Article pubs.acs.org/JPCC
Microstructural Evolution Of Iron Oxyfluoride/Carbon Nanocomposites Upon Electrochemical Cycling M. Sina,†,‡ N. Pereira,†,§ G. G. Amatucci,†,§ and F. Cosandey*,† †
Department of Materials Science and Engineering, Rutgers University, Piscataway, New Jersey 08854, United States Energy Storage Research Group (ESRG), Rutgers University, North Brunswick, New Jersey 08902, United States
§
S Supporting Information *
ABSTRACT: High electrochemical performance iron oxyfluoride conversion electrode undergoes complex electrochemical reaction mechanisms upon cycling. In this work, a combination of selected area electron diffraction (SAED) and scanning transmission electron microscopy/electron energy loss spectroscopy (STEM/EELS) analysis techniques have been used to understand the conversion-reconversion mechanisms of FeO0.7F1.3/C upon cycling. Considerable changes have been observed with cycling. For the fully delithiated electrodes, the nanometer scale intermixing of amorphous rutile and nanocrystalline rocksalt phases is stable up to 20 cycles; however, upon further cycling the amount of amorphous rutile phase decreased and amount of rocksalt phase increased gradually, implying incomplete reconversion reactions with increasing cycle number. In addition, a progressive growth of solid electrolyte interphase (SEI) layer was observed with cycling, which is mainly composed of LiF. Interestingly, Fe2+ and Fe nanoparticles were found trapped in the SEI layer with increasing cycle number. Upon cycling, the combined progressive increase in Fe2+ content and insulating LiF (from SEI and conversion product) give rise to the observed capacity loss.
I. INTRODUCTION Transition metal fluoride/carbon nanocomposites have been under extensive investigation in the recent years due to their high theoretical capacity in the range of 500 to 800 mAh/g.1,2 These electrode materials possess high capacity because upon full discharge, the transition metals are reduced to their metallic state via a conversion reaction and, in the process, transfer all their valence electrons.1,3−7 In fluoride-based compounds with trivalent cations, the conversion and reconversion reactions are complex, and numerous studies have been performed in various systems such as FeF3,5,6,8−10 BiF3,11,12 and FeOF.5,6,13−15 Among the iron fluoride systems, FeOF nanocomposite shows the most promise as a high capacity cathode material because of its higher capacity retention upon cycling and lowest intrinsic charge−discharge voltage hysteresis of 0.7 V (measured by reverse step PITT at C/1000) as compared to 1.3 V for FeF3 and FeF2 .16 In addition, observations made by X-ray diffraction,6 pair distribution function (PDF),14 and electron diffraction15 reveal a reduction in crystallinity from the original rutile structure transforming upon delithiation into a complex multiphase microstructure consisting of a (Li−Fe−O-F) rocksalt and an amorphous phase with rutile type nearest neighbors.14,15 Although capacity fading has been reported upon electrochemical cycling, relatively little is known on the microstructural changes that are occurring upon cycling. In the course of initial TEM investigations, we observed that the metastable intermixed rocksalt and amorphous phases, formed upon delithiation, crystallize under the electron beam into the stable initial rutile structure.15 It is now well © XXXX American Chemical Society
documented that the electron beam can lead to irradiation effects such as heating, electrostatic charging, ionization (radiolysis), and knock-on atomic displacement.17 These irradiation effects are generally associated with material degradation such as decomposition, material loss by sputtering, and phase transformations.18 Knock-on displacement originates from electron elastic scattering with the nucleus of the atom, which can displace the atoms to interstitial positions or to sputtering of surface atoms. Knock-on displacement occurs when the incident electron energy is higher than the displacement energy which is dependent on the atomic number and bond strength.17,19,20 Electron irradiation, however, is not always detrimental, and under a small electron dose, it can promote a phase transition from nonequilibrium to equilibrium states, such as a transition from amorphous to crystalline states. These transitions occur whenever electron irradiation by knockon displacement provides sufficient atomic displacement to move the atoms into their stable crystalline configuration, which corresponds to a net decrease in Gibbs free energy.21−23 Such a transition could also be induced thermally via activated diffusion processes by annealing at high temperatures to more than half the melting point,24−28 while beam-induced crystallization occurs without any local beam heating.21,22,28,29 As diffusion rate is often controlled by vacancies, this type of radiation damage has been described in terms of vacancy Received: April 5, 2016 Revised: June 5, 2016
A
DOI: 10.1021/acs.jpcc.6b03485 J. Phys. Chem. C XXXX, XXX, XXX−XXX
Article
The Journal of Physical Chemistry C enhanced displacements.30 Several studies have shown that electron beam induced crystallization is a general phenomenon and can occur in a wide range of metastable materials such as for amorphous Si,19 ZrO2,31 MgAl2O4,32 FeF3,33 and metallic glasses.22,23,34 Saifullah et al.33 have found that under the influence of electron beam irradiation, amorphous FeF3 thin films rearrange their randomly distributed FeF6 octahedra to form crystalline FeF3. More recently, electron beam irradiation has been used to study the conversion reactions that are occurring during lithiation of amorphous MoOx thin films35 or to induce a spinel to rocksalt phase transformation in Li4Ti5O12 anodes.36 In the present work, we study the microstructural evolution occurring upon cycling of FeO0.7F1.3/C by combined scanning transmission electron microscopy (STEM), selected area electron diffraction (SAED), and electron energy loss spectroscopy (EELS) in order to understand capacity fading mechanisms. Additionally, we have used the formation of long-range order induced by electron beam irradiation to study the metastability of phases formed during delithiation of FeO0.7F1.3/C cathode materials. Using this methodology of electron beam induced crystallization, we have also followed the effects of electrochemical cycling on the formation and stability of metastable phases. The results of this study on FeOF are compared with results obtained on FeF237 and the beneficial roles of oxygen on cycling stability are presented.
dispersed in anhydrous DMC (BASF), and a small droplet of the dispersion was cast on a lacy carbon grid and then transferred into the TEM under controlled atmosphere with a vacuum transfer TEM holder (Gatan). SAED patterns and EELS spectra were obtained with a JEOL 2010F microscope at 197 kV equipped with a Gatan Image Filter (GIF-200) spectrometer. The energy resolution was 0.9 eV with EELS data obtained using a collection angle (β) of 27 mrad and convergence angle (α) of 10 mrad. The Fe valence state was determined using the Fe L3/L2 ratio methods and calibrated against known standards.5 The uniformly irradiated areas were obtained in TEM mode by over focusing the condenser lens. The SAED patterns and EELS spectra were each recorded as a function of time at a dose rate of 1.42 C/cm2/s (1 C/cm2 = 6.25 × 104 e/nm2). Total electron doses from 20 to 780 C/cm2 were used in this study. Diffraction patterns were quantified using Process Diffraction38 by first radially averaging the pattern followed by background removal and finally obtaining diffraction pattern intensity profiles.
III. RESULTS AND DISCUSSION III.1. Electrochemistry of FeO0.7F1.3/C Cathode Material upon Cycling. The representative voltage profiles of FeOF for four particular cycles (3, 20, 40, and 60 cycles) under a constant current of 50 mA/g at 60 °C are shown in Figure 1a. It
II. EXPERIMENTAL SECTION II.1. Materials Synthesis. The nanostructured iron oxyfluoride was synthesized by a solution process from reacting iron metal with fluorosilicic acid aqueous solutions (H2SiF6). In this process a mixture of dissolved iron metal in a 20−25 wt % fluorosilisic acid in water was heated at 40−45 °C for 12 h and then heated at 110 °C until the dry powder (FeSiF6·6H2O) is formed. After formation of FeSiF6·6H2O, the powder is annealed in the temperature range from 150 to 300 °C under a controlled atmosphere or in air to form FeOxF2−x. The oxygen content x can be controlled via oxygen partial pressure of the annealing gas. Further details can be found in the original paper on FeOxF2−x synthesis by Pereira et al.6 II.2. Electrochemistry. Electrodes were fabricated from nanocomposite powders with and without polymer binder. The FeO0.7F1.3/C nanocomposite powder was prepared by highenergy ball milling with 15 wt % activated carbon black (Asupra, Norit) for 1 h under helium atmosphere. The plastic tape electrodes were fabricated using the Bellcore developed process consisting of mixing the active materials with poly(vinylidene fluoride-cohexafluoropropylene) (Kynar 2801,Elf Atochem) binder, carbon black (Super P, MMM), and dibutyl phthalate (Aldrich) plasticizer. The resulting electrodes consisted of 57.5% active material, 12.5% carbon black and 30% binder, after extraction of the plasticizer. For electrochemical testing, coin cells were assembled in a He-filled drybox using glass fiber separators saturated with 1 M LiPF6 in ethylene carbonate/dimethyl carbonate (EC/DMC) electrolyte (BASF). The cells were tested versus lithium metal in galvanostatic mode in a Series 4000 Maccor cycler. The cells were cycled under a constant current of 50 mA/g between 1.5 and 4.5 V at 60 °C.6 II.3. TEM Analysis. The cycled cells were disassembled in a He glovebox and the FeO0.7F1.3/C electrode material was extracted. The electrodes with polymer binder were scratched and converted into powder form. The powder was then
Figure 1. (a) Voltage profiles after 3, 20, 40, and 60 cycles and (b) discharge capacity as a function of cycle numbers for FeO0.7F1.3/C cathode cycled at a rate of 50 mA/g at 60 °C.
is now well-documented5,6,14−16,39 that the voltage profile consists of two distinct reduction steps involving first an intercalation process (stage I) with a reduction of Fe3+ to Fe2+, followed by a second plateau (stage II) at approximately 2 V, corresponding to the partial conversion of Fe2+ to metallic Fe. The specific capacity for the 3rd and 20th cycle samples show B
DOI: 10.1021/acs.jpcc.6b03485 J. Phys. Chem. C XXXX, XXX, XXX−XXX
Article
The Journal of Physical Chemistry C similar voltage profiles with a capacity, as measured between 1.5 and 4.5 V, of about 360 mAh/g. However, the voltage drops significantly for the 60th cycle sample, with a capacity of only 80 mAh/g. Up to 20 cycles, the electrochemical result reveal a reduction plateau at approximately 2 V. Interestingly, the reduction plateau regions decreased and almost disappeared for the 40th and 60th cycle data, which is accompanied by a significant capacity drop. The discharge capacity of FeO0.7F1.3/ C as a function of cycle numbers is shown in Figure 1b. After an initial drop after the first cycle, the capacity remains constant at about 360 mAh/g of nanocomposite up to 20 cycles after which it drops to 80 mAh/g at 60 cycles. III.2. Phase Evolution in Delithiated FeO0.7F1.3/C upon Cycling. In order to understand the observed electrochemical changes upon cycling, the cycled samples were studied by a combination of electron diffraction and STEM/EELS analysis techniques. The SAED pattern of a delithiated sample after three cycles (Figure 2a) contains diffuse rings corresponding to the presence of amorphous rutile and rocksalt-type phases similar to what has been reported after the first cycle.15 However, more intense rings are observed for delithiated samples after 60 cycles (Figure 2b) associated with the presence of crystalline LiF and Fe phases. Despite the overlap between the reflections, the presence of crystalline LiF phase can be asserted from the appearance of 111 and 311 reflections. In addition, Fe nanoparticles, about 2 nm in size, are observed in the ADF-STEM images, as shown in Figure 2c. The corresponding SAED intensity profiles of delithiated FeO0.7F1.3/C after various cycle numbers up to 60 cycles are shown in Figure 3. From one cycle up to 20 cycles, the microstructure of the fully delithiated samples consist of a mixture of rocksalt-type and amorphous (rutile-type) phases. The IS and IR peak intensities are associated with the disordered rutile phase with short-range order. Upon cycling, the reflections marked I1 and I2 (rocksalt-type phase) became more intense, whereas the amorphous reflections (IR and IS) decreased in intensity upon cycling. The amorphous reflections (IR and IS) are almost nonexistent for 42 cycles and higher. Similarly, the individual sharp diffraction spots now appearing in the SAED shown in Figure 2b obtained after 60 cycles are indicative of increased crystallinity and particle size of the rocksalt-type phase, as well as from the existence of Fe nanoparticles, which are observed directly in the ADF-STEM image (Figure 2c). Furthermore, STEM-EELS analysis was used to investigate the chemical and valence state changes of the cycled cathode. The EELS spectra of the delithiated samples as a function of cycle numbers are shown in Figure 4. Only Fe-M edge can be observed up to 20 cycles. But after 20 cycles, the Li−K edge characteristics of LiF could already be detected in some areas, which grew in intensity with a further increase in cycle number. These changes in Li−K edge are accompanied by the gradual growth of F−K and O−K edges as well as with the gradual disappearance of O−K prepeak (marked by an arrow in Figure 4b). The progressive increase in LiF content from 20 cycles and higher can be attributed in part to unreacted LiF and to the development of LiF containing solid electrolyte interface (SEI) caused by decomposition of the electrolyte.12,37 It is worth noting that these structural changes do not occur uniformly throughout the electrode in which fully reconverted and partially reconverted areas have been observed as shown in Figure 5. The fully reconverted areas (area I) show a strong O−K prepeak as well as a small F−K prepeak (marked by an arrow in Figure 5a) characteristic of Fe in a high valence
Figure 2. SAED pattern of delithiated FeO0.7F1.3/C after (a) 3 cycles and (b) 60 cycles; (c) the corresponding ADF-STEM image of the delithiated sample after 60 cycles.
state (2.5 and higher).5 The measured Fe valence state from area I changes only slightly upon cycling, decreasing from Fe2.7+ to Fe2.5+ after 60 cycles. In contrast, a smaller O−K prepeak and higher F content are observed in areas II, which are observed starting at 20 cycles and higher. The average Fe valence state from Area II is also smaller decreasing from Fe1.8+ to Fe1.5+. This smaller Fe valence state is most likely formed by a mixture of Fe2+ in the rocksalt phase with Fe0 as metallic nanoparticles. The microstructure from area II is similar to the one represented in Figure 2c, revealing the presence of Fe nanoparticles, while the ADF-STEM images from area I do not show any contrast variations. This result would indicate that the content of unreacted Fe nanoparticles after delithiation increases with number of cycles. In some areas, regions with about 10−30 nm in thickness and lower contrast have been C
DOI: 10.1021/acs.jpcc.6b03485 J. Phys. Chem. C XXXX, XXX, XXX−XXX
Article
The Journal of Physical Chemistry C
Figure 3. SAED intensity profiles of delithiated FeO0.7F1.3/C as a function of number of cycles with also location of diffraction reflections for Fe, rocksalt, and rutile phases. The lines marked I1 and I2 are associated with the strongest reflections of the rocksalt phase, while IR and IS mark the diffuse reflections associated with amorphous rutile phase.
observed in our ADF-STEM images, as shown in Figure 6a. The corresponding EELS spectra depicted of Figures 6b and 6c indicate that this layer contains mainly Li and F with some O and C (not shown here), which differ from the electrode material fine structures. An extra peak at 708 eV, marked by an arrow, is observed within the F−K edge. Upon deconvolution of this F−K edge with F−K edge from a LiF standard, two peaks appear separated by 12.3 eV, as shown in Figure 6d. The energy position and separation between these peaks are indicative of Fe2+ and correspond to the Fe−L3 and Fe−L2 transitions.9,40 These EELS results are identical to our previous report on SEI evolution in FeF2 upon cycling, suggesting the dissolution of active materials in the electrolyte.37 However, these features appear only after a few cycles in FeF2 (after 10 cycles) while they start to appear later after more than 20 cycles in FeOF. Additional aspects of phase evolutions have been analyzed further upon probing relative phase stability by electron beam
Figure 5. (a) O−K, F−K and Fe−L edges EELS spectra of fully delithiated FeO0.7F1.3/C after 20 cycles taken from two different areas. An F−K edge pre-peak marked by an arrow is associated with Fe in high valence state, (b) Fe valence state evolution from these two areas as a function of cycle number.
induced phase transformation as described in the following chapter. III.3. Phase Metastability and Electron Beam Induced Phase Transformation. As mentioned in the introduction, electron beam irradiation can induce crystallization of metastable materials even at low electron dose.24−28 To
Figure 4. EELS spectra from the fully delithiated FeO0.7F1.3/C after various cycles (a) Fe−M edge and Li−K edge which is marked by an arrow and (b) O−K, F−K and Fe−L3,2 edges. In (b), the oxygen pre-peak, marked by an arrow, decreases in intensity as a function of cycle number. D
DOI: 10.1021/acs.jpcc.6b03485 J. Phys. Chem. C XXXX, XXX, XXX−XXX
Article
The Journal of Physical Chemistry C
Figure 6. (a) ADF-STEM image of delithiated FeOF after 20 cycles and characteristic EELS from area marked A with (b) Li6K edge, (c) O−K edge and F−K edges and (d) enlarged view of F−K with the F−K from LiF normalized with respect to the first peak and subtracted signal revealing the presence of two peaks separated by 12.3 eV and identified as Fe-L3 and Fe-L2 transitions.
simultaneous Li removal from the amorphous rutile and rocksalt-type phases, along with an increase in the Fe oxidation state from Fe2+ to Fe3+.15 At full delithiation, the Fe molar fraction, as determined by PDF, is 30% in the rocksalt phase, with the remaining 70% in the amorphous rutile phase.14 This result seems to indicate that Fe in a high valence state is required for the occurrence of electron beam crystallization to rutile phase and such a transformation is more pronounced with increasing fraction of Fe3+, that is, at the end of delithiation. Indeed, as the Fe 3+ content increases a corresponding amount of Li+ is removed from the structure, thus, increasing the quantity of cation vacancies, which should ease the process of knock on displacement. The effect of electron dose (780 C/cm2) on the microstructure of the fully delithiated FeO0.7F1.3/C is shown in Figure 7b. The intensity profile of the first cycle sample, which contains rocksalt and amorphous reflections (IR and IS) recrystallized to a rutile-type structure. This is observed from the appearance of the most intense (110) and (211) reflections of P42/mnm rutile structure. The recrystallization to rutile-type structure is observed up to 20 cycles, which corresponds to the cycled samples with limited capacity loss (cf. Figure 1). However, for the higher cycle numbers, the presence of stable and more intense cubic reflections (I1 and I2) associated with a coarsened rocksalt-type microstructure is more stable under the electron beam. Thus, capacity retention upon cycling occurs only whenever the intimate rocksalt and amorphous rutile phase mixture is maintained corresponding to the metastable rutile type structure.
determine the stable equilibrium structure of the delithiated samples for various Li contents and cycle numbers, we performed first a series of experiments at various electron doses from 20 to 780 C/cm2 (Figure S1). The electron dose of 780 C/cm2 has been found to induce full transformation to rutile structure while limiting the fluorine loss to less than 1% (Figure S2). The evolution of SAED intensity profiles of the delithiated FeO0.7F1.3/C samples ranging from 1.68 to 0.18 Li content after 1 cycle obtained after an electron dose at 780 C/ cm2 is shown in Figure 7a. The electron beam irradiation does not affect the intensity profiles of the delithiated samples with Li content from 1.68 to 0.69 Li. This is the Li concentration range where the microstructure consists of an intimate mixture of rocksalt, LiF and Fe nanoparticles. In the fully lithiated state, the molar fraction of Fe as nanoparticles is 0.5 as measured by electron diffraction15 and PDF analysis.14 In the Li concentration range from 0.69 to 0.18, the electron beam irradiation induces a gradual crystalline rutile structure. This phenomenon is visible from the appearance of the strongest (110) and (211) reflections of rutile at 0.329 and 0.172 nm, respectively. These results imply that delithiated samples with 0.44 Li and below contain a metastable amorphous-nanocrystalline microstructure, which transform back to the more stable crystalline rutile structure after electron beam irradiation. In addition, crystallization in the fully delithiated sample (0.18 Li) is fastest and requires a lower electron dose (270 C/cm2) compared to crystallization of the 0.44 Li or 0.32 Li samples, which require 550 and 780 C/cm2, respectively (Figure S1). In the second stage of delithiation (from 0.69 to 0.18 Li), Fe nanoparticles are no longer present and delithiation involves E
DOI: 10.1021/acs.jpcc.6b03485 J. Phys. Chem. C XXXX, XXX, XXX−XXX
Article
The Journal of Physical Chemistry C
and fluorine environment; however, the rutile structure, which is stabilized after electron beam induced atomic rearrangement, consists of Fe octahedral (Fe(OF)6) environment.42 This octahedral rearrangement from rocksalt to rutile structure requires both structural transformation as well as a volume change. Such an atomic rearrangement does not occur electrochemically and leads to the formation of an amorphous phase with only short-range order. An interesting point to note is that long-range diffusion of Fe is not necessary during rocksalt to rutile transformation as Fe atoms in the rocksalt and rutile unit cells, forms face center cubic (FCC) and body centered tetragonal (BCT) sublattices, respectively (see Figure S3). This FCC to BCT transformation can be obtained, without displacing the Fe atoms. In such a diffusionless transformation, known as the Bain transformation,43 only some lattice strain is required to accommodate a rocksalt to rutile transformation.
IV. DISCUSSION The results of this investigation show that capacity fade does occur in FeO0.7F1.3/C with a mechanism similar to FeF2 involving Fe dissolution and formation of excess LiF.37 However, FeO0.7F1.3/C, as opposed to FeF2, contains rocksalt-type phase throughout the lithiation−delithiation range, and it is most likely that the presence of this rocksalt phase that surrounds the Fe nanoparticles minimizes the dissolution rate and loss of Fe, resulting in an initial higher cycling performance. For the cycled FeO0.7F1.3/C, the presence of rocksalt phase reduces the possibility of Fe immediate contact with the electrode/electrolyte interface, thereby offering some protection against Fe nanoparticle-induced catalytic reactions with electrolyte. However, upon further cycling, some more insulating LiF inevitably forms, which prevent the delithiation to proceed. Additional protection with further enhancement in cycling performance can be envisioned with the addition of an ultrathin protecting oxide surface layer or coating. For instance, enhanced cycling stability has been reported for FeOF nanorods after ex situ coating with a thin conducting (PEDOT) polymer.44 The formation of a surface protective coating can be done also in situ by exposing FeO0.7F1.3/C nanocomposites to air, resulting in a microstructure consisting of FeO0.7F1.3/C nanoparticles encapsulated by a thin (1−2 nm) O-rich rocksalt layer.45 This rocksalt layer was found to be electrochemically inactive, as Fe did not change valence state upon cycling, but provides a protective layer, minimizing reaction with electrolyte. In situ formation of Fe2O3 was also explored for FeF3 conversion cathode material following air annealing at 500 °C.46 The resulting FeF3 with a thin 5 nm Fe2O3 oxide surface layer shows enhanced capacity retention upon cycling as compared to pristine FeF3. In a recent investigation on the role of oxygen on nanocomposite synthesis, we had already explored the cycling performance of FeF2 and FeOxF2−x nanocomposites prepared by mechanical milling in He or air environment, and the results are reproduced in Figure 9.39 The beneficial effects on cycling performance are more pronounced for FeOxF2−x than for FeF2 with no fading (albeit with lower initial capacity) for FeOxF2−x annealed in air after ball milling in air. The microstructural details are not known yet, but the results of this investigation point toward the importance of surface modification for enhancing cycling performance of FeOxF2 conversion materials by protecting the active material from the electrolyte and
Figure 7. (a) SAED intensity profiles of delithiated FeO0.7F1.3/C electrode for various Li content after an electron dose 780 C/cm2. Electron irradiation induced amorphous to crystalline transition occurs for Li content below 0.69 Li and (b) SAED intensity profiles of cycled delithiated FeO0.7F1.3/C as a function of number of cycles after electron irradiation with an electron dose of 780 C/cm2.
An interesting point to note is the similarity between the 60 cycles SAED intensity profile (Figure 7b) with the fully lithiated 1.68 Li (Figure 7a). This is an indication that delithiation is no longer effectively occurring and the electrode remains in effect in a discharged state with a microstructure consisting mostly of Fe and LiF. This is partly the result of the destruction of the fine electronically conduction percolated network (and increase distance between Fe nanoparticles) caused by Fe loss via dissolution in the electrolyte and higher content of insulating LiF layer (cf. Figure 4) that impedes electronic and ion mobility. An important point to note here is that the reconversion processes associated with a transformation from cubic rocksalt to rutile phase requires only short-range structural rearrangement of octahedral FeO6 with an expansion of the unit cell33,41 of about 8%. Such short-range atomic rearrangements can be achieved easily via the electron beam irradiation.33,41 Figure 8 shows the atomic arrangements of rocksalt and rutile structures, where in both structures Fe occupies octahedral interstitial sites surrounded with six anions (O and F). In addition, Fe atom sublattice is face-center-cubic (FCC) and body-center-tetragonal (BCT) in rocksalt and rutile structures, respectively (Figure S3). In the lithiated phase with disordered-rocksalt structure, Li and Fe reside randomly in octahedral sites with an FCC oxygen F
DOI: 10.1021/acs.jpcc.6b03485 J. Phys. Chem. C XXXX, XXX, XXX−XXX
Article
The Journal of Physical Chemistry C
Figure 8. (a, b) Rocksalt crystal unit cell and (c, d) rutile crystal structure, showing the octahedral arrangements (Fe(OF)6) in rocksalt and rutile unit cells.
addition, a Fe−L edge has been detected in the SEI EELS spectrum suggesting the loss of active material. The continuous increases of Fe dissolution and LiF retention upon cycling give rise to capacity fade. On the other hand, the presence of a rocksalt phase in the conversion reaction of FeO0.7F1.3/C tends to mitigate these deleterious effects and therefore, results in increased cycling stability as compared to FeF2.
■
ASSOCIATED CONTENT
S Supporting Information *
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcc.6b03485. Additional figures: S1, Phase evolution as a function of electron dose; S2, F/Fe concentration as a function of electron dose; S3, schematic diagram of Bain FCC to BT transformation (PDF).
Figure 9. Comparison of discharged capacity as a function of cycle number for FeF2 and FeOxF2−x nanocomposites prepared under various oxygen containing environments with curve 1, FeF2 milled in He; curve 2, FeF2 milled in air; curve 3, FeOxF2−x milled in He; curve 4, FeOxF2−x milled in air; curve 5, FeOxF2−x milled in He, followed by annealing in air; and curve 6, FeOxF2−x milled in air followed by annealing in air. (Data taken from Pereira et al.39)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Tel.: 848 445 4942.
minimizing nanoparticle induced catalytic reactions as well as loss of active material via Fe2+ dissolution.
Present Address ‡
Department of NanoEngineering University of California, San Diego La Jolla, CA 92093−0448, U.S.A.
V. SUMMARY AND CONCLUSIONS In this study, the degradation mechanism of FeO0.7F1.3/C nanocomposite electrode upon electrochemical cycling has been investigated. Our TEM/STEM-EELS analyses revealed that in the delithiated samples, the rocksalt-type phase became more crystalline and the amount of LiF increased with progressive cycling number. In addition, the amount of amorphous rutile-type phase in the conversion reaction was found to decrease. The ADF-STEM images and EELS spectra from the cycled samples indicate the growth of a SEI layer which consists mostly of LiF phase and carbon species. In
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported as part of the NorthEast Center for Chemical Energy Storage (NECCES), an Energy Frontier Research Center funded by the U.S. Department of Energy (DOE), Office of Science, Basic Energy Sciences, under Awards DE-SC0001294 and DE-SC0012583 G
DOI: 10.1021/acs.jpcc.6b03485 J. Phys. Chem. C XXXX, XXX, XXX−XXX
Article
The Journal of Physical Chemistry C
■
(17) Egerton, R. F.; Li, P.; Malac, M. Radiation damage in the TEM and SEM. Micron 2004, 35, 399−409. (18) Egerton, R. F. Mechanisms of radiation damage in beamsensitive specimens, for acceleration voltages between 10 and 300 kV. Microsc. Res. Tech. 2012, 75, 1550−1556. (19) Libera, M. Local amorphous thin-film crystallization induced by focused electron-beam irradiation. Appl. Phys. Lett. 1996, 68, 331−333. (20) Nagase, T.; Umakoshi, Y. Effect of irradiation temperature on the electron irradiation induced nanocrystallization behavior in Fe88.0Zr9.0B3.0 amorphous alloy. Mater. Sci. Eng., A 2003, 347, 136− 144. (21) Qin, W.; Szpunar, J. A.; Umakoshi, Y. Electron or ion irradiation-induced phase-change mechanism between amorphous and crystalline state. Acta Mater. 2011, 59, 2221−2228. (22) Nino, A.; Nagase, T.; Umakoshi, Y. Electron irradiation induced crystallization and amorphization in Fe77Nd4.5B18.5 metallic glass. Mater. Sci. Eng., A 2007, 449-451, 1115−1118. (23) Liu, X. J.; Chen, G. L.; Hou, H. Y.; Hui, X.; Yao, K. F.; Lu, Z. P.; Liu, C. T. Atomistic mechanism for nanocrystallization of metallic glasses. Acta Mater. 2008, 56, 2760−2769. (24) Lu, K. Nanocrystalline metals crystallized from amorphous solids. Nanocrystallization, structure, and properties. Mater. Sci. Eng., R 1996, 16, 161−221. (25) Nagase, T.; Hosokawa, T.; Umakoshi, Y. Electron irradiationinduced phase transition of an amorphous phase and face-centered cubic solid solutions in Zr66.7Pd33.3 metallic glass. Metall. Mater. Trans. A 2007, 38, 223−235. (26) Motta, A. T. Amorphization of intermetallic compounds under irradiation - A review. J. Nucl. Mater. 1997, 244, 227−250. (27) Nagase, T.; Umakoshi, Y. Electron irradiation induced crystallization of the amorphous phase in Zr-Cu based metallic glasses with various thermal stability. Mater. Trans. 2004, 45, 13−23. (28) Nagase, T.; Umakoshi, Y. Electron irradiation induced crystallization behavior in Zr66.7Cu33.3 and Zr65.0Al7.5Cu27.5 amorphous alloys. Mater. Sci. Eng., A 2003, 352, 251−260. (29) Meldrum, A.; Boatner, L. A.; Ewing, R. C. Electron-irradiationinduced nucleation and growth in amorphous LaPO4, ScPO4, and zircon. J. Mater. Res. 1997, 12, 1816−1827. (30) Muller, D. D.; Silcox, J. Radiation damage of Ni3Al by 100 keV electrons. Philos. Mag. A 1995, 71, 1375−1387. (31) Roddatis, V. V.; Su, D. S.; Jentoft, F. C.; Schlogl, R. Temperature- and electron-beam-induced crystallization of zirconia thin films deposited from an aqueous medium: a transmission electron microscopy study. Philos. Mag. A 2002, 82, 2825−2839. (32) Ning, Y.; Li, C. Structural inhomogeneity and crystallization behavor of aerosol-reacted MgAl2O4 powders. Mater. Lett. 1992, 15, 84−88. (33) Saifullah, M. S. M.; Botton, G. A.; Boothroyd, C. B.; Humphreys, C. J. Electron energy loss spectroscopy studies of the amorphous to crystalline transition in FeF3. J. Appl. Phys. 1999, 86, 2499−2504. (34) Nagase, T.; Sanda, T.; Nino, A.; Qin, W.; Yasuda, H.; Mori, H.; Umakoshi, Y.; Szpunar, J. A. MeV electron irradiation induced crystallization in metallic glasses: Atomic structure, crystallization mechanism and stability of an amorphous phase under the irradiation. J. Non-Cryst. Solids 2012, 358, 502−518. (35) Chen, W.; Zhang, H.; Wang, Y.; Ma, Z.; Li, Z. In-situ microstructural investigation by electron-beam irradiation indiced crystallization of amorphous MoOx thin films with high performance for Li-ion storage. Electrochim. Acta 2014, 144, 369−375. (36) Su, D.; Wang, F.; Ma, C.; Jiang, N. Engineering nano-composite Li4Ti5O12 anodes via scanning electron-probe fabrication. Nano Energy 2013, 2, 343−350. (37) Sina, M.; Thorpe, R.; Rangan, S.; Pereira, N.; Bartynski, R. A.; Amatucci, G. G.; Cosandey, F. Investigation of SEI Layer Formation in Conversion Iron Fluoride Cathodes by Combined STEM/EELS and XPS. J. Phys. Chem. C 2015, 119, 9762−9773.
REFERENCES
(1) Amatucci, G. G.; Pereira, N. Fluoride based electrode materials for advanced energy storage devices. J. Fluorine Chem. 2007, 128, 243− 262. (2) Cabana, J.; Monconduit, L.; Larcher, D.; Palacin, M. R. Beyond Intercalation-Based Li-Ion Batteries: the State of the Art and Challenges of Electrode Materials Reacting Through Conversion Reactions. Adv. Mater. 2010, 22, E170−E192. (3) Badway, F.; Bervas, M.; Pereira, N.; Mansour, A. N.; Yoon, W. S.; Al-Sharab, J.; Plitz, I.; Tong, W.; Cosandey, F.; Amatucci, G. G. Reversible conversion metal fluoride nanocomposites: new options for next generation high energy lithium batteries. Proc. Power Sources Conf. 2006, 42nd, 277−280. (4) Badway, F.; Mansour, A. N.; Pereira, N.; Al-Sharab, J. F.; Cosandey, F.; Plitz, I.; Amatucci, G. G. Structure and Electrochemistry of Copper Fluoride Nanocomposites Utilizing Mixed Conducting Matrices. Chem. Mater. 2007, 19, 4129−4141. (5) Cosandey, F.; Su, D.; Sina, M.; Pereira, N.; Amatucci, G. G. Fe valence determination and Li elemental distribution in lithiated FeO0.7F1.3/C nanocomposite battery materials by electron energy loss spectroscopy (EELS). Micron 2012, 43, 22−29. (6) Pereira, N.; Badway, F.; Wartelsky, M.; Gunn, S.; Amatucci, G. G. Iron Oxyfluorides as High Capacity Cathode Materials for Lithium Batteries. J. Electrochem. Soc. 2009, 156, A407−A416. (7) Wang, F.; Robert, R.; Chernova, N. A.; Pereira, N.; Omenya, F.; Badway, F.; Hua, X.; Ruotolo, M.; Zhang, R.-G.; Wu, L.-J.; Volkov, V.; Su, D.; Key, B.; Whittingham, M. S.; Grey, C. P.; Amatucci, G. G.; Zhu, Y.-M.; Graetz, J. Conversion Reaction Mechanisms in Lithium Ion Batteries: Study of the Binary Metal Fluoride Electrodes. J. Am. Chem. Soc. 2011, 133, 18828−18836. (8) Badway, F.; Pereira, N.; Cosandey, F.; Amatucci, G. G. Carbonmetal fluoride nanocomposites - Structure and electrochemistry of FeF3: C. J. Electrochem. Soc. 2003, 150, A1209−A1218. (9) Cosandey, F.; Al-Sharab, J. F.; Badway, F.; Amatucci, G. G.; Stadelmann, P. EELS spectroscopy of iron fluorides and FeFx/C nanocomposite electrodes used in Li-ion batteries. Microsc. Microanal. 2007, 13, 87−95. (10) Yamakawa, N.; Jiang, M.; Key, B.; Grey, C. P. Identifying the Local Structures Formed during Lithiation of the Conversion Material, Iron Fluoride, in a Li Ion Battery: A Solid-State NMR, X-ray Diffraction, and Pair Distribution Function Analysis Study. J. Am. Chem. Soc. 2009, 131, 10525−10536. (11) Bervas, M.; Mansour, A. N.; Yoon, W. S.; Al-Sharab, J. F.; Badway, F.; Cosandey, F.; Klein, L. C.; Amatucci, G. G. Investigation of the lithiation and delithiation conversion mechanisms of bismuth fluoride nanocomposites. J. Electrochem. Soc. 2006, 153, A799−A808. (12) Gmitter, A. J.; Badway, F.; Rangan, S.; Bartynski, R. A.; Halajko, A.; Pereira, N.; Amatucci, G. G. Formation, dynamics, and implication of solid electrolyte interphase in high voltage reversible conversion fluoride nanocomposites. J. Mater. Chem. 2010, 20, 4149−4161. (13) Chevrier, V. L.; Hautier, G.; Ong, S. P.; Doe, R. E.; Ceder, G. First-principles study of iron oxyfluorides and lithiation of FeOF. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 87, 094118. (14) Wiaderek, K. M.; Borkiewicz, O. J.; Castillo-Martinez, E.; Robert, R.; Pereira, N.; Amatucci, G. G.; Grey, C. P.; Chupas, P. J.; Chapman, K. W. Comprehensive Insights into the Structural and Chemical Changes in Mixed-Anion FeOF Electrodes by Using Operando PDF and NMR Spectroscopy. J. Am. Chem. Soc. 2013, 135, 4070−4078. (15) Sina, M.; Nam, K.-W.; Su, D.; Pereira, N.; Yang, X.-Q.; Amatucci, G. G.; Cosandey, F. Structural phase transformation and Fe valence evolution in FeOxF2−x/C nanocomposite electrodes during lithiation and de-lithiation processes. J. Mater. Chem. A 2013, 1, 11629−11640. (16) Ko, J. K.; Wiaderek, K. M.; Pereira, N.; Kinnibrugh, T. L.; Kim, J. R.; Chupas, P. J.; Chapman, K. W.; Amatucci, G. G. Transport, Phase Reactions, and Hysteresis of Iron Fluoride and Oxyfluoride Conversion Electrode Materials for Lithium Batteries. ACS Appl. Mater. Interfaces 2014, 6, 10858−10869. H
DOI: 10.1021/acs.jpcc.6b03485 J. Phys. Chem. C XXXX, XXX, XXX−XXX
Article
The Journal of Physical Chemistry C (38) Labar, J. L. Consistent indexing of a (set of) SAED pattern(s) with the ProcessDiffraction program. Ultramicroscopy 2005, 103, 237− 249. (39) Pereira, N.; Wartelsky, M.; Karavadi, R.; Cosandey, F.; Amatucci, G. G. Impact of Oxygen in Iron Fluoride Cathode Materials for Lithium Batteries. 216th ECS Meeting, Vienna, Austria, 2009. (40) Colliex, C.; Manoubi, T.; Ortiz, C. Electron-energy-lossspectroscopy near-edge fine stuctures in the iron-oxygen system. Phys. Rev. B: Condens. Matter Mater. Phys. 1991, 44, 11403−11411. (41) Vijayakumar, M.; Kerisit, S.; Wang, C.; Nie, Z.; Rosso, K. M.; Yang, Z.; Graff, G.; Liu, J.; Hu, J. Effect of Chemical Lithium Insertion into Rutile TiO2 Nanorods. J. Phys. Chem. C 2009, 113, 14567−14574. (42) Chevrier, V. L.; Hautier, G.; Ong, S. P.; Doe, R. E.; Ceder, G. First-principles study of iron oxyfluorides and lithiation of FeOF. Phys. Rev. B: Condens. Matter Mater. Phys. 2013, 87, 094118/094111− 094118/094119. (43) Roldan Cuenya, B.; Doi, M.; Lobus, S.; Courths, R.; Keune, W. Observation of the fcc.-to-bcc. Bain transformation in epitaxial Fe ultrathin films on Cu3Au(001). Surf. Sci. 2001, 493, 338−360. (44) Fan, X.; Luo, C.; Lamb, J.; Zhu, Y.; Xu, K.; Wang, C. PEDOT encapsulated FeOF nanorod cathodes for high energy lithium-ion batteries. Nano Lett. 2015, 15, 7650−7656. (45) Kim, S.-W.; Pereira, N.; Chernova, N. A.; Omenya, F.; Gao, P.; Whittingham, M. S.; Amatucci, G. G.; Su, D.; Wang, F. Structure Stabilization by mixed anions in oxyfluoride cathodes for high-energy lithium batteries. ACS Nano 2015, 9, 10076−10084. (46) Zhang, W.; Ma, L.; Yue, H.; Yang, Y. Synthesis and characterization of in situ Fe2O3-coated FeF3 cathode materials for rechargeable lithium batteries. J. Mater. Chem. 2012, 22, 24769− 24775.
I
DOI: 10.1021/acs.jpcc.6b03485 J. Phys. Chem. C XXXX, XXX, XXX−XXX
