based on the positions of M−M peaks in the PDF data. Overlap of the peaks corresponding to two MTd-MOh distances, d3(i) = 3.28 Å and d3(ii) = 3.42 Å (Figure 5c), results in the third peak in the PDF data of Z-Na0.1[Mn0.66Fe0.20Cu0.14]O2 (Figure 5a,b). Thus, the intensity ratio of peak 3: peak 2 (which is related to the concentration of M cations in the tetrahedral sites in the interlayer space) in the PDF profile of ZNa0.1[Mn0.66Fe0.20Cu0.14]O2 is lower than Z-Na0.1[Mn0.5Fe0.5]O 2 , but higher than the previously reported ZNa0.1[Mn0.65Fe0.20Ni0.15]O214(Figure S3a). In summary, the transition-metal migration into the interlayer space at high voltage is decreased in the Cu-substituted composition compared to the parent composition Z-Na0.1[Mn0.5Fe0.5]O2, but is higher than the Ni-substituted material (Figure S3b). In oxidized Nax[Mn0.65Fe0.20Ni0.15]O2 and Nax[Mn0.5Fe0.5]O2, Fe4+ ions are stabilized by transfer of electrons from oxygen orbitals to iron orbitals. This ligand to metal charge transfer (LMCT), promotes weakening of the bonding of neighboring atoms, corresponding to partial oxidation of the oxide ions, and enabling the migration of transition metals out of the MO2 layers.14 The higher population of migrated M cations in the Cu-substituted vs Ni-substituted composition may be due to a higher concentration of Fe4+ ions in the charged state of the former, considering that the Cu redox couple contributes less to the redox capacity than Ni (see below). Moreover, strong hybridization of O-2p orbitals with Cu-3d orbitals in Cu3+containing oxides, specifically high-temperature Cu-based superconductors, is well-known.49−52 Therefore, LMCT stabilization of Cu3+ ions, similar to Fe4+ ions, may be the origin of the behavior at high sodium vacancy content. The contribution of oxide ions to the redox process in Nax[Mn0.66Fe0.20Cu0.14]O2 is also supported by XAS experiments (see below). However, it should be noted that PDF analyses were performed on chemically oxidized samples, and the structural features in these samples are not exactly the same as in those that are slowly charged in electrochemical cells. Electrochemical Performance. Na0.67[Mn0.66Fe0.20Cu0.14]O2 delivers a high specific capacity of ≈176 mAh·g−1 in the wide voltage range of 1.5−4.3 V, cycled at a rate of C/20 (13 mAh·g−1). The smooth galvanostatic charge/discharge profile (Figure 6a) indicates facile (de)intercalation of sodium ions. This is in sharp contrast to the more stepped profile exhibited by the Ni analogue (Na0.67[Mn0.66Ni0.20Cu0.14]O2,),40 suggesting that detrimental Na-vacancy ordering is largely suppressed in the Fe composition. However, the two phase transitionsin particular the P2−Z transitionnonetheless result in fading of the average voltage upon cycling and thus the delivered energy (Figure 6b). Voltage fading is also a critical challenge with highenergy density Li2MnO3-based electrode materials and is proposed to be induced by irreversible structural changes.53 The large voltage polarization observed at high voltage disappears when the P2−Z phase transition is avoided by limiting the upper cutoff voltage (Figure 6c). When Na0.67[Mn0.66Fe0.20Cu0.14]O2 is cycled within its P2-stability window of 2.1−4.1 V (determined from operando XRD analysis), a reversible capacity of 94 mAh·g−1 is obtained. The voltage profile shows good stability over 100 cycles (Figure 6d). In terms of specific energy, the cells exhibit an initial specific energy of ≈272 Wh·kg−1 when cycled in this narrower window, with specific energy retention of 84% after 100 cycles (Figure 6e). Conversely, the initial specific energy of ≈491 Wh·kg−1 obtained in the wide 1.5−4.3 V window at a rate of C/20 shows only 40% retention after 100 cycles. With respect to rate

Figure 8. Mn K-edge pre-edge normalized XANES spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 electrode collected during (a) the first charge and (b) the first discharge. The insets show the voltage profile and the points at which each scan was started.

compared to oxygen. Importantly, contraction of M−O and M−M distances in Nax[Mn0.5Fe0.5]O2 after chemical oxidation is clearly observable. In the oxidized samples, a new peak emerges at ∼3.5 Å, whereas the intensity of the peak at ∼3 Å decreases: i.e., the second peak in the PDF curve of the P2 structure splits in the Z-phase. The splitting of the first M−M peak in Z-Na0.1[Mn0.5Fe0.5]O2 phase along with the disappearance of interlayer M−M correlation peaks (>5 Å, not shown here) was previously explained by the migration of a fraction of transition metals into the interlayer space and destruction of P2 stacking.14 We proposed that the transfer of electrons from oxygen orbitals to Fe4+ cations weakens the bonding of neighboring atoms, resulting in the migration of Fe3+ (and possibly some Mn4+) ions out of the octahedral sites of the MO2 layer. The interaction of the displaced transition metals with the adjacent oxygen layers causes contraction of the interlayer distance, in agreement with the trend observed by operando XRD. The migration of transition metal cations to the interlayer space and the high concentration of sodium vacancies - results in the glide of adjacent MO2 layers and formation of a bilayer with an O2 stacking scheme that accommodates the migrated metal cations in the interlayer tetrahedral sites, and sodium ions in the octahedral sites. A direct comparison of the effect of metal substitution is also of interest. The interlayer distance of each bilayer in ZNa0.1[Mn0.66Fe0.20Cu0.14]O2 was set so that both MOh-O and MTd-O distances are equal to the value d1 = 1.91 Å, the position of the first peak in Figure 5a. The distance between centers of bilayers (10.2 Å) was determined from the position of the first peak in the XRD pattern (Figure S2), assuming that it represents a (00l) reflection. The in-plane distances were set 6690


Article


Figure 9. Cu K-edge XAS spectra. (a) Voltage profile of a Nax[Mn0.66Fe0.20Cu0.14]O2 electrode from which the X-ray absorption spectra at the Cu Kedge were collected. The marks on the graph show the points at which data collection was started; (b−d) normalized XANES spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 at the Cu K-edge during the first charge (b), first discharge (c), and the comparison of the initial state and the end of discharge (d).

compensation mechanisms induced by insertion and extraction of sodium ions. X-ray Absorption Spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 at the Mn K-Edge. Figure 7a shows the first cycle charge/ discharge profile of a Nax[Mn0.66Fe0.20Cu0.14]O2 electrode (V vs Na metal). The XANES spectra of the sample at the Mn Kedge were measured at various sodium contents corresponding to the marks in the figure; these are the points at which each scan was started. Those marks are color coded with the corresponding normalized XANES spectra presented in Figure 7b−d and Figure S6b, and the first derivative of the normalized XANES spectra is presented in Figure S7 b-f. The XANES spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 at the Mn K-edge progressively shifted to higher energy from scan M1 to M5 over the first charge (Figure 7b), indicating the oxidation of Mn ions within this voltage range. The capacity obtained up to the point that the scan M5 was complete (at 3.4 V; t = 3.94 h, equivalent to Δx ≈ 0.20 in NaxMO2, see Figure S5) is in excellent agreement with the oxidation of all the Mn3+ in Na0.67[Mn3+0.19Mn4+0.47Fe3+0.20Cu2+0.14]O2 to Mn4+, assuming that no parasitic reaction occurs. The Cu and Fe ions are not electrochemically active in this voltage range up to 3.4 V. The Mn K-edge spectra did not change significantly from scan M5 to M10, and therefore, Mn ions do not participate in the redox process above 3.4 V. On discharge, the spectra were the same from scan M10 to M13, but shifted to lower energy from scan M13 to M17, indicating the reduction of Mn4+ in this region (Figure 7c). Figure 7d shows that the edge position in scan M17 is at lower energy than scan M1 (the pristine composition.

behavior, Na0.67[Mn0.66Fe0.20Cu0.14]O2 exhibits relatively good rate capability when cycled within the P2 stability window; retaining 93% of the initial specific energy when the rate is increased from C/20 (13 mA·g−1) to C/10 (26 mA·g−1) and from C/10 to C/5 as also shown in Figure 6e. However, when a large specific current of 190 mA·g−1 (C/1.4) is applied to the cell, rapid energy fading results. A comparison of specific energy retention for Na0.67[Mn0.66Fe0.2Cu0.14]O2, Na0.67[Mn0.65Fe0.20Ni0.15]O2, and Na0.67[Mn0.5Fe0.5]O2 cycled at two different voltage windows and rates is presented in Figure S4 (the data for the last two compositions is adapted from ref 14). In the wide voltage range, Na0.67[Mn0.5Fe0.5]O2 shows much more severe fading over the initial five cycles. However, all three compositions subsequently exhibit similar retention. Operando X-ray Absorption Spectroscopy. Figure S5 shows a comparison of the first and second galvanostatic charge/discharge profile of P2−Na0.67[Mn0.66Fe0.20Cu0.14]O2 cycled vs Na metal, obtained from a conventional coin cell and a cell designed for operando XAS measurement, at a current rate of C/20. The data collected from the cell modified for operando XAS experiments is very similar to the standard coin cell; the former has a slightly higher voltage polarization, which is expected. Lack of uniform stack pressure and conductivity in the modified coin cell designs for operando experiments affect the performance.54 The XANES spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 at the Mn, Fe, and Cu K-edges were collected during the first charge and discharge in order to gain insight into local structural changes and the charge 6691


Article


Figure 10. Fe K-edge XAS spectra. (a) Voltage profile of a Nax[Mn0.66Fe0.20Cu0.14]O2 electrode from which the X-ray absorption spectra at the Fe Kedge were collected. The marks on the graph show the points at which data collection was started. (b−d) Normalized XANES spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 at the Fe K-edge during the first charge (b), first discharge (c), and the comparison of the initial state and the end of discharge (d).

to quadrupole electric transitions to t2g and eg orbitals.30,31,55 The intensity of the Mn pre-edge peaks increased from scan M1 to scan M8 and remained unchanged from scan M8 to M10, the point at which the material adopts the Z-phase (Figure 8a). The positions of the pre-edge peaks of Mn K-edge spectra do not shift upon charge, although the continuous shift of the edge from scan M1 to M5 indicates increase in the Mn oxidation state. The pre-edge peak intensity is inversely related to the number of 3d electrons.55,56 However, the pre-edge peak intensity of the initial composition was not recovered at the end of discharge (Figure 8b). The pre-edge peak intensity depends significantly on the coordination number; the pre-edge peak intensities of four-coordinate transition-metal compounds are much higher than those of six-coordinate compounds.55 Migration of transition metals into the interlayer tetrahedral site occurs in some LixMO253,57 and NaxMO213,14 (M: 3d metal) at high voltage as discussed earlier in the PDF section. However, the increase in the intensity of pre-edge peak of this material at voltages lower than 4.1 V cannot be attributed to the migration of manganese ions into a tetrahedral site. The tetrahedral site in the transition-metal layer is too small to accommodate any manganese ions (a distance of ≈1.7 Å between X and O in XO4 tetrahedron). Formation of O-type stacking faults with available tetrahedral sites does not occur because the operando X-ray diffraction analysis of the material did not show any notable peak broadening. An increase in the Mn pre-edge peak intensity during charge was also observed in Li1.2[NixMnyCoz]O2.30−32 An initial Mn4+ oxidation state was reported for those compositions and remained unchanged

This shift indicates that the concentration of Mn3+ ions in the discharged material has increased and explains the orthorhombic distortion of Nax[Mn0.66Fe0.20Cu0.14]O2 at low voltage, revealed by operando XRD analysis. The changes in the first derivative curves are in perfect agreement with the modifications observed in the normalized XANES spectra. Mn Pre-Edge Features. The Mn K-edge XANES spectrum of Na0.67[Mn0.66Fe0.20Cu0.14]O2 also shows a well-resolved doublet pre-edge peak at 6541 and 6543 eV (Figure 7b−d). The pre-edge peak in a K-edge XANES spectrum is the consequence of two primary contributions.55 One is the electric dipole transition from the 1s orbital to the p component of p-d hybridized orbitals. The other is a 1s−3d transition, which is forbidden in the electric dipole transition mechanism, but can occur due to the electric quadrupole transition. The probability of the latter is much lower than that of the former, meaning that the electric quadrupole transition results in a much lower intensity of the pre-edge peak. An electric quadrupole transition can occur for any symmetry, whereas p−d hybridizationand thus an electric dipole transition from 1s to the p−d hybridized orbitalsdoes not occur for ideal octahedral symmetry. However, the distortion of octahedral symmetry causes mixing of 3d and 4p orbitals, resulting in an increase in the intensity of the pre-edge peak. The pre-edge peaks in K-edge XANES spectra of transition metals are sensitive to, and therefore provide information about, the coordination number, symmetry, and the number of electrons of the absorbing atom. A doublet pre-edge peak in the K-edge XANES spectrum of a 3d transition metal with an octahedral environment is assigned 6692


Article


Figure 11. Fe K-edge XAS spectra. Pre-edge spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 electrode charge and (b) the first discharge collected at insets show the voltage profile and the points at started.

more prominent. It may be that not all the copper ions are oxidized in Nax[Mn0.66Fe0.20Cu0.14]O2 during charge. Another mechanism that might be involved in the change of the local Cu ion structure is electron transfer from oxide ions to copper ions. The Cu K-edge spectrum of Na0.67[Mn0.66Fe0.20Cu0.14]O2 shows a very weak single pre-edge peak (Figure S10). A single pre-edge peak is expected for Cu2+ (d9) because the t2g orbital is fully occupied and only an electric quadrupole transition to the eg orbital can occur. The X-ray absorption spectra of the later elements of the first-row transition metals, particularly Cu, show much weaker intensity of the pre-edge peak compared to those of early transition metals in accord with what we observe. However, their pre-edge peak intensities increase significantly by changing from a 6-fold coordination environment to a 4fold.55 The lack of change of the Cu pre-edge peak over cycling shows that the octahedral environment of copper ions is well preserved. X-ray Absorption Spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 at the Fe K-Edge. The normalized XANES spectra of Fe ions in Nax[Mn0.66Fe0.20Cu0.14]O2 demonstrate no discernible change in the rising edge of absorption spectra from scan F1 to F6 during charge, but small modifications are observed in the absorption maxima (Figure S11). The most noticeable evolution in the Fe K-edge spectra during charge appears at scan F7 (which started at 4.06 V, see Figure 10a): the peak maximum intensity clearly decreases (Figure 10b). Similar behavior was observed for the Fe K-edge XANES spectrum of P2−Na2/3[Mn1/2Fe1/2]O2 on charging from 3.8 to 4.2 V.4 Scan F8 and F9 overlap perfectly and change slightly at the absorption maximum compared to scan F7. However, scan F10collected at open circuit voltage after charging to 4.3 V shows a slight change of the absorption edge and maximum compared with scans F8 and F9. During discharge, the Fe K-edge XANES spectra remain virtually the same from scan F10 to scan F12, but they change from scan F12−F13 (Figure 10c). Comparison of scan F1 and F13 demonstrates an irreversible modification in the shape of the spectrum, at the absorption maximum and higher energy region, caused by charge and discharge (Figure 10d). No shift is observed in the Fe K-edge position in the first derivative curves of the spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 at various compositions (x) during charge and discharge, although the intensity changes at high voltage (Figure S12). This behavior is different from the other transition metals measured in this study, but similar to that of Nax[Mn1/2Fe1/2]O24 and NaxFeO2,13 in which the oxidation of Fe3+ to Fe4+ was demonstrated by Mössbauer spectroscopy. The changes in the shape and the intensity of absorption maxima in Fe K-edge XANES spectra seem to originate from local structure modification at the Fe centers. If the modifications in the shape and intensity of the spectra are attributed to the oxidation of iron ions, then they contribute only a very small fraction of the capacity during the charge of the Nax[Mn0.66Fe0.20Cu0.14]O2/Na cell (see Figure S5), assuming that the Fe3+/4+ redox process dominates within the narrow voltage window of 3.9 to 4.1 V (corresponding to scan F6 to F7). It was reported14 that the Fe4+ ion fraction in the oxidized states of Nax[Mn0.5+yNiyFe0.5−2y]O2 (x ≈ 0.15, y = 0, 0.1, 0.15) measured by Mössbauer spectroscopy were systematically lower than the values expected from electrochemical measurements. We previously proposed redox activity of oxide ions based on a combination of structural characterization, Mössbauer spec-

normalized XANES during (a) the first the Fe K-edge. The which each scan was

during cycling. However, the progressive distortion of octahedral symmetry of manganese ions in Li1.2[NixMnyCoz]O2 during charge and discharge was correlated with the change in intensity of pre-edge peaks. The higher intensity of Mn preedge peaks compared to those of the Fe and Cu pre-edge peaks of the materials in our study (see below) can be attributed to the lower number of 3d electrons in Mn3+ and Mn4+. The increase of the pre-edge intensity in the Mn K-edge spectra probably originates from the Mn ion’s octahedral symmetry distortion that precludes exact description, similar to the case of some Li-rich transition-metal oxides.30−32 X-ray Absorption Spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 at the Cu K-Edge. Figure 9 and Figures S8−S9 show the normalized XANES spectra of Nax[Mn0.66Fe0.20Cu0.14]O2 and their first derivative curves measured at the Cu K-edge during first charge and discharge, along with the charge/discharge profile of the cell (Figure 9a). The spectra do not show any significant changes from scan C1 to scan C5 because here, Mn3+ oxidation predominates. A small shift to higher energy is observed in the position of the XANES spectra from scan C5 (3.6 V) to scan C7 (4.0 V; Figure 9b), which corresponds to oxidation of an additional Δx ≈ 0.12 (Figure S5) beyond the Mn3+/4+ couple that is complete at ∼3.5 V. Following that, there is no important change in the position or shape of the spectra, indicating that oxidation of Cu2+ to Cu3+ is complete around 4.0 V. During discharge, the K-edge peak shifts to lower energy from scan C11 to C12, in agreement with reduction of Cu3+ back to Cu2+. We note that the shift in the position of the absorption edge and modification of the Cu K-edge peak features reported for Nax[Cu1/3Mn2/3]O226 on charge were 6693


Article

Chemistry of Materials troscopy, and electrochemical measurements. In another study,58 about 20% of Fe4+ ions formed by charging O3NaFeO2 to 3.6 V were reduced back to Fe3+ during the storage of the charged cell at open circuit voltage. It was suggested that although conventional organic electrolytes are expected to be stable over this voltage range, the catalytic effect of Fe4+ ions facilitates the oxidation of the electrolyte. However, this is not expected to affect the spectra in this study because of the realtime nature of the operando characterization. Fe Pre-Edge Features. The shape of the pre-edge peak in the Fe K-edge spectrum of Na0.67[Mn0.66Fe0.20Cu0.14]O2 is noisy due to the interference from Mn ion X-ray absorption. The preedge peak intensity increases from scan F7 to F9 (Figure 11a), in the voltage region which the Z-phase is formed. It was reported that the intensity of the pre-edge peak similarly increased in the Fe K-edge XANES spectra of O3−Na1−xFeO2 on charge to 4 V vs Na.13 Moreover, their EXAFS data showed a decrease in the distance of the first coordination shell (Fe−O) and the intensity of the second coordination shell (Fe−Fe). Significant improvement is found in the capacity retention and the voltage polarization of O3-NaFeO2 when the cutoff voltage is limited to 3.4 V. Based on the evolution of Fe K-edge XANES data, irreversible migration of some of Fe3+ ions into tetrahedral sites in the O3 interlayer space at high voltage was suggested as the origin of the electrochemical degradation.13 As we discussed above, the increase in the intensity of pre-edge peak in our Fe K-edge XANES spectra supports the migration of a small fraction of Fe ions into interlayer P2-tetrahedral sites at high voltage (>4.1 V) based on the PDF analysis of this material and other Fe-containing P2-type sodium metal oxides.14 The intensity of the pre-edge peak did not recover to its initial value during discharge to 3.4 V, where scan F13 was collected (Figure 11b). However, the reversibility of the highvoltage phase transition in P2-type Fe-containing oxides is different from O3-type materials as evidenced by their electrochemical performance; a P2−Na0.67[Mn0.66Fe0.20Cu0.14]O2/Na cell delivers ≈96% of its initial capacity in the fifth discharge when cycled within 1.5−4.3 V range, whereas an O3NaFeO2/Na cell shows less than 20% of its initial capacity after 5 cycles when cycled within the 2.5−4.0 V range. Moreover, the first discharge capacity of O3-NaFeO2 is decreased by increasing the upper limit of the cycling voltage from 3.5 to 4.5 V.

(1) The operando XRD data shows that the P2 phase initiates a transition to a low-crystalline phase at 4.1 V, and the interlayer distance continually decreases to the end of charge (4.3 V). (2) Ex situ PDF analysis of a chemically oxidized sample demonstrates there is migration of the transition-metal ions into the tetrahedral sites of the interlayer space (with an O2-stacking), explaining the shrinkage of the interlayer distance on charge above 4.1 V. (3) Electrochemical analysis shows that charging above 4.1 V results in high-voltage polarization, capacity fading, and voltage fading. (4) The XANES spectra of the transition metals show no discernible evolution in the peak edges and maxima during charge above 4.1 V. (5) The combination of the first 3 observations above clearly shows that the electrode material in the bulk evolves on during charge at high voltage above 4.1 V. However (i.e., point 4), XANES spectra do not show any redox activity of transition metals in that voltage range, implying the redox activity of oxide anions occurs simultaneously with metal migration. The role of ligand to metal charge transfer was previously proposed in some sodium metal oxides, such as P2− Na0.67[Mn0.5Fe0.5]O213,14 and O3-NaFe0.3Ni0.7O2.15 The redox chemistry of oxide anions in Li2MO3-based oxides (M: transition metal), such as Li1+x[NiyCozMn1−x−y−z]O2 and Li2[Ru1−yMy]O3, has gained much attention recently.57,59−64 Lithium-excess manganese-rich transitional metal oxides deliver high capacities (beyond the values expected from transitionmetal redox reactions) along with a characteristic voltage plateau during the first charge. Oxygen evolution from the structure is considered an important mechanism involved in the first charge oxidation process of Li2MnO3-based oxides.65,66 It was reported that the oxygen participation into the first charge process of Li1.20[Mn0.54Co0.13Ni0.13]O260 is in the form of irreversible O2 gas evolution at the surface, whereas within the bulk, oxide ions are partially oxidized and are reduced back at the following discharge. On the other hand, a combination of X-ray photoelectron spectroscopy, transmission electron microscopy, and density functional theory (DFT) calculations studies of Li2[Ru1−yMy]O3 revealed the reversible oxidation of oxide ions to peroxo-like O22− species, concomitant with the migration of transition metals within the metal and lithium layers.57 In a recent study,67 a significant contribution of oxygen to the redox reaction of O3-NaFe0.5Ni0.5O2 was determined by a combination of X-ray spectroscopy, Mössbauer spectroscopy, and density functional theory calculations. Our operando XANES study of P2− Na0.67[Mn0.66Fe0.20Cu0.14]O2 strongly supports the reversible participation of oxide anions in the redox process. Electrolyte decomposition cannot be the main mechanism that accounts for the charge obtained at voltages higher than 4.1 V, because the accompanying phase transition implies an evolution of the structure. The evolution of oxygen from the structure cannot be the principal oxidation mechanism in this system as demonstrated by the good reversibility of the redox reaction. The migration of transition-metal ions suggested by our pair distribution function analysis appears to be concomitant with, or triggered by, possible oxide anion redox. Further study of anionic redox chemistry in this material, such as O K-edge and 3d metal L-edge XANES measurement, is underway in our lab

■

CONCLUSIONS Structural and electrochemical investigation of Na0.67[Mn3+0.19Mn4+0.47Fe3+0.20Cu2+0.14]O2 as a positive electrode demonstrates that ≈ 0.45−0.50 mol Na+ per one mole of active material are electrochemically extracted in the first charge up to 4.3 V. The operando XANES study of Nax[Mn0.66Fe0.20Cu0.14]O2 collected at the Mn, Fe and Cu Kedges indicates the contribution of each transition-metal ion to the oxidation reaction in different voltage regimes as demonstrated by the shift in the absorption edges. By correlation to the operando electrochemical profile on charge, we see that the oxidation of Mn3+, Fe3+ and Cu2+ to Mn4+, Fe3+,4+, and Cu2+,3+, respectively, compensate the charge for only the extraction of Δx ≈ 0.35 mol Na+ up to 4.1 V (x ≈ 0.2 is assigned to Mn3+/4+ oxidation; and x ≈ 0.15 to a combination of Cu2/3+ and Fe3/4+ redox). A summary of our observations on Nax[Mn0.66Fe0.20Cu0.14]O2 when charged above 4.1 V is as follows: 6694


Article


Nax[Fe0.5Mn0.5]O2 made from earth-abundant elements for rechargeable Na batteries. Nat. Mater. 2012, 11, 512−517. (5) Mortemard de Boisse, B.; Carlier, D.; Guignard, M.; Delmas, C. Structural and Electrochemical Characterizations of P2 and New O3NaxMn1‑yFeyO2 Phases Prepared by Auto-Combustion Synthesis for Na-Ion Batteries. J. Electrochem. Soc. 2013, 160, A569−A574. (6) Thorne, J. S.; Dunlap, R. A.; Obrovac, M. N. Structure and Electrochemistry of NaxFexMn1‑xO2 (1.0 ≤ x ≤ 0.5) for Na-Ion Battery Positive Electrodes. J. Electrochem. Soc. 2013, 160, A361−A367. (7) Sharma, N.; Gonzalo, E.; Pramudita, J. C.; Han, M. H.; Brand, H. E. A.; Hart, J. N.; Pang, W. K.; Guo, Z.; Rojo, T. The Unique Structural Evolution of the O3-Phase Na2/3Fe2/3Mn1/3O2 during High Rate Charge/Discharge: A Sodium-Centred Perspective. Adv. Funct. Mater. 2015, 25, 4994−5005. (8) Mortemard de Boisse, B.; Carlier, D.; Guignard, M.; Bourgeois, L.; Delmas, C. P2-NaxMn1/2Fe1/2O2 Phase Used as Positive Electrode in Na Batteries: Structural Changes Induced by the Electrochemical (De)intercalation Process. Inorg. Chem. 2014, 53, 11197−11205. (9) Mortemard de Boisse, B.; Cheng, J. H.; Carlier, D.; Guignard, M.; Pan, C. J.; Bordere, S.; Filimonov, D.; Drathen, C.; Suard, E.; Hwang, B. J.; Wattiaux, A.; Delmas, C. O3-NaxMn1/3Fe2/3O2 as a positive electrode material for Na-ion batteries: structural evolutions and redox mechanisms upon Na+ (de) intercalation. J. Mater. Chem. A 2015, 3, 10976−10989. (10) Takeda, Y.; Nakahara, K.; Nishijima, M.; Imanishi, N.; Yamamoto, O.; Takano, M.; Kanno, R. Sodium Deintercalation from Sodium Iron-Oxide. Mater. Res. Bull. 1994, 29, 659−666. (11) Okada, S.; Takahashi, Y.; Kiyabu, T.; Doi, T.; Yamaki, J.; Nishida, T. Layered Transition Metal Oxides as Cathodes for Sodium Secondary Battery. ESC Abstr. 2006, 201−202. (12) Yabuuchi, N.; Yoshida, H.; Komaba, S. Crystal Structures and Electrode Performance of Alpha-NaFeO2 for Rechargeable Sodium Batteries. Electrochemistry 2012, 80, 716−719. (13) Yabuuchi, N.; Komaba, S. Recent research progress on iron- and manganese-based positive electrode materials for rechargeable sodium batteries. Sci. Technol. Adv. Mater. 2014, 15, 043501. (14) Talaie, E.; Duffort, V.; Smith, H. L.; Fultz, B.; Nazar, L. F. Structure of the high voltage phase of layered P2-Na2/3‑z[Mn1/2Fe1/2]O2 and the positive effect of Ni substitution on its stability. Energy Environ. Sci. 2015, 8, 2512−2523. (15) Wang, X. F.; Liu, G. D.; Iwao, T.; Okubo, M.; Yamada, A. Role of Ligand-to-Metal Charge Transfer in O3-Type NaFeO2-NaNiO2 Solid Solution for Enhanced Electrochemical Properties. J. Phys. Chem. C 2014, 118, 2970−2976. (16) Kim, D.; Lee, E.; Slater, M.; Lu, W. Q.; Rood, S.; Johnson, C. S. Layered Na[Ni1/3Fe1/3Mn1/3]O2 cathodes for Na-ion battery application. Electrochem. Commun. 2012, 18, 66−69. (17) Yoshida, H.; Yabuuchi, N.; Komaba, S. NaFe0.5Co0.5O2 as high energy and power positive electrode for Na-ion batteries. Electrochem. Commun. 2013, 34, 60−63. (18) Li, X.; Wu, D.; Zhou, Y. N.; Liu, L.; Yang, X. Q.; Ceder, G. O3type Na(Mn0.25Fe0.25Co0.25Ni0.25)O2: A quaternary layered cathode compound for rechargeable Na ion batteries. Electrochem. Commun. 2014, 49, 51−54. (19) Liu, L.; Li, X.; Bo, S. H.; Wang, Y.; Chen, H. L.; Twu, N.; Wu, D.; Ceder, G. High-Performance P2-Type Na2/3(Mn1/2Fe1/4Co1/4)O2 Cathode Material with Superior Rate Capability for Na-Ion Batteries. Adv. Energy Mater. 2015, 5, 1500944. (20) Han, M. H.; Gonzalo, E.; Sharma, N.; Lopez del Amo, J. M.; Armand, M.; Avdeev, M.; Saiz Garitaonandia, J. J. S.; Rojo, T. HighPerformance P2-Phase Na2/3Mn0.8Fe0.1Ti0.1O2 Cathode Material for Ambient-Temperature Sodium-Ion Batteries. Chem. Mater. 2016, 28, 106−116. (21) Han, M. H.; Gonzalo, E.; Singh, G.; Rojo, T. A comprehensive review of sodium layered oxides: powerful cathodes for Na-ion batteries. Energy Environ. Sci. 2015, 8, 81−102. (22) Kundu, D.; Talaie, E.; Duffort, V.; Nazar, L. F. The Emerging Chemistry of Sodium Ion Batteries for Electrochemical Energy Storage. Angew. Chem., Int. Ed. 2015, 54, 3431−3448.

to achieve an in-depth understanding of the involved redox reactions, which is necessary for the design of high-performance electrode materials for Na-ion batteries.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemmater.7b01146. Schematic representation of the operando XAS cell; PDF data for Z-Na0.1[Mn 0.66 Fe 0.2Cu 0.14 ]O2 and the fit compared to Z-Na0.1[Mn0.66Fe0.2Ni0.15]O2 RD patterns, electrochemical performance of energy density as a function of cycling for Z-Na0.1[Mn0.66Fe0.2Cu0.14]O2 compared to other related electrode materials, cell performance for a coin cell and operando cell, and additional XANES data (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Linda F. Nazar: 0000-0002-3314-8197 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge BASF SE for ongoing support through the BASF Research Network in Electrochemistry and Batteries, and the Natural Sciences and Engineering Council of Canada for partial financial support of this work through their Discovery and Canada Research Chair programs, and the Canadian Light Source (CLS). X-ray absorption spectroscopy experiments in this research were performed at the Canadian Light Source (HXMA beamline), which is supported by the Canada Foundation for Innovation, Natural Sciences and Engineering Research Council of Canada, the University of Saskatchewan, the Government of Saskatchewan, Western Economic Diversification Canada, the National Research Council Canada, and the Canadian Institutes of Health Research. E.T. and S.Y.K. thank the Canadian Light Source for graduate student travel support. The PDF data for ZNa0.1[Mn0.5Fe0.5]O2 and Z-Na0.1[Mn0.65Fe0.20Ni0.15]O2, collected and analyzed by Dr. Victor Duffort, was taken from our previous study (ref 14). We thank Dr. Duffort for establishing the groundwork for the PDF analysis, the bilayer model and for very helpful discussions.

■

REFERENCES

(1) Larcher, D.; Tarascon, J. M. Towards greener and more sustainable batteries for electrical energy storage. Nat. Chem. 2015, 7, 19−29. (2) de la Llave, E.; Borgel, V.; Park, K. J.; Hwang, J. Y.; Sun, Y. K.; Hartmann, P.; Chesneau, F. F.; Aurbach, D. Comparison between NaIon and Li-Ion Cells: Understanding the Critical Role of the Cathodes Stability and the Anodes Pretreatment on the Cells Behavior. ACS Appl. Mater. Interfaces 2016, 8, 1867−1875. (3) Yabuuchi, N.; Kubota, K.; Dahbi, M.; Komaba, S. Research Development on Sodium-Ion Batteries. Chem. Rev. 2014, 114, 11636− 11682. (4) Yabuuchi, N.; Kajiyama, M.; Iwatate, J.; Nishikawa, H.; Hitomi, S.; Okuyama, R.; Usui, R.; Yamada, A.; Komaba, S. P2-type 6695


Article


Power and Long-Life Sodium-Ion Batteries. ACS Appl. Mater. Interfaces 2016, 8, 31661−31668. (41) Shannon, R. D. Revised Effective Ionic-Radii and Systematic Studies of Interatomic Distances in Halides and Chalcogenides. Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr. 1976, 32, 751−767. (42) Duffort, V.; Talaie, E.; Black, R.; Nazar, L. F. Uptake of CO2 in Layered P2-Na0.67Mn0.5Fe0.5O2: Insertion of Carbonate Anions. Chem. Mater. 2015, 27, 2515−2524. (43) Stephens, P. W. Phenomenological model of anisotropic peak broadening in powder diffraction. J. Appl. Crystallogr. 1999, 32, 281− 289. (44) Sharma, N.; Tapia-Ruiz, N.; Singh, G.; Armstrong, A. R.; Pramudita, J. C.; Brand, H. E. A.; Billaud, J.; Bruce, P. G.; Rojo, T. Rate Dependent Performance Related to Crystal Structure Evolution of Na0.67Mn0.8Mg0.2O2 in a Sodium-Ion Battery. Chem. Mater. 2015, 27, 6976−6986. (45) Sharma, N.; Han, M. H.; Pramudita, J. C.; Gonzalo, E.; Brand, H. E. A.; Rojo, T. A comprehensive picture of the current rate dependence of the structural evolution of P2-Na2/3 Fe2/3Mn1/3O2. J. Mater. Chem. A 2015, 3, 21023−21038. (46) Parant, J.-P.; Olazcuaga, R.; Devalette, M.; Fouassier, C.; Hagenmuller, P. Sur quelques nouvelles phases de formule NaxMnO2 (x ≤ 1). J. Solid State Chem. 1971, 3, 1−6. (47) Burns, P. C.; Hawthorne, F. C. Static and dynamic Jahn-Teller effects in Cu2+ oxysalt minerals. Can. Mineral. 1996, 34, 1089−1105. (48) Kubota, K.; Ikeuchi, I.; Nakayama, T.; Takei, C.; Yabuuchi, N.; Shiiba, H.; Nakayama, M.; Komaba, S. New Insight into Structural Evolution in Layered NaCrO2 during Electrochemical Sodium Extraction. J. Phys. Chem. C 2015, 119, 166−175. (49) Nucker, N.; Fink, J.; Fuggle, J. C.; Durham, P. J.; Temmerman, W. M. Evidence for Holes on Oxygen Sites in the High-Tc Superconductors La2‑xSrxCuO4 and YBa2Cu3O7‑y. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37, 5158−5163. (50) Choudhury, D.; Rivero, P.; Meyers, D.; Liu, X.; Cao, Y.; Middey, S.; Whitaker, M. J.; Barraza-Lopez, S.; Freeland, J. W.; Greenblatt, M.; Chakhalian, J. Anomalous charge and negative-charge-transfer insulating state in cuprate chain compound KCuO2. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 92, 201108. (51) Nimkar, S.; Sarma, D. D.; Krishnamurthy, H. R. ElectronicStructure of NaCuO2. Phys. Rev. B: Condens. Matter Mater. Phys. 1993, 47, 10927−10930. (52) Goodenough, J. B.; Zhou, J. S.; Allan, K. Charge Fluctuations and an Ionic Covalent Transition in La2‑xSrxCuO4. J. Mater. Chem. 1991, 1, 715−724. (53) Mohanty, D.; Li, J. L.; Abraham, D. P.; Huq, A.; Payzant, E. A.; Wood, D. L.; Daniel, C. Unraveling the Voltage-Fade Mechanism in High-Energy-Density Lithium-Ion Batteries: Origin of the Tetrahedral Cations for Spinel Conversion. Chem. Mater. 2014, 26, 6272−6280. (54) Borkiewicz, O. J.; Wiaderek, K. M.; Chupas, P. J.; Chapman, K. W. Best Practices for Operando Battery Experiments: Influences of Xray Experiment Design on Observed Electrochemical Reactivity. J. Phys. Chem. Lett. 2015, 6, 2081−2085. (55) Yamamoto, T. Assignment of pre-edge peaks in K-edge x-ray absorption spectra of 3d transition metal compounds: electric dipole or quadrupole? X-Ray Spectrom. 2008, 37, 572−584. (56) Iwasawa, Y. X-ray Absorption Fine Structure for Catalysts and Surfaces; World Scientific Publishing Co. Pte. Ltd.: Singapore, 1996. (57) Sathiya, M.; Abakumov, A. M.; Foix, D.; Rousse, G.; Ramesha, K.; Saubanere, M.; Doublet, M. L.; Vezin, H.; Laisa, C. P.; Prakash, A. S.; Gonbeau, D.; VanTendeloo, G.; Tarascon, J. M. Origin of voltage decay in high-capacity layered oxide electrodes. Nat. Mater. 2015, 14, 230−238. (58) Lee, E.; Brown, D. E.; Alp, E. E.; Ren, Y.; Lu, J.; Woo, J. J.; Johnson, C. S. New Insights into the Performance Degradation of FeBased Layered Oxides in Sodium-Ion Batteries: Instability of Fe3+/Fe4+ Redox in alpha-NaFeO2. Chem. Mater. 2015, 27, 6755−6764. (59) Sathiya, M.; Rousse, G.; Ramesha, K.; Laisa, C. P.; Vezin, H.; Sougrati, M. T.; Doublet, M. L.; Foix, D.; Gonbeau, D.; Walker, W.;

(23) Ono, Y.; Yui, Y.; Hayashi, M.; Asakura, K.; Kitabayashi, H.; Takahashi, K. I. Electrochemical Properties of NaCuO2 for Sodium-ion Secondary Batteries. ECS Trans. 2014, 58, 33−39. (24) Xu, S. Y.; Wu, X. Y.; Li, Y. M.; Hu, Y. S.; Chen, L. Q. Novel copper redox-based cathode materials for room-temperature sodiumion batteries. Chin. Phys. B 2014, 23, 118202. (25) Mu, L. Q.; Hu, Y. S.; Chen, L. Q. New layered metal oxides as positive electrode materials for room-temperature sodium-ion batteries. Chin. Phys. B 2015, 24, 038202. (26) Mason, C. W.; Lange, F.; Saravanan, K.; Lin, F.; Nordlund, D. Beyond Divalent Copper: A Redox Couple for Sodium Ion Battery Cathode Materials. ECS Electrochem. Lett. 2015, 4, A41−A44. (27) Mu, L. Q.; Xu, S. Y.; Li, Y. M.; Hu, Y. S.; Li, H.; Chen, L. Q.; Huang, X. J. Prototype Sodium-Ion Batteries Using an Air-Stable and Co/Ni-Free O-3-Layered Metal Oxide Cathode. Adv. Mater. 2015, 27, 6928−6933. (28) Li, Y. M.; Yang, Z. Z.; Xu, S. Y.; Mu, L. Q.; Gu, L.; Hu, Y. S.; Li, H.; Chen, L. Q. Air-Stable Copper-Based P2-Na7/9Cu2/9Fe1/9Mn2/3O2 as a New Positive Electrode Material for Sodium-Ion Batteries. Adv. Sci. 2015, 2, 1500031. (29) Zheng, L.; Li, J.; Obrovac, M. Crystal Structures and Electrochemical Performance of Air-Stable Na2/3Ni1/3‑xCuxMn2/3O2 in Sodium Cells. Chem. Mater. 2017, 29, 1623−1631. (30) Oishi, M.; Fujimoto, T.; Takanashi, Y.; Orikasa, Y.; Kawamura, A.; Ina, T.; Yamashige, H.; Takamatsu, D.; Sato, K.; Murayama, H.; Tanida, H.; Arai, H.; Ishii, H.; Yogi, C.; Watanabe, I.; Ohta, T.; Mineshige, A.; Uchimoto, Y.; Ogumi, Z. Charge compensation mechanisms in Li1.16Ni0.15Co0.19Mn0.5O2 positive electrode material for Li-ion batteries analyzed by a combination of hard and soft X-ray absorption near edge structure. J. Power Sources 2013, 222, 45−51. (31) Buchholz, D.; Li, J.; Passerini, S.; Aquilanti, G.; Wang, D. D.; Giorgetti, M. X-ray Absorption Spectroscopy Investigation of LithiumRich, Cobalt-Poor Layered-Oxide Cathode Material with High Capacity. ChemElectroChem 2015, 2, 85−97. (32) Koga, H.; Croguennec, L.; Menetrier, M.; Mannessiez, P.; Weill, F.; Delmas, C.; Belin, S. Operando X-ray Absorption Study of the Redox Processes Involved upon Cycling of the Li-Rich Layered Oxide Li1.20Mn0.54Co0.13Ni0.13O2 in Li Ion Batteries. J. Phys. Chem. C 2014, 118, 5700−5709. (33) Karan, N. K.; Slater, M. D.; Dogan, F.; Kim, D.; Johnson, C. S.; Balasubramanian, M. Operando Structural Characterization of the Lithium-Substituted Layered Sodium-Ion Cathode Material P2Na0.85Li0.17Ni0.21Mn0.64O2 by X-ray Absorption Spectroscopy. J. Electrochem. Soc. 2014, 161, A1107−A1115. (34) Wu, X. H.; Xu, G. L.; Zhong, G. M.; Gong, Z. L.; McDonald, M. J.; Zheng, S. Y.; Fu, R. Q.; Chen, Z. H.; Amine, K.; Yang, Y. Insights into the Effects of Zinc Doping on Structural Phase Transition of P2Type Sodium Nickel Manganese Oxide Cathodes for High-Energy Sodium Ion Batteries. ACS Appl. Mater. Interfaces 2016, 8, 22227− 22237. (35) Rodriguez-Carvajal, J. Recent Advances in Magnetic-Structure Determination by Neutron Powder Diffraction. Phys. B 1993, 192, 55−69. (36) Le Bail, A. Whole powder pattern decomposition methods and applications: A retrospection. Powder Diffr. 2005, 20, 316−326. (37) Juhas, P.; Davis, T.; Farrow, C. L.; Billinge, S. J. L. PDFgetX3: a rapid and highly automatable program for processing powder diffraction data into total scattering pair distribution functions. J. Appl. Crystallogr. 2013, 46, 560−566. (38) Farrow, C. L.; Juhas, P.; Liu, J. W.; Bryndin, D.; Bozin, E. S.; Bloch, J.; Proffen, T.; Billinge, S. J. L. PDFfit2 and PDFgui: computer programs for studying nanostructure in crystals. J. Phys.: Condens. Matter 2007, 19, 335219. (39) Ravel, B.; Newville, M. ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron Radiat. 2005, 12, 537−541. (40) Kang, W. P.; Yu, D. Y. W.; Lee, P. K.; Zhang, Z. Y.; Bian, H. D.; Li, W. Y.; Ng, T. W.; Zhang, W. J.; Lee, C. S. P2-Type NaxCu0.15Ni0.20Mn0.65O2 Cathodes with High Voltage for High6696


Article

Chemistry of Materials Prakash, A. S.; Ben Hassine, M.; Dupont, L.; Tarascon, J. M. Reversible anionic redox chemistry in high-capacity layered-oxide electrodes. Nat. Mater. 2013, 12, 827−835. (60) Koga, H.; Croguennec, L.; Menetrier, M.; Douhil, K.; Belin, S.; Bourgeois, L.; Suard, E.; Weill, F.; Delmas, C. Reversible Oxygen Participation to the Redox Processes Revealed for Li1.20Mn0.54Co0.13Ni0.13O2. J. Electrochem. Soc. 2013, 160, A786−A792. (61) Foix, D.; Sathiya, M.; McCalla, E.; Tarascon, J. M.; Gonbeau, D. X-ray Photoemission Spectroscopy Study of Cationic and Anionic Redox Processes in High-Capacity Li-Ion Battery Layered-Oxide Electrodes. J. Phys. Chem. C 2016, 120, 862−874. (62) Grimaud, A.; Hong, W. T.; Shao-Horn, Y.; Tarascon, J. M. Anionic redox processes for electrochemical devices. Nat. Mater. 2016, 15, 121−126. (63) Seo, D. H.; Lee, J.; Urban, A.; Malik, R.; Kang, S.; Ceder, G. The structural and chemical origin of the oxygen redox activity in layered and cation-disordered Li-excess cathode materials. Nat. Chem. 2016, 8, 692−697. (64) Perez, A. J.; Batuk, D.; Saubanere, M.; Rousse, G.; Foix, D.; McCalla, E.; Berg, E. J.; Dugas, R.; van den Bos, K. H. W.; Doublet, M. L.; Gonbeau, D.; Abakumov, A. M.; Van Tendeloo, G.; Tarascon, J. M. Strong Oxygen Participation in the Redox Governing the Structural and Electrochemical Properties of Na-Rich Layered Oxide Na2IrO3. Chem. Mater. 2016, 28, 8278−8288. (65) La Mantia, F.; Rosciano, F.; Tran, N.; Novak, P. Quantification of Oxygen Loss from Li1+x(Ni1/3Mn1/3Co1/3)1‑xO2 at High Potentials by Differential Electrochemical Mass Spectrometry. J. Electrochem. Soc. 2009, 156, A823−A827. (66) Rana, J.; Stan, M.; Kloepsch, R.; Li, J.; Schumacher, G.; Welter, E.; Zizak, I.; Banhart, J.; Winter, M. Structural Changes in Li2MnO3 Cathode Material for Li- Ion Batteries. Adv. Energy Mater. 2014, 4, 1300998. (67) Nanba, Y.; Iwao, T.; Mortemard de Boisse, B.; Zhao, W. W.; Hosono, E.; Asakura, D.; Niwa, H.; Kiuchi, H.; Miyawaki, J.; Harada, Y.; Okubo, M.; Yamada, A. Redox Potential Paradox in NaxMO2 for Sodium-Ion Battery Cathodes. Chem. Mater. 2016, 28, 1058−1065.

6697


Structural Evolution and Redox Processes Involved in the

Recommend Documents