Reversible Insertion in AFeF3 (A = K+, NH4+) Cubic Iron Fluoride

Subscriber access provided by CARLETON UNIVERSITY

Functional Inorganic Materials and Devices

Reversible Insertion in AFeF3 (A = K+, NH4+) Cubic Iron Fluoride Perovskites Andréa MARTIN, Enrique Soto Santiago, Erhard Kemnitz, and Nicola Pinna ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.9b10659 • Publication Date (Web): 20 Aug 2019 Downloaded from pubs.acs.org on August 21, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Reversible Insertion in AFeF3 (A = K+, NH4+) Cubic Iron Fluoride Perovskites Andréa Martin, Enrique S. Santiago, Erhard Kemnitz, Nicola Pinna* Institut für Chemie and IRIS Adlershof, Humboldt-Universität zu Berlin, Brook-Taylor-Str. 2, D-12489 Berlin, GERMANY

Keywords: KFeF3, NH4FeF3, metal fluoride perovskites, lithium-ion battery, sodium-ion battery, reversible insertion.

ABSTRACT. The search for new cathode materials is primordial for alkali-ion battery systems, which are facing a constantly growing demand for high energy density storage devices. In quest of more performant active compounds on the positive side, anhydrous iron(III) fluoride demonstrated to be a good compromise in terms of high capacity, operating voltage and low cost. However, its reaction towards lithium leads to complicated insertion/conversion reactions, which hinder its performances in Li-Ion cells. Cycling this material against larger ions such as sodium and potassium is hard or simply impossible due to the size of the channels of the FeF3 framework impeding ions diffusion. Herein, we propose a strategy, based on the use of cubic perovskite AFeF3 (A = K+, NH4+) as starting materials, allowing the straightforward insertion (after a first disinsertion of the alkali and/or NH4+ ion) of lithium within the structure and enabling the cycling towards larger alkali ions such as sodium and potassium. For example, a cubic KFeF3

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

perovskite, produced by a facile synthesis method, shows superior rate capability towards lithium retaining a capacity of up to 132mAh.g-1 at 5C, or of 120mAh.g-1 at 5C towards sodium and enabling cycling towards potassium. Moreover, cubic NH4FeF3 perovskite is discussed for the first time as suitable cathode material for alkali-ion batteries.

1. INTRODUCTION Lithium-ion battery (LIB) has imposed itself as the main storage energy technology over the past decades for the mobile device market, due to its excellent storage capacity and its cost effective aspect. However, because of the growing consumption of mobile devices and the onset of the implementation of such technology into electrical vehicles, the electrochemical performances of the LIBs need to be greatly improved

1,2,3.

Moreover, an effective and cheaper approach is

needed to store intermittent energy produced by wind turbines and solar panels, for example. Sodium-ion battery (NIB) could be a good alternative to the more expensive LIBs for stationary battery application, but at the moment the best configuration of active materials for the electrode formulations is still unclear

4,5.

For the further development of these two technologies, new

materials need to be considered in order to further increase the energy and power density of these devices. To this end, a better understanding of the mechanisms taking place during the electrochemical reactions of active electrode materials is needed. In the search for new materials, iron(III) fluoride emerged as good candidate due to its high operating voltage, because of its wide bandgap (4.49 eV), high theoretical capacity, abundance, and environmental friendliness. To overcome the high insulating character of the metal-fluorine bond, strategies such as the downsizing of the particle size and the coupling with a conductive


2

Page 3 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


matrix have been successful

6,7,8.

FeF3 crystalizes under different polymorphs depending on the

water content 9,10. Anhydrous FeF3 adopts a compact hexagonal conformation close to the one of the ReO3 with a slight rotation of the octahedrons of the framework (Fe-F-Fe bond angles close to 150°). The reaction of FeF3 with one lithium ion per formula unit passes through complicated insertion/conversion mechanisms, influencing the stability, reversibility and the rate capability of the fluoride material 11. By preventing these detrimental conversion reactions upon cycling, the performances of iron fluoride-based compounds could be drastically improved. Our previous study showed that after the disinsertion of the sodium from the orthorhombic NaFeF3 perovskite, the structure adopted a cubic FeF3 conformation (ReO3-structural type with Fe-F-Fe bond angles close to 180°), which is normally not stable at room temperature. We found out that this cubic conformation allowed the reversible insertion of lithium and sodium ions into the structure, which would not be the case if the compound would endorse the hexagonal structure (Fe-F-Fe bond angles close to 150°). Furthermore, this cubic conformation enabled the extremely fast cycling of the NaFeF3 towards lithium exhibiting a reversible capacity of 100mAh.g-1 at 25C 12. This extraordinary rate capability for such a compound is believed to be due to the straightforward insertion of lithium ions within the cubic framework, which is not the case when hexagonal iron(III) is used as starting material. However, the orthorhombic structure of the starting NaFeF3 appears to be not perfectly suitable for the fast diffusion of ions since the channels present in the framework are too narrow. Recently, Han et al. proposed the use of iron fluoride compounds crystallizing under different phases stabilized by potassium ions to improve the diffusivity of large ions such as sodium

13.

Even though the materials presented larger

tunnels, the performances achieved were still below the ones we reported for the NaFeF3 phase. Yi et al. and Cao et al. reported interesting insertion performances of the cubic KFeF3 phase, but


3


Page 4 of 24

not as good as recently published for the NaFeF3 14,15. Both studies started from micron-sized particles subsequently downsized by ball-milling in harsh conditions for long times, which could affect the organization of the crystalline structure and create strain in the composite, as a result, a non-complete extraction of the potassium ions was observed hindering the performance of the materials 16,17. On the other hand, compounds presenting ammonium ions in their structure have been only seldom studied as active materials for batteries, and are generally used in K-ion systems where the ammonium ions act as a structure expander due to its large ionic radius

18.

However, only a few of these compounds undergo the electrochemical disinsertion of this cation along the cycling. Very recently, Wu et al. proposed for the first time a “rocking-chair” system using the ammonium ions as ionic driving force

19.

Since then, different compounds have been

reported to exhibit a reversible ammonium disinsertion, but only in aqueous systems 20,21. In this article, we describe a new approach to cycle iron(III) fluoride towards insertion of alkali ions by starting from different cubic perovskites structures, able to achieve very high performance in terms of capacity and rate capability. The starting fluoride materials, namely the KFeF3 and NH4FeF3 and their intermediate phases, are synthesized through a simple fluorolytic sol-gel method using ammonium fluoride

17,22,2317, 23.

This synthesis approach based on soft

chemistry allows the control of the morphology of the produced nanoparticles and enables the use of mild ball-milling conditions. Although the potassium-containing phase has been already proposed for alkali-ion batteries, only poor performances have been reported. On the other hand, the cubic perovskite fluoride containing ammonium ions has never been reported so far for the insertion of alkali ions. This structure benefits from the lightweight and mobile NH4+ cations and preserves the perovskite structure along the cycling. The cubic perovskite framework is kept after the first disinsertion of the ammonium ions and allows the fast diffusion of alkali ions


4

Page 5 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


within the structure. The stabilization of such a conformation leads to a high rate cyclability and high capacity. Herein, syntheses and performances of these materials are discussed, as well as the mechanisms involved in the charge/discharge behavior by ex-situ X-ray. 2. RESULTS AND DISCUSSION In our previous study, we introduced the synthesis of nanosized NaFeF3 material by the fluorination/sodiation in solution by microwave heating of very small FeF2 nanoparticles stabilized by oleylamine 12. The high surface area of the FeF2 rutile precursor allowed the total conversion of the material into the orthorhombic NaFeF3 perovskite structure. In contrast to this two-step synthesis, the use of transition metal acetates and ammonium fluoride as the only source of fluorine in benzyl alcohol, in the presence of an alkali precursor or not, enables the synthesis of the cubic perovskite structure by a mild heating process in only one step (cf. experimental section). Figure 1a and b display typical images of the KFeF3 and NH4FeF3 materials synthesized by the fluorolytic sol-gel process under soft conditions. Both compounds consist of pseudo-cubic-shaped nanoparticles with a narrow size distribution forming small aggregates. Those materials adopt a cubic perovskite structure belonging to the Pm-3m space group with no visible impurity based on X-ray powder diffraction (Figure 1c). The differences between the two respective diffractograms (mostly relative intensity of the reflections) are due to the exchange of the cation located in the (0.5, 0.5, 0.5) crystalline lattice position, which represents the A-site of the perovskite structure. NH4+ has a larger ionic radius resulting in a slightly larger lattice parameter (a(NH4FeF3) = 4.23 Å and rNH4+ = 1.41 Å) and a smaller atomic scattering factor, compared to K+ ions (a(KFeF3) =4.13 Å rK+ = 1.33 Å)

24.

The whole range of

compositions K1-x(NH4)xFeF3 ( 0 ≤ x ≤ 1 ) can be synthesized by varying the amount of potassium ethoxide used as reactant. The final compositions, as verified by EDX analysis,


5


Page 6 of 24

correspond to the K+/NH4+ ratio chosen (Figure S1b). On the other hand, a small excess of KOEt regarding the equivalent amount of reagent ensures the obtaining of a pure KFeF3. Only a few studies report the direct synthesis NH4M(+2)F3 (M = transition metals) phases, avoiding a high temperature thermal treatment of typically (NH4)1+xM(+3)F4+x ( 0 ≤ x ≤ 2 ) precursor phases 25,26,27. The use of the very simple synthesis route reported herein can be applied to other transition metals and is not limited only to potassium, but also to other B-type cations (Fig.S2 a and b). The perovskite crystalline framework adopting a cubic structure is well suited for facilitating the A-type ion diffusion through the large channels along the three-basis vector of the cubic lattice (Figure 1 d). Although the as-synthesized particles present a homogeneous nanosized morphology, which shortens the electrons and ions paths, they still present low conductivity due to the highly ionic bond-type. Therefore, a mild ball-milling has been used to combine active materials with carbon black, leading to an increase in the conductivity of the electrode material without dramatically affecting the crystalline framework (Figure S3a and b), which is also kept after cycling (Figure S3c and d).


6

Page 7 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. (a) TEM image of the KFeF3 and (b) of the NH4FeF3 compounds. (c) Diffraction patterns of the KFeF3 (black), NH4FeF3 (red) and their respective reference patterns, (d) graphic representation of the crystalline structure of the two compounds belonging to the space group Pm-3m. The central location is occupied either by the K atoms in the case of the KFeF3 or by the ammonium cations in the case of the NH4FeF3.


7


Page 8 of 24

The cycling experiments of the perovskite compounds were carried out in half cell set up at room temperature using metallic sodium and lithium as references and counter electrodes. Figure 2.a displays the galvanostatic profile of the KFeF3 cycled towards sodium. The first charge process representing the first disinsertion of the K+ shows a different behavior compared to the subsequent ones, due to the different nature of the cation extracted from the framework. This behavior can also be observed during the cycling towards lithium. Additional capacity is observed when the system is cycled up to 4.5V at C/10 (17.6 mA.g-1) compared to what is needed to extract one K+ per formula unit. This is due to a side reaction occurring at high cutoff potential involving the decomposition of the electrolyte at the surface of the metal fluoride compound. The next discharge takes place in the 3V region with a slightly pronounced plateau and a capacity in line with the theoretical one (176 mAh.g-1). The potential is stopped at 2V at the end of the discharge to avoid any conversion reaction that would hinder the subsequent insertion of alkali ions. The subsequent charges/discharges display obvious plateaus located at 3.3V and 3V as we already reported in the case of the NaFeF3 compound cycled towards Na 12. The cyclic voltammetry waves are in accordance with the results observed by galvanostatic experiments with broad signals during the first cycle, followed by well-defined redox peaks at the already mentioned potentials (Figure S5a). An extended peak can be observed after 4.0 V, probably due to the decomposition of the electrolyte. The rate capability is remarkable with about 70% of the theoretical capacity recovered after the first discharge at 5C. On the other side, the capacity retention is poor (Figure 2c), this is due to the high cutoff potential causing the decomposition of the electrolyte, or other phenomena related to the alkalis dis/insertion processes, as this does not depend on the C-rate used. The use of lower cutoff potentials results


8

Page 9 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


in lower discharge capacity (Figure S4a and b), caused by the incomplete disinsertion of the potassium ions from the crystalline structure.

Figure 2. Electrochemical performances of KFeF3. Galvanostatic charge and discharge curves of KFeF3 at the 1st, 10th and 50th cycles and cycled at C/10 (a) towards Na, (b) towards Li. Capacity retention along the cycles of the KFeF3 at C/10, 1C, 5C cycled (c) towards Na and (d) towards Li.

The galvanostatic charge and discharge curves of the potassium iron fluoride when cycled towards lithium are shown in Figure 2b. After disinsertion of the potassium ions characterized


9


Page 10 of 24

by a first slopped curve, a single clear plateau can be observed between 3.5V and 2.8V, supported by cyclic voltammetry experiments (Figure S5b). Similar to the cycling towards Na, the theoretical capacity is reached during the first discharge at C/10 towards Li. Then, after 100 subsequent cycles, the capacity decreases down to 63% of its initial value (Figure 2d). The columbic efficiency of the first cycle is 82% and then stabilized at 94% after 10 cycles. The material shows again a very good rate capability with up to 75% of the theoretical capacity delivered during the discharge at 5C. At even higher current densities, the compound shows outstanding performances, with ~60mAh.g-1 at 40C (Figure S6). This extremely high rate capability, even higher compared to the one we reported previously for the NaFeF3, can be attributed to the open framework presented by KFeF3. Unfortunately, the potential window needed to obtain the full capacity of the material drives to low capacity retention, certainly due to the instability of the electrolyte towards the metal fluoride compound or/and irreversible phenomena occurring during the charge/discharge processes.


10

Page 11 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3. Electrochemical performances of NH4FeF3. Galvanostatic charge and discharge of the NH4FeF3 (a) of the 1st, 10th and 50th cycles and cycled at C/2 towards Na, (b) of the 1st, 10th and 50th cycles and cycled towards Li. Capacity retention along the cycles of the NH4FeF3 (c) towards Na at C/2, 1C, 2C and 5C and (d) towards Li at C/10, 1C and 5C.

NH4FeF3 as material for Li-Ion battery has been reported only one time, but as an anode material reacting through conversion reaction, as it is usually the case for FeF3 and other MF2 materials 26. However, cycling of NH4FeF3 towards alkali ions through insertion reactions has never been investigated so far. Figure 3a and b display galvanostatic charge and discharge cycles of NH4FeF3 towards Na and Li, respectively. The first step is the disinsertion of the ammonium cations from the crystalline framework represented by a slopped plateau between 3.5 and 4.5 V


11


Page 12 of 24

vs. Na/Na+ and 3.3 and 4.5V vs. Li/Li+. This first step shows additional capacity in the case of lithium as well as for the sodium system (Cth(NH4FeF3 ) = 206mAh.g-1), which is probably due either from side reactions occurring when metal fluorides are used at high voltage cutoff or by introducing NH4+ cations within the electrolyte. It is quite surprising that ammonium ions can follow a redox disinsertion from the perovskite structure, similarly to alkali ions. It should be stressed that this is the first time that such a phenomenon is reported for non-aqueous alkali-ion systems. After the first charge process, the insertion of alkali ions occurs as for other insertion materials. The discharge potential is stopped at 2V to avoid any conversion reaction of the perovskite crystalline framework. Towards sodium, the discharge profile does not present a pronounced plateau compared to the charge where small features can be observed around 3V (due to the crystallinity decrease of the compound after ammonium ions extractions, cf. below). The cycling voltammetry shows very broad redox waves over 20 cycles, in agreement with the not well-pronounced plateau observed in the galvanostatic experiments (Figure S5a). The capacity retention over 100 cycles is about 85% with a first discharge capacity of 180mAh.g-1 (Figure 3a), which roughly represents 0.9 sodium ion inserted per formula unit. The rate capability of the NH4FeF3 phase in the sodium system achieves 100mAh.g-1 when cycled at 5C. The good rate capability is ascribed to the large channels that the crystalline cubic perovskite framework offers. Towards lithium, the theoretical capacity is achieved at a current density of C/2 with a relatively well-defined plateau around 3V, supported by cyclic voltammetry experiments (Figure S5b), with a 94% capacity retention after 100 cycles (Figure 3b). The material presents poor columbic efficiency for the 10 first cycles at C/2, but it increases above 90% afterward. The NH4FeF3 can undergo current densities as high as 5C while maintaining


12

Page 13 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


around half of the theoretical capacity with good capacity retention, namely 74% of retention after 300 cycles (Figure S7). It is interesting to note that the cycling of such cubic perovskite structures is also possible towards potassium. The large channels of the cubic structure can easily host the larger potassium ions. The KFeF3 compound can deliver 126mAhg-1 at C/10 with 50% of capacity retention over 50 cycles (Figure S8a and b). A large hysteresis between the charge and discharge is observed, causing poor columbic efficiency. The NH4FeF3 compound displays also relative good capacity cycled under the same conditions (Figure S8c). These results show that the cubic perovskite fluoride compounds could stand as a candidate for positive electrode in K-ions batteries. On the other hand, the tight crystalline framework of anhydrous hexagonal iron(III) fluoride does not allow the reaction towards potassium as denoted by the low discharge capacity (

Reversible Insertion in AFeF3 (A = K+, NH4+) Cubic Iron Fluoride

Recommend Documents