Electrochemical Zinc Intercalation in Lithium ... - ACS Publications

Article pubs.acs.org/cm

Electrochemical Zinc Intercalation in Lithium Vanadium Oxide: A High-Capacity Zinc-Ion Battery Cathode Muhammad H. Alfaruqi,†,‡ Vinod Mathew,†,‡ Jinju Song,† Sungjin Kim,† Saiful Islam,† Duong Tung Pham,† Jeonggeun Jo,† Seokhun Kim,† Joseph Paul Baboo,† Zhiliang Xiu,† Kug-Seung Lee,§ Yang-Kook Sun,∥ and Jaekook Kim*,† †

Department of Materials Science and Engineering, Chonnam National University, 300 Yongbongdong, Bukgu, Gwangju 500-757, South Korea § Nanoprobe XAFS Beamline, Pohang Accelerator Laboratory, Pohang 790-784, South Korea ∥ Energy Storage & Conversion Laboratory, Hanyang University, Seoul 133-791, South Korea S Supporting Information *

ABSTRACT: Rechargeable zinc-ion batteries (ZIBs) with high energy densities appear promising to meet the increasing demand for safe and sustainable energy storage devices. However, electrode research on this low-cost and green system are faced with stiff challenges of identifying materials that permit divalent ion-intercalation/deintercalation. Herein, we present layered-type LiV 3O 8 (LVO) as a prospective intercalation cathode for zinc-ion batteries (ZIBs) with high storage capacities. The detailed phase evolution study during Zn intercalation using electrochemistry, in situ XRD, and simulation techniques reveals the large presence of a singlephase domain that proceeds via a stoichiometric ZnLiV3O8 phase to reversible solid−solution ZnyLiV3O8 (y > 1) phase. The unique behavior, which is different from the reaction with lithium, contributes to high specific capacities of 172 mAh g−1 and amounts to 75% retention of the maximum capacity achieved in 65 cycles with 100% Coulombic efficiency at a current density of 133 mA g−1. The remarkable performance makes the development of this low-cost and safe battery technology very promising, and this study also offers opportunities to enhance the understanding on electrochemically induced metastable phases for energy storage applications.

■

INTRODUCTION The growing demand for green and sustainable energy has brought rechargeable lithium-ion batteries (LIBs) to the fore for potential electric vehicle, intelligent grid, and renewable energy storage applications.1−3 Notwithstanding that this battery technology has clearly dominated the portable electronics market, challenges of making lithium affordable due to constrains in long-term availability and lowering the toxicity in the LIB components for large-scale applications have still remained.4 Sodium-ion, potassium-ion, magnesium-ion, and zinc-ion batteries are cost effective and ecofriendly due to the utilization of naturally abundant elements.5−17 Of these, the most studied is the sodium-ion battery as various electrodes including phosphates, oxides, sulfates, sulfides, organic materials, carbon, and metal electrodes have been identified.5−9 The higher molar mass of sodium leads to the disadvantage of a relatively low energy density and thereby makes sodium-ion batteries appealing for stationary storage applications. However, potassium-ion batteries are increasingly studied, although diffusion barriers of the potassium ion with higher ionic radii remain.10 Rechargeable batteries employing divalent ions can © XXXX American Chemical Society

deliver higher storage capacities due to the advantage of the two electron transfer, but this comes at the cost of slow insertion kinetics. In particular, the well-known magnesium battery is faced with the stiff obstacle of developing electrolytes that can support reversible deposition/dissolution of magnesium.11 The recently introduced zinc-ion battery (ZIB) is a low-cost and efficient technology due to its cheap electrode components, ecofriendly electrolyte with relatively high electrical conductivity (higher by 2 orders of magnitude than in organic solutions),18 simple fabrication, and the ability to deliver high rate capacities and specific power. In comparison with its counterparts, the ZIB technology has been the least studied since only a few cathodes like MnO2, Na3V2(PO4)3, and V2O5 have been developed so far.12−17 Hence, the effective development of electrodes facilitating the storage/release of divalent Zn ions has become critical for the realization of ZIBs as a next-generation device. Received: November 30, 2016 Revised: February 1, 2017

A

DOI: 10.1021/acs.chemmater.6b05092 Chem. Mater. XXXX, XXX, XXX−XXX

Article

Chemistry of Materials

Figure 1. (a) Schematic of the aqueous rechargeable Zn−LVO cell. (b) Rietveld refinement of the XRD pattern of the LVO prepared by solid-state reaction (inset, crystal structure): (c) FE-SEM, (d) FE-TEM, (e) HR-TEM images, and (e) SAED pattern of LVO.

electrochemistry was investigated in 1981, LVO has been studied for cathode applications in lithium batteries due to its relatively easy synthesis, high ion diffusion, structural stability, and cycle life.20−27 Also, excess vacant sites in the interlayer of the layered structure can support the occupation of more Li ions. Further, the weak Li−O coordination, which arises from the lower electron density of the O2− ions due to the increased M−O bonds in the lattice, permits the hopping of Li+ ions and hence fast diffusivity in LVO. The chemical diffusion coefficient

The layered-type structure and the feasibility of vanadium to exist in high oxidation states make Li1+xV3O8 (LVO) an interesting candidate to promote intercalation/deintercalation of ions with relatively larger ionic radii. The monoclinic LVO system is comprised of two edge-shared octahedral (VO6) and trigonal bipyramid (VO5) units that form a “string” and “ribbon” arrangement (via corner sharing) of (V3O8)− layers along the (100) plane, the layers being linked by Li+ ions in the interstitial octahedral and tetrahedral sites.19 Ever since its B


Article

Chemistry of Materials (DLi) of LVO has been determined to be ∼10−11−10 −12 cm2 s−1, this value being lowered by 2 orders of magnitude (∼10−13−10 −15 cm2 s−1) in the Li-rich (Li4V3O8) phase.26,27 The electrochemical behavior of LVO with lithium reveals the existence of two-phase as well as single-phase domains with either specific or solid−solution compositions, as observed from the typical voltage profile comprised of sloping curves/ potential drops and plateaus. The room-temperature phase behavior study of Li1+xV3O8 during lithium intercalation reactions revealed the accommodation of up to 4 Li ions per formula unit under controlled conditions and is described as follows:24−27 a solid−solution phase in the 0 < x < 1.5 range, the 1.5 < x < 3.2 domain corresponds to a two-phase reaction comprising of LiV3O8 and Li4V3O8, and the 3.2 < x < 4 window is defined by the single-phase reaction of Li4V3O8. Kawakita et al. claimed that lithium insertion beyond x = 3.2 increases the cell volume of the monoclinic unit cell phase.24,25 Considering the intercalation of three Li ions per unit cell, the maximum theoretical specific capacity of LVO is estimated as ∼280 mAh g−1, which is considerably higher than those achieved by other LIB cathodes. During early charging/delithiation even when only a small amount of lithium is extracted or when the lithium content is apparently high (x = 2.7, Li3.7V3O8), the Li4V3O8 phase disappears and delithiation proceeds via the single phase of LiV3O8 until the starting material is recovered after complete lithium extraction. Thus, it is worth noting that the lithium content is crucial in influencing the Li+ distribution, the formation of solid−solution/biphasic domains with or without structural modulations, and hence the chemical properties of the LVO. More importantly, the layered structure of LiV3O8 was reported to undergo very slight or insignificant modification upon Li-ion intercalation/deintercalation.26,27 Therefore, the possibilities of hosting excess guest ions coupled with the wide charge-balancing property of vanadium redox couples (V5+/V4+/V3+) while simultaneously maintaining structural integrity makes the layered-type LVO an interesting candidate to probe for divalent ion insertion. In particular, the fact that the ionic radii of Li+ (0.74 Å) and Zn2+ (0.76 Å) are comparable motivates the study of LVO for electrochemical Zn-ion insertion. Although the LVO consists of lithium and vanadium elements, which are not considered as abundant elements, the present study may pave the way for the realization of layered-type host structures for multivalent battery applications. In the present work, we demonstrate the facile electrochemical Zn intercalation and hence highly reversible storage capacities in a LVO cathode for prospective zinc storage applications. Metallic zinc and 1 M ZnSO4 solution was used as the anode and electrolyte, respectively (as illustrated in Figure 1a). This battery configuration delivered a maximum discharge capacity of 200 mAh g−1 under 133 mA g−1 current density with almost 100% Coulombic efficiencies for 65 cycles. In situ synchrotron X-ray diffraction, ex situ XANES, simulation XRD, and ex situ XRD studies were performed to clearly understand the phase evolution of LVO. During early discharge, Zn intercalation in LVO is described by a solid−solution behavior, while at the intermediate discharge, it is presumed that a trend related to more than one two-phase reaction involving the stoichiometric ZnLiV3O8 phase is followed. During the last stage of the discharge reaction, the solid−solution behavior of the ZnyLiV3O8 (y ≥ 1) phase is observed. On the consecutive Zn-deintercalation process, the ZnyLiV3O8 phase transforms to LiV3O8 via a single-phase behavior. Interestingly, the Zn-

intercalation mechanism is slightly different from the electrochemical lithiation process in layered-type LVO, wherein the Li-intercalation process is mostly explained by a two-phase reaction over a wide discharge potential window while the Li extraction is attributed to a single-phase reaction mechanism. Besides, to the authors’ best knowledge, this is the first instance that the formation of a stoichiometric ZnLiV3O8 phase during electrochemical reaction is suggested. The present study thus not only motivates further exploration of these materials to understand their physicochemical properties but also calls for their utilization in various applications including energy storage.

■

EXPERIMENTAL SECTION

Material Preparation. Commercially available lithium carbonate (Li2CO3, 98% Daejung) and vanadium oxide (V2O5, 98% Aldrich) powders were ball milled with acetone at a rotation speed of 180 rpm for 48 h. The mixture was then dried in a vacuum evaporator at 120 °C before heating at 400 °C for 12 h in air. After heating, the furnace was allowed to cool to room temperature without any external cooling. The final powder obtained after thorough grinding was used for further characterization. X-ray Diffraction (XRD) Studies. The XRD studies of the powder and ex situ electrodes were performed on an X’Pert PANalytical Model high-resolution X-ray diffractometer with Cu Kα radiation (λ = 1.54056 Å). The scan was recorded at 40 kV and 30 mA in the 2θ range between 10° and 80° with a resolution of 0.01°. The electrodes were separated from the coin cells upon completion of electrochemical discharge/charge reactions and rinsing them with distilled water before drying under vacuum conditions at room temperature. Electron Microscopy Studies. The surface morphology was studied by using a S-4700 Hitachi model and Philips Tecnai F20 model operating at 200 keV (KBSI, Chonnam National University, South Korea) to record the FE-SEM and FE-TEM images, respectively. For TEM imaging, the sample was dispersed in ethanol using ultrasonic vibration before dropping onto copper grids. Electrochemical Insertion of Zn2+. The paste made by mixing 70, 20, and 10 wt % of active material, Ketjen black, and teflonated acetylene black (TAB), respectively, was pressed onto a stainless-steel mesh before vacuum drying at 120 °C for 12 h and thereby formed the cathode. Zn metal foil and a 1 M ZnSO4 (pH 4.0) solution were used as the anode and electrolyte, respectively. While preparing the 2032type coin cells, a glass fiber separator soaked with the electrolyte was pressed between the prepared cathode and the anode in open air conditions and aged for 12 h before performing electrochemical discharge/charge measurements using a BTS 2004H model (NAGANO KEIKI Co., LTD, Ohta-ku, Tokyo, Japan) battery tester at different current densities between 0.6 and 1.2 V vs Zn2+/Zn. For the cyclic voltammetry (CV) scans, an AUTOLAB PGSTAT302N potentiostat model workstation was used. Synchrotron XRD Studies. The synchrotron beamline at 1D KIST-PAL (Pohang Accelerator Laboratory) was used for in situ XRD measurements. The beamline setup uses a MAR345 image plate detector that operates at 2.5 GeV with 200 mA maximum storage current. A double-bounce Si(111) monochromator utilizes the focused X-ray beam from a toroidal mirror to produce a monochromatic radiation of 12.4016 keV (0.9997 Å). For high-resolution configuration in the reciprocal space, a Si(111) monochromator and a Si(111) analyzer crystal are employed. The XRD patterns were recorded with respect to the wavelength of 1.00076 Å and replotted after conversion of the 2θ values with respect to Cu Ka radiation (λ = 1.5414 Å). The electrode for in situ measurement was prepared by making a film using the active material:carbon black:TAB ratio mentioned earlier and casting onto a stainless-steel mesh. This electrode was used to assemble the test cell with Kapton tapes covering the apertures on two sides of the outer cases. The fabricated spectroelectrochemical cell was tested under a constant discharge/charge cycling rate of 16 mA g−1, while the XRD data was collected in situ. C


Article


Figure 2. (a) Cyclic voltammetry measurements performed at 0.5 mV s−1 scan rate. (b) Initial five voltage profiles of the LVO electrode at a current density of 16 mA g−1. (c) Cycle performance and corresponding Coulombic efficiencies between 0.6 and 1.2 V vs Zn2+/Zn at 133 mA g−1. (d) Rate performance of the prepared LVO cathode.

are in good agreement with those reported previously.21−23 As observed from Figure 1b (inset), the monoclinic structure of LVO drawn using VESTA software28 consists of octahedral VO6 and distorted trigonal bipyramidal VO5 units linked by corner-sharing oxygen atoms to form sheets parallel to the {100} facets, and the sheets appear to be “pinned” together by the Li+ ions between them.21−23 It is observed from the XRD pattern in Figure 1b that the intensity of the (100) diffraction line is higher than that of the (−111) plane in the present LVO powder. This indicates that the dimensions of the LVO crystal normal specific to the (100) plane are large; thus, the present LVO particles may provide more access for guest ion into available sites.29 It is anticipated that these beneficial features in the present LVO can contribute to a higher electrochemical reactivity and hence zinc storage properties. Electron microscopy studies were performed to study the morphology and structural characteristics of the present LVO, and the results are presented in Figure 1c−f. The FE-SEM and FETEM images shown in Figure 1c and 1d, respectively, illustrate that the LVO powder is composed of flake-like particles with sizes in the range of 200−500 nm. The inset of Figure 1e shows a single flake-like particle, while Figure 1e, which shows the high-resolution-TEM (HR-TEM) image, reveals distinct lattice

Synchrotron XAS Studies. The synchrotron X-ray absorption spectroscopy (XAS) measurements on the cathodes of the test cells were carried out ex situ at the BL7D beamline of the Pohang Light Source (PLS) in the 2.5 GeV storage ring with a ring current of 120− 180 mA. The X-ray photon energy was tuned by a Si (111) doublecrystal monochromator. Specifically, the monochromator was detuned to lower the incident X-ray beam intensity by approximately 20% to facilitate the elimination of unwarranted higher order harmonics. The XAS data was collected in the transmission mode at room temperature, while ATHENA software was used to convert the obtained data to the absorbance mode.

■

RESULTS AND DISCUSSION The high-resolution XRD pattern of the LVO powder prepared by a solid state reaction was refined using the Rietveld method, and the results are presented in Figure 1b. The Rietveld refinement was performed using the FULLPROF software, and the corresponding refinement data in detail are also provided (Table S1, Supporting Information). The peaks of the sample can be well indexed on the basis of a pure monoclinic LVO phase with the P21/m space group (JCPDS 72-1193). Obviously, the XRD pattern reveals the high purity of the prepared sample. The unit cell parameters were calculated to be a = 6.6463 Å, b = 3.5963 Å, and c = 12.0046 Å, and these values D


Article


Figure 3. (a) Electrochemical discharge profile of the fabricated spectroelectrochemical cell cycled within 1.2−0.6 V at a current rate of 16 mA g−1. Solid spheres (dark blue, discharging; cyan, charging) present the locations at which the synchrotron XRD scans were initiated. In situ XRD patterns obtained within selected scanning angle (2θ) domains of (b) 13.2−18°, (c) 20−29.5°, (d) 30−39.5°, and (e) 39.5−45° are presented.

Under the galvanostatic condition of 16 mA g−1 current density, the electrochemical curves representing zinc intercalation into the parent LiV3O8 material from the OCV of 1.3 V down to 0.6 V vs Zn/Zn2+ followed by zinc deintercalation up to 1.2 V and the subsequent four discharge/charge curves are presented in Figure 2b. As observed in Figure 2b, there are two discharge plateau-like regions which are in congruence with the redox peaks observed during repeated cycling in the cyclic voltammograms (Figure 2a). The small variation in the discharge and charge curve behaviors indicates that the mechanisms of intercalation and deintercalation, respectively, are not exactly similar, as observed from the CV results. However, the striking resemblances of the consecutive discharge (charge) curves clearly reveal a reversible Znintercalation (deintercalation) reaction. However, in comparison, the electrochemical lithiation in LVO reveals more than three plateau-like region within the potential range of 2.0−4.0 V vs Li/Li+ (Figure S1, Supporting Information). For the LiV3O8 phase with lithium in the Li1 octahedral site, three vacant tetrahedral sites, namely, Li2, Li3, and Li4, with different energies are available in the interstitial layers for Li-ion insertion during electrochemical discharge reaction.23,24 On the other hand, as observed from the discharge/charge profile in Figure 2b, the Zn ions intercalated during first discharge are slightly higher than those deintercalated during first charge possibly due to the trapping of some relatively small amount of Zn2+ ions in the host structure. This behavior supports the findings in electrodes, such as MnO2 and V2O5, when tested for ZIB applications.12,35 From the second cycle, almost all intercalated Zn ions are deintercalated (Figure 2b), and this trend is observed for almost 65 cycles under a higher current denisty of 133 mAg−1, as observed in Figure 2c. Further, the cycle performance plot indicates that the specific capacities gradually increase, and the electrode experiences periodic

fringes and thereby demonstrates the high crystallinity of the present LVO sample. The interplanar spacings of 0.63 and 0.34 nm are consistent with the (100) and (011) planes of LVO, respectively, and are well matched with the XRD results in Figure 1b. The selected area electron diffraction (SAED) pattern of the sample, depicted in Figure 1f, can be well indexed to the monoclinic LVO crystal along the [01−1] zone axis. Cyclic Voltammetry and Electrochemical Studies. Cyclic voltammetry (CV) measurements performed on the fabricated Zn/LVO cell in a 1 M ZnSO4 electrolyte at 0.5 mV s−1 within 0.6−1.2 V vs Zn2+/Zn is presented in Figure 2a. The two distinctly consistent peaks at around 0.8 and 0.65 V in the forward scan and slightly different values, namely, 0.79 and 1.1 V, in the reverse scan, respectively, can be attributed to a twostep reaction associated with zinc intercalation and extraction processes through the LVO host. For example, Lee et al. assigned the intense peak and shoulder at around 1.2 and 0.9 V, respectively, observed during the cathodic scan of a Zn/αMnO2 cell to stable zinc-ion insertion into the α-MnO2 host.30 Similarly, zinc-ion insertion into a Prussian blue analog of zinc hexacyanoferrate (ZnHCF), which was investigated by Zhang et al., revealed a strong 1.7 V CV peak along with a shoulder at 1.2 V.31 Also, it is worth noting that the observation of such peaks is typical of divalent or trivalent ion intercalation, including Zn2+, Mg2+, Ca2+, Sr2+, and Al3+, into the cathodic host.12−15,31−34 The enhancing peak currents on repeated cycling are attributed to gradual electrode activation. Thus, the overall electrochemical reaction of the Zn/LVO cell can be written as Anode: Zn ↔ Zn(2+) + 2e−

(1)

Cathode: y Zn(2+) + LiV3O8 ↔ Zn yLiV3O8

(2) E


Article


Figure 3b, that the (100) reflection plane is shifted to higher 2θ position (from 14.06° to 14.82°), thereby indicating a corresponding decrease (by ∼0.3 Å) in the respective interplanar distance (or a value). This reduction indicates an improvement in the structural coordination due to the strong electrostatic interaction between the intercalated zinc ions and the (V3O8)− layers. Concordantly, a negative shift in the 2θ position (from 23.14° to 22.31°) of the (003) peak signifies a slight increase (by ∼0.11 Å) in the “c” parameter. This expansion arises from the reduction of V5+ ions and the resultant increase in the M−M bond distance between the layers. Similarly, the interplanar distance of the (−111) plane expands from 3.25 to 3.40 Å (Figure 3c). Overall, the characteristics of shifting reflection lines accompanied by the contraction/expansion of the unit cell parameters is typical of layered-type host structures undergoing guest-ion intercalation. Interestingly, the rate of shift in the 2θ position of the (100) peak varies during the entire discharge reaction. In the initial discharge region (1.28−0.82 V), the (100) diffraction peak reveals narrow/slight 2θ shifts (Figure 3b). This is indicative of a single-phase domain displaying a solid−solution behavior, and the corresponding discharge cycle response demonstrates a gradually sloping potential curve, whereas at intermediate discharge within the 0.81−0.7 V domain, wide 2θ shifts accompanied by splitting of the (100) diffraction peak are revealed, and this trend is different from the usual two-phase reaction represented by the peak-splitting feature at a fixed 2θ position in the XRD pattern and the flat discharge plateau in the 1.5 < x < 3.2 potential domain during lithium intercalation in LixV3O8.24−27 This implies that, in the present case, a complicated trend representing the formation of two successive two-phase reactions may explain this unique trend in the intermediate discharge region.36 The marginal shifting of 2θ corresponding to the (100) peak in the final discharge (0.69− 0.6 V) region represents a single-phase domain. Thus, the electrochemical Zn insertion or discharge mechanism can be represented by three regions, as indicated by the blue (Region I), pink (Region II), and red (Region III) colored lines, respectively, in Figure 3b. More importantly, the evolution of a new set of four peaks, viz., 31.5°, 32.8°, 33.5°, and 34.7°, within the range of 31−35° and at 38°, 38.5°, and 39° (marked by purple solid diamond in Figure 3c and 3d) with gradually increasing intensities during the continuous discharge reaction are observed. The new peaks coupled with the features of peak shifting, splitting, and broadening of the characteristic (100) line in the intermediate stage (Region II) during the first discharge (Zn insertion) clearly suggests the evolution of a new phase with well-defined composition. Further, sharp reflection lines at 2θ = 43−45° corresponding to the stainless-steel current collector used in electrode fabrication make it difficult to distinguish active material peaks within the domain (Figure 3e). Nevertheless, unlike the first discharge reaction, during subsequent charging, the characteristic LVO diffraction lines are restored via just solid−solution behavior with no peak-splitting feature and confirms that the discharge and charge mechanism of Zn-intercalation and -deintercalation reactions, respectively, in LVO tend to be slightly different. Interestingly, the XRD pattern after complete charging is quite similar to that of the parent/pristine electrode before electrochemical reaction, thereby suggesting that Zn insertion (extraction) into (from) the layered-type LVO host is feasible and reversible. It is worth noting that unlike the phase change behavior during Li insertion in LiV3O8, no biphasic domains representative of

capacity activation processes during the cycling. It is possible that this phenomenon can be related to structural evolution during electrochemical Zn insertion. However, the XRD study for the electrode recovered after 65 cycles was performed in order to understand the structural properties of the LVO electrode after Zn insertion. The corresponding results (Figure S2, Supporting Information) indicate that the LVO structure is fairly maintained, though further detailed studies are required to confirm this finding. Interestingly, when checked for cycling stability at a lower current density (∼16 mAh g−1), the electrode did not reveal any such activation process, though only limited cycling was feasible (Figure S3, Supporting Information). Hence, it is mostly probable that the activation process can be related to the applied high current density and/ or the chemical ion kinetics involved during the electrochemical reaction. Precisely, at high current densities the insertion of divalent charge carriers like Zn2+ ions with comparatively sluggish ion kinetics can facilitate such activation processes. On the other hand, the limited cycling capacity observed at low current densities can be due to vanadium dissolution associated with the prolonged time duration of the discharge/charge cycling. However, further works are required to identify the exact cause and possible measures to address the capacity fading in this electrode. In the rate performance study, when cycled for five times under each progressive current densities, viz., 16, 33, 66, 133, 266, 533, 1066, and 1666 mA g−1, respectively, as shown in Figure 2d, the LVO electrode exhibits average discharge capacities of 256, 230, 211, 188, 148, 79, 47, and 29 mAh g−1. Interestingly, after cycling at the highest current density of 1666 mA g−1, the average discharge capacity recovers to 191 mAh g−1 as the initial current density of 16 mA g−1 is applied. This implies that the present LVO electrode delivers considerable rate capabilities. In Situ Synchrotron XRD Studies. Voltage plateaus in the corresponding electrochemical profile are typical of first-order transitions representing a two-phase reaction, while the gradually sloping/abruptly falling potential curves are trademarks of single-phase domains. To confirm the phase behavior of LVO, synchrotron XRD studies were conducted in situ using a customized 2032-type coin cell (spectroelectrochemical cell) with windows on both faces covered by transparent Kapton tapes. This window feature enables the acquisition of XRD data as the initial electrochemical discharge/charge reaction occurs. The initial discharge/charge profiles of the Zn//LVO system along with the sequentially numbered solid spheres which indicate the specific depth of discharge (DOD, navy blue colored) and state of charge (SOC, dark cyan colored) during cycling are provided in Figure 3a. Nineteen scans were recorded during discharge cycling, while the remaining 19 scans were obtained in the subsequent charge cycle, as shown in Figure 3b−e. The lower half (of Figure 3b−e) corresponds to the spectra recorded during the discharge cycling, while the upper half represents those obtained during charge cycling. The periodic scan numbers have also been provided for easy identification. Scan 1 (recorded at OCV/before the electrochemical reaction) reveals the preferred characteristic peaks at 14°, 23°, 25.9°, 28°, 30.7°, 39°, 40.5°, and 50.5° corresponding to the (100), (003), (011), (−111), (103), (−114), (−301), and (020) reflections of the LVO phase, respectively. During the discharge reaction from the OCV to complete first discharge (0.6 V), the variations in the (100) diffraction line are significant because the generation of lattice parameter “a” is hugely influenced by the zinc content. It is observed, from F


Article

Chemistry of Materials peak-splitting features with fixed 2θ positions are observed through the entire discharge/charge reaction. These results thus clearly indicate that the electrochemical Zn insertion is slightly different from the case of lithium intercalation in LixV3O8. To further validate the dissolution and deposition reactions of zinc in LVO, ex situ XRD analysis of the anode from the test cell was performed before and after electrochemical reaction. The patterns display just metallic zinc peaks and are devoid of byproduct/impurity peaks corresponding to either ZnO or Zn(OH)2, which are usually related to the discharge mechanism in Zn/MnO2 alkaline batteries (Figure S4, Supporting Information). Ex Situ XANES Studies. In order to study the change in vanadium oxidation states while discharging/charging (or Zn insertion/extraction) the Zn//LVO test cell, ex situ synchrotron XANES was performed for the LVO cathode recovered at various initial DOD (0.83 and 0.6 V) and SOC (0.95 and 1.2 V) conditions between 1.2 and 0.6 V, and the results are presented in Figure 4a and 4b, respectively. For comparison,

evaluation of whether oxidation states as V3+, V4+, and V5+ ions can exist in different symmetries, namely, tetrahedral (Td), octahedral (Oh), and square pyramidal (Py) coordination. Precisely, V3+ displays Oh symmetry, and V4+ exists in Oh and Py coordination, whereas all three symmetries are common for V5+ (electronic configuration 3s23p63d0).37 In general, Td symmetry contributes to high pre-edge intensities, whereas apparently low intensities are observed for Oh and Py coordination.38 Hence, in the present case, the lower pre-edge intensity of LVO is due to the distorted octahedral vanadium coordination, while the square pyramidal symmetry is associated with the higher preedge intensity of V5+ ion in V2O5. More importantly, the main absorption edge negatively shifts toward lower binding energies at progressive DOD’s (Figure 4a), confirming that the vanadium oxidation state is reduced due to Zn intercalation. Further, the reducing pre-edge intensities (shown in Figure 4a, inset) at lower discharge potentials suggest a higher octahedral symmetry during the final stages of the electrochemical discharge reaction.39 On complete discharge (0.6 V, DOD), the V K-edge spectrum is consistent with that for standard V4+ ion, as evidenced from the similarities in the rising edge pattern (and hence the main absorption edge) and the reduced preedge intensities (Figure 4a, inset). Although there appears to be a similarity in the pre-edge width on comparing the two spectra, the higher intensity in the reduced LVO (or the spectrum at 0.6 V, DOD) suggests a slightly more distorted octahedral V4+ coordination compared to that in standard V2O4. This pre-edge intensity variation (also observed when comparing the spectra of parent LVO and V2O5 with ∼V5+ oxidation states in Figure 4a) can be explained based on the slight change in the distortion around the vanadium atom, which ultimately influences the overall spread of V−O distances specific to each of these compounds.40 Further, the unique trend of the reference V3+ (V2O3, dark gray line in Figure 4a) pattern with almost perfect octahedral symmetry provides further confirmation on the assignment of the V4+ state to the reduced LVO pattern after complete discharge reaction. Reversely, on consecutive charge or Zn-deintercalation reaction, the recorded V K-edge patterns (Figure 4b) reveal that the main absorption peak continuously shifts toward higher energies (or positive shift) and the pre-edge intensity increases, suggesting an increase in the average vanadium oxidation state. In other words, as the Zn-deintercalation reaction progresses, the Kedge shift varies linearly with the average oxidation state of vanadium. At complete charging (1.2 V, SOC), the V K-edge spectrum matches well with that of the original LVO cathode, as observed from the magnified view in Figure 4b (inset). This implies that the average vanadium oxidation state of the LVO cathode reverts back from around V4+ to approximately V5+ after the charge reaction. Overall, the XANES result lends support to the premise that reduction and oxidation of the vanadium oxidation states occur during the zinc insertion and extraction across the LiV3O8 electrode, respectively. XRD Simulation Studies. Our attempts to refine the in situ XRD patterns by the Rietveld method and thereby understand the phase evolution of LVO were unsuccessful probably due to the apparently insufficient intensities of the characteristic reflection lines. Therefore, for further understanding on the Zn-inserted phase formed during discharge reaction, XRD patterns pertaining to Zn occupation at various Li sites in the LiV3O8/Li4V3O8 structures were simulated using the X’PERT HighScore Plus program and compared with the fully discharged (0.6 V, DOD) in situ XRD pattern. Also, a

Figure 4. Comparison of normalized V K-edge XANES spectra of initial state (LVO) and the standard vanadium oxides with those obtained (a) at different depth-of-discharge (DOD) conditions during discharge cycling and (b) at different state-of-charging (SOC’s) conditions during charging.

the OCV and reference vanadium (V3+/V4+/V5+) spectra are also provided. As expected, the main edge features of the fresh electrode (before electrochemical reaction) in Figure 4a are consistent with that of standard V5+ in V2O5. For example, the trends in the rising edge of the main absorption and the high pre-edge peak intensity are similar (Figure 4a, inset). However, a careful evaluation of the pre-edge features is necessary for G


Article


ZnyLiV3O8 (y > 1) phase is also feasible. The structural data of this phase has also been provided (Figure S6; Table S3, Supporting Information). Nevertheless, on comparison with the characteristic LiV3O8 pattern, the simulated ZnLiV3O8 pattern reveals shifts in the (100) and (−111) diffraction peaks toward higher and lower 2θ angles, respectively, in addition to the peak evolution at higher 2θ angles (Figure 5). These observations of peak shift and peak evolution in the simulated ZnLiV3O8 pattern are consistent with those observed for the completely discharged (0.6 V, DOD) in situ XRD pattern (Figure 5). For example, the new peaks observed in the scanning angle domains of 31−35°, 37−39°, and 24.1° in the simulated ZnLiV3O8 pattern match well with those observed in the fully discharged in situ XRD pattern. These results imply that all diffraction peaks after complete discharge can be assigned to a ZnyLiV3O8 phase. However, the small deviations in the 2θ positions of the shifted peaks can be related to the slight variation in the zinc content present in the LVO. Further, these observations clearly confirm that the monoclinic structural framework undergoes slight variation throughout the Zn-intercalation process as in the case of lithiation in LVO. Also interesting to observe is that the unique patterns of simulated ZnLiV3O8 and Li4V3O8 indicate that the variations may be related to the presence of Li and their positioning at various sites in the monoclinic unit cell. It is worth noting that the use of XRD simulation to support or confirm the observed experimental data has also been used for Li-ion/Na-ion insertion in less crystalline insertion hosts like MnO2.41,42 However, further theoretical studies utilizing DFT calculations and modeling will aid in more clearly establishing the structural parameters of the ZnyLiV3O8 phase. Electrochemical Mechanism of Zn Insertion. Overall, the in situ XRD patterns clearly reveal the shifting of characteristic LVO peaks, and emergence of new peaks as the discharge reaction proceeds. The XANES study clearly confirmed the transition between V5+ and V4+ oxidation states accompanying Zn intercalation/deintercalation in LVO. The simulation XRD results indicate that the formed Zn-inserted LVO phase corresponds to ZnyLiV3O8, its structure being in close resemblance to Li4V3O8. From these results, the electrochemical mechanism occurring in LVO during Zn insertion/deinsertion can be explained based on the illustration in Figure 6: The initial stage of Zn intercalation is attributed to the single-phase reaction of LiV3O8 wherein the Zn ions begin to occupy vacant lithium sites with different energies. It is highly possible that the initial electrochemical Zn insertion in LVO is attributed to the occupation of Zn in the Li(2) sites to form the stoichiometric ZnLiV3O8 phase. In the intermediate stage, although the peak-splitting feature of the (100) diffraction line appears to suggest the coexistence of two phases probably belonging to LiV3O8 and ZnLiV3O8, the accompanying shifting of 2θ positions specific to this domain points to a mechanism that remains complicated to be resolved.39 Nevertheless, further Zn insertion probably leads to the occupation of Li(3) sites by Zn ions. In other words, the final stage at proximities to complete discharge confirms the single-phase behavior of the new Zn-inserted phase (ZnyLiV3O8, y > 1), as observed from the shifted (100) line and the newly evolved peaks (Figure 3c). Therefore, based on the change of the unit cell parameters during electrochemical Zn-ion insertion, Zn ions tend to diffuse along the b axis in the LVO host structure (Figure S2, Supporting Information). Overall, unlike the case of lithiation, the electrochemical Zn

comparative study of the simulated ZnLiV3O8 phase and Li4V3O8 patterns was performed to understand the present electrochemical reaction from the perspective of lithium insertion in LVO. Initially, XRD patterns of LiV3O8 and Li4V3O8 were generated using the standard crystallographic data (corresponding crystallographic information files from inorganic crystal structure database (ICSD)), and the resulting patterns (Figure 5) concur well with those reported.20−23 Both

Figure 5. Simulated XRD patterns of LiV3O8 and Li4V3O8 and the pattern obtained by assuming Zn ions to occupy the 2e Wyckoff sites in the monoclinic unit cell of LVO in comparison to the in situ XRD pattern of the Zn-inserted LVO or fully discharged pattern.

phases have an almost similar structure with negligent variations. In LiV3O8, lithium occupies the Li(1) octahedral site, while in Li4V3O8, additional lithiums are found in three other vacant tetrahedral sites, namely, Li(2), Li(3), and Li(4), with different site energies. According to Sarkar et al., who studied in detail the electrochemical lithiation in LiV3O8, the lithium at the Li(1) site can be retained in the same site while the incoming Li ions occupy the Li(2) and Li(3) sites.23 In accordance, it was assumed that the zinc ions Wyckoff sites of Li(2) in LiV3O8 and the XRD pattern of the stoichiometric ZnLiV3O8 phase were simulated. The structural data of this stoichiometric ZnLiV3O8 phase has also been provided (Figure S5; Table S2, Supporting Information). Moreover, on the basis of the theoretical consideration that more vacant sites in LVO are available, further Zn insertion to Li(3)/Li(4) sites is feasible. Therefore, considering that the most probable site for Zn ions to occupy is the Li(3) site, the realization of the H


Article


Figure 6. Schematic of the Zn-intercalation mechanism in the present LiV3O8 cathode.

be realized. Although further studies on characterizing the Zninserted phase using Rietveld refinement are required, it is worth noting that vanadium-based oxides with layered-type structures are beneficial for ion insertion since this structure promotes facile accommodation and transport of guest ions within its lattice. More importantly, the feasibility of vanadium to exist in several oxidation states and thereby maintain charge compensation makes such layered-type oxides accommodate the storage and release of even divalent charge carriers. These advantages of the present LiV3O8 cathode ultimately contribute to the realization of high energy densities in ZIB applications. However, detailed studies are required further to understand and overcome the limited cycling capabilities of the present LVO electrode to facilitate practical applications. On the other hand, considering the fact that recent studies of electrochemical Zn insertion have been predominantly focused on tunnel-based manganese oxide cathodes, this study not only showcases LVO as a potential cathode host for ZIB applications but also paves the way for furthering the development of this promising, safe, and low-cost technology in the near future. Furthermore, although further studies are required to understand the ZnyLiV3O8 phase, this study clearly puts forward the possibility for the existence of such possible phases, which can tend to be useful for a variety of applications.

insertion in LVO can be described mainly by a storage mechanism indicating a progressive single-phase transition that appears to proceed to the phase of ZnyLiV3O8 (Figure 6). On the other hand, the early stage of the electrochemical charging/ Zn deinsertion is represented by the single-phase reaction of ZnyLiV3O8 that gradually proceeds toward the recovery of LiV3O8. Also, the almost superimposable XRD patterns of the completely charged product and the original electrode indicate that the LVO structure undergoes negligible variation after the electrochemical reaction and that the Zn insertion is almost completely reversible. Although the overall structural framework undergoes very slight modification during electrochemical reaction of LVO with zinc or lithium, the storage mechanism of zinc is different from the coexisting two-phase domain observed for electrochemical lithiation in LiV3O8.43−45 Interestingly, the initial discharge capacity of 267 mAh g−1 achieved by the present LVO electrode under a low current density of 16 mA g−1 (Figure 2b) amounts to almost 95% of the calculated capacity (280 mAh g−1) provided a three electron transfer corresponding to a Zn-inserted phase with stoichiometry of Zn1.5LiV3O8 is assumed. This assumption could explain the slight shift between the peak positions of characteristic (100) and (−111) diffraction planes of the simulated ZnLiV3O8 and fully discharged patterns in Figure 5. Nevertheless, at a high current density of 133 mA g−1 (Figure 2c), an average discharge capacity of almost 180 mAh g−1, corresponding to 1 mol of zinc insertion per LVO (or ZnyLiV3O8, y = 1) is achieved during the initial few cycles. Also, this practical capacity concurs well with the assumption in our simulation studies. As expected, the lower Zn insertion in LVO can be attributed to the high current density applied. However, on continuous cycling, the experimental specific capacities are even slightly higher (∼200 mAh g−1) than the calculated values. This implies that more sites become available in the layered-type LVO host for Zn insertion during prolonged cycling. Thus, the possibility of insertion of more than 1 mol of zinc per unit formula of LVO by the accompanying reduction of vanadium oxidation state can

■

CONCLUSION In summary, the present study reports on the electrochemical Zn-insertion feasibility and the corresponding phase evolution in LVO by using electrochemistry, in situ XRD, and simulation techniques. The layered-type structure and the wide oxidation state variation of vanadium appears to facilitate the accommodation and transportation of Zn ions during the electrochemical reaction. The flake-type LiV3O8 cathode delivered an average discharge capacity of 172 mAh g−1 after 65 cycles with almost 100% Coulombic efficiencies at 133 mA g−1 current density. In addition, phase evolution of LVO related to zinc intercalation in aqueous ZIB applications was examined I


Article


(5) Singh, P.; Shiva, K.; Celio, H.; Goodenough, J. B. Eldfellite, NaFe(SO4)2: an intercalation cathode host for low-cost Na-ion batteries. Energy Environ. Sci. 2015, 8, 3000−3005. (6) Hwang, J.-Y.; Oh, S.-M.; Myung, S.-T.; Chung, K. Y.; Belharouak, I.; Sun, Y.-K. Radially aligned hierarchical columnar structure as a cathode material for high energy density sodium-ion batteries. Nat. Commun. 2015, 6, 6865. (7) Barpanda, P.; Liu, G.; Ling, C. D.; Tamaru, M.; Avdeev, M.; Chung, S. C.; Yamada, Y.; Yamada, A. Na2FeP2O7: A safe cathode for rechargeable sodium-ion batteries. Chem. Mater. 2013, 25, 3480−3487. (8) Kitajou, A.; Yamaguchi, J.; Hara, S.; Okada, S. Discharge/charge reaction mechanism of a pyrite-type FeS2cathode for sodium secondary batteries. J. Power Sources 2014, 247, 391−395. (9) Luo, W.; Allen, M.; Raju, V.; Ji, X. An organic pigment as a highperformance cathode for sodium-ion batteries. Adv. Ener. Mat 2014, 4, 1400554. (10) Jian, Z.; Luo, W.; Ji, X. Carbon electrodes for K-ion batteries. J. Am. Chem. Soc. 2015, 137, 11566−11569. (11) Song, J.; Sahadeo, E.; Noked, M.; Lee, S. B. Mapping the challenges of Magnesium battery. J. Phys. Chem. Lett. 2016, 7, 1736− 1749. (12) Xu, C.; Li, B.; Du, H.; Kang, F. Energetic zinc ion chemistry: The rechargeable zinc ion battery. Angew. Chem., Int. Ed. 2012, 51, 933−935. (13) Alfaruqi, M. H.; Mathew, V.; Gim, J.; Kim, S.; Song, J.; Baboo, J. P.; Choi, S. H.; Kim, J. Electrochemically induced structural transformation in a γ-MnO2 cathode of a high capacity zinc-ion battery system. Chem. Mater. 2015, 27, 3609−3620. (14) Alfaruqi, M. H.; Gim, J.; Kim, S.; Song, J.; Jo, J.; Kim, S.; Mathew, V.; Kim, J. Enhanced reversible divalent zinc storage in a structurally stable α-MnO2 nanorod electrode. J. Power Sources 2015, 288, 320−327. (15) Lee, J.; Ju, J. B.; Cho, W.; Cho, B.; Oh, S. Todorokite-type MnO2 as a zinc-ion intercalating material. Electrochim. Acta 2013, 112, 138−143. (16) Li, G.; Yang, Z.; Jiang, Y.; Huang, Y. Towards polyvalent ion batteries: A zinc-ion battery based on NASICON structured Na3V2(PO4)3. Nano Energy 2016, 25, 211−217. (17) Kundu, D.; Adams, B. D.; Duffort, V.; Vajargah, S. H.; Nazar, L. F. A high-capacity and long-life aqueous rechargeable zinc battery using a metal oxide intercalation cathode. Nat. Energy 2016, 1, 16119. (18) Wang, H.; Zeng, Y.; Huang, K.; Liu, S.; Chen, L. Improvement of cycle performance of lithium ion cell LiMn2O4/LixV2O5 with aqueous solution electrolyte by polypyrrole coating on anode. Electrochim. Acta 2007, 52, 5102−5107. (19) Wadsley, A. D. Crystal chemistry of non-stoichiometric pentavalent vandadium oxides: crystal structure of Li1+xV3O8. Acta Crystallogr. 1957, 10, 261−267. (20) Ohzuku, T.; Ueda, A. Why transition metal (di) oxides are the most attractive materials for batteries. Solid State Ionics 1994, 69, 201− 211. (21) Sarkar, S.; Banda, H.; Mitra, S. High capacity lithium-ion battery cathode using LiV3O8 nanorods. Electrochim. Acta 2013, 99, 242−252. (22) Wang, H.; Ren, Y.; Wang, Y.; Wang, W.; Liu, S. Synthesis of LiV3O8 nanosheets as a high-rate cathode material for rechargeable lithium batteries. CrystEngComm 2012, 14, 2831−2836. (23) Sarkar, S.; Bhowmik, A.; Bharadwaj, M. D.; Mitra, S. Phase transition, electrochemistry, and structural studies of high rate LixV3O8 cathode with nanoplate morphology. J. Electrochem. Soc. 2014, 161, A14−A22. (24) Kawakita, J.; Miura, T.; Kishi, T. Lithium insertion and extraction kinetics of Li1+xV3O8. J. Power Sources 1999, 83, 79−83. (25) Kawakita, J.; Katayama, Y.; Miura, T.; Kishi, T. Structural properties of Lil+xV3O8 upon lithium insertion at ambient and high temperature. Solid State Ionics 1998, 107, 145−152. (26) Pan, A.; Liu, J.; Zhang, J.-G.; Cao, G.; Xu, W.; Nie, Z.; Jie, X.; Choi, D.; Arey, B. W.; Wang, C.; Liang, S. Template free synthesis of LiV3O8 nanorods as a cathode material for high-rate secondary lithium batteries. J. Mater. Chem. 2011, 21, 1153−1161.

by using electrochemistry, in situ XRD, and simulation techniques. The phase behavior studies confirmed the presence of two single-phase domains with low and high zinc contents (ZnyLiV3O8), respectively, and the possible coexistence of two adjacent two-phase reactions involving stoichiometric ZnLiV3O8. During the consecutive Zn extraction, the ZnyLiV3O8 phase transforms to LiV3O8 via a single-phase behavior. Thus, although the layered-type structure of LVO undergoes slight modification during the electrochemical reaction with zinc or lithium, the phase behavior is different from that observed during lithiation. Therefore, the present study encourages the utilization of such an approach to understand the phase evolution in layered-type systems but also demonstrates vanadium-based intercalation hosts as a promising cathode for ZIBs. Furthermore, the present work motivate studies not only on developing electrodes for this low-cost and safe battery technology but also to probe the use of electrochemically induced metastable materials for various applications including energy storage.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemmater.6b05092. Zn2+ diffusion pathway, cyclability data, ex situ XRD, and structural information (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Yang-Kook Sun: 0000-0002-0117-0170 Jaekook Kim: 0000-0002-6638-249X Author Contributions ‡

M.H.A. and V.M. contributed equally for this work. The manuscript was written through contributions of all authors. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Global Frontier Program through the Global Frontier Hybrid Interface Materials (GFHIM) of the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning (2013M3A6B1078875 or 2013-073298). This work was also supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIP) (2014R1A2A1A10050821).

■

REFERENCES

(1) Chu, S.; Majumdar, A. Opportunities and challenges for a sustainable energy future. Nature 2012, 488, 294−303. (2) Goodenough, J. B.; Park, K.-S. The Li-ion rechargeable battery: a perspective. J. Am. Chem. Soc. 2013, 135, 1167−1176. (3) Larcher, D.; Tarascon, J. − M. Towards greener and more sustainable batteries for electrical energy storage. Nat. Chem. 2015, 7, 19−29. (4) Goodenough, J. B.; Kim, Y. Challenges for rechargeable Li batteries. Chem. Mater. 2010, 22, 587−603. J


Article

Chemistry of Materials (27) de Picciotto, L. A.; Adendorff, K. T.; Liles, D. C.; Thackeray, M. M. Structural characterization of Lil+xV3O8 insertion electrodes by single-crystal X-ray diffraction. Solid State Ionics 1993, 62, 297−307. (28) Momma, K.; Izumi, F. VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data. J. Appl. Crystallogr. 2011, 44, 1272−1276. (29) Shi, Q.; Liu, J.; Hu, R.; Zeng, M.; Dai, M.; Zhu, M. An amorphous wrapped nanorod LiV3O8 electrode with enhanced performance for lithium ion batteries. RSC Adv. 2012, 2, 7273−7278. (30) Lee, B.; Yoon, C. S.; Lee, H. R.; Chung, K. Y.; Cho, B. W.; Oh, S. H. Electrochemically-induced reversible transition from the tunneled to layered polymorphs of manganese dioxide. Sci. Rep. 2014, 4, 6066−6074. (31) Zhang, L.; Chen, L.; Zhou, X.; Liu, Z. Morphology-dependent electrochemical performance of Zinc hexacyanoferrate cathode for zinc-ion battery. Sci. Rep. 2015, 5, 18263−18274. (32) Wang, R. Y.; Wessells, C. D.; Huggins, R. A.; Cui, Y. Highly reversible open framework nanoscale electrodes for divalent ion batteries. Nano Lett. 2013, 13, 5748−5752. (33) Liu, S.; Pan, G. L.; Li, G. R.; Gao, X. P. Copper hexacyanoferrate nanoparticles as cathode material for aqueous Al-ion batteries. J. Mater. Chem. A 2015, 3, 959−962. (34) Zhang, L.; Chen, L.; Zhou, X.; Liu, Z. Towards high-voltage aqueous metal-ion batteries beyond 1.5 V: The zinc/zinc hexacyanoferrate system. Adv. Energy Mater. 2015, 5, 930−935. (35) Senguttuvan, P.; Han, S.-D.; Kim, S.; Lipson, A. L.; Tepavcevic, S.; Fister, T. T.; Bloom, I. D.; Burrell, A. K.; Johnson, C. S. A high power rechargeable nonaqueous multivalent Zn/V2O5 battery. Adv. Energy Mater. 2016, 6, 1600826−1600832. (36) Zhang, L.; Chen, L.; Zhou, X.; Liu, Z. Towards high-voltage aqueous metal-ion batteries beyond 1.5 V: The zinc/zinc hexacyanoferrate system. Adv. Energy Mater. 2015, 5, 930−935. (37) Schindler, M.; Hawthorne, F. C.; Baur, W. H. Crystal chemical aspects of vanadium: Polyhedral geometries, characteristic bond valences, and polymerization of (VOn) polyhedral. Chem. Mater. 2000, 12, 1248−1259. (38) Chaurand, P.; Rose, J.; Briois, V.; Salome, M.; Proux, O.; Nassif, V.; Olivi, L.; Susini, J.; Hazemann, J.-L.; Bottero, J.-Y. New methodological approach for the Vanadium K-edge X-ray absorption near-edge structure interpretation: Application to the speciation of vanadium in oxide phases from steel slag. J. Phys. Chem. B 2007, 111, 5101−5110. (39) Tossici, R.; Marassi, R.; Berrettoni, M.; Stizza, S. Study of amorphous and crystalline Li1+xV3O8 by FTIR, XAS and electrochemical techniques. Solid State Ionics 1992, 57, 227−234. (40) Wong, J.; Lytle, F. W.; Messmer, R. P.; Maylotte, D. H. X-edge absorption spectra of selected vanadium compounds. Phys. Rev. B: Condens. Matter Mater. Phys. 1984, 30, 5596−5610. (41) Jiao, F.; Bruce, P. G. Mesoporous crystalline β-MnO2−a reversible positive electrode for rechargeable lithium batteries. Adv. Mater. 2007, 19, 657−660. (42) Su, D.; Ahn, H.-J.; Wang, G. β-MnO2 nanorods with exposed tunnel structures as high-performance cathode materials for sodiumion batteries. NPG Asia Mater. 2013, 5, e70. (43) Mansour, A. N.; Smith, P. H.; Baker, W. M.; Balasubramanian, M.; McBreen, J. In situ XAS investigation of the oxidation state and local structure of vanadium in discharged and charged V2O5 aerogel cathodes. Electrochim. Acta 2002, 47, 3151−3161. (44) Pistoia, G.; Panero, S.; Tocci, M.; Moshtev, R. V.; Manev, V. Solid solutions Li1+xV3O8 as cathodes for high rate secondary Libatteries. Solid State Ionics 1984, 13, 311−318. (45) Pistoia, G.; Pasquali, M.; Tocci, M.; Moshtev, R. V.; Manev, V. Li/Lil+xV308 secondary b batteries. III. Further characterization of the mechanism of Li+ insertion and of the cycling behavior. J. Electrochem. Soc. 1985, 132, 281−284.

K
