In situ techniques with high temporal, spatial and chemical resolution are key to understand ubiquitous solid-state phase transformations, which are crucial to many technological applications. Hard X-ray spectro-imaging can visualize electrochemically driven phase transformations but demands considerably large samples with strong absorption signal so far. Here we show a conceptually new data analysis method to enable operando visualization of mechanistically relevant weakly absorbing samples at the nanoscale and study electrochemical reaction dynamics of iron fluoride, a promising high-capacity conversion cathode material. In two specially designed samples with distinctive microstructure and porosity, we observe homogeneous phase transformations during both discharge and charge, faster and more complete Li-storage occurring in porous polycrystalline iron fluoride, and further, incomplete charge reaction following a pathway different from conventional belief. These mechanistic insights provide guidelines for designing better conversion cathode materials to realize the promise of high-capacity lithium-ion batteries.
Solid-state phase transformations are ubiquitous and important to numerous applications1. They are not only the primary way to make technologically important materials (such as martensitic steel, inorganic ceramics and thin-film absorbers for solar cells), but also lie at the heart of electrochemical energy storage, in which the insertion and extraction of charge storage ions (such as Li+ and Na+) are always accompanied by electrochemically driven solid-state phase transformations2. There has been significant interest in monitoring and probing these transformation processes over the last few decades in the hope of gaining mechanistic understanding to guide further optimization and bring technological benefits. The most commonly used techniques to analyse characteristics of phase transformations include in situ X-ray diffraction3, neutron diffraction4, X-ray absorption spectroscopy (XAS)5,6 and solid-state nuclear magnetic resonance6. These techniques enable valuable insights into changes in long-range structure, local bonding and chemical environment, elemental composition and oxidation state. However, they only reveal average information over a large sample volume (micrometre to millimetre scale). Recently, the remarkable advances in in situ transmission electron microscopy (TEM) have made it possible to probe phase transformation down to nanoscale7,8,9,10. When coupled with electron diffraction (ED) and/or electron energy-loss spectroscopy, in situ TEM serves as a perfect diagnostic tool to track nanoscale morphological and structural changes as well as new phase nucleation and propagation with nanoscale spatial resolution. However, TEM-based experiments require restrictively thin samples (100 nm or below) and must be compatible with the high vacuum environment inside the microscope. These limitations complicate the experimental design and the straightforward correlation of the observed phenomena to what actually occur under realistic conditions. Furthermore, TEM-based techniques are inherently incapable of probing the bulk of a working device, such as a battery or a fuel cell electrode, which is typically made of multiple components assembled at micro-length or larger length scales.
Hard X-ray spectro-imaging provides an innovated approach to visualize electrochemically driven solid-state phase transformations at the nanoscale11. Taking advantage of the strong and deeply penetrating hard X-rays generated by synchrotron radiation and the chemical and elemental sensitivity with a full-field imaging capability provided by the transmission X-ray microscopy (TXM) coupled with X-ray absorption near-edge structure spectroscopy (XANES)12,13, progression of a electrochemical reaction in a realistic battery electrode can be visualized in a large (tens of micrometres) field-of-view (FOV) with nanoscale spatial resolution. Unlike in situ TEM experiments that often have to be carried out using specialized sample holders in electrochemical conditions drastically different from those in a practical battery7,8,9, hard X-ray-based experiments can be conveniently performed in relevant and realistic conditions using a much simpler cell design14,15,16,17. Proof-of-concept operando hard X-ray spectro-imaging experiments have been recently demonstrated by using large microparticles18,19,20,21 or aggregates of small particles20 (several micrometres in total size) with strong X-ray absorption signal, in which cases chemical phase maps could be generated using simple data processing methods by approximating the X-ray absorption of materials under study to the experimentally determined total X-ray absorption. However, those approaches ignored background absorption and thus cannot accurately determine actual X-ray absorption of smaller and weakly absorbing samples at single-particle level under operando conditions. To study small particles, which, in fact, are more electrochemically active and mechanistically relevant, we herein propose and develop a conceptually new data analysis method to reliably determine X-ray absorption of study materials and for the first time realized operando studies of phase evolution in a high-capacity Li-ion battery conversion cathode with nanoscale chemical resolution.
Here the conversion reaction of iron fluoride (FeF3) was chosen as a demonstration example for operando mechanistic studies. FeF3 is a promising prototype conversion cathode material with extremely high Li-storage capacity (712 mAh g−1)22,23,24,25,26,27,28,29, four or five times higher than current intercalation cathode materials such as LiCoO2 (∼140 mAh g−1) and LiFePO4 (∼170 mAh g−1). This is achieved by utilizing all of the oxidation states of Fe through an electrochemical conversion reaction to enable multiple electron transfer and Li-ion storage.
Consequently, batteries based on a FeF3 conversion cathode (theoretical voltage ∼2.7 V)23 and a suitable lithium-containing anode, such as a protected lithium metal anode30, hold the promise to significantly increase the energy density of current Li-ion battery technology22,31. However, such promise has yet to be realized because of challenges associated with the significant phase transformation and structural rearrangement during cycling. Two prominent ones are capacity decay and a large voltage hysteresis24,26,32,33,34. Solving these challenges requires a better understanding of the electrochemical reaction mechanism under operando conditions, especially the recharge reaction, which has been surprisingly under-researched compared with the discharge reaction35,36. Here we use the improved operando hard X-ray spectro-imaging to track the phase evolution of FeF3 cathodes during cycling and reveal how electrochemical reactions progress kinetically and spatially, which provides insights essential to rationally designing electrode microstructure to achieve fast kinetics and high reversibility. We also discover evidences across different length scales that suggest a charge reaction pathway different from the traditional belief, which advances the understanding on the causes of capacity decay and voltage hysteresis.
Improved experimental setup and new data analysis method
In the operando experiments (Fig. 1a), synchrotron monochromatic X-rays are directed to transmit through a perforated 2032-type coin-cell (Fig. 1b,c) containing the FeF3 cathode and all the other key components of a realistic battery, such as carbon black, polymeric binder and a liquid electrolyte (Fig. 1d). The holes in the coin-cell, which are sealed using Kapton tape (Fig. 1c,d), need to be small to ensure a small cell impedance. The resulted absorption-contrast images are projected onto a lens-coupled full-field CCD detector using a zone-plate and recorded.
Figure 2 illustrates general procedures to realize chemical phase mapping; a detailed comparison between previous methods (Fig. 2a–d) and our new method (Fig. 2a, b, e, f) are discussed later in the manuscript. First, a series of images are collected by scanning the energy across the Fe K-edge (7,112 eV) in a step size of 2 eV, one image at each energy (Fig. 2a). Then, the XANES spectra at each pixel are constructed by plotting normalized X-ray absorption versus energy (Fig 2b,c, or b,e). They are fitted to reference spectra collected from standard compounds (Fe3+F3, Fe2+F2 and Fe) to determine ratio between different Fe-containing phases so that red-green-blue (RGB) colours can be accordingly assigned to generate a phase map (Fig 2d or f). As Fe of various oxidation states interact with X-ray differently, their spatial distribution at different states of discharge/charge reveals progression of the electrochemical reaction in the FeF3 cathode.
We have developed a conceptually new data analysis method to enable operando spectro-imaging of small samples with weak absorption signal (see Methods and more details available in Supplementary Methods). In operando experiments, the X-rays are not just absorbed by the Fe-containing-active material, but also attenuated by all other battery components in the pathway of the beam, such as the carbon black, polymeric binder, separator and liquid electrolyte, which may be considered as an internal background as a whole. Previous methods depend heavily upon strong X-ray absorption of large samples (tens of micrometres in size) and consider that the X-ray absorption of materials under study is approximately equal to the experimentally determined total X-ray absorption18,19,20,21, in which background absorption is omitted. This approximation is no longer valid for smaller and weakly absorbing samples, such as the porous microwires (MWs) examined here (effective thickness <1 μm when porosity is considered). It results in improperly normalized XANES spectra (Fig. 2c) and consequently unsatisfactory chemical phase maps (Fig. 2d and see an example of complete comparison in Supplementary Fig. 1). We propose and show that the X-ray absorption of the Fe-containing-active material can be determined more accurately by subtracting an internal background spectrum from the total X-ray attenuation over the whole imaging area of 512 × 512 pixels. We use the X-ray attenuating information readily available from an area in the same FOV, where the X-rays only pass through the other battery components but not the Fe-containing-active material (Fig. 2a, black box), to more accurately represent the internal background absorption spectrum (Fig. 2b, black circles). We then subtract such background from the total X-ray absorption spectrum (Fig. 2b, red circles) at each pixel to determine the actual X-ray absorption of the study material. We also note that background absorption can be mathematically fitted to allow high-quality normalization within a single XANES spectrum37,38, but such methods are extremely calculation-intensive to implement when dealing with a considerably large number of spectra (512 × 512 spectra) in spectro-imaging. After internal background removal using our new method, the XANES spectrum (Fig. 2e) at each pixel can be correctly normalized and fitted to a linear combination of standard reference spectra (Supplementary Fig. 2), enabling high-quality chemical phase mapping under operando battery conditions (Fig. 2f). Note that even though the intrinsic spatial resolution is dictated by the current instrumentation (∼25 nm for camera binning 1)13, the new data analysis method helps determine the X-ray absorption of weakly absorbing samples more accurately at single-particle level so that many more image pixels that contain meaningful chemical information are preserved. These improvements enable the chemical phase mapping of samples that are smaller in size (hundreds of nanometres) than those reported before (several micrometres in total size)18,19,20,21.
FeF3 model samples and in situ electrochemical cell
Our operando experiments also benefit from two specially designed FeF3 samples with well-defined morphologies of polyhedron (Fig. 3a) and MWs (Fig. 3b). These two samples are synthesized for the first time by rationally controlling supersaturation26,39,40,41 (see Methods for synthetic details) and are quite different in microstructure, porosity and electrochemical activity. The FeF3 polyhedra are single-crystalline (as proven by the ED pattern in Fig. 3a inset), non-porous (Supplementary Fig. 3) and can only reach approximately one-third of the theoretical Li-storage capacity in the conventional battery test (Supplementary Fig. 5), whereas the MWs are polycrystalline (Fig. 3b inset), mesoporous, grain-boundary-rich (Supplementary Fig. 4) and almost reach full capacity (Supplementary Fig. 5). We made an electrode containing both the polyhedra and MWs to enable comparative study and reveal the relation between structure and electrochemical properties. Porous carbon paper (∼110 μm in thickness) was used as the current collector for this electrode. It is quite transparent to hard X-rays but still robust enough for handling, which is critical to the operando spectro-imaging experiment. More details on the electrode and cell preparation can be found in Supplementary Methods. The operando cell was discharged at a constant current of ∼1/15 C (∼47.5 mA g−1) to 1.5 V (Fig. 3c). TXM images were recorded in dynamic conditions at different states of discharge/charge in two different locations. The data were processed by our new approach and fitted to the standard reference spectra (Fig. 3d) to generate the chemical phase maps shown in Fig. 3e (FOV 1) and 3f (FOV 2). Similar operando studies were also carried out in an electrode containing only FeF3 MWs cycled in potentiostatic mode (Supplementary Fig. 6).
Operando visualization of FeF3 electrochemical reaction
Enabled by the technological advances in spectro-imaging, we first visualized the progression of electrochemical discharge reaction in the FeF3 conversion cathode. Two consecutive phase transformations consistent with the sequential lithiation reaction of FeF3 (refs 26, 32, 33) were observed in both the polyhedra and MWs (Fig 3e,f). First, from the open-circuit voltage 3.24 to 1.62 V, red pixels were gradually replaced by green pixels (Fig 3e,f, map 1 to 3), which is related to the initial Li+ intercalation into FeF3 with Fe (+III, red colour) being reduced to Fe (+II, green colour). Then, in the sloping voltage plateau between 1.62 and 1.5 V, blue pixels appeared at the expense of the green pixels (Fig 3e,f, map 4 to 5), indicating the formation of metallic Fe (blue colour) through the conversion reaction. However, although the first reduction could proceed to completion in both FeF3 samples, the progression of the second reduction is clearly quite different. The conversion from Fe (+II) to metallic Fe was incomplete for the polyhedra (Fig. 3e, map 5). Compared with the reference spectra (Fig. 3d), selected-area XANES spectra taken on the polyhedron reveals that Fe (+II) and metallic Fe co-existed in the end (Fig. 3g, top panel). The contrast in electrochemical activity is further highlighted in Fig. 3f, in which one MW and one polyhedron are situated side by side. The MW became mostly metallic Fe (mostly blue), whereas the polyhedra did not react completely (still a lot of green) at the end of discharge (Fig. 3f, map 5). The change in XANES spectra taken on the MWs (Fig. 3g) is consistent with the successive reduction of Fe (+III) to Fe (+II) and then to metallic Fe. Quantitative comparison of the reaction progress based on spectrum fitting (Fig. 3h) also confirms that the MWs react faster and more completely than the polyhedra. These observations not only explain the difference in discharge capacity in our conventional battery tests (Supplementary Fig. 5) but also reveal the importance of porous and grain-boundary-rich structure to promote the complete three-Li+ storage for FeF3 conversion cathodes. During discharge, electron transport may become less of a concern once metallic iron precipitates out and starts propagating to form a conductive network within the solid8,26,35,42. Therefore, the key to achieving high capacity likely depends on efficient Li+ transport to trigger the Fe and LiF precipitation after the structure is saturated with Li+ (ref. 8). This process could be facilitated by the porous and grain-boundary-rich structure of the MWs.
Furthermore, we were able to visualize the spatial dynamics of the electrochemical reaction over a large area (tens of micrometres) thanks to the unique capability of TXM-XANES in spatially resolving chemical identification in a large FOV. As shown by the concurrent colour change in the ensemble of polyhedra (Fig. 3e) and MWs (Supplementary Fig. 6), the electrochemically driven phase transformations in the FeF3 cathode are relatively homogenous, indicating the absence of preferential reaction sites during discharge. All particles were actively discharging, although some local regions on individual particles appeared to react more slowly. Because these experiments were carried out on both single-crystalline and polycrystalline FeF3 samples under operando conditions, we believe this observation reveals the inherent reaction characteristics of FeF3 conversion cathodes. Interestingly, this behaviour is different from that of the LiFePO4 intercalation cathode material, in which significant inhomogeneities (that is, active particle fraction<100%) were observed and believed to be a major limiting factor for further improving high-rate performance43,44. Therefore, this operando visualization of FeF3 (active particle fraction=100%) suggests that it could be feasible for conversion cathode material to achieve better rate capability than what was traditionally believed22,24, likely via further nanostructure engineering26.
Our operando experiments further enabled the investigation of the charge reaction of the FeF3 cathode. This unique charge (delithiation) study has not been possible so far using in situ TEM techniques due to the instability of organic electrolyte under electron beam and/or the difficulty of applying a controlled constant current. The operando cell was charged at the same rate of ∼1/15 C to 4.5 V (Fig. 4a) and chemical phase maps at four different states of charge are shown in Fig. 4b. The electrochemical reaction occurred uniformly everywhere in the sample, similar to what was observed during discharge. The overall phase transformation during charge is in agreement with metallic Fe gradually converting into Fe (+II) as the voltage increases, but not reaching Fe (+III). This can be clearly seen from the chemical phase maps as well as their corresponding XANES spectra (Fig. 4c) taken from a selected area of the MW. The relative mole fraction of different Fe oxidation states during both discharge and charge was determined by spectrum fitting and shown in Fig. 4d. When the charge process finished at the cutoff voltage of 4.5 V, the dominant oxidation state was Fe (+II) (∼70%, Fig. 4d), which is consistent with what was suggested by the previous ex situ TEM-electron energy-loss spectroscopy studies on the recharged FeF3/C electrodes45. There was also some metallic Fe that did not react during recharge (∼30%, Fig. 4d). The operando studies in voltammetric mode showed the same phase transformation behaviour (Supplementary Fig. 6).
Phase transformations over a large area of the electrode
To confirm what we observed locally in the two FOVs represents the global changes that occurred in the whole electrode, we collected additional data from a few other areas at the end of discharge and charge. The resulting chemical phase maps give consistent pictures (Supplementary Fig. 7). Furthermore, we carried out an operando XAS experiment on a FeF3 MW electrode cycled at ∼1/12 C rate. The larger size X-ray beam (∼0.5−1.0 mm) in the spectroscopy experiment allows the tracking of phase transformation over a much larger area of the electrode. The observed changes in XAS spectra collected at different states of discharge (Fig. 5a) and charge (Fig. 5b) during continuous cycling (Fig. 5c) are in agreement with what we observed locally in a single MW (Figs 3g and 4c) and the corresponding chemical phase maps (Figs 3f and 4b), indicating that the electrochemical reaction is relatively homogeneous over a large area of the electrode. We further performed spectrum fitting to determine the ratio between different Fe oxidation states during cycling (Fig. 5d). Fitting details and representative best fits are shown in Supplementary Table 1 and Supplementary Fig. 8, respectively. The fitting result corroborates our findings in the spectro-imaging experiments (Fig. 4d). The charge reaction went through Fe (+II) containing intermediate phases (green bar in Fig. 5d) and Fe (+III, red bar in Fig. 5d) was barely recovered when recharged to 4.5 V, but could indeed be formed if a constant-voltage charging step was applied following the constant-current charging. Ex situ XAS spectra (Supplementary Fig. 9) were also collected from electrodes that were cycled to different states of discharge/charge. The result is consistent with those obtained under operando conditions.
Our investigation of the charge reaction across different length scales provides a plausible explanation for the first cycle capacity loss, which is a problem commonly observed for FeF3 cathodes24,26,32 but poorly understood. We found that the capacity loss was caused by the incomplete reconversion of Fe to Fe (+III) during charge, which appears to be a kinetically limited process considering the slow diffusion of iron and fluoride ions. Another important implication is the indication of a different charge reaction pathway from the previous understanding. First-principles calculations previously suggested that a number of Fe (+III) containing compounds form sequentially during charge, which seemingly offered a plausible explanation for the voltage hysteresis between discharge and charge33. However, the operando investigation herein shows Fe (+II)-containing compounds formed first and Fe (+II) was slowly converted to Fe (+III) at high voltage during charge. A thorough (re)investigation using integrated experimental and theoretical approaches would be useful to address this discrepancy between our results and the previous simulation.
We have developed an operando X-ray spectro-imaging technique empowered by a new data analysis method to more accurately account for the internal background absorption and determine the X-ray absorption of weakly absorbing samples to enable the first visualization of the electrochemical reaction in high-capacity FeF3 cathodes at the nanoscale. These studies reveal the importance of porous nanostructure in achieving fast conversion reaction kinetics and high capacity for FeF3 cathodes, which could be a structural design principle generally applicable to other conversion electrode materials22. Further, more research efforts should focus on the charge reaction, which appears to be kinetically slow even for a highly active porous nanostructured FeF3 cathode and is confirmed as the bottleneck to utilize the full capacity of conversion electrode materials upon cycling. This work also provides guidelines in experimental design of in situ electrochemical cell fabrication, background subtraction and data analysis method to facilitate operando mechanistic studies of other electrode materials22,46,47 to elucidate reaction pathways and diagnose possible failure mechanisms. The temporal and spatial resolution of the chemical phase mapping will be further improved with the full commissioning of brighter synchrotron light sources, together with the development of lensless imaging methods48 and better data processing algorithms. This will ultimately allow operando studies of other complex solid-state phase transformations that are not electrochemically driven and lead to beneficial solutions towards many technological applications.
Synthesis of FeF3 MWs and polyhedra
FeF3 MWs were prepared by thermal dehydration of α-FeF3·3H2O MWs at 350 °C for 2.5 h in argon atmosphere. The precursor α-FeF3·3H2O MWs were synthesized by reacting Fe(NO3)3·9H2O and HF aqueous solution in ethanol with a concentration ratio of c(Fe3+):c(HF):c(H2O) ≈53.2 mM:500 mM:11,575 mM at 60 °C for 18 h. Rhombohedral phase FeF3 polyhedra were prepared by thermal conversion of the metastable cubic phase FeF3 polyhedra at 350 °C for 30 min in argon atmosphere. The precursor cubic phase FeF3 polyhedra were synthesized by reacting Fe(NO3)3·9H2O and HF aqueous solution in ethanol with a concentration ratio of c(Fe3+):c(HF):c(H2O) ≈ 20 mM:100 mM:4,670 mM at 60 °C for 24 h. More synthetic details can be found in the Supplementary Methods.
Scanning electron microscopy images were acquired using a LEO 55 VP scanning electron microscope at 5 kV. TEM images and ED patterns were acquired using either a Tecnai T-12 (120 kV) or a FEI Titan TEM (200 kV). Powder X-ray diffraction data were collected on a Bruker D8 diffractometer using Cu Kα radiation. The Brunauer–Emmet–Teller surface area and pore size distribution of the FeF3 MWs were calculated from nitrogen adsorption–desorption isotherms measured by a Quantachrome Autosorb-1 gas sorption analyser. Ex situ electrochemical measurements were performed on electrodes made of 70 wt% active material, 20 wt% carbon black and 10 wt% binder. CR2032-type coin cells were assembled in an argon-filled glovebox, using Li metal as the counter/quasi-reference electrode, 1 M LiPF6 in EC/DMC (1/1 by volume) as the electrolyte and electrolyte-soaked polyethylene-polypropylene films as the separator. Electrochemical impedance spectroscopy and galvanostatic cycling were performed using either a Biologic SP-200 or a VMP-3 Potentiostat/Galvanostat controlled by EC-Lab software.
Operando hard X-ray spectro-imaging and chemical map construction
The operando hard X-ray spectro-imaging experiments were performed using the full-field TXM at beamline X8C, National Synchrotron Light Source, Brookhaven National Laboratory, using a perforated 2032-type coin cell with holes on both sides sealed by Kapton tapes. The holes were sealed using Kapton tapes and need to be small to ensure a small cell impedance. The operando measurements were performed on electrodes made of 30 wt% FeF3-active material (polyhedra and MWs 1:1 by weight, or MWs only), 50 wt% carbon black and 20 wt% binder (see a representative scanning electron microscopy image in Supplementary Fig. 10). Note that both carbon black and binder are transparent to hard X-rays. Thin aluminum foils (∼8 μm thickness) or carbon papers (∼110 μm thickness) were used as current collectors for the FeF3 electrodes. The cell was put into a custom-built holder mounted on a motorized X, Y, Z, θ stage and aligned so that the X-ray beam could transmit through. A FOV of 40 × 40 μm2 with a 2,048 × 2,048 CCD camera was used. The cell was continuously cycled in galvanostatic or potentiostatic mode and absorption-contrast images (X-ray transmitted through the sample) and reference background images (X-ray passing through air) were collected in sequence under dynamic conditions. To track the phase transformations in the electrode, a full series of TXM images were collected at each state of discharge and charge. Each TXM image series was collected by scanning across the Fe K-edge (7,112 eV) from 7,091 to 7,285 eV, with a step size of 2 eV and taking one TXM image at each energy step, which contains 512 × 512 XANES spectra when using 4 × 4 binned camera binning. The exposure time for each image was 4 s. Each chemical phase map took ∼8 min to finish. After collection each set of data, the area of study (FOV 1) was allowed to rest for ∼16 min (not exposed to X-rays) to minimize any potential impact induced by the X-ray beam, during which a new set of data was taken in another area of study (FOV 2) and background reference images (X-ray passing through air) were also recorded after that. The output pixel size is ∼80 nm (camera binning 4).
The XANES spectrum at each pixel was normalized using our new method and then fitted with the linear combination of standard reference spectra collected from FeF3, FeF2 and Fe powders under the same conditions using TXM. The rutile FeF2 was used to represent all the possible rutile-related Fe2+-containing phases. This is a reasonable approximation because it was reported that the LixFeF3 (when x≈1.0) phase contains structural features that are found in the rutile FeF2 structure23. The spectrum fitting was carried out by minimizing the R value (a measure of misfit) for each spectrum at each pixel, which is defined as:
where Ei is 7,091 eV, Ef is 7,285 eV, dataE is the normalized spectrum at each pixel for the given energy E, and refE is the possible fitting reference value that is a linear combination of X-ray attenuation of FeF3, FeF2 and Fe. R values were minimized at each pixel to find the best-matched phase combination of different Fe oxidation states so that Red–Green–Blue (Red: Fe3+, Green: Fe2+, Blue: Fe) colours can be assigned accordingly to generate the chemical phase maps. We applied an R-value filter (misfit filter) to the resulting phase map and only pixels with R<0.08 were displayed in order to give the most accurate chemical phase information. See full details in Supplementary Methods.
The operando X-ray absorption experiments were performed at beamline X18A, National Synchrotron Light Source, Brookhaven National Laboratory, using a perforated 2032-type coin cell with holes on both sides sealed by Kapton tapes. The measurements were performed on electrodes made of 70 wt% FeF3-active material, 20 wt% carbon black and 10 wt% binder. The measurements were performed in transmission mode using a Si (111) double-crystal monochromator, which was detuned to ∼35% of its original maximum intensity to eliminate the high order harmonics in the beam. A reference X-ray absorption spectrum of Fe (K-edge 7,112 eV) was simultaneously collected using a standard Fe foil. Energy calibration was done using the first inflection point of the Fe K-edge spectrum as the reference point. The X-ray absorption data were processed and analysed using IFEFFIT-ATHENA37. Standard reference spectra from FeF3, FeF2 and Fe powders were collected to carry out spectrum fitting and determine the ratio between different Fe oxidation states. Ex situ spectra were also collected from electrodes cycled to different states of discharge and charge. The electrodes were recovered from coin cells disassembled in the glovebox and sealed in between two pieces of Kapton tape.
How to cite this article: Li L. et al, Visualization of electrochemically driven solid-state phase transformations using operando hard X-ray spectro-imaging. Nat. Commun, 6:6883 doi: 10.1038/ncomms7883 (2015).
Kostorz, G. Phase Transformations in Materials Wiley-VCH (2001).
Chiang, Y.-M. Building a better battery. Science 330, 1485–1486 (2010).
Liu, H. et al., Capturing metastable structures during high-rate cycling of LiFePO4 nanoparticle electrodes. Science 344, 1251817-1–1251817-7 (2014).
Sharma, N. et al., Structural changes in a commercial lithium-ion battery during electrochemical cycling: An in situ neutron diffraction study. J. Power Sources 195, 8258–8266 (2010).
Yu, X. et al., Understanding the rate capability of high-energy-density Li-rich layered Lli1.2Ni0.15Co0.1Mn0.55O2 cathode materials. Adv. Energy Mater 4, 1300950 (2014).
Hu, Y.-Y. et al., Origin of additional capacities in metal oxide lithium-ion battery electrodes. Nat. Mater 12, 1130–1136 (2013).
Huang, J. Y. et al., In situ observation of the electrochemical lithiation of a single SnO2 nanowire electrode. Science 330, 1515–1520 (2010).
Wang, F. et al., Tracking lithium transport and electrochemical reactions in nanoparticles. Nat. Commun 3, 1201 (2012).
Gu, M. et al., Demonstration of an electrochemical liquid cell for operando transmission electron microscopy observation of the lithiation/delithiation behavior of Si nanowire battery anodes. Nano Lett. 13, 6106–6112 (2013).
Liao, H.-G. et al., Facet development during platinum nanocube growth. Science 345, 916–919 (2014).
Andrews, J. C. & Weckhuysen, B. M. Hard X-ray spectroscopic nano-imaging of hierarchical functional materials at work. ChemPhysChem 14, 3655–3666 (2013).
Meirer, F. et al., Three-dimensional imaging of chemical phase transformations at the nanoscale with full-field transmission X-ray microscopy. J. Synchrotron. Radiat. 18, 773–781 (2011).
Wang, J. et al., Automated markerless full field hard X-ray microscopic tomography at sub-50 nm 3-dimension spatial resolution. Appl. Phys. Lett. 100, 143107 (2012).
Ebner, M., Marone, F., Stampanoni, M. & Wood, V. Visualization and quantification of electrochemical and mechanical degradation in Li ion batteries. Science 342, 716–720 (2013).
Nelson, J. et al., In operando X-ray diffraction and transmission X-ray microscopy of lithium sulfur batteries. J. Am. Chem. Soc. 134, 6337–6343 (2012).
Chao, S.-C. et al., Study on microstructural deformation of working Sn and SnSb anode particles for Li-ion batteries by in situ transmission X-ray microscopy. J. Phys. Chem. C 115, 22040–22047 (2011).
Weker, J. N. et al., In situ nanotomography and operando transmission X-ray microscopy of micron-sized Ge particles. Energy Environ. Sci 7, 2771–2777 (2014).
Wang, J., Chen-Wiegart, Y.-c. K. & Wang, J. In situ chemical mapping of a lithium-ion battery using full-field hard X-ray spectroscopic imaging. Chem. Commun. 49, 6480–6482 (2013).
Yang, F. et al., Nanoscale morphological and chemical changes of high voltage lithium–manganese rich NMC composite cathodes with cycling. Nano Lett. 14, 4334–4341 (2014).
Wang, J., Chen-Wiegart, Y.-c. K. & Wang, J. In operando tracking phase transformation evolution of lithium iron phosphate with hard X-ray microscopy. Nat. Commun 5, 4570 (2014).
Yu, Y.-S. et al., Nonequilibrium pathways during electrochemical phase transformations in single crystals revealed by dynamic chemical imaging at nanoscale resolution. Adv. Energy Mater 5, 1402040 (2015).
Cabana, J., Monconduit, L., Larcher, D. & Palacín, M. R. Beyond intercalation-based Li-ion batteries: the state of the art and challenges of electrode materials reacting through conversion reactions. Adv. Mater. 22, E170–E192 (2010).
Li, H., Balaya, P. & Maier, J. Li-storage via heterogeneous reaction in selected binary metal fluorides and oxides. J. Electrochem. Soc. 151, A1878–A1885 (2004).
Badway, F., Cosandey, F., Pereira, N. & Amatucci, G. G. Carbon metal fluoride nanocomposites: High-capacity reversible metal fluoride conversion materials as rechargeable positive electrodes for Li batteries. J. Electrochem. Soc. 150, A1318–A1327 (2003).
Badway, F., Pereira, N., Cosandey, F. & Amatucci, G. G. Carbon-metal fluoride nanocomposites: Structure and electrochemistry of FeF3:C. J. Electrochem. Soc. 150, A1209–A1218 (2003).
Li, L., Meng, F. & Jin, S. High-capacity lithium-ion battery conversion cathodes based on iron fluoride nanowires and insights into the conversion mechanism. Nano Lett. 12, 6030–6037 (2012).
Li, H., Richter, G. & Maier, J. Reversible formation and decomposition of LiF clusters using transition metal fluorides as precursors and their application in rechargeable Li batteries. Adv. Mater. 15, 736–739 (2003).
Li, C., Mu, X., van Aken, P. A. & Maier, J. A high-capacity cathode for lithium batteries consisting of porous microspheres of highly amorphized iron fluoride densified from its open parent phase. Adv. Energy Mater 3, 113–119 (2013).
Kim, S.-W. et al., Fabrication of FeF3 nanoflowers on CNT branches and their application to high power lithium rechargeable batteries. Adv. Mater. 22, 5260–5264 (2010).
Zheng, G. et al., Interconnected hollow carbon nanospheres for stable lithium metal anodes. Nat. Nanotech 9, 618–623 (2014).
Armand, M. & Tarascon, J. M. Building better batteries. Nature 451, 652–657 (2008).
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. 131, 10525–10536 (2009).
Doe, R. E., Persson, K. A., Meng, Y. S. & Ceder, G. First-principles investigation of the Li-Fe-F phase diagram and equilibrium and nonequilibrium conversion reactions of iron fluorides with lithium. Chem. Mater. 20, 5274–5283 (2008).
Liu, P. et al., Thermodynamics and kinetics of the Li/FeF3 reaction by electrochemical analysis. J. Phys. Chem. C 116, 6467–6473 (2012).
Wang, F. et al., Conversion reaction mechanisms in lithium ion batteries: Study of the binary metal fluoride electrodes. J. Am. Chem. Soc. 133, 18828–18836 (2011).
Zhang, W. et al., In situ electrochemical XAFS studies on an iron fluoride high-capacity cathode material for rechargeable lithium batteries. J. Phys. Chem. C 117, 11498–11505 (2013).
Ravel, B. & Newville, M. Athena, Artemis, Hephaestus: data analysis for X-ray absorption spectroscopy using IFEFFIT. J. Synchrotron. Radiat. 12, 537–541 (2005).
Weng, T.-C., Waldo, G. S. & Penner-Hahn, J. E. A method for normalization of X-ray absorption spectra. J. Synchrotron. Radiat. 12, 506–510 (2005).
Morin, S. A., Bierman, M. J., Tong, J. & Jin, S. Mechanism and kinetics of spontaneous nanotube growth driven by screw dislocations. Science 328, 476–480 (2010).
Meng, F., Morin, S. A., Forticaux, A. & Jin, S. Screw dislocation driven growth of nanomaterials. Acc. Chem. Res. 46, 1616–1626 (2013).
Li, L. et al., Facile solution synthesis of α-FeF3·3H2O nanowires and their conversion to α-Fe2O3 nanowires for photoelectrochemical application. Nano Lett. 12, 724–731 (2012).
Parkinson, M. F. et al., Effect of vertically structured porosity on electrochemical performance of FeF2 films for lithium batteries. Electrochim. Acta. 125, 71–82 (2014).
Chueh, W. C. et al., Intercalation pathway in many-particle LiFePO4 electrode revealed by nanoscale state-of-charge mapping. Nano Lett. 13, 866–872 (2013).
Li, Y. et al., Current-induced transition from particle-by-particle to concurrent intercalation in phase-separating battery electrodes. Nat. Mater 13, 1149–1156 (2014).
Cosandey, F. et al., EELS spectroscopy of iron fluorides and FeFx/C nanocomposite electrodes used in Li-ion batteries. Microsc. Microanal. 13, 87–95 (2007).
Hua, X. et al., Comprehensive study of the CuF2 conversion reaction mechanism in a lithium ion battery. J. Phys. Chem. C 118, 15169–15184 (2014).
Li, L., Caban-Acevedo, M., Girard, S. N. & Jin, S. High-purity iron pyrite (FeS2) nanowires as high-capacity nanostructured cathodes for lithium-ion batteries. Nanoscale 6, 2112–2118 (2014).
Shapiro, D. A. et al., Chemical composition mapping with nanometre resolution by soft X-ray microscopy. Nat. Photon 8, 765–769 (2014).
This research is supported by NSF grant DMR-1106184 for the synthesis and structural characterization of the materials, and the UW-Madison WEI Seed Grant and Research Corporation SciaLog Award for the X-ray spectro-imaging and electrochemical studies. The operando hard X-ray spectro-imaging experiments and XAS experiments were performed at beamline X8C and X18A, respectively, at the National Synchrotron Light Source, Brookhaven National Laboratory, which are supported by the US Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-98CH10886. P.G. and F.W. were supported by the Laboratory Directed Research and Development (LDRD) program at Brookhaven National Laboratory. J.C. and Y.-S.Y. were 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, Offi ce of Science, Basic Energy Sciences under Award # DE-SC0012583. L.L also thanks Vilas Research Travel Awards for partially supporting the travel to the synchrotron facilities.
The authors declare no competing financial interests.
About this article
Cite this article
Li, L., Chen-Wiegart, Yc., Wang, J. et al. Visualization of electrochemically driven solid-state phase transformations using operando hard X-ray spectro-imaging. Nat Commun 6, 6883 (2015). https://doi.org/10.1038/ncomms7883
Local oxidation states of Ni, Co, and Mn atoms within pristine and charged Li Ni0.88Co0.08Mn0.04O2 primary particles
Journal of Physics and Chemistry of Solids (2021)
Advanced Materials (2021)
Nature Materials (2021)
Materials Today (2021)
Advanced Materials (2021)