Skip to main content
NCBI home page
ACS Omega logoLink to ACS Omega
. 2021 Jan 8;6(3):1917–1929. doi: 10.1021/acsomega.0c04675

Synthesis and Characterization of a Novel Hydrated Layered Vanadium(III) Phosphate Phase K3V3(PO4)4·H2O: A Functional Cathode Material for Potassium-Ion Batteries

Tristram Jenkins 1, Jose A Alarco 1,*, Ian D R Mackinnon 1
PMCID: PMC7841776  PMID: 33521432

Abstract

graphic file with name ao0c04675_0012.jpg

Hydrated vanadium(III) phosphate, K3V3(PO4)4·H2O, has been synthesized by a facile aqueous hydrothermal reaction. The crystal structure of the compound is determined using X-ray diffraction (XRD) analysis aided by density functional theory (DFT) computational investigation. The structure contains layers of corner-sharing VO6 octahedra connected by corner and edge-sharing PO4 tetrahedra with a hydrated K+ ion interlayer. The unit cell is assigned to the orthorhombic system (space group Pnna) with a = 10.7161(4) Å, b = 20.8498(10) Å, and c = 6.5316(2) Å. Earlier studies of this material report a K3V2(PO4)3 stoichiometry with a NASICON structure (space group R3®c). Previously reported XRD and electrochemical data on K3V2(PO4)3 are critically evaluated and we suggest that they display mixed phase compositions of K3V3(PO4)4·H2O and known electrochemically active phases KVP2O7 and K3V(PO4)2. In the present study, the synthesis conditions, structural parameters, and electrochemical properties (vs K/K+) of K3V3(PO4)4·H2O are clarified along with further physical characterization by scanning electron microscopy (SEM), energy-dispersive X-ray (EDX), X-ray fluorescence (XRF), Raman spectroscopy, Fourier transform infrared (FT-IR), and thermogravimetric analysis (TGA).

1. Introduction

Vanadium phosphate materials exhibit a complex and useful crystal chemistry due to multiple oxidation states of vanadium (+2, +3, +4, +5) that allow for various vanadium–oxygen (V–O) coordination conditions including octahedral, square pyramidal, and tetrahedral. This combination results in a range of edge and corner-sharing arrangements for V–O polyhedra and PO4 tetrahedra.1−5 In recent years, V(III) phosphates have been evaluated for use as electroactive materials for electrochemical energy storage applications because their crystal structures allow for reversible and stable interstitial storage. In addition, these compounds show fast ionic conduction during intercalation/deintercalation of alkali cations.6−9 Examples include Na3V2(PO4)3 (NVP) (rhombohedral, R3®c), and Li3V2(PO4)3 (LVP) (monoclinic, C2/c), which are both considered fast-ion conducting solids for their respective alkali cations. As a result, these compounds have been applied successfully as both positive and negative electrode materials in sodium and lithium-ion battery applications.6,10,11 When considering the effectiveness of V(III) phosphates as alkali metal insertion hosts, it is reasonable to search for similar potassium-based compounds for potassium-ion batteries.

Hence, the compound K3V2(PO4)3 (KVP) has been recently synthesized and investigated as a potassium analogue to NVP and LVP. First published in 2015 by Zhang et al.,12 KVP was originally synthesized and applied as a candidate cathode material for sodium-ion batteries where the study showed promising electrochemical performance (119 mAh g–1 at 200 mA g–1) and good long-term cyclability (96% capacity retention after 2000 cycles). KVP has subsequently inspired further studies where it has been applied as both cathode and anode in potassium-ion batteries.12−14 Unfortunately, this research on apparent KVP compounds suggests poor Coulombic efficiency, a lowered capacity, and low cycling stability when applied as an electrode in potassium-ion batteries.12−14

In the original publication, the crystal structure of KVP was not identified by the authors but suggested to be a “possible new crystalline phase” as it could not be indexed to any known phase from the JCPDS database. Subsequent literature citations identified KVP only by visual match of collected X-ray diffraction (XRD) patterns to the pattern presented in the original publication with limited, or no, structural refinement.13,14 More recently, a 2019 study described the structure of K3V2(PO4)3 using synchrotron XRD as assignable to the rhombohedral crystal system with an R3®c space group, similar to Na3V2(PO4)3.15 This description was partially based on the observed shift in peak positions and intensities that occurred during an in situ synchrotron XRD study. Because the structural evolution during the synchrotron study was similar to that of a Na3V2(PO4)3 cathode during Na+ extraction in sodium-ion batteries, the potassic analogue was assumed.12 While cell dimensions were reported for the refinement of synchrotron data, the free atom positions, reliability factors, and further crystallographic details of K3V2(PO4)3 were not documented.12

In this article, we report our findings on the crystal structure refinement of orthorhombic K3V3(PO4)4·H2O, a new hydrated layered vanadium(III) phosphate phase. During our study, candidate unit cells of K3V2(PO4)3 are used during XRD refinement and tested for validity using density functional theory (DFT) geometry optimization. However, refinement results showed the best agreement with an orthorhombic unit cell (space group Pnna) and formula stoichiometry of K3V3(PO4)4·H2O. This crystallographic determination is supported by detailed spectroscopic, microscopic, and bulk chemical analyses for optimized synthesis conditions. Electrochemical data vs K/K+ are collected and contrasted with previous studies on K3V2(PO4)3. In conjunction with XRD, this comparison indicates that the electrochemically active components of previously reported K3V2(PO4)3 compounds are in fact likely to contain the recently identified phase K3V3(PO4)4·H2O along with known phases KVP2O7 and K3V(PO4)2.

2. Results

2.1. Synthesis and Characterization of K3V3(PO4)4·H2O

The powder sample of K3V3(PO4)4·H2O when synthesized is a fine-grained, light green powder. Scanning electron microscopy (SEM) images of the washed powder sample display primary particles of diameter ∼2–10 μm and a thickness of 0.5–10.0 μm. These primary particles were generally observed in larger, nonuniform aggregates with wide-ranging diameters from 6 to 60 μm. The primary particles show an anisotropic platelike morphology where a layer format can be clearly observed. Based on agreement with the layers of the refined unit cell (see below), we assume that the face of each plate corresponds to the (010) plane (Figure 1).

Figure 1.

Figure 1

Open in a new tab

SEM images displaying the morphology of (a) singular primary particles and (b) aggregated primary particles.

Inductively coupled plasma optical emission spectrometry (ICP-OES) results of the unwashed bulk sample present a K/V/P atom ratio of 3.03/2.00/2.97, agreeing with the proposed stoichiometry of K3V2(PO4)3. However, following the identification of phase impurities by XRD (see Figure S1), sample washing is performed to remove impurities and the atom ratios are reanalyzed using X-ray fluorescence (XRF). These XRF data show a 2.96/3.06/3.98 atomic ratio for K/V/P, thus suggesting notional atom ratios consistent with K3V3(PO4)4. This bulk stoichiometry is also confirmed through energy-dispersive spectrometry (EDS) mapping where the particle atom ratios of K/V/P are measured to be 3.07/3.03/3.89, respectively (see Figure S2). Based on these data, the charge balance determines the oxidation state of vanadium as +3. To confirm the oxidation state of vanadium, X-ray photoelectron spectroscopy (XPS) analysis is performed on the pristine K3V3(PO4)4·H2O sample. Figure 2 shows the V 2p core level XPS spin–orbit spectra of K3V3(PO4)4·H2O with binding energy peaks centered at 516.4 and 523.9 eV corresponding to the V 2p3/2 and V 2p1/2 splitting values, respectively. These values agree well with previously reported V 2p peak positions for Li3V2(PO4)316−18 and Na3V2(PO4)3,19 ascribed to vanadium in its trivalent V3+ state. Based on XPS and charge balance, it can therefore be concluded that vanadium exists in a predominantly V3+ oxidation state within the pristine K3V3(PO4)4·H2O sample.

Figure 2.

Figure 2

Open in a new tab

V 2p3/2 and V 2p1/2 peak XPS spectra of pristine K3V3(PO4)4·H2O.

To assess the thermal stability and water content, thermogravimetric analysis (TGA) of the K3V3(PO4)4·H2O sample is performed from room temperature to 800 °C (see Figure 3). Upon reaching 110 °C, the sample temperature is held for 2 h to remove any adsorbed water from the sample. Above 110 °C, the sample shows a gradual mass loss of ∼2.8%, equivalent to 1 formula unit of H2O between 100 and 500 °C. At higher temperatures, a short plateau for the sample mass occurs from 500 to 550 °C indicating that the material is fully dehydrated. This plateau is followed by a continuous mass gain of ∼2.2% from 550 to 800 °C.

Figure 3.

Figure 3

Open in a new tab

TGA curve for K3V3(PO4)4·H2O.

2.2. Computational Investigation of K3V3(PO4)4·H2O

Computational investigations of the structures for K3V2(PO4)3 and K3V3(PO4)4·H2O are performed using DFT geometry optimization of candidate unit cells provided in the literature. We develop the initial structure of K3V2(PO4)3 using the R3®c unit cell and lattice parameters for the K3V2(PO4)3 structure reported by Zhang et al.12 Atomic coordinates for the unit cell are taken from isostructural Na3V2(PO4)3.15 During optimization, lattice parameter values reduce substantially along the a, b, and c axes to values similar to literature values for Na3V2(PO4)3.15 The unit cell maintains an R3®c space group suggesting that the proposed unit-cell geometry is reasonable for K3V2(PO4)3. Table 1 shows key parameters for K3V2(PO4)3 reported by Zhang et al.12 and the geometry optimized structure using Materials Studio calculations. The calculated XRD patterns based on the optimized rhombohedral unit cell, as well as the structure reported by Zhang et al.12 did not match the experimental XRD data collected from the synthesized sample. Few experimental XRD peaks could be indexed to either rhombohedral unit cell. We show a comparison of the simulated powder XRD patterns for both the literature and geometry optimized K3V2(PO4)3 structures in Figure S1 of the Supporting Information.

Table 1. Cell Parameters of Candidate Structures for the KVP Compound.

  reference space group a (Å) b (Å) c (Å) γ (deg) V (Å3)
K3V2(PO4)3 Zhang et al.12 R3®c 18.133 18.133 31.812 120 9058.57
K3V2(PO4)3 computed (this work) R3®c 9.3799 9.3799 22.962 120 1749.62
K3Fe3(PO4)4·H2O Lii et al.20 Pnna 10.072 20.823 6.5296 90 1457.15
K3V3(PO4)4·H2O computed (this work) Pnna 10.832 21.26 6.616 90 1523.77
Open in a new tab

The experimental XRD pattern was further investigated using the PDF4+ database where all known phases for elements K, P, O, H, V, and Fe are included during the search match process. A successful match is found for the known iron(III) phosphate phase, K3Fe3(PO4)4·H2O (PDF no. 01404-012-4816). K3Fe3(PO4)4·H2O is assigned an orthorhombic cell with the space group Pnna.20 Database hkl positions and relative peak intensities of K3Fe3(PO4)4·H2O are consistent with the experimental XRD pattern for the optimum synthesized sample.

The DFT computed structure for K3V3(PO4)4·H2O is generated based on the starting structure, K3Fe3(PO4)4·H2O, where Fe(1) and Fe(2) atom sites are replaced with V(1) and V(2) atom sites before performing DFT geometry optimization. The final K3V3(PO4)4·H2O unit cell consists of 14 K, 12 V, 20 P, 70 O, and 8 H atoms and shows a slight expansion in lattice parameters while maintaining the Pnna space group. The computed lattice parameters for DFT geometry optimized K3V3(PO4)4·H2O are shown in Table 2. The experimentally determined lattice parameters and atomic coordinates for K3V3(PO4)4·H2O and the computed K3V3(PO4)4·H2O structure are compared in Table S1 of the Supporting Information.

Table 2. Structural Parameters Determined for K3V3(PO4)4·H2O.

crystal system orthorhombic
 space group
 Pnna  
unit-cell dimensions a = 10.7161(4) Å, b = 20.8498(10) Å, c = 6.5316(2) Å
α = β = γ = 90°
unit-cell volume 1459.35(10) Å3
        
atom Wyckoff position x y z occupancy Beq
K2 8e 0.50572 0.08971 0.36069 1 0.031
V2 8e 0.25951 0.33737 0.50606 1 0.561
P1 8e 0.24553 0.30149 0.00361 1 0.482
P2 8e 0.00257 0.10212 0.13875 1 0.497
O7 8e 0.32731 0.17927 0.27626 1 0.821
O8 8e 0.34783 0.24946 0.61005 1 0.608
O9 8e 0.22615 0.15976 0.66097 1 0.742
O10 8e 0.13098 0.23667 0.02243 1 0.663
O11 8e 0.131 0.11605 0.00826 1 0.782
O12 8e 0.36854 0.40076 0.52101 1 0.853
O13 8e 0.5103 0.53241 0.24749 1 0.979
O14 8e 0.00412 0.15911 0.29749 1 0.742
H1 8e 0.20372 0.00714 0.72271 1 3.948
K1 4c 0.25 0 0.16427 1 2.242
O15 4c 0.25 0 0.58698 1 1.418
V1 4d 0.00716 0.25 0.25 1 0.49
conditions for data collection
radiation source Cu Kα (λ = 1.5406 Å)
scan mode powder capillary
temperature 297 K
angular range 5° ≤ 2θ ≤ 140°
step scan increment (2θ) 0.02°
step scan rate (1) 2.053° min–1 (5–35°), (2) 1.013° min–1 (35–65°), (3) 0.503° min–1 (65–90°), (4) 0.200° min–1 (90–140°)
sample displacement (2θ) 0.01(1)°
no. of reflections 1395
anisotropic strain parameters S(400) = 0.04(7), S(040) = 0.26(4), S(004) = −0.086(1), S(220) = 1.65(7), S(202) = 17.87(3), S(022) = 0.55(12), and η = 0.59(6)
conventional Rietveld R factors
Rwp = 5.17%, Rexp = 4.81%, GOF = 1.07
Open in a new tab

2.3. Structure Refinement of K3V3(PO4)4·H2O

With a suitable starting structure based on the DFT computed structure for K3V3(PO4)4·H2O, Rietveld refinements were performed on the experimental XRD pattern of the KVP sample. During refinement, contributions from the instrument function are described and fixed via profile fitting to the NIST quartz standard 640C using a combination of Lorentzian, Hat, Gaussian, and Circle functions. In addition, anisotropic microstrain is included by using the phenomenological model described by Stephens.21 For this model, six Shkl independent parameters (S400, S040, ...) and a mixing factor, η, are required to describe the Stephens model for the orthorhombic system. These Shkl parameters allow the model to accommodate deviation of interplanar distances across a family of crystallographic planes representative of the crystal system. During refinement, the atom positions and thermal parameters of K and V are refined and restrained prior to refinement of the P, O, and H positions.

The results of the refinement using the K3Fe3(PO4)4·H2O structure reported by Lii et al.20 are shown in the Supporting Information (Figure S3). The K3Fe3(PO4)4·H2O refinement model demonstrates an adequate fit with a reasonably low difference and reliability factors, but clear mismatches in the intensity and minor differences in peak shapes are evident. Using the DFT computed starting structure for K3V3(PO4)4·H2O, a high-quality Rietveld refinement of the experimental XRD pattern is obtained. For this refinement as shown in Figure 4, previous mismatches for peak positions are well explained and the Iobs – Icalc difference plot is exceptional for K3V3(PO4)4·H2O. The high quality of the refinement is also validated by a goodness-of-fit (GOF) value (GOF = 1.07) and small differences and reliability factors (Rwp = 5.17%, and Rexp = 4.81% for 80 refined parameters). These outcomes suggest an adequate description of the structural model for K3V3(PO4)4·H2O. Rietveld refinement results are listed in Table 1 and K–O, V–O, H–O, and P–O bond lengths are listed in Table S2 in the Supporting Information. The cell parameters were determined to be a = 10.7161(4) Å, b = 20.8948(10) Å, and c = 6.5316(2) Å with standard deviations in parentheses.

Figure 4.

Figure 4

Open in a new tab

Rietveld refinement results for K3V3(PO4)4·H2O with the background fit and Bragg positions of key reflections for the Pnna space group. The curved background at a lower angle originates from the capillary.

2.4. Crystal Structure of K3V3(PO4)4·H2O

Figure 5 shows the structure of K3V3(PO4)4·H2O that contains layers of corner-sharing VO6 octahedra connected by PO4 tetrahedra parallel to the x–z plane. In detail, V(2)O6 octahedra and P(1)O4 tetrahedra share a common edge across O(8)–O(10) and share corners of O(7) and O(9) to form a distorted repeating chain structure with the V(2)O6 octahedra. Another repeating link is found by the corner sharing of O(14) between the V(1)O6 octahedra and P(1)O4 tetrahedra along the y axis. Both these repeat chains are connected along the x axis via O(8) and O(10) corner sharing to form the V3(PO4)4 slab layers along the x–z plane. This V3(PO4)4 slab stacks alternately along the y axis with an interlayer slab containing two different potassium-ion sites (K(1) at 4c and K(2) at 8e), with K(1) possessing sevenfold oxygen coordination and K(2) having eightfold coordination (see Figure S4). The water molecule, comprising H(1)–O(15)–H(1), is coordinated to both K(1) and K(2) where O(15) links once to the K(1) atom position and twice to the K(2) atom position.

Figure 5.

Figure 5

Open in a new tab

Crystal structure of K3V3(PO4)4·H2O with the unit cell outlined along the (a) a axis [100], (b) b axis [010], and (c) c axis [001] and (d) illustration of the V3(PO4)4 chain layer with oxygen positions outlined.

2.5. Vibrational Spectroscopy of K3V3(PO4)4·H2O

To confirm the character of the synthesized sample of K3V3(PO4)4·H2O, both Raman and Fourier transform infrared (FT-IR) spectra are obtained and are presented in Figure 6. Both spectra display a similar grouping of intense bands in the 1400–700 cm–1 region and a group of slightly weaker bands in the 700–450 cm–1 region. These bands agree reasonably well with previous reports for related metal phosphate compounds22−24 that describe the respective PO43– intramolecular stretching and bending character.

Figure 6.

Figure 6

Open in a new tab

(a) Fourier transform infrared and (b) Raman spectra with inserted H–O–H bending and stretching regions of K3V3(PO4)4·H2O.

The 1400–700 cm–1 region shows six bands across the Raman spectrum and a further three bands are evident in the infrared spectrum. We assign these bands to the PO43– intramolecular stretching modes (ν1 and ν3). The 450–700 cm–1 region shows a further seven Raman bands and five complementary infrared bands. These bands are assigned to the PO43– intramolecular bending vibrations (ν2 and ν4). From Raman spectroscopy, another grouping of strong bands is observed between 100 and 400 cm–1. This low-frequency region is tentatively attributed to external lattice vibrations. The presence of graphitic carbon is also identified by Raman spectroscopy from the broad bands apparent at 1300 and 1583 cm–1. These bands are ascribed to the D and G bands of graphite, respectively. The presence of residual carbon within the sample can be attributed to the use of oxalic acid during synthesis, which converts to graphite during the final calcination stage.

Lattice water in the structure of K3V3(PO4)4·H2O did not elicit strong bands in either infrared or Raman spectra; however, its presence is still observed. Within the infrared spectrum two minor bands are present in the 1600–1650 and 3100–3600 cm–1 regions (i.e., at 1606.1, 1641.0, 3411.0, and 3561.4 cm–1). The infrared bands in these regions are assigned to the H–O–H bending mode and H–O–H stretching vibrations, respectively. Within the Raman spectrum, evidence of a low intensity, broad band is also found across the 3300–3600 cm–1 region which is assigned to the hydroxyl (O–H) stretching vibrations.

To further investigate the characteristic vibrational frequencies for K3V3(PO4)4·H2O, phonon dispersion calculations for the refined unit cell were implemented via DFT. The optically active modes of K3V3(PO4)4·H2O are distributed among the following irreducible representations of the D2h6 point group.

2.5.

Theoretical Raman and IR active vibrational frequencies of K3V3(PO4)4·H2O are identified via group theory analysis. 348 normal modes are identified, of which 174 are described as Raman active and 129 as IR active. A full list of DFT calculated frequencies is provided in Table S5.

2.6. Electrochemical Investigation of K3V3(PO4)4·H2O

The electrochemical behavior of K3V3(PO4)4·H2O was investigated between 2.5 and 4.0 V (vs K/K+). Initially, cyclic voltammetry (CV) was performed across the first five charge–discharge cycles to determine the anodic and cathodic regions associated with the removal and reinsertion of K+ ions during electrochemical cycling. Figure 7a shows the CV curve for K3V3(PO4)4·H2O where a broad anodic (oxidative) peak region can be observed between 3.7 and 3.9 V (associated with the removal of K+ ions) and a corresponding sharp cathodic (reducing) peak region at 3.6 V (associated with K+ reinsertion into the structure). These peak regions correspond well with the observed charge–discharge profile where a sloping plateau region can be observed from 3.7 to 4.0 V during charge and from 3.6 to 3.4 V on discharge (see Figure 7b). As such, these redox reactions result in an initial charge–discharge capacity of 38.4 and 24.4 mAh g–1, respectively, at a rate of 0.5 C (1 C = 120.3 mAh g–1, based on the molecular weight of K3V3(PO4)4·H2O and assuming the 3 K+ ion redox process).

Figure 7.

Figure 7

Open in a new tab

(a) Cyclic voltammetry curve of the K3V3(PO4)4·H2O electrode from the first to the fifth cycle at 0.1 mV s–1, (b) charge–discharge curve of K3V3(PO4)4·H2O from the 1st to the 100th cycle, (c) long-term cycling performance of K3V3(PO4)4·H2O at 0.2 C (1 C = 120 mA g–1), and (d) rate capability testing of K3V3(PO4)4·H2O at 0.2, 0.5, 1, and 2 C (cell was cycled 20 cycles at 0.2 C initially to stabilize the initial poor Coulombic efficiency).

The capacity retention and long-term cycle stability of the K3V3(PO4)4·H2O electrode was measured across 100 cycles at 0.5 C. After cycling of the electrode, its discharge capacity on the 100th cycle demonstrated a reasonable capacity retention of 77.7%, corresponding to 18.7 mAh g–1. During the long cycle, K3V3(PO4)4·H2O displayed a poor initial Coulombic efficiency of 60.9%, but over subsequent cycles the round-trip efficiency improved to 90.7% on the 50th cycle and stabilized at 90% for the remainder of the experiment. The electrochemical rate capability of K3V3(PO4)4·H2O was also measured across a range from 0.2 to 2 C where discharge capacities of 21, 16, 13, and 10 mAh g–1 at 0.2, 0.5, 1, and 2 C, respectively, were delivered. In addition, the sample maintained a discharge capacity of 20 mAh g–1 when returned to 0.2 C.

3. Discussion

Our approach to crystal structure determination in this work is similar in philosophy to an earlier study on precise determination of the site occupancy, structure, and space group for bismuth-based pyrochlore photocatalysts.25 In the pyrochlore series, substitution of a larger ion (e.g., Cd) can lead to changes in symmetry from cubic to rhombohedral with consequent implications for experimental and theoretical determinations of band gap values.25 In this study, clarity on the structure and composition of the KVP phase influences our models on potassium transport in electrode materials for batteries. We elaborate below on synthesis conditions for the optimum production of K3V3(PO4)4·H2O, key influences on ion transport, and spectroscopic interpretation of this important phase.

3.1. Synthesis Conditions

In this study, a range of synthesis conditions were evaluated including various stoichiometric ratios between 3:2:3 and 3:3:4 for K/V/PO4 precursors. The primary and minor phases observed during synthesis were identified using powder XRD data and the results are summarized in Table S3 of the Supporting Information. Based on phase impurities, crystallization of K3V3(PO4)4·H2O appears highly dependent on the maintenance of a reducing environment during both hydrothermal treatment and final calcination stages.

In the XRD of the unwashed sample at a ratio of 3:2:3, minor peaks at 9.79 and 13.49° are matched to the monoclinic impurity phase K3V(PO4)2 (PDF no. 04-011-3486). At a ratio of 3:2.2:3.2 and higher, the monoclinic impurity phase, KVP2O7, was observed for all syntheses. Above ratios of 3:2.4:3.4, a secondary phase identified as VO(H2PO2)2·H2O also occurs under these conditions. Neither of these phases could be removed by alternate washing with DI water and ethanol after synthesis. The VO(H2PO2)2·H2O phase is also found in higher proportions as the ratio of precursors approaches 3:3:4. The increasing proportion of phase impurities is expected to be due to the increased concentration of the phosphate precursor producing a more oxidative local environment during the hydrothermal and final calcination steps.

The KVP2O7 impurity is also observed during attempts to perform final calcination under pure Argon gas compared to the use of Ar/H2(5%). For the stoichiometric ratio 3:3:4 in a pure argon environment, the K3V3(PO4)4·H2O phase fails to crystallize completely. This sensitivity to the reducing environment during synthesis has been similarly documented for the synthesis of LiFePO4, where comparable trends in formation of phase impurities, such as LiFeP2O7, is observed in less reducing environments.26

3.2. Comparison of Vibrational Spectra

DFT calculated values have been used to assign 21 Raman and 12 infrared active bands for observed spectra from K3V3(PO4)4·H2O. These observed bands with their respective symmetry assignments are presented alongside the corresponding DFT calculated band frequencies in Table 3.

Table 3. Raman and Infrared Modes for K3V3(PO4)4·H2O with Symmetry Assignments.

Raman
 FT-IR
observed DFT mode observed DFT mode
114.4 116.1 B2g 481.6 486.1 B1u
184.9 184.7 Ag 539.5 539.9 B2u
225.6 225.2 Ag 558.3 563.9 B3u
254.8 253.9 B3g 665.8 661.3 B1u
293.6 295.1 B2g 690.9 678.1 B2u
327.4 327.2 B1g 870.7 866.5 B1u
353.1 355.1 B1g 930.0 937.9 B3u
393 391.5 B2g 1097.3 1105.7 B1u
443.9 443.7 Ag 1167.2 1132.1 B1u
467.6 466.0 B3g      
521.1 523.7 B1g 1606.1 1471.7 B1u
561.8 564.9 B1g 3411.0 3355.54 B2u
593 595.8 Ag 3561.4 3355.7 B3u
665.1 666.1 Ag      
676.7 671.9 B1g      
930.1 937.7 Ag      
983.7 984.1 Ag      
1013.4 1002.1 Ag      
1084.1 1057.4 B1g      
1111.9 1112.5 B3g      
1174.5 1168.5 Ag      
Open in a new tab

The experimental Raman and infrared spectra present 33 Raman and IR active bands vs 348 from DFT calculations (Table S5). DFT calculations show good agreement with experimental results where general groupings of active band regions are reasonably estimated in both Raman and FT-IR spectra (see Figure 8). This agreement suggests that more active bands than identified from the experimental spectra may exist in closely grouped regions but are not visible as a result of band overlap.

Figure 8.

Figure 8

Open in a new tab

Observed (top) and calculated (bottom) Raman (left) and FT-IR (right) spectra for K3V3(PO4)4·H2O in the 100–4000 cm–1 wavenumber range.

3.3. Water Content and Thermal Stability

The water content of K3V3(PO4)4·H2O was substantiated through XRD refinement of its isomorphic structure, K3Fe3(PO4)4·H2O, and verified through TGA analysis. However, similar to literature findings regarding Na3Fe3(PO4)4,27 line broadening of diffraction lines along the a axis was observed. This broadening suggests strain effects parallel to the layer directions in the structure (S220 = 1.65%, S022 = 0.55%). This strain effect may be a result of the rehydration of potassium ions in the interlayer space upon cooling. As the sample is synthesized at 900 °C, it is unlikely that interlayer water is present during the initial phase of crystallization. Hydration during cooling will introduce interlayer structural defects to varying degrees across the bulk samples. Inhomogeneous hydration may also account for the apparent weak water character of K3V3(PO4)4·H2O observed with both Raman and infrared spectroscopy.

DFT phonon dispersion calculations suggest the presence of two Raman and two infrared active water bands in the 3100–3300 cm–1 regions, indicative of strong hydrogen bonding in water molecules and the existence of at least two different hydrogen bonded states.28 The experimental data presents bands in the 3400–3600 cm–1 region. The presence of the O–H stretching vibrations in the 3400–3600 cm–1 region suggests that weak hydrogen bonding, rather than strong bonding, is present in the water molecule of K3V3(PO4)4·H2O.

In this study, while TGA is primarily used to confirm the water content of the sample, the wide range of temperature for the water loss and regain of weight is unusual. Good thermal stability of crystal water in K3V3(PO4)4·H2O can be assumed between room temperature and 110 °C as the apparent mass loss during further heating to 550 °C equates to 1 formula unit of H2O, matching the stoichiometry. Nevertheless, the evidence of a phase change or reaction is displayed by the mass gain behavior above 550 °C. This mass gain behavior has also been previously observed during TGA analysis of the K3V2(PO4)3 compound in an air environment and was attributed to oxidation.29 Although the TGA experiment presented in this study is conducted in a N2 environment, the observed mass gain may be attributed to oxidation of K3V3(PO4)4·H2O due to the possible protection gas impurity. We believe this oxidation is likely attributed to vanadium, given that the gain is quite small (∼2.2%) and the sample color transitioned from green to yellow following TGA analysis, suggesting a subtle structural change that is yet to be determined.

3.4. K3V3(PO4)4·H2O: Implications for Battery Applications

Similar to recent findings for the sodium vanadium(III) phosphate phase, Na3V3(PO4)4,30 the structure of this material is found to be isomorphic with a previously reported iron(III) analogue, K3Fe3(PO4)4·H2O (PDF no. 04-012-4816).20 This new K3V3(PO4)4·H2O phase exhibits a layer structure parallel to the x-axis consisting of corner-sharing VO6 octahedra connected by corner and edge-sharing PO4 tetrahedra with a hydrated K+ ion interlayer. This layer structure is similar to many well-documented battery electrode intercalation materials31 and should allow K+ ion transport in the coordinated K+ layer along the x–z plane. Aside from diffusion paths parallel to the x-z basal plane, two other possible K+ ion diffusion tunnels can be seen along the (011) plane perpendicular to the (041) and (041̅) planes (see Figure 9). This suggests that K3V3(PO4)4·H2O possesses a set of three-dimensional (3D) diffusion pathways that make it suitable for intercalation/deintercalation and ionic transport of K+ ions.

Figure 9.

Figure 9

Open in a new tab

Perspective views of K3V3(PO4)4·H2O potential K-ion diffusion channels parallel to the (a) x–z basal plane [001] and along the [011] plane, perpendicular to (b) [041] and (c) [041̅].

Due to an absence of documentation regarding XRD peak positions and associated lattice parameters, it is difficult to confirm whether previous battery studies have reported results from the pure phase of K3V3(PO4)4·H2O, or in fact, a mixture of K3V3(PO4)4·H2O and other impurity phases identified in this work. However, based on observed ∼3:2:3 K/V/P atom ratios found from ICP-OES and EDS analyses12−14 conducted in these studies, we infer that varying ratios of electrochemically active phase impurities were present across the reported KVP compounds. For instance, the impurity phase KVP2O7 was present within our synthesized samples when K/V/P ratios exceeded 3:2.2:3.2 or when the calcination was performed in a less reducing atmosphere (i.e., pure argon). KVP2O7 has been previously reported by Park et al.32 as a promising high-energy cathode material for K+ ion batteries. KVP2O7 was found to possess a good discharge capacity (60 mAh g–1 at 0.25 C) corresponding to the reversible intercalation of 0.6 K+ ions accompanied by a phase transition from monoclinic P21/c KVP2O7 to triclinic P1̅ K1–xVP2O7 (x ≈ 0.6). Unfortunately, the presence of this phase cannot be easily identified in past studies due to the characteristic XRD peak overlap between KVP2O7 and K3V3(PO4)4·H2O at the 15–16, 22–23, and 28–31° 2θ regions.

Monoclinic K3V(PO4)2 was also found during sample preparation. The crystal structure of K3V(PO4)2 (monoclinic, space group 14, P121/c1) possesses an open polyhedral framework of PO4 and VO4 tetrahedra that could potentially serve as diffusion channels for the reversible insertion/extraction of K+ ions. This material has been studied recently by Bodart et al.33 and is found to be electrochemically active (vs K/K+), with a working potential of 3.5–4.0 V and a practical reversible capacity of 101 mAh g–1 (at 15 mA g–1 current density).33 Similarly to KVP2O7, the presence of the K3V(PO4)2 phase in the previous structural work is difficult to discern due to the XRD peak overlap at 28–31° 2θ, but minor peaks characteristic of the K3V(PO4)2 phase at ∼10 and ∼13° 2θ can be observed in the XRD patterns of KVP reported in previous studies (see circled areas in Figure 10a,c,e). The presence of these peaks suggests that the K3V(PO4)2 phase is present in significant quantities in previous characterized KVP samples, potentially contributing to their observed electrochemical performance.

Figure 10.

Figure 10

Open in a new tab

Previously published XRD patterns and CV curves for K3V2(PO4)3 compounds vs K/K+ reported by (a, b) Han et al. (adapted from ref (14) with permission from the Royal Society of Chemistry), (c, d) Zheng et al. (adapted with permission from ref (13). Copyright 2019 Elsevier B.V. All rights reserved), and (e, f) Zhang et al. (adapted from ref (34) under Creative Commons Attribution License) compared with (g, h) XRD and CV curve of K3V3(PO4)4·H2O. Characteristic XRD peaks for the K3V(PO4)2 impurity phase are circled in red.

The presence of electrochemically active phase impurities in previously reported KVP compounds can also be inferred when contrasting previous electrochemical measurements of K3V2(PO4)3 and the work presented in this article. Cyclic voltammetry of KVP shows inconsistencies between previous studies, where the presence and the absence of an anodic peak at 3.7 V and a cathodic peak at 3.4 V are seen in addition to the 3.7–3.9 and 3.6 V regions attributed to K3V3(PO4)4·H2O in this study (see Figure 10b,d,f,h). Therefore, the presence of KVP2O7 or K3V(PO4)2 as impurities may have significantly impacted the electrochemical performance characteristics and reported capacity contributions in previous K3V2(PO4)3 electrochemical studies.

Previous electrochemical testing of KVP materials has also often focused on electrochemical optimization via nanosizing,14 morphology control,29 and carbon additives34 to shorten ionic diffusion lengths and to improve electronic conductivity. These approaches have resulted in clear electrochemical performance improvement compared to “bulk” KVP reference samples.12,14 Nonetheless, even in optimized KVP samples from previous studies, there remains inherently poor Coulombic efficiency, hysteresis, and early cycle capacity fading; characteristics that remain a present issue for KVP materials within nonaqueous potassium-ion batteries.

When considering the possible sources of these electrochemical performance constraints in relation to the hydrated layer structure of K3V3(PO4)4·H2O, a number of possibilities become apparent. For instance, structural phase transitions during insertion and removal of K+ within K3V3(PO4)4·H2O may induce irreversible stacking sequence changes of the VO6–PO4 layers,35 which would account for the exhibited poor Coulombic efficiency and charge capacity loss during the initial cycles. In addition, the H2O molecule in the interlayer of K3V3(PO4)4·H2O has the potential to dissociate from the structure during initial cycles and may then be partially or fully replaced by K+ ions from the nonaqueous electrolyte. This scenario is similar to the potassium-ion battery study conducted by Hyoung et al. for the material VOPO4·2H2O,36 where interlayer H2O removal and replacement with K+ ions caused irreversible lattice contraction and structural change, leading to poor ionic diffusion kinetics and preventing the reinsertion of a portion of K+ into the VOPO4·2H2O structure. Another last consideration is that removal of interlayer H2O into a nonaqueous electrolyte during electrochemical cycling may lead to deleterious side reactions with the electrolyte salt, potassium reference foil, or even the dehydrated K3V3(PO4)4·H2O electrode itself, again, contributing to an exhibited poor electrochemical performance.

A phenomenon similar to VOPO4·2H2O may be present in K3V3(PO4)4·H2O whereby lattice contraction from H2O removal during long cycling leads to poor ionic diffusion kinetics and a lowered K+ intercalation capacity. Different to VOPO4·2H2O, the discharge capacity fades slowly and the Coulombic efficiency of K3V3(PO4)4·H2O actually improves over long cycling (60.9%(1st)–90.7%(50th)). This is the result of greater losses in charge capacity (54.68% capacity retention), than in discharge capacity (77.7% capacity retention) between the 1st and 100th cycles. This possibly indicates that K3V3(PO4)4·H2O undergoes a slow phase transition during long cycling, which results in a lowered K+ intercalation capacity, but allows for a more stable and reversible intercalation process during electrochemical cycling.

During the study by Hyoung et al.,36 the key to stable cyclability of VOPO4·2H2O was the introduction of water into the electrolyte where potassium ions (90%) and hydronium ions (10%) were present in the wet electrolyte. Thus, the stable hydrated structure of K3V3(PO4)4·H2O suggests that it may have wider applicability as a potential intercalation host for aqueous or mixed aqueous/nonaqueous K+ ion electrochemical systems. To further confirm the presence of these potential sources of poor electrochemical performance, the in/ex situ study of the structural change/evolution and the electrochemical performance of K3V3(PO4)4·H2O in aqueous and mixed electrolyte systems is recommended for future studies.

4. Conclusions

We have reinvestigated the primary phase of the reported compound K3V2(PO4)3 and identified this as a hydrated layered vanadium(III) phosphate phase, K3V3(PO4)4·H2O. The compound is synthesized as a powder sample and prepared using a facile aqueous hydrothermal method similar to previous methods with careful attention to reducing conditions. The crystal structure investigations and analyses are aided by DFT geometry optimization of the proposed crystal structures. The DFT geometry optimized starting structure for K3V3(PO4)4·H2O leads to very good agreement between the proposed structure and collected XRD powder patterns (Rwp = 5.17% and Rexp = 4.80% for 80 refined parameters).

A layered, platelike morphology is observed by SEM and K/V/P atomic ratios of 3:3:4 were confirmed by ICP-OES and XRF analysis. The water content of the structure as 1 H2O molecule per formula unit was also confirmed by TGA analysis. The crystal structure and thermal stability of this hydrated phase suggest that it may be a suitable alkali cation intercalation host for K+ ions and alternative alkali ions. This suggestion is supported by the exhibited electrochemical activity of K3V3(PO4)4·H2O vs K/K+, displaying characteristic charge and discharge plateaus at 3.7–3.9 and 3.6 V, respectively. Furthermore, the stable hydrated structure suggests wider applicability as a potential intercalation host for aqueous electrochemical systems based on K+ or multivalent charge carrier ions (i.e., Zn2+, Al3+), though further electrochemical studies of K3V3(PO4)4·H2O will be required to challenge this hypothesis.

5. Experimental Section

A series of aqueous hydrothermal syntheses were undertaken to optimize the yield and quality of the final product. Data presented here summarizes the optimum approach to obtain K3V3(PO4)4·H2O. This method follows that of Jiang et al.37 but with modifications to hydrothermal conditions and the final calcination time and temperature to successfully crystallize the target phase.

5.1. Synthesis Method

Analytical grade K2CO3, V2O5, C2H2O4·2H2O, and NH4H2PO4 obtained from Sigma-Aldrich were used for all experiments. Initially, 0.747 g of V2O5 (4.0 mmol) and 1.531 g of oxalic acid dihydrate (C2H2O4·2H2O) (12.1 mmol) were dissolved in 20 mL of DI water and stirred at 600 rpm for 30 min at 70 °C until a homogeneous dark blue solution was obtained. This color change indicates a change in the oxidation state to V (IV). Concurrently, 0.839 g of K2CO3 (6.0 mmol) and 1.397 g of NH4H2PO4 (12.1 mmol) were dissolved in 20 mL of DI water and then slowly added to the initial solution. Following this procedure, the solution was transferred to a Teflon-lined stainless-steel autoclave and heated for 16 h at 160 °C. After hydrothermal treatment, the resulting solution was agitated ultrasonically for 15 min, transferred to a Petri dish and then dried on a hotplate overnight at 60 °C. The resulting light green compound was collected and ground before heating at 350 °C for 2 h in air. The material was then reground and calcined at 900 °C for 20 h under an Ar/H2 (5%) atmosphere at a flow rate of 10 mL min–1, a heating rate of 5 °C min–1, and a cooling rate of 5 °C min–1. The sample was reground and washed with ethanol and DI water repeatedly using vacuum filtration to remove minor K3V(PO4)2 and KOH·H2O impurities.

5.2. Characterization Methods

Powder XRD patterns were collected using a Rigaku SmartLab diffractometer (Cu Kα radiation (λ = 1.5406 Å), a fixed divergent slit of 1°, 40 kV, and 40 mA at 25 °C, operating in powder capillary mode using a quartz glass capillary, ⌀ ∼ 0.3 mm). Data were obtained for 5° < 2θ < 140° using a variable count time (VCT) scheme at a constant step size of 0.02° 2θ and the following scan rates: (1) 2.053° min–1 (5–35°), (2) 1.013° min–1 (35–65°), (3) 0.503° min–1 (65–90°), and (4) 0.200° min–1 (90–140°). The XRD search and match procedure was performed using Bruker AXS DIFFRAC.Eva, and all database XRD patterns were taken from the PDF4+ database.

Synthesized powders were examined using a Zeiss Sigma field emission scanning electron microscope (FESEM) equipped with an Oxford Instruments EDS detector. Samples were coated with gold to minimize charging in the SEM. Bulk chemical compositions were determined using inductively coupled plasma spectrometry (ICP-OES) and X-ray fluorescence (XRF). XPS analysis was performed on a Kratos AXIS Supra X-ray photoelectron spectrometer and peaks were fitted in CasaXPS using a Shirley background extended to the O 1s peak. The water content and thermal stability of the sample were determined by thermogravimetric analysis (TGA) using a NETZSCH STA 499 F3 Jupiter. For TGA, samples were heated at a rate of 5 °C min–1 in flowing nitrogen. Initially, the sample was heated from room temperature to 100 °C and was held for 2 h to remove adsorbed water, after which the sample was heated to 800 °C. Raman scattering spectra were collected using a Renishaw inVia Raman microscope and Fourier transform infrared (FT-IR) spectroscopy spectra were collected using a Nicolet iS50 FT-IR spectrometer. Phase purities of all syntheses were confirmed via Raman spectroscopy.

5.3. Model Calculations

Model calculations were conducted with the density functional theory (DFT) framework using the CASTEP module of Materials Studio. For geometry optimization the generalized gradient approximation (GGA)-Perdew–Burke–Ernzerhof (PBE) exchange–correlation functional was used with the BFGS minimization scheme. The electronic wave functions for each k-point were expanded utilizing a plane-wave basis set and norm-conserving pseudopotentials and density mixing schemes. An energy cutoff of 900 eV and a k-point separation of 0.04 Å–1 were used. Phonon dispersion calculations were also conducted for the final refined unit cell using the linear response method. The local-density approximation (LDA) exchange–correlation functional with ultrasoft pseudopotentials was utilized for the initial geometry optimization with otherwise identical run conditions to those stated above.

Rietveld refinement of XRD data was performed using Bruker AXS Topas v6.38 During Rietveld refinement, the initial crystal system and space group indexing was performed using candidate cells collected from the literature, the PDF4+ database, and DFT structural calculation. The full refinement of K3V3(PO4)4·H2O consisted of 80 parameters including unit-cell parameters (3), scaling factor (1), crystallite size (1), Stephens anisotropic strain model factors (7), atom site positions (42 with 6 special positions), and thermal parameters (16). Crystal structure visualizations and bond lengths were generated using CrystalMaker X version 10.5.3 software using Shannon–Prewitt (S&P) ionic radii. Distortion index values were generated using Vesta software.39 All simulated XRD powder diffraction patterns were generated using the Reflex module of the Materials Studio 2018 package.

5.4. Electrochemical Measurements

Electrochemical measurements were performed as half-cells using 2032 type coin cells assembled in an argon-filled glove box. The coin cells consisted of a potassium metal foil counter electrode, glass fiber separator, and an electrolyte consisting of 1 mol L–1 KPF6 in a mixture of ethylene carbonate (EC) and dimethyl carbonate (DMC) (1:1 by volume). Electrochemical charge–discharge cycling was conducted at 24, 60, 120, and 240 mA g–1 for galvanostatic measurements with a fixed potential window between 2.5 and 4.0 V. Cyclic voltammetry was performed at a fixed scan rate of 0.1 mV s–1 within the same fixed potential range. All testing was performed at room temperature utilizing a battery test system (Biologic, VMP-300).

Acknowledgments

The authors would like to acknowledge that this activity received funding from ARENA as part of ARENA’s Research and Development Programme: Renewable Hydrogen for Export (Contract no. 2018/RND012). The views expressed herein are not necessarily the views of the Australian Government, and the Australian Government does not accept responsibility for any information or advice contained herein. We would also like to thank the staff and provisions from the Central Analytical Research Facility (CARF), The Institute for Future Environments (IFE), and the High Performance Computing (HPC) facilities at QUT for access to the equipment and facilities used in this work. Special thanks are given to Dr. Tony Wang for his assistance and guidance with the XRD powder diffraction data collection approach and analysis, and to Dr. Llew Rintoul for his insight during collection and analysis of the Raman and FT-IR spectroscopic data.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsomega.0c04675.

  • Brief summary of the rationale for choosing a DFT-assisted approach to structural refinement along with all supporting figures and tables (PDF)

  • Crystallographic structural information file for K3V3(PO4)4·H2O (CIF)

The authors declare no competing financial interest.

Supplementary Material

ao0c04675_si_001.pdf (630.3KB, pdf)
ao0c04675_si_002.cif (14.8KB, cif)

References

  1. Tao D.; Wang S.; Liu Y.; Dai Y.; Yu J.; Lei X. Lithium vanadium phosphate as cathode material for lithium ion batteries. Ionics 2015, 21, 1201–1239. 10.1007/s11581-015-1405-3. [DOI] [Google Scholar]
  2. Fang Y.; Zhang J.; Xiao L.; Ai X.; Cao Y.; Yang H. Phosphate Framework Electrode Materials for Sodium Ion Batteries. Adv. Sci. 2017, 4, 1600392 10.1002/advs.201600392. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. He G.; Kan W. H.; Manthiram A. Delithiation/lithiation behaviors of three polymorphs of LiVOPO4. Chem. Commun. 2018, 54, 13224–13227. 10.1039/C8CC07446A. [DOI] [PubMed] [Google Scholar]
  4. Kim J.; Yoon G.; Kim H.; Park Y.-U.; Kang K. Na3V(PO4)2: A New Layered-Type Cathode Material with High Water Stability and Power Capability for Na-Ion Batteries. Chem. Mater. 2018, 30, 3683–3689. 10.1021/acs.chemmater.8b00458. [DOI] [Google Scholar]
  5. Boudin S.; Guesdon A.; Leclaire A.; Borel M. M. Review on vanadium phosphates with mono and divalent metallic cations: syntheses, structural relationships and classification, properties. Int. J. Inorg. Mater. 2000, 2, 561–579. 10.1016/S1466-6049(00)00074-X. [DOI] [Google Scholar]
  6. Zhang X.; Rui X.; Chen D.; Tan H.; Yang D.; Huang S.; Yu Y. Na3V2(PO4)3: an advanced cathode for sodium-ion batteries. Nanoscale 2019, 11, 2556–2576. 10.1039/C8NR09391A. [DOI] [PubMed] [Google Scholar]
  7. Rui X.; Yan Q.; Skyllas-Kazacos M.; Lim T. M. Li3V2(PO4)3 cathode materials for lithium-ion batteries: A review. J. Power Sources 2014, 258, 19–38. 10.1016/j.jpowsour.2014.01.126. [DOI] [Google Scholar]
  8. Hautier G.; Jain A.; Ong S. P.; Kang B.; Moore C.; Doe R.; Ceder G. Phosphates as Lithium-Ion Battery Cathodes: An Evaluation Based on High-Throughputab InitioCalculations. Chem. Mater. 2011, 23, 3495–3508. 10.1021/cm200949v. [DOI] [Google Scholar]
  9. Wang Q.; Xu J.; Zhang W.; Mao M.; Wei Z.; Wang L.; Cui C.; Zhu Y.; Ma J. Research progress on vanadium-based cathode materials for sodium ion batteries. J. Mater. Chem. A 2018, 6, 8815–8838. 10.1039/C8TA01627E. [DOI] [Google Scholar]
  10. Mao W.-f.; Tang H.-q.; Tang Z.-y.; Yan J.; Xu Q. Configuration of Li-Ion Vanadium Batteries: Li3V2 (PO4) 3 (cathode)a∥ Li3V2 (PO4) 3 (anode). ECS Electrochem. Lett. 2013, 2, A69–A71. 10.1149/2.008307eel. [DOI] [Google Scholar]
  11. Jian Z.; Sun Y.; Ji X. A new low-voltage plateau of Na3V2(PO4)3 as an anode for Na-ion batteries. Chem. Commun. 2015, 51, 6381–6383. 10.1039/C5CC00944H. [DOI] [PubMed] [Google Scholar]
  12. Zhang L.; Zhang B.; Wang C.; Dou Y.; Zhang Q.; Liu Y.; Gao H.; Al-Mamun M.; Pang W. K.; Guo Z.; Dou S. X.; Liu H. K. Constructing the best symmetric full K-ion battery with the NASICON-type K3V2(PO4)3. Nano Energy 2019, 60, 432–439. 10.1016/j.nanoen.2019.03.085. [DOI] [Google Scholar]
  13. Zheng S.; Cheng S.; Xiao S.; Hu L.; Chen Z.; Huang B.; Liu Q.; Yang J.; Chen Q. Partial replacement of K by Rb to improve electrochemical performance of K3V2(PO4)3 cathode material for potassium-ion batteries. J. Alloys Compd. 2020, 815, 152379 10.1016/j.jallcom.2019.152379. [DOI] [Google Scholar]
  14. Han J.; Li G. N.; Liu F.; Wang M.; Zhang Y.; Hu L.; Dai C.; Xu M. Investigation of K3V2(PO4)3/C nanocomposites as high-potential cathode materials for potassium-ion batteries. Chem. Commun. 2017, 53, 1805–1808. 10.1039/C6CC10065A. [DOI] [PubMed] [Google Scholar]
  15. Zatovsky I. V. NASICON-type Na(3)V(2)(PO(4))(3). Acta Crystallogr., Sect. E: Struct. Rep. Online 2010, 66, i12. 10.1107/S1600536810002801. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Yang X.-S.; Wang Y.-Y.; Zhong Y.-J.; Zhong B.-H.; Tang Y. Macroporous network Li3V2(PO4)3/C cathode material with excellent high-rate performance for lithium-ion batteries. RSC Adv. 2015, 5, 77637–77642. 10.1039/C5RA14542B. [DOI] [Google Scholar]
  17. Membreño N.; Park K.; Goodenough J. B.; Stevenson K. J. Electrode/Electrolyte Interface of Composite α-Li3V2(PO4)3 Cathodes in a Nonaqueous Electrolyte for Lithium Ion Batteries and the Role of the Carbon Additive. Chem. Mater. 2015, 27, 3332–3340. 10.1021/acs.chemmater.5b00447. [DOI] [Google Scholar]
  18. Ren M.; Zhou Z.; Li Y.; Gao X. P.; Yan J. Preparation and electrochemical studies of Fe-doped Li3V2(PO4)3 cathode materials for lithium-ion batteries. J. Power Sources 2006, 162, 1357–1362. 10.1016/j.jpowsour.2006.08.008. [DOI] [Google Scholar]
  19. Klee R.; Aragón M. J.; Lavela P.; Alcántara R.; Tirado J. L. Na3V2(PO4)3/C Nanorods with Improved Electrode–Electrolyte Interface As Cathode Material for Sodium-Ion Batteries. ACS Appl. Mater. Interfaces 2016, 8, 23151–23159. 10.1021/acsami.6b07950. [DOI] [PubMed] [Google Scholar]
  20. Lii K. H. K3Fe3(PO4)4·H2O: An Iron(III) Phosphate with a Layer Structure. Eur. J. Solid State Inorg. Chem. 1995, 917–926. [Google Scholar]
  21. Stephens P. W. Phenomenological model of anisotropic peak broadening in powder diffraction. J. Appl. Crystallogr. 1999, 32, 281–289. 10.1107/S0021889898006001. [DOI] [Google Scholar]
  22. Frost R. L.; López A.; Scholz R.; Xi Y.; Lana C. The molecular structure of the phosphate mineral beraunite Fe2+Fe53+(PO4)4(OH)5·4H2O – A vibrational spectroscopic study. Spectrochim. Acta, Part A 2014, 128, 408–412. 10.1016/j.saa.2014.02.198. [DOI] [PubMed] [Google Scholar]
  23. Burba C. M.; Frech R. Vibrational spectroscopic studies of monoclinic and rhombohedral Li3V2(PO4)3. Solid State Ionics 2007, 177, 3445–3454. 10.1016/j.ssi.2006.09.017. [DOI] [Google Scholar]
  24. Frost R. L.; Weier M. L.; Erickson K. L.; Carmody O.; Mills S. J. Raman spectroscopy of phosphates of the variscite mineral group. J. Raman Spectrosc. 2004, 35, 1047–1055. 10.1002/jrs.1251. [DOI] [Google Scholar]
  25. Perenlei G.; Talbot P. C.; Martens W. N.; Riches J.; Alarco J. A. Computational prediction and experimental confirmation of rhombohedral structures in Bi1.5CdM1.5O7 (M = Nb, Ta) pyrochlores. RSC Adv. 2017, 7, 15632–15643. 10.1039/C6RA27633D. [DOI] [Google Scholar]
  26. Ong S. P.; Wang L.; Kang B.; Ceder G. Li–Fe–P–O2 Phase Diagram from First Principles Calculations. Chem. Mater. 2008, 20, 1798–1807. 10.1021/cm702327g. [DOI] [Google Scholar]
  27. Trad K.; Carlier D.; Croguennec L.; Wattiaux A.; Lajmi B.; Ben Amara M.; Delmas C. A Layered Iron(III) Phosphate Phase, Na3Fe3(PO4)4: Synthesis, Structure, and Electrochemical Properties as Positive Electrode in Sodium Batteries. J. Phys. Chem. C 2010, 114, 10034–10044. 10.1021/jp100751b. [DOI] [Google Scholar]
  28. Frost R. L. An infrared and Raman spectroscopic study of the uranyl micas. Spectrochim. Acta, Part A 2004, 60, 1469–1480. 10.1016/j.saa.2003.08.013. [DOI] [PubMed] [Google Scholar]
  29. Wang X.; Niu C.; Meng J.; Hu P.; Xu X.; Wei X.; Zhou L.; Zhao K.; Luo W.; Yan M.; Mai L. Novel K3V2(PO4)3/C Bundled Nanowires as Superior Sodium-Ion Battery Electrode with Ultrahigh Cycling Stability. Adv. Energy Mater. 2015, 5, 1500716 10.1002/aenm.201500716. [DOI] [Google Scholar]
  30. Liu R.; Liu H.; Sheng T.; Zheng S.; Zhong G.; Zheng G.; Liang Z.; Ortiz G. F.; Zhao W.; Mi J.; Yang Y. Novel 3.9 V Layered Na3V3(PO4)4 Cathode Material for Sodium Ion Batteries. ACS Appl. Energy Mater. 2018, 1, 3603–3606. 10.1021/acsaem.8b00889. [DOI] [Google Scholar]
  31. Kaufman J.; Vinckevičiu̅tė J.; Kolli S.; Goiri J.; Ven A. Understanding intercalation compounds for sodium-ion batteries and beyond. Philos. Trans. R. Soc., A 2019, 377, 20190020 10.1098/rsta.2019.0020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Park W. B.; Han S. C.; Park C.; Hong S. U.; Han U.; Singh S. P.; Jung Y. H.; Ahn D.; Sohn K.-S.; Pyo M. KVP2O7 as a Robust High-Energy Cathode for Potassium-Ion Batteries: Pinpointed by a Full Screening of the Inorganic Registry under Specific Search Conditions. Adv. Energy Mater. 2018, 8, 1703099 10.1002/aenm.201703099. [DOI] [Google Scholar]
  33. Bodart J.; Eshraghi N.; Carabin T.; Vertruyen B.; Boschini F.; Mahmoud A. In Spray-Dried K3V(PO4)2 and K3V(PO4)2/Carbon as New Cathode Materials for Potassium-Ion Batteries, LE STUDIUM Conference—Towards Futuristic Energy Storage; Paving Its Way through Supercapacitors, Li-Ion Batteries and Beyond, France, 2020.
  34. Zhang X.; Kuang X.; Zhu H.; Xiao N.; Zhang Q.; Rui X.; Yu Y.; Huang S. Hybrid Cathodes Composed of K3V2(PO4)3 and Carbon Materials with Boosted Charge Transfer for K-Ion Batteries. Surfaces 2020, 3, 1–10. 10.3390/surfaces3010001. [DOI] [Google Scholar]
  35. Radin M. D.; Alvarado J.; Meng Y. S.; Van der Ven A. Role of Crystal Symmetry in the Reversibility of Stacking-Sequence Changes in Layered Intercalation Electrodes. Nano Lett. 2017, 17, 7789–7795. 10.1021/acs.nanolett.7b03989. [DOI] [PubMed] [Google Scholar]
  36. Hyoung J.; Heo J. W.; Chae M. S.; Hong S.-T. Electrochemical Exchange Reaction Mechanism and the Role of Additive Water to Stabilize the Structure of VOPO4·2 H2O as a Cathode Material for Potassium-Ion Batteries. ChemSusChem 2019, 12, 1069–1075. 10.1002/cssc.201802527. [DOI] [PubMed] [Google Scholar]
  37. Jiang X.; Yang L.; Ding B.; Qu B.; Ji G.; Lee J. Y. Extending the cycle life of Na3V2(PO4)3 cathodes in sodium-ion batteries through interdigitated carbon scaffolding. J. Mater.Chem. A 2016, 4, 14669–14674. 10.1039/C6TA05030A. [DOI] [Google Scholar]
  38. Coelho A. TOPAS and TOPAS-Academic: an optimization program integrating computer algebra and crystallographic objects written in C++. J. Appl. Crystallogr. 2018, 51, 210–218. 10.1107/S1600576718000183. [DOI] [Google Scholar]
  39. Momma K.; Izumi F. VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data. J. Appl. Crystallogr. 2011, 44, 1272–1276. 10.1107/S0021889811038970. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ao0c04675_si_001.pdf (630.3KB, pdf)
ao0c04675_si_002.cif (14.8KB, cif)

Articles from ACS Omega are provided here courtesy of American Chemical Society

RESOURCES