High-entropy oxides (HEOs) are single phase solid solutions where five or more metals share the same sublattice, giving rise to unexpected features in various fields of applications. Recently, HEOs have emerged as an alternative conversion electrode anode material for next-generation Li-ion batteries, where the combination of several different elements in a single solid solution can synergistically act to overcome some of its main drawbacks, improving performance. Due to their chemical complexity, x-ray absorption spectroscopy (XAS) emerges as an appropriate technique to study the electronic (x-ray absorption near edge structure, XANES) and local structure (extended x-ray absorption fine structure, EXAFS) of these compounds as a function of cycling. This work aims to highlight the capabilities of XAS as an element-specific probe to understand a material’s structure at the atomistic level through EXAFS modeling of (MgFeCoNiCuZn)O high-entropy system and how to extract valuable information about the bond distance, number of near neighbors, and local disorder, which are crucial to a full understanding of the electrochemical reaction mechanisms of such battery electrodes.

High-entropy oxides (HEOs) are disordered solid solutions where five or more metals randomly share the cationic position in the oxide lattice while oxygen remains in the other. In this case, the configurational entropy overcomes any enthalpic contribution to the Gibbs free energy, playing the main role in the compound’s structural stabilization. Due to the temperature dependence of the entropic contribution, a single solid solution phase stably exists at only high temperatures.1–4 This stabilization mechanism has been exhaustively used recently to prepare different compounds other than oxides, having B, C, N, and even F as the anionic species and having a variety of structures, from the symmetric rock-salt and fluorite to the more complex spinels and perovskites.5–8 This random cationic distribution enables the so-called cocktail effect, where different metals can synergistically act to overcome individual and detrimental characteristics of the individual metals alone.9,10

Over the past few years, high-entropy oxides have emerged as an electrode material for Li-ion batteries. The continuous push for materials, which are less expensive, more abundant, and more environment-friendly than those of conventional electrodes, has led to extensive efforts to develop new Li-ion battery chemistries. Exploiting the entropy stabilization effect significantly increased the compositional space of chemical mixtures.11–23 However, the inherent chemical disorder increases the analytical complexity of those systems, making structural analysis difficult. X-ray absorption spectroscopy (XAS) is, therefore, an ideal tool for electronic and local structural characterization of disordered complex systems, such as high-entropy oxides.24–27 As an element-specific technique, it can probe local ordering at short range, revealing valuable information about coordination, bond distance, local disorder, and neighboring atoms, isolating individual contributions during HEO electrochemical operation. Using a combination of XANES and EXAFS, we aim to fundamentally understand the (MgFeCoNiCuZn)O short-range ordering arrangement and electrochemical reaction mechanism, integrating high-entropy oxides and energy storage through absorption spectroscopy.

The six-element high-entropy oxide (Mg 1 / 6Fe 1 / 6Co 1 / 6Ni 1 / 6Cu 1 / 6Zn 1 / 6)O, denoted (MgFeCoNiCuZn)O, was prepared by solid-state reaction from equimolar amounts of the individual precursor metal oxides: MgO, Fe 2O 3, CoO, NiO, CuO, and ZnO, at 1150 ° C followed by rapid quenching to room temperature. The crystal structure and short-range arrangement were analyzed by x-ray diffraction (XRD) and extended x-ray absorption spectroscopy (XAS/EXAFS), respectively. XRD data were taken in a Bruker D2 Phaser diffractometer in Bragg–Brentano geometry with a Ni filter and LynxEye detector, and Pawley refinement was used to obtain information about the lattice. The refinement is based on the (MgCoNiCuZn)O rock-salt structure, and details can be found elsewhere.14, Ex situ XAS was measured in the continuous scan mode at the Materials Research Collaborative Access Team (MRCAT) beamline 10-BM, at the Advanced Photon Source, in the fluorescence mode using a five-grid ion chamber.28 Although synthesized as a single rock-salt phase as revealed by x-ray diffraction [Fig. 1(a)], the local structure around the different metal atoms present in its composition reveals significant differences in the first two coordination shells, as can be seen by the Fourier transform of the absorption fine structure (EXAFS), in Fig. 1(b). In this case, x-ray diffraction averages the structure over all highly symmetric metallic sites in the crystal lattice, even though those metals have a lower individual symmetry, giving an averaged structure over the unit cells, thus limiting the local structure description.29,30

FIG. 1.

(a) X-ray diffraction (XRD) confirms (MgFeCoNiCuZn)O as a single rock-salt phase (Fm-3m), sintered at 1150 °C. (b) Extended x-ray absorption spectroscopy (EXAFS) at different metallic edges shows the rock-salt structural signature at the local level, consisting of two independent atomic shells of oxygen and metallic cations, respectively.

FIG. 1.

(a) X-ray diffraction (XRD) confirms (MgFeCoNiCuZn)O as a single rock-salt phase (Fm-3m), sintered at 1150 °C. (b) Extended x-ray absorption spectroscopy (EXAFS) at different metallic edges shows the rock-salt structural signature at the local level, consisting of two independent atomic shells of oxygen and metallic cations, respectively.

Close modal

Despite the fact that EXAFS shows the same rock-salt signature for all metallic edges, the position and shape of these features differ among the metallic species pointing to inconsistencies such as the bond distance, number of near neighbors, and local ordering that are usually imperceptible by XRD. For example, the Fe K-edge EXAFS of (MgFeCoNiCuZn)O [Fig. 1(b), in purple] shows the first oxygen shell at around 1.45 Å, which is slightly bigger than the Cu–O (1.41 Å), and smaller than Co–O (1.56 Å), Ni–O (1.56 Å), and Zn–O (1.58 Å) environments. It is important to note that the Fourier transformed (FT) EXAFS is considered a pseudo-radial distribution function and the distances mentioned above are not really the metal–oxygen bond distances, being shifted by 0.5 Å due to the effective scattering factor and phase-shift of surrounding atoms, but it is a qualitative indication of different local neighboring environments. The amplitude and shape of the first oxygen shell can also reveal differences in the local arrangement around the transition metals, which can be related to coordination number, and local disordering. Careful investigation of the second shell points in the same direction, where differences in the bond distance, coordination, and disorder are visible. In this case, the second shell is a random and uniform mixture of the six metal cations due to the entropy stabilization process. The FT EXAFS is useful for qualitative understanding of the system but can be inaccurate and misleading. Thus, quantitative data analysis relying on the k-space modulated scattering spectra should be used to assess those differences.

In an XAS experiment, the x-ray absorption of the sample is measured as a function of the incident x-ray energy. At energies above the edge, the interference between the outgoing photoelectron wave and the wave backscattered from neighboring atoms modulates the absorption cross section of the absorbing element, containing precise information of short-range ordering, known as EXAFS (extended x-ray absorption fine structure). The x-ray absorption near edge structure, or XANES, provides information about the electronic structure and local geometry and is located within a few eVs of the element’s absorption edge. There is no clear boundary between XANES and EXAFS regions, but the latter conventionally starts at 20–30 eV above the edge position, where the core hole excitations are less pronounced.31,32

The physical description of the EXAFS region, which essentially are the oscillations above the edge, accounts for the multiple photoelectron scattering processes around the absorbing atom due to the neighboring environment. Thus, the EXAFS can be simply interpreted as the sum of the individual contributions from each scattering atom, and it is given by the EXAFS equation [Eq. (1)]. In the EXAFS equation, the effective scattering amplitude f ( k ) and phase-shift δ ( k ) are properties of the neighboring atoms, while the path degeneracy/coordination number ( N), bond distance ( R), and Debye–Waller factor ( σ 2) describe the local geometry of the absorber–scatterer interaction. The latter is the root mean square deviation over the absorber–scatterer distance, accounting for thermal and local disordering. The effective scattering amplitude and phase-shift depend on the atomic number and thus contain the necessary information about the nature of the scattering atoms. S 0 2 in Eq. (1) accounts for the inelastic loss process due to multi-electron excitations at the absorbing atom. The typical distance that a photoelectron can travel before it scatters inelastically and loses coherence combined with the core-hole lifetime is called the mean-free path, λ ( k ), and is one of the reasons that EXAFS is a local structure probe, not able to see much further than 5 Å,33–35 
χ ( k ) = j S 0 2 N j f j ( k ) e 2 R j / λ ( k ) e 2 k 2 σ j 2 k R j 2 sin [ 2 k R j + δ j ( k ) ] .
(1)

The structural information is encoded in the amplitude, shape, phase, and frequency of the EXAFS oscillations, χ ( k ). In order to extract the local structural terms ( N, R, and σ 2), it is necessary to calculate the parameters that define the scattering process f ( k ), δ ( k ), and λ ( k ). Ab initio calculations using the muffin-tin formalism of model compounds have been a reliable source of local structural modeling representing the EXAFS as a sum of geometric contributions of scattering paths.36–41 

Figure 2(a) shows the (MgFeCoNiCuZn)O EXAFS model of the Fe K-edge. Scattering paths were calculated using the rock-salt (Fm-3m) structure in the IFEFFIT computing engine in Artemis.42,43 In this model, the first coordination shell consists of the octahedral oxygen environment, being a single scattering path with a six-fold degeneracy. In high-entropy oxides, the metal cations randomly and homogeneously share the same lattice position and are cooperatively responsible for the second coordination shell in the FT EXAFS. In this case, six different single scattering paths with degeneracy 2, one for each metal present in the cation sublattice were used, corresponding to a total of 12 atoms in the second coordination shell as expected from a rock-salt structure.

FIG. 2.

EXAFS model of (MgFeCoNiCuZn)O high-entropy oxide. The model consists of a Fe–O path (in green) referent to the first coordination shell, and six Fe–M paths where M = Mg, Fe, Co, Ni, Cu, and Zn cations. All paths are single scattering and the degeneracy of the Fe–M is 2 for each cation in M, considering the random and homogeneous nature of the high-entropy solid solution. The fitting range is marked with a Hanning window function (blue dashed line) having Δ R = 0.2 Å, isolating the first two shells in this analysis.

FIG. 2.

EXAFS model of (MgFeCoNiCuZn)O high-entropy oxide. The model consists of a Fe–O path (in green) referent to the first coordination shell, and six Fe–M paths where M = Mg, Fe, Co, Ni, Cu, and Zn cations. All paths are single scattering and the degeneracy of the Fe–M is 2 for each cation in M, considering the random and homogeneous nature of the high-entropy solid solution. The fitting range is marked with a Hanning window function (blue dashed line) having Δ R = 0.2 Å, isolating the first two shells in this analysis.

Close modal

Ideally, the model should account for the different scattering factors of Mg, Fe, Co, Ni, Cu, and Zn atoms; however, f ( k ) and δ ( k ) depend weakly on the identity of the scattering atom and are relatively insensitive to small differences in Z. This distinction becomes even more subtle considering that five out of the six cations in (MgFeCoNiCuZn)O lie within Z ± 4, making them almost indistinguishable also for x-ray scattering.40,44,45 Keeping this limitation in mind, the second scattering path was also modeled as consisting of only Fe, Co, Ni, Cu, or Zn ( N = 12) and compared to the ideal model, called Me herein. The purpose is to simplify the overall EXAFS analysis, by decreasing the variability of N combinations of 6 different scatterers in 12 atomic positions and reducing the necessity of using a large number of variable parameters. It is important to note that this approach disregards any short-range ordering contribution to the EXAFS model and the homogeneity of the high-entropy solid solution. Mg is also disregarded in this simplified model as it has a considerably different scattering factor and is not as strong as the transition metals, accounting for only 1/6 of the cations in the high-entropy composition.

Figure 2(b) shows the quality of the fit in terms of the R-factor by the type of near neighbors in the second path for all the transition metal K-edges in the high-entropy compound. The neighbor Me denotes the sum of the six single paths for each of the cations with N = 2 for each. In the models, bond distance ( R TM-O) and Debye–Waller factor ( σ TM-O 2) of the first oxygen coordination shell are allowed to vary in order to capture the local structure around the different absorbers. R TM-M and σ TM-M 2 are the same and allowed to vary for all six TM–Me scattering paths. N TM-O and N TM-M are set to 6 and 12, respectively, maintaining the consistency of the rock-salt structure without assuming oxygen vacancies. As can be seen from Fig. 2(b), the quality of the fits slightly improves when a single element from Fe to Zn is used as the sole representative of the second scattering path, instead of a homogeneous mixture (Me), and it has its minimum for Fe, Co, and Ni scattering species. This improvement compared to the ideal model (Me) is probably due to the limited k-range (2–12 Å 1), the similarity of scattering factors of the transition metals, and the limitation to only two scattering distances, that cannot capture the complexity of the entropic mixture but can easily explain the differences in the short-range structure. The EXAFS fit results using the six different models for the Fe K-edge of (MgFeCoNiCuZn)O are shown in Table I and the corresponding fits for the Co, Ni, Cu, and Zn edges are shown in the supplementary material (Tables SI–SIV).

TABLE I.

(MgFeCoNiCuZn)O Fe K-edge EXAFS fit results for bond lengths (R), and σ2 for the six different EXAFS models showing the consistency of the resulting model parameters.

ModelRFe-O (Å) σ Fe-O 2 (10−3 Å2)RFe-M (Å) σ Fe-M 2 (10−3 Å2)R-factor (%)
12Fe 1.98 ± 0.01 9.7 ± 1.2 3.03 ± 0.01 16.3 ± 1.1 1.7 
12Co 1.98 ± 0.01 9.8 ± 1.0 3.02 ± 0.01 15.9 ± 1.0 1.3 
12Ni 1.98 ± 0.01 9.8 ± 1.0 3.01 ± 0.01 15.6 ± 0.9 1.2 
12Cu 1.97 ± 0.01 9.8 ± 1.2 3.01 ± 0.01 14.9 ± 1.0 1.5 
12Zn 1.97 ± 0.01 9.9 ± 1.3 3.00 ± 0.02 14.7 ± 1.1 1.9 
12Me 1.99 ± 0.02 9.8 ± 1.5 3.00 ± 0.02 13.4 ± 1.2 2.4 
ModelRFe-O (Å) σ Fe-O 2 (10−3 Å2)RFe-M (Å) σ Fe-M 2 (10−3 Å2)R-factor (%)
12Fe 1.98 ± 0.01 9.7 ± 1.2 3.03 ± 0.01 16.3 ± 1.1 1.7 
12Co 1.98 ± 0.01 9.8 ± 1.0 3.02 ± 0.01 15.9 ± 1.0 1.3 
12Ni 1.98 ± 0.01 9.8 ± 1.0 3.01 ± 0.01 15.6 ± 0.9 1.2 
12Cu 1.97 ± 0.01 9.8 ± 1.2 3.01 ± 0.01 14.9 ± 1.0 1.5 
12Zn 1.97 ± 0.01 9.9 ± 1.3 3.00 ± 0.02 14.7 ± 1.1 1.9 
12Me 1.99 ± 0.02 9.8 ± 1.5 3.00 ± 0.02 13.4 ± 1.2 2.4 

Values of R Fe-O, σ Fe-O 2, R Fe-M, and σ Fe-M 2 are consistent throughout the models tested here, although the R-factor changes considerably, showing that a simplified model can be used for short-range ordering even when cationic disordering is not considered. Table II shows the fit results for the model with Co as the single scatterer in the second metallic shell [12Co in Fig. 2(b)]. Values of bond distance for the first coordination shell of Fe and Cu are considerably smaller than Co, Ni, and Zn, and deviate considerably from the (MgFeCoNiCuZn)O lattice parameter [ a = 4.24 Å from Fig. S1(a), supplementary material]. Cu–O also shows a Jahn–Teller distortion, previously found in these systems.12,44 These distortions disappear in the second shell, where values of bond distance do not show significant variation between the metallic edges. Large σ 2 values illustrate not only a high degree of local disorder around the Fe, mostly due to the ionic size difference, but also the deviation from our simplified models and the real high-entropy oxide local structure. Nonetheless, this model was chosen to be representative of the second scattering path and will be used here in further analysis of (MgFeCoNiCuZn)O in a battery system.

TABLE II.

EXAFS fit results using 12Co model for all metallic edges of (MgFeCoNiCuZn)O in terms of bond distance and σ2 showing short-range distortions not captured by XRD.

PathFe K-edgeCo K-edgeNi K-edgeCu K-edgeZn K-edge
RTM-O (Å) 1.98 ± 0.01 2.07 ± 0.02 2.07 ± 0.01 1.98/2.27 ± 0.02a 2.05 ± 0.02 
σ TM-O 2 (10−3 Å29.6 ± 1.2 7.7 ± 2.0 5.0 ± 1.2 7.0/11.5 ± 0.9a 11.1 ± 4.9 
RTM-12Co (Å) 3.01 ± 0.01 3.01 ± 0.02 3.00 ± 0.01 2.99 ± 0.02 3.02 ± 0.02 
σ TM-12Co 2 (10−3 Å215.9 ± 1.1 11.0 ± 1.2 10.6 ± 0.9 14.7 ± 1.0 15.2 ± 1.5 
PathFe K-edgeCo K-edgeNi K-edgeCu K-edgeZn K-edge
RTM-O (Å) 1.98 ± 0.01 2.07 ± 0.02 2.07 ± 0.01 1.98/2.27 ± 0.02a 2.05 ± 0.02 
σ TM-O 2 (10−3 Å29.6 ± 1.2 7.7 ± 2.0 5.0 ± 1.2 7.0/11.5 ± 0.9a 11.1 ± 4.9 
RTM-12Co (Å) 3.01 ± 0.01 3.01 ± 0.02 3.00 ± 0.01 2.99 ± 0.02 3.02 ± 0.02 
σ TM-12Co 2 (10−3 Å215.9 ± 1.1 11.0 ± 1.2 10.6 ± 0.9 14.7 ± 1.0 15.2 ± 1.5 
a

The Cu–O environment exhibits two different bond distances related to the Jahn–Teller distortion.

Bond distance ( R), number of near neighbors ( N), and local disorder ( σ 2) are all parameters that can be calculated from an EXAFS model and which are important for the analysis of electrochemical systems. During Li-ion battery operation, for example, Li ions shuttle between the electrodes using the cathode and anode materials as their hosts. This insertion and extraction of ions produces structural changes that can be captured by the EXAFS modeling, providing insights into the electrochemical behavior of the electrode materials. Recently, high-entropy oxides have been tested as both cathode and anode materials for Li-ion applications.11–23 The former usually relies on the intercalation of Li ions, while the latter are generally conversion materials. (MgFeCoNiCuZn)O falls in the conversion anode category, where electrons are extracted from the system through the reduction of the transition metal oxides and creation of Li 2O, following Eq. (2), below:
MO + 2 e + 2 Li M + Li 2 O .
(2)

Due to the extra amount of electrons retrieved from the conversion reaction compared to a simple intercalation, this class of electrodes can achieve the high-energy densities required for today’s energy storage demands.46–49 The change in the local structure inherent from the reaction, and the complexity of the high-entropy configuration makes x-ray absorption spectroscopy a unique technique to characterize electrochemical performance, providing information about electronic and local structural changes during battery operation.

Herein, we focus on the (MgFeCoNiCuZn)O compound to demonstrate the potential of XAS in characterizing different elements in the high-entropy composition, and how a simple and reliable EXAFS model can be used for local structural characterization. The compound was electrochemically tested by cyclic-voltammetry (CV) and galvanostatically cycled in a coin cell using a half-cell configuration (Li-metal as the counter electrode), with the results shown in Fig. 3. From the CV [Fig. 3(a)] and voltage profile [Fig. 3(b)], it is possible to notice the redox reactions during the discharge and charging steps, and the potentials where they take place. Although it is not possible to assign which of the metals are electrochemically active agents as their redox peaks are superimposed in the same voltage range, it is clear from the difference between the first two cycles that there is a strong irreversible reaction at 0.5 V during the initial lithiation. Figure 3(c) shows (MgFeCoNiCuZn)O specific capacity as a function of the number of cycles. It is notable that the irreversibility persists for nearly eight cycles until it stabilizes at 450 mAh/g. These results are consistent with our previous work on high-entropy rock-salt and spinel structures.14,50 The (MgFeCoNiCuZn)O theoretical capacity reaches 611 mAh/g assuming Mg inactivity and Fe, Co, Ni, Cu, and Zn full conversion ( + 2 to 0), which is considerably higher than 450 mAh/g achieved after initial cycling, indicating an irreversibility of the electrochemically active species during the lithiation/delithiation process. The calculated theoretical capacity is also substantially lower than the first lithiation process [1099 mAh/g from Fig. 3(b)], suggesting side reactions taking place at the electrode, such as SEI formation and space charge effects.14 

FIG. 3.

Electrochemical performance of (MgFeCoNiCuZn)O in terms of cyclic-voltammetry (a) and galvanostatic cycling (b) for the first two cycles, at a scan rate of 0.2 mV/s, and current density of 50 mA/g; (c) cycling capacity and coulombic efficiency at different current densities, ranging from 50 to 1000 mA/g.

FIG. 3.

Electrochemical performance of (MgFeCoNiCuZn)O in terms of cyclic-voltammetry (a) and galvanostatic cycling (b) for the first two cycles, at a scan rate of 0.2 mV/s, and current density of 50 mA/g; (c) cycling capacity and coulombic efficiency at different current densities, ranging from 50 to 1000 mA/g.

Close modal

In order to analyze the electrochemical reaction mechanism of (MgFeCoNiCuZn)O and assign the electrochemically active species in this material, two cycled electrodes at the first full lithiation (0.1 V vs Li/Li +) and delithiation (3.0 V Li/Li +) states were extracted from a coin cell and compared to the uncycled (MgFeCoNiCuZn)O electrode. XAS spectra were collected at the Fe (7112 eV), Co (7708 eV), Ni (8331 eV), Cu (8980 eV), and Zn (9660 eV) K-edges. To avoid any undesired side reaction not related to the battery operation, the electrodes were extracted inside an Ar-filled glovebox, and the remaining electrolyte (1M LiPF 6, EC:EMC) cleaned with dimethyl carbonate (DMC) before being wrapped in Kapton tape and vacuum-sealed in polyethylene bags. Those two layers of protection are necessary to prevent oxidation of the highly reactive metallic nanoclusters that originate from the conversion reaction and are present in the fully lithiated state.14,50 The typical XAS experiment involves measuring the incident x-ray intensity before and after the sample and using Beer’s law to calculate the absorption coefficient.51 Depending on factors such as concentration, thickness, and even geometry, the sample’s fluorescence can also be used to measure the XAS. Herein, the transition metal edges were all measured in the continuous scan fluorescence mode due to the low active material loading present in the coin cell electrodes.

Figure 4 shows the Fourier transforms of the Fe K-edge EXAFS and XANES spectra for the uncycled, first fully lithiated (0.1 V vs Li/Li +) and delithiated (3.0 V Li/Li +) electrodes. Significant changes in the local and electronic structure can be seen due to the lithium conversion reaction. During the first lithiation, the two EXAFS peaks characteristic of the rock-salt phase are replaced by a new shell at 1.97 Å. This shell completely disappears during the first delithiation, giving place to a peak at 1.43 Å  that can be assigned to the return of the Fe–O bonds. Similarly, the Fe K-edge XANES changes from characteristic (MgFeCoNiCuZn)O oxide to a metallic state during the first lithiation and returns to an oxide state upon delithiation, suggesting the metallic nature of the new shell seen in EXAFS. XANES is not only sensitive to the coordination environment (its shape) but to the element’s oxidation state. The first is expressed in terms of the allowed photoelectric transitions giving rise to distinctive features (pre-edge) together with the absorption process characteristic of the coordination, while the latter is manifested by a shift of the rising edge energy. This shift can be easily explained in terms of a simple electrostatic model, where atoms with higher oxidation states should have higher charge and thus require more energy to participate in the photoelectric process. On the contrary, reduced species require less energy to eject electrons, shifting the absorption edge to lower energies. Both XAS regions are governed by the same phenomena but XANES accounts for longer absorber-scatter interactions due to the low kinetic energy of the photoelectron at this region, being more sensitive to multiple scattering and complicating the XANES physical description. Nevertheless, this region has characteristic features that can be used for compound identification through spectral matching.

FIG. 4.

Fourier transforms of the EXAFS spectra (a) and XANES region (b) of Fe K-edge at uncycled, first lithiated, and first delithiated states.

FIG. 4.

Fourier transforms of the EXAFS spectra (a) and XANES region (b) of Fe K-edge at uncycled, first lithiated, and first delithiated states.

Close modal

For instance, Fe K-edge XANES of the (MgFeCoNiCuZn)O electrode in uncycled, lithiated, and delithiated states, are compared with metallic Fe, FeO, and Fe 3O 4 oxides, in Fig. 4. The first lithiated state (in red), in turn, is similar to metallic Fe (in green), which can be used to identify the metallic nature of the new lithiated shell at 1.97 Å in the EXAFS. As can be seen, the Fe K-edge of the uncycled electrode resembles the Fe 3O 4 spinel more than the expected FeO rock-salt. In fact, a linear combination fit of those reference oxides estimates approximately 82.4% of the iron present in this structure has a spinel structure while 17.6% has a rock-salt structure. This subtlety was not captured by XRD but can be explained by a spinel short-range ordering around Fe cations having, most likely, also tetrahedral arrangements in addition to the rock-salt octahedral. In fact, Fig. S2 (supplementary material) compares the Fe K-edge EXAFS of FeO (Fm-3m), Fe 3O 4 (Fd-3m), and the uncycled (MgFeCoNiCuZn)O (Fm-3m). Qualitatively, the (MgFeCoNiCuZn)O first shell resembles the Fe 3O 4 spinel, not only by its shape but also radial distance, 1.44 and 1.45 Å, respectively. On the other hand, the second shell looks more like FeO rock-salt not only because of similar bond distances but also the absence of the left shoulder at 3.1 Å, characteristic of the spinel. This indicates a localized spinel short-range ordering, which disappears already by the second shell.

The EXAFS modeling paths and fit results of the battery electrodes are also shown in Fig. 4. For all three states, Fe–O (in green) and Fe–M (in cyan) paths were used, with M consisting of only Co atoms as the metallic representative of the simplified model, as stated previously. This same model is also used for Co, Ni, Cu, and Zn K-edges. The many-body amplitude reduction factor, S 0 2, used in the fit was obtained from the Fe, Co, Ni, Cu, and Zn metallic reference foils, and it is equal to 0.64, 0.65, 0.76, 0.86, and 0.74, respectively. A better description of the model and the electrode preparation can be found elsewhere.14,50 Figure S3 and Table SV (supplementary material) show the EXAFS fits and the numerical values of the calculated parameters. The changes in the number of near neighbors and bond distances of the oxygen and metallic paths are shown in Fig. 5 for all the metallic edges and can be used to assess the degree of electrochemical activity of the different cations in the (MgFeCoNiCuZn)O compound.

FIG. 5.

EXAFS fit results: number of near neighbors (a) and (b) and bond distance (c) and (d) for Fe, Co, Ni, Cu, and Zn K-edges at uncycled (black), fully lithiated (red), and delithiated (blue) states.

FIG. 5.

EXAFS fit results: number of near neighbors (a) and (b) and bond distance (c) and (d) for Fe, Co, Ni, Cu, and Zn K-edges at uncycled (black), fully lithiated (red), and delithiated (blue) states.

Close modal

Initially, all five transition metal cations show an octahedral oxygen environment in the uncycled state [Fig. 5(a), in black]. During the first lithiation [Fig. 5(a), in red] the oxygen bond, represented by the number of near neighbor oxygens, completely disappears for Fe, Co, Ni, and Cu, while decreases considerably for Zn. At the same time, a metallic bond appears for all cations, illustrating the conversion reaction described above, in Fig. 5(b) (in red). Upon delithiation, the oxygen bond returns to the cations in different degrees [Fig. 5(a), in blue]. For Fe and Zn, the metallic bond completely disappears during delithiation, while it remains for Co, Ni, and Cu, indicating a stronger electrochemical activity of the former. It is important to note that the number of oxygens near neighbors in the final delithiated state never returns to six (octahedral environment), indicating a tetrahedral tendency for Fe and Zn.

Similarly, the change in bond distance of oxygen and metallic paths, in Figs. 5(c) and 5(d), can be used to track the nature of the reacted compounds, by simply comparing their values with other known structures. For example, it is easy to notice that the metallic shell in the lithiated state of all edges on Fig. 5(b) (in red) is not the same as the second metallic shell of the uncycled state (in black), as the bond distance of the former is considerably shorter than the latter, being, respectively, 2.43 and 3.01 Å, for Fe. This same trend can be noticed for all the cations, where the new metallic shell formed during the first lithiation is the product of the conversion reaction, consisting of a different phase as can be seen by the different value of bond distance. Upon delithiation, those values remain roughly the same for Co, Ni, and Cu, showing the irreversible nature of these newly created metallic structures. The irreversible nature of Co, Ni, and Cu electrochemical reactions can directly explain the discrepancy between theoretical and experimental specific capacity results (Fig. 3), jeopardizing the full potential of this high-entropy electrode.

X-ray absorption spectroscopy is a powerful tool for the electronic structure (XANES) and local geometry (EXAFS) characterization of compositionally complex systems, such as high-entropy oxides. Coupled with the ability of analyzing one specific element at a time without interference of other species makes it unique for electrochemical characterization, where the oxidation and structure of individual elements are directly responsible for cycling performance. The EXAFS region can be modeled to describe the short-range disorder and distortions present in a high-entropy structure, providing valuable information about bond distance, number of near neighbors, and local disorder, which are crucial to understanding the electrochemical reaction mechanism of such battery electrodes. It has been shown here that assumptions about the nature of the scattering paths can help to simplify the EXAFS analysis without jeopardizing the quality of the fits, helping to understand the electrochemical roles of different elements in the (MgFeCoNiCuZn)O electrode.

The supplementary material contains figures of the Pawley fits to the x-ray diffraction data; tables with the fit parameters for each of the two-shell models tested; Fe K-edge EXAFS compared to iron oxide standards; and tables of fit parameters and figures for the final EXAFS fits of Fe, Co, Ni, Cu, and Zn K-edges of (MgFeCoNiCuZn)O described in the paper.

This work was supported by the Illinois Institute of Technology Duchossois Leadership Professors Program. MRCAT operations are supported by the Department of Energy and the MRCAT member institutions. This research used resources of the Advanced Photon Source; a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357.

The authors have no conflicts to disclose.

Otavio J. Marques: Conceptualization (equal); Formal analysis (equal); Investigation (equal); Writing – review & editing (equal). Carlo U. Segre: Conceptualization (equal); Investigation (equal); Supervision (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
C. M.
Rost
,
E.
Sachet
,
T.
Borman
,
A.
Moballegh
,
E. C.
Dickey
,
D.
Hou
,
J. L.
Jones
,
S.
Curtarolo
, and
J.-P.
Maria
, “
Entropy-stabilized oxides
,”
Nat. Commun.
6
,
8485
(
2015
).
2.
B. L.
Musicó
,
D.
Gilbert
,
T. Z.
Ward
,
K.
Page
,
E.
George
,
J.
Yan
,
D.
Mandrus
, and
V.
Keppens
, “
The emergent field of high entropy oxides: Design, prospects, challenges, and opportunities for tailoring material properties
,”
APL Mater.
8
,
040912
(
2020
).
3.
G. N.
Kotsonis
,
S. S. I.
Almishal
,
F.
Marques dos Santos Vieira
,
V. H.
Crespi
,
I.
Dabo
,
C. M.
Rost
, and
J.
Maria
, “
High-entropy oxides: Harnessing crystalline disorder for emergent functionality
,”
J. Am. Ceram. Soc.
106
,
5587
5611
(
2023
).
4.
S. S.
Aamlid
,
M.
Oudah
,
J.
Rottler
, and
A. M.
Hallas
, “
Understanding the role of entropy in high entropy oxides
,”
J. Am. Chem. Soc.
145
,
5991
6006
(
2023
).
5.
Y.
Ma
,
Y.
Ma
,
Q.
Wang
,
S.
Schweidler
,
M.
Botros
,
T.
Fu
,
H.
Hahn
,
T.
Brezesinski
, and
B.
Breitung
, “
High-entropy energy materials: Challenges and new opportunities
,”
Energy Environ. Sci.
14
,
2883
2905
(
2021
).
6.
Z.-Y.
Liu
,
Y.
Liu
,
Y.
Xu
,
H.
Zhang
,
Z.
Shao
,
Z.
Wang
, and
H.
Chen
, “
Novel high-entropy oxides for energy storage and conversion: From fundamentals to practical applications
,”
Green Energy Environ.
8
,
1341
1357
(
2023
).
7.
Y.
Wang
,
J.
Mi
, and
Z.-S.
Wu
, “
Recent status and challenging perspective of high entropy oxides for chemical catalysis
,”
Chem. Catal.
2
,
1624
1656
(
2022
).
8.
M.
Fu
,
X.
Ma
,
K.
Zhao
,
X.
Li
, and
D.
Su
, “
High-entropy materials for energy-related applications
,”
iScience
24
,
102177
(
2021
).
9.
Y.
Chen
,
H.
Fu
,
Y.
Huang
,
L.
Huang
,
X.
Zheng
,
Y.
Dai
,
Y.
Huang
, and
W.
Luo
, “
Opportunities for high-entropy materials in rechargeable batteries
,”
ACS Mater. Lett.
3
,
160
170
(
2020
).
10.
C.
Oses
,
C.
Toher
, and
S.
Curtarolo
, “
High-entropy ceramics
,”
Nat. Rev. Mater.
5
,
295
309
(
2020
).
11.
A.
Sarkar
,
L.
Velasco
,
D.
Wang
,
Q.
Wang
,
G.
Talasila
,
L.
de Biasi
,
C.
Kübel
,
T.
Brezesinski
,
S. S.
Bhattacharya
,
H.
Hahn
, and
B.
Breitung
, “
High entropy oxides for reversible energy storage
,”
Nat. Commun.
9
,
3400
(
2018
).
12.
N.
Qiu
,
H.
Chen
,
Z.
Yang
,
S.
Sun
,
Y.
Wang
, and
Y.
Cui
, “
A high entropy oxide (Mg 0.2Co 0.2Ni 0.2Cu 0.2Zn 0.2O) with superior lithium storage performance
,”
J. Alloys Compd.
777
,
767
774
(
2019
).
13.
A.
Sarkar
,
Q.
Wang
,
A.
Schiele
,
M. R.
Chellali
,
S. S.
Bhattacharya
,
D.
Wang
,
T.
Brezesinski
,
H.
Hahn
,
L.
Velasco
, and
B.
Breitung
, “
High-entropy oxides: Fundamental aspects and electrochemical properties
,”
Adv. Mater.
31
,
1806236
(
2019
).
14.
O. J.
Marques
,
C.
Chen
,
E. V.
Timofeeva
, and
C. U.
Segre
, “
Local structure and conversion chemistry of high-entropy oxides as Li-ion anodes
,”
J. Power Sources
564
,
232852
(
2023
).
15.
K.
Wang
,
W.
Hua
,
X.
Huang
,
D.
Stenzel
,
J.
Wang
,
Z.
Ding
,
Y.
Cui
,
Q.
Wang
,
H.
Ehrenberg
,
B.
Breitung
,
C.
Kübel
, and
X.
Mu
, “
Synergy of cations in high entropy oxide lithium ion battery anode
,”
Nat. Commun.
14
,
1487
(
2023
).
16.
J.
Wang
,
Y.
Cui
,
Q.
Wang
,
K.
Wang
,
X.
Huang
,
D.
Stenzel
,
A.
Sarkar
,
R.
Azmi
,
T.
Bergfeldt
,
S. S.
Bhattacharya
,
R.
Kruk
,
H.
Hahn
,
S.
Schweidler
,
T.
Brezesinski
, and
B.
Breitung
, “
Lithium containing layered high entropy oxide structures
,”
Sci. Rep.
10
,
18430
(
2020
).
17.
Y.
Pei
,
Q.
Chen
,
M.
Wang
,
P.
Zhang
,
Q.
Ren
,
J.
Qin
,
P.
Xiao
,
L.
Song
,
Y.
Chen
,
W.
Yin
,
X.
Tong
,
L.
Zhen
,
P.
Wang
, and
C.-Y.
Xu
, “
A medium-entropy transition metal oxide cathode for high-capacity lithium metal batteries
,”
Nat. Commun.
13
,
6158
(
2022
).
18.
A.
Amiri
and
R.
Shahbazian-Yassar
, “
Recent progress of high-entropy materials for energy storage and conversion
,”
J. Mater. Chem. A
9
,
782
823
(
2021
).
19.
B.
Ouyang
and
Y.
Zeng
, “
The rise of high-entropy battery materials
,”
Nat. Commun.
15
,
973
(
2024
).
20.
P.
Ghigna
,
L.
Airoldi
,
M.
Fracchia
,
D.
Callegari
,
U.
Anselmi-Tamburini
,
P.
D’Angelo
,
N.
Pianta
,
R.
Ruffo
,
G.
Cibin
,
D. O.
de Souza
, and
E.
Quartarone
, “
Lithiation mechanism in high-entropy oxides as anode materials for Li-ion batteries: An operando XAS study
,”
ACS Appl. Mater. Interfaces
12
,
50344
50354
(
2020
).
21.
B.
Xiao
,
G.
Wu
,
T.
Wang
,
Z.
Wei
,
Y.
Sui
,
B.
Shen
,
J.
Qi
,
F.
Wei
, and
J.
Zheng
, “
High-entropy oxides as advanced anode materials for long-life lithium-ion batteries
,”
Nano Energy
95
,
106962
(
2022
).
22.
X.
Zou
,
Y.-R.
Zhang
,
Z.-P.
Huang
,
K.
Yue
, and
Z.-H.
Guo
, “
High-entropy oxides: An emerging anode material for lithium-ion batteries
,”
Chem. Commun.
59
,
13535
13550
(
2023
).
23.
Z.
Lun
,
B.
Ouyang
,
D.-H.
Kwon
,
Y.
Ha
,
E. E.
Foley
,
T.-Y.
Huang
,
Z.
Cai
,
H.
Kim
,
M.
Balasubramanian
,
Y.
Sun
,
J.
Huang
,
Y.
Tian
,
H.
Kim
,
B. D.
McCloskey
,
W.
Yang
,
R. J.
Clément
,
H.
Ji
, and
G.
Ceder
, “
Cation-disordered rocksalt-type high-entropy cathodes for Li-ion batteries
,”
Nat. Mater.
20
,
214
221
(
2021
).
24.
G. H. J.
Johnstone
,
M. U.
González-Rivas
,
K. M.
Taddei
,
R.
Sutarto
,
G. A.
Sawatzky
,
R. J.
Green
,
M.
Oudah
, and
A. M.
Hallas
, “
Entropy engineering and tunable magnetic order in the spinel high-entropy oxide
,”
J. Am. Chem. Soc.
144
,
20590
20600
(
2022
).
25.
E.
Krysko
,
L.
Min
,
Y.
Wang
,
N.
Zhang
,
J. P.
Barber
,
G. E.
Niculescu
,
J. T.
Wright
,
F.
Li
,
K.
Burrage
,
M.
Matsuda
,
R. A.
Robinson
,
Q.
Zhang
,
R.
Katzbaer
,
R.
Schaak
,
M.
Terrones
,
C. M.
Rost
, and
Z.
Mao
, “
Studies on the structure and the magnetic properties of high-entropy spinel oxide (MgMnFeCoNi)Al 2O 4
,”
APL Mater.
11
,
101123
(
2023
).
26.
Y.
Pu
,
D.
Moseley
,
Z.
He
,
K. C.
Pitike
,
M. E.
Manley
,
J.
Yan
,
V. R.
Cooper
,
V.
Mitchell
,
V. K.
Peterson
,
B.
Johannessen
,
R. P.
Hermann
, and
P.
Cao
, “
(Mg,Mn,Fe,Co,Ni)O: A rocksalt high-entropy oxide containing divalent Mn and Fe
,”
Sci. Adv.
9
,
eadi8809
(
2023
).
27.
A. R.
Mazza
,
X.
Gao
,
D. J.
Rossi
,
B. L.
Musico
,
T. W.
Valentine
,
Z.
Kennedy
,
J.
Zhang
,
J.
Lapano
,
V.
Keppens
,
R. G.
Moore
,
M.
Brahlek
,
C. M.
Rost
, and
T. Z.
Ward
, “
Searching for superconductivity in high entropy oxide Ruddlesden–Popper cuprate films
,”
J. Vac. Sci. Technol. A
40
,
013404
(
2021
).
28.
A. J.
Kropf
,
J.
Katsoudas
,
S.
Chattopadhyay
,
T.
Shibata
,
E. A.
Lang
,
V. N.
Zyryanov
,
B.
Ravel
,
K.
McIvor
,
K. M.
Kemner
,
K. G.
Scheckel
,
S. R.
Bare
,
J.
Terry
,
S. D.
Kelly
,
B. A.
Bunker
,
C. U.
Segre
,
R.
Garrett
,
I.
Gentle
,
K.
Nugent
, and
S.
Wilkins
, “
The new MRCAT (sector 10) bending magnet beamline at the advanced photon source
,”
AIP Conf. Proc.
1234
,
299
302
(
2010
).
29.
B. D.
Cullity
,
Elements of X-ray Diffraction
(
Addison-Wesley
,
1956
).
30.
A. R.
West
,
Solid State Chemistry and its Applications
(
John Wiley & Sons
,
2022
).
31.
E. A.
Stern
,
Y.
Ma
,
O.
Hanske-Petitpierre
, and
C. E.
Bouldin
, “
Radial distribution function in X-ray-absorption fine structure
,”
Phys. Rev. B
46
,
687
694
(
1992
).
32.
J. E.
Penner-Hahn
, “X-ray absorption spectroscopy,” in Comprehensive Coordination Chemistry II, Vol. 2, edited by A. Lever (Elsevier, Amsterdam, 2003), Chap. 2.13, pp. 159–186.
33.
E. A.
Stern
, “
Theory of the extended X-ray-absorption fine structure
,”
Phys. Rev. B
10
,
3027
3037
(
1974
).
34.
P. A.
Lee
and
J. B.
Pendry
, “
Theory of the extended X-ray absorption fine structure
,”
Phys. Rev. B
11
,
2795
2811
(
1975
).
35.
C. A.
Ashley
and
S.
Doniach
, “
Theory of extended X-ray absorption edge fine structure (EXAFS) in crystalline solids
,”
Phys. Rev. B
11
,
1279
1288
(
1975
).
36.
M.
Roy
and
S. J.
Gurman
, “
Amplitude reduction in EXAFS
,”
J. Synchrotron. Radiat.
6
,
228
230
(
1999
).
37.
M.
Newville
, “
Fundamentals of XAFS
,”
Rev. Mineral. Geochem.
78
,
33
74
(
2014
).
38.
J.
Rehr
,
J. M.
de Leon
,
S.
Zabinsky
, and
R.
Albers
, “
Theoretical X-ray absorption fine structure standards
,”
J. Am. Chem. Soc.
113
,
5135
5140
(
1991
).
39.
J. J.
Rehr
,
R. C.
Albers
, and
S. I.
Zabinsky
, “
High-order multiple-scattering calculations of X-ray-absorption fine structure
,”
Phys. Rev. Lett.
69
,
3397
3400
(
1992
).
40.
J. J.
Rehr
and
R. C.
Albers
, “
Theoretical approaches to X-ray absorption fine structure
,”
Rev. Mod. Phys.
72
,
621
654
(
2000
).
41.
J. J.
Rehr
,
J. J.
Kas
,
M. P.
Prange
,
A. P.
Sorini
,
Y.
Takimoto
, and
F.
Vila
, “
Ab initio theory and calculations of X-ray spectra
,”
C. R. Phys.
10
,
548
559
(
2009
).
42.
M.
Newville
, “
IFEFFIT: Interactive XAFS analysis and FEFF fitting
,”
J. Synchrotron. Radiat.
8
,
322
324
(
2001
).
43.
B.
Ravel
and
M.
Newville
, “
ATHENA, ARTEMIS, HEPHAESTUS: Data analysis for x-ray absorption spectroscopy using IFEFFIT
,”
J. Synchrotron. Radiat.
12
,
537
541
(
2005
).
44.
P. J.
Brown
,
A. G.
Fox
,
E. N.
Maslen
,
M. A.
O’Keefe
, and
B. T. M.
Willis
, “Intensity of diffracted intensities,” in International Tables for Crystallography (International Union of Crystallography, 2006), pp. 554–595.
45.
C.
Colliex
,
J. M.
Cowley
,
S. L.
Dudarev
,
M.
Fink
,
J.
Gjønnes
,
R.
Hilderbrandt
,
A.
Howie
,
D. F.
Lynch
,
L. M.
Peng
,
G.
Ren
,
A. W.
Ross
,
V. H.
Smith
,
J. C. H.
Spence
,
J. W.
Steeds
,
J.
Wang
,
M. J.
Whelan
, and
B. B.
Zvyagin
, “Electron diffraction techniques,” in International Tables for Crystallography (International Union of Crystallography, 2006), pp. 259–429.
46.
P.
Poizot
,
S.
Laruelle
,
S.
Grugeon
,
L.
Dupont
, and
J.-M.
Tarascon
, “
Nano-sized transition-metal oxides as negative-electrode materials for lithium-ion batteries
,”
Nature
407
,
496
499
(
2000
).
47.
S.
Fang
,
D.
Bresser
, and
S.
Passerini
, “
Transition metal oxide anodes for electrochemical energy storage in lithium- and sodium-ion batteries
,”
Adv. Energy Mater.
10
,
1902485
(
2020
).
48.
K.
Cao
,
T.
Jin
,
L.
Yang
, and
L.
Jiao
, “
Recent progress in conversion reaction metal oxide anodes for Li-ion batteries
,”
Mater. Chem. Front.
1
,
2213
2242
(
2017
).
49.
M.
Zheng
,
H.
Tang
,
L.
Li
,
Q.
Hu
,
L.
Zhang
,
H.
Xue
, and
H.
Pang
, “
Hierarchically nanostructured transition metal oxides for lithium-ion batteries
,”
Adv. Sci.
5
,
1700592
(
2018
).
50.
O. J. B. J.
Marques
,
M. D.
Walter
,
E. V.
Timofeeva
, and
C. U.
Segre
, “
Effect of initial structure on performance of high-entropy oxide anodes for Li-ion batteries
,”
Batteries
9
,
115
(
2023
).
51.
G. F.
Lothian
, “
Beer’s law and its use in analysis. A review
,”
The Analyst
88
,
678
685
(
1963
).