betaFeedback

WorksPhysics and Astronomy

Operando real-space imaging of a structural phase transformation in a high-voltage electrode

Abstract

Discontinuous solid-solid phase transformations play a pivotal role in determining properties of rechargeable battery electrodes. By leveraging operando Bragg Coherent Diffractive Imaging (BCDI), we investigate the discontinuous phase transformation in LixNi0.5Mn1.5O4 within a fully operational battery. Throughout Li-intercalation, we directly observe the nucleation and growth of the Li-rich phase within the initially charged Li-poor phase in a 500 nm particle. Supported by the microelasticity model, the operando imaging unveils an evolution from a curved coherent to planar semi-coherent interface driven by dislocation dynamics. We hypothesize these dislocations exhibit a glissile motion that facilitates interface migration without diffusion of host ions, leaving the particle defect-free post-transformation. Our data indicates negligible kinetic limitations impacting the transformation kinetics, even at discharge rates as fast as C/2. This study underscores BCDI's capability to provide operando insights into nanoscale phase transformations, offering valuable guidance for electrochemical materials design and optimization.

Manuscript

Introduction

Over a century ago, Gibbs classified phase transformations into two fundamentally different types based on variations in the order parameter[1]. Continuous transformations exhibit variations that are “small in degree but may be great in its extent in space”. The initial phase is unstable to infinitesimal fluctuations, resulting in a continuous change of the order parameter across large regions, which can be monitored by measuring macroscopic properties. For example, the order-disorder phase transformation in alloys is studied by measuring diffraction averaged over a large volume[2]. In contrast, discontinuous transformations display variations that are “initially smaller in extent but great in degree”[2]. An energy barrier stabilizes the system against infinitesimally small fluctuations until nucleation occurs at a localized region, causing a disruptive change in the order parameter. A classic example is the martensitic transformation in steel[3]. Capturing intermediate stages of a discontinuous phase transformation is significantly more challenging, as it requires time- and spatially-resolved measurements to observe the nascent nucleus (often only a few nanometers large) and the subsequent interface propagation during growth. Yet, discontinuous phase transformations are of utmost importance in materials science because properties can be tuned by balancing nucleation and growth to achieve the desired microstructure[4, 5].

Many electrode materials in rechargeable batteries undergo discontinuous solid-solid phase transformations[6]. These are driven by extraction and insertion of ionic species during charge and discharge. The discontinuous phase transformations are deemed undesirable because the internal stresses at the migrating interface can impede kinetics and lead to mechanical degradations7. Nevertheless, many durable, high-rate cathode materials experience phase separation from the discontinuous phase transformation during cycling8–10. One example is the disordered spinel phase of LixNi0.5Mn1.5O4\mathrm{Li}_x\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_4 (0<x<10<x<1) (LNMO), which recently regained significant attention as a cathode material for electric cars because it is cobalt-free and high-voltage. The material shows great potential to outperform its commercial opponents in terms of stability and cost. Additionally, the low nickel content in LixNi0.5Mn1.5O4\mathrm{Li}_x\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_4 is particularly appealing as nickel raises environmental and cost concerns similar to cobalt11,12.

Understanding the mechanisms underpinning phase separation in electrode materials and other functional materials under real operating conditions has become critical of today’s materials science: it affects materials systems ranging from quantum materials to energy storage and conversion materials. Addressing this issue requires a three-fold approach. (1) Operando measurements with setup that resembles relevant conditions in a working device to capture the transformation. (2) Adequately high temporal resolution to resolve intermediates stages of the transformation. (3) Sufficiently fine spatial resolution in three dimension to distinguish the spatial extent of the nucleated phase and its interface with respect to the parent phase. It can be further complicated by the nano-sized morphology of intercalation systems, which is a way to combat slow diffusion of ions through the host structure[5].

X-ray powder diffraction is a pillar for operando characterization of phase transformations in electrochemical systems with sufficient time resolution. During a phase transformation, the diffraction condition differs between the coexisting phases, and each crystalline phase produces a distinct Debye-Scherrer ring13–15. Yet powder diffraction averages across many particles within the battery and lacks the spatial resolution to resolve the local structure of the nucleus and the interface during subsequent growth. Recent advances in high-brilliance synchrotron sources have enabled operando measurements to capture diffraction signals from individual sub-micron particles16. Furthermore, Bragg Coherent Diffractive Imaging (BCDI) provides detailed 3D structural information by inverting coherent diffraction data to real space via an iterative phase retrieval algorithm17. This technique measures the strain distribution and buried defects within individual particles18–21. In principle, BCDI offers an unparalleled opportunity to image nucleation and growth in individual electrode particles under operating conditions, all without the need for specialized sampling environments. Nevertheless, the challenge of inverting diffraction data comprised of separated diffraction peaks has hindered the realization of real space imaging of phase transformations22,23.

Here, we use operando BCDI to image a discontinuous phase transformation induced by Li-intercalation during galvanostatic battery discharge. In a single 500 nm large LixNi0.5Mn1.5O4\mathrm{Li}_x\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_4 particle embedded in a fully operational cell, we directly observe the nucleation and growth of the Li-rich phase inside the initial, fully charged Li-poor phase. Operando imaging also reveals the transformation of a curved coherent interface into a planar semi-coherent interface, driven by the introduction of dislocations. These dislocations move with the interface and exit the particle after the phase transformation completes, leaving a defect free crystal. Supported by the microelasticity model, our data show that the dislocation array at the semi-coherent interface reorients the interface during operation. Our data is consistent with a conservative motion of dislocations at the interface.

Consequently, we see no evidence of structural degradation or kinetic limitation from the semi-coherent interface on the phase transformation, even at discharge rates as fast as C/2. Our study highlights BCDI’s potential as a robust tool for operando insights into structural phase transformations in nanoscale systems, whether induced by electrochemical processes, temperature fluctuations, optical light exposure, or electronic excitation.

Results

Figure 1a shows the experimental setup for the operando measurement (see Methods and Figure S1–Figure 2). During discharge, Li-ion intercalation induces a structural phase transformation from the Li-poor phase (smaller lattice constant, dcd_c) to the Li-rich phase (larger lattice constant, ddd_d)13,24. The different lattice spacing results in two separate Bragg reflections, G111,c\mathbf{G}_{111,c} and G111,d\mathbf{G}_{111,d}. When the two phases coexist inside a single crystal, both Bragg reflections are present simultaneously (Figure 1b). Due to the illumination from spatially coherent x-rays on a single particle, the scattering amplitude is decorated by an interference pattern18. When the coherent x-rays illuminate the particle with coexisting phases, the superposition of the interference patterns captures the spatial distribution and the relative crystallographic registry of both phases, as well as the structure of the interface. To measure the 3D interference patterns, we recorded a series of 2D sections of the Ewald sphere across the reciprocal space by rocking the operando cell (Figure S3)25.

Operando Bragg Coherent Diffraction of a single Li$_x$Ni$_{0.5}$Mn$_{1.5}$O$_4$ nanoparticle during the discontinuous solid-solid phase transformation induced by electrochemical Li-insertion

Figure 1. Operando Bragg Coherent Diffraction of a single Lix_xNi0.5_{0.5}Mn1.5_{1.5}O4_4 nanoparticle during the discontinuous solid-solid phase transformation induced by electrochemical Li-insertion. a) Experimental setup showing the operando coin cell, illuminated by coherent X-rays at 9 keV with a focus size of 800 nm, and diffraction around the 111 Bragg peak recorded on an area detector. During discharge, the Lix_xNi0.5_{0.5}Mn1.5_{1.5}O4_4 particle undergoes a discontinuous phase transformation via phase coexistence of the Li-poor phase (in blue) and Li-rich phase (in red), separated by the interface (in green). b) Schematic illustrating the Ewald sphere construction mapping out the interference profile surrounding the reciprocal lattice points G111,dG_{111,d} (Li-rich phase) and G111,cG_{111,c} (Li-poor phase). The area detector records a segment of the Ewald sphere. As it intersects both reciprocal space vectors, the detector image shows a split peak. c) – j) Cross-sections of the 3D diffraction pattern for the same Lix_xNi0.5_{0.5}Mn1.5_{1.5}O4_4 nanoparticle at various depths of discharge. The vertical axis shows the lattice spacing, dd, determined as 2π/q2\pi/q, where qq is the coordinate in reciprocal space.

Operando visualization of the coexisting phases during the structural phase transformation under lithium intercalation

Figure 2. Operando visualization of the coexisting phases during the structural phase transformation under lithium intercalation. a) – p) The 3D strain field inside a single LixNi0.5Mn1.5O4\mathrm{Li}_x\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_4 nanoparticle during the two-phase reaction induced by Li intercalation. The strain maps are extracted by inverting coherent x-ray diffraction data shown in Figure 1 c-Figure 1j and Figure S4. The semi-transparent isosurface portrays the particle shape, while the colored slices display the strain distribution, ε111(r)\varepsilon_{111}(\mathbf{r}), illustrated on three chosen planes. We show the strain field, ε111(r)\varepsilon_{111}(\mathbf{r}), as the local lattice constant compared to the average lattice constant, dad_a, of the (111) planes between the fully charged, Li-poor (dcd_c) and the partially discharged, Li-rich (ddd_d) phase, where da=(dc+dd)/2d_a = (d_c + d_d)/2 and ε111(r)=d(r)/da−1\varepsilon_{111}(r) = d(r)/d_a - 1. At 0% DoD in a), the particle has a uniform negative strain shown in blue, corresponding to the Li-poor phase with a small lattice constant, dcd_c. At the end of the two-phase reaction around 68.1% DoD in p), the particle presents a uniform positive strain shown in red, corresponding to the Li-rich phase with a large lattice constant ddd_d.

Crystal structure of the disordered $Li_xNi_{0.5}Mn_{1.5}O_4$

Figure S1. Crystal structure of the disordered LixNi0.5Mn1.5O4Li_xNi_{0.5}Mn_{1.5}O_4. The disordered LixNi0.5Mn1.5O4Li_xNi_{0.5}Mn_{1.5}O_4 has the cubic spinel structure with space group Fd3ˉmFd\bar{3}m. The Ni and Mn reside within the purple octahedra with Ni occupying 25% of the sites and Mn occupying 75% of the sites. The Li resides within the green tetrahedra. The red atoms are oxygens.

Ewald sphere construction showing the Ewald sphere, 2D detector, incident and diffracted wavevectors $k_i$ and $k_f$, scattering vector $Q$, and lattice plane

Figure S3. Ewald sphere construction. Rocking the crystal, shown as lattice plane in red, results in the Ewald sphere slicing the vicinity of the Bragg peaks at different positions in the reciprocal space. This is equivalent to shifting the detector perpendicular to the Ewald sphere around the Bragg peaks. The nearly parallel slices were combined to obtain a 3D pattern.[1]

Figure 1c-Figure 1j presents the central slice of the measured operando 3D diffraction data as a function of depth of discharge (DoD) for the 111 Bragg peak of both phases converted to lattice spacing dd (see Figure S4 for the full dataset with 50 measurements). At 0% DoD, corresponding to the fully charged state, only one Bragg peak surrounded by interference fringes is visible at larger dd (Figure 1c). During the initial stages of discharge, spanning from 0% to 35% DoD, the peak position is static, yet variations in the interference fringes suggest local structural changes within the nanoparticle (Figure 1d). Starting at 41.5% DoD (Figure 1e), a secondary peak emerges at larger dd, steadily gaining intensity at the cost of diminishing intensity in the primary peak (Figure 1e-Figure 1i). This is the hallmark of a discontinuous structural phase transformation, characterized by a substantial variation in the order parameter (namely, the lattice constant). The presence of both diffraction peaks emanating from an individual sub-micrometer crystal indicates phase coexistence. Notably, the presence of the coexisting peaks in the diffraction data occurs at the voltage plateau in the electrochemical data26 (Figure S5), consistent with previous operando x-ray diffraction13,20.

Full set of the operando diffraction data from a single $\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4}$ particle undergoing a structural phase transformation during discharge

Figure S4. Full set of the operando diffraction data from a single LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} particle undergoing a structural phase transformation during discharge. Diffraction data shows the decrease of intensity of the lithium-poor phase (larger 2θ2\theta) and the increase of intensity of the lithium-rich phase (smaller 2θ2\theta). During the transition, two peaks coexist, indicating the presence of both phases inside the LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} nanoparticle. The two-phase reaction ends around when the battery is 68% discharged. The scalebar is 0.08 nm−1\mathrm{nm}^{-1}. The diffraction data is inverted using the correlated phase retrieval algorithm.

Electrochemical and diffraction data of a cell

Figure S5. Electrochemical data of the cell (top) and diffraction data of a single LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} particle within the cell (bottom). The voltage plateau in the electrochemical data coincides with the two-peak coexistence region in the diffraction data. This indicates that the electrochemistry behavior of the LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} particle that we selected for reconstruction is representative for all active material. When the specific capacity reaches around 80 mAh/g, the cell enters the solid-solution regime, which is reflected as one diffraction peak continuously shifting its 2θ2\theta position.

To interpret the measurements, we use the recently developed correlated phase retrieval algorithm23 to invert the diffraction data into real-space 3D images. Critical for the success of the algorithm is inverting a series of operando measurements simultaneously, while assuming an approximately static shape of the particle across different charge states (Figure S6). We argue that the assumption is true here because the material is stable for hundreds of charge-discharge cycles, suggesting retention of the particle shape during a single discharge. The algorithm reconstructs the particle shape and the displacement field of the nanoparticle along the scattering vector (here, [111])20,21,27. Subsequently, we derive the 3D strain distribution from this displacement field through numerical differentiation along the scattering vector28. The reconstructed diffraction pattern exhibits good agreement with the measured diffraction data (Figure S7), affirming the success of the phase retrieval process.

Illustration of the correlated phase retrieval algorithm, showing scans, Trial 1, Trial 2, Trial 29, and Trial 30 reconstructions, selection of the Best 5 trails, and iterative support reconstruction

Figure S6. Illustration of the correlated phase retrieval algorithm. The illustration shows one iteration of the algorithm. A set of scans (we used 10 scans) that include both single-peak and two-peak diffraction are reconstructed with alternating Error Reduction (ER)[2] and Relaxed Averaged Alternating Reflections (RAAR)[3] phase retrieval algorithms separately 30 times. Each individual reconstruction is called a trail. Out of the 30 trails for each scan, we select the best 5 trails. The support of the reconstruction for next iteration is then calculated by averaging the shape of the best 5 trails for all 10 scans. During the selection of the best 5 trails, the particle shape can appear inverted (two solutions (S(r)S(r) and S∗(−r)S^{*}(-r) are indistinguishable, where rr is the coordinate and ∗* denotes complex conjugate). To determine if the reconstruction is inverted, we use cross-correlate strain among different reconstructions.

Side-by-side comparison of the central theta slice from the measured diffraction and the Fourier transform of reconstructions

Figure S7. A side-by-side comparison of the central theta slice from the measured diffraction (top) and the Fourier transform of reconstructions (bottom). The close alignment between the reconstructed results and the diffraction data is indicative of successful phase retrieval. The false colors are identical to Figure S4.

Figure 2 illustrates the operando strain evolution within a LixNi0.5Mn1.5O4\mathrm{Li}_x\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_4 nanoparticle, as obtained from the phase retrieval on the operando diffraction data (see Figure S8 for the full dataset of 50 images). Overall, the imaging data portrays the nucleation and growth of the Li-rich phase at the expense of the Li-poor phase through interface advancement (similar behavior is observed in another LixNi0.5Mn1.5O4\mathrm{Li}_x\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_4 particle, see Figure S11-12). At the onset of discharge, the nanoparticle consists of an almost homogeneous Li-poor phase (negative strain, depicted in blue) (Figure 2a-Figure 2c). As the electrochemical lithiation proceeds, an inclusion of the Li-rich phase (positive strain, depicted in red) nucleates at the bottom right corner of the nanoparticle (Figure 2d). At this stage, the imaging suggests the presence of multiple nucleation sites (Figure S9); however, as the growth proceeds, only a single inclusion becomes visible (Figure 2e). The merging is likely driven by surface tension, akin to Ostwald ripening or coarsening²⁹. Throughout the subsequent lithiation, the Li-rich phase grows at the expense of the Li-poor phase via interface propagation (Figure 2e-Figure 2o). We approximate the interface propagation velocity to be around 0.13 nm/s, notably slower than the mean velocity of Li-ions due to diffusion (Figure S10). Consequently, the Li ions have sufficient time to equilibrate concentration gradients within each phase, likely resulting in a sharp Liconcentration gradient at the interface30. This is consistent with the interface width between the coexisting structures in Figure 2i-Figure 2j, which is less than 100 nm, approaching the spatial resolution of operando BCDI21. By the end of the process, the nanoparticle comprises entirely of the Li-rich phase (Figure 2p).

Operando imaging of structural phase transformation during discharge

Figure S8. Full set of the operando imaging of the structural phase transformation during discharge. For each state of discharge, the chosen planes at the same position are imaged for the visualization of strain distribution inside the entire particle. The particle starts with a uniform negative strain (Li-poor phase) and ends the two-phase reaction with a uniform positive strain (Li-rich phase). The intermediate stages show that particle has both the red and blue phases, where the red phase grows at the expanse of blue phase, defining a nucleation and growth mechanism for the two-phase reaction. The particle is about 500 nm large.

2D slice of the 3D strain distribution during the nucleation of the Li-rich phase

Figure S9. 2D slice of the 3D strain distribution during the nucleation of the Li-rich phase. The arrows indicate the nucleation sites, of the new Li-rich phase (red) inside the Li-poor phase (blue). When the Li-rich phase nucleates, it starts around the edge of the particle and can occur simultaneously at multiple locations. As the Li-rich phase continues to grow, the nucleated sites grow and, in the meantime, coalesce. Later, the nucleated sites merge for a total reduction of interface area.

Interface velocity measured during the operando imaging experiment

Figure S10. Interface velocity measured during the operando imaging experiment. Looking at specifically the region when a linear interface exists, we approximate the distance that the interface travels to be around 65 nm in 494 seconds. In three-dimensional diffusion in spinel LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4}, the movement of ions can be described with mean square displacement ⟨x2⟩=6Dt\langle x^{2} \rangle= 6Dt, where DD is the diffusion coefficient and tt is the time. Using bulk diffusion coefficient of Li in LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4}, 2×10−12 cm2/s2 \times10^{-12}\ \mathrm{cm}^{2}/\mathrm{s} [4], we calculate the average distance for the given time, x=6Dt=6∗2∗10−12 cm2/s∗494 s=7.7×10−5 cm=770 nmx = \sqrt{6Dt} = \sqrt{6 * 2 * 10^{-12}\ \mathrm{cm}^{2}/\mathrm{s} * 494\ \mathrm{s}} = 7.7 \times10^{-5}\ \mathrm{cm} = 770\ \mathrm{nm}. This is one order of magnitude larger than the 65 nm we observe. We conclude that at the discharge rate, the interface propagation is not limited by Li diffusion in the particle.

Additional electrochemical and diffraction data

Figure S11. Additional electrochemical data of another cell (top) and diffraction data of another LixNi0.5Mn1.5O4\mathrm{Li}_x\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_4 particle that was discharged at C/10C/10 (bottom). Similar to the particle discussed above, this supplementary particle also displays an extended period of two-phase coexistence during the discharge, visible as the peak splitting in the diffraction. The two-phase region is again coincident with the voltage plateau in the electrochemical data.

In addition to visualizing in real time the phase distribution and the interface propagation during the phase transformation, operando BCDI provides insights into the morphological evolution of the interface. Our data shows an initially curved interface (Figure 2e-Figure 2h) that subsequently transforms into a planar configuration (Figure 2j-Figure 2k). These morphological changes are likely associated with the microstructure of the interface between the Li-poor and the Li-rich phases at different DoDs. Heterointerfaces between two distinct crystalline structural phases are typically classified into three types: coherent interfaces maintaining complete continuity of the lattice (Figure 3a), semi-coherent interfaces with piecewise continuity separated by misfit dislocations (Figure 3d), and incoherent interfaces with no registry between the two phases31. In our single particle diffraction data, the coexisting phases generate diffraction peaks in proximity (Figure 1e-Figure 1h). Therefore, the two crystalline phases are closely aligned, ruling out a fully incoherent interface, which typically occurs when the crystal planes are misaligned by more than 15 degrees32.

Microelasticity theory for coherent and semi-coherent interface

Figure 3. Microelasticity theory for coherent and semi-coherent interface in LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4}. a) Illustration showing a coherent interface. Each plane of blue atoms representing the Li-poor phase is connected to a plane of the red atoms representing the Li-rich phase. The lattice distortion gradually intensifies from the left to the right. b) Isosurface plot of the elastic strain energy as a function of normal direction for a coherent interface. The red arrows indicate the energy minima directions, ⟨100⟩\langle100\rangle. c) 2D parametric plot of the strain energy illustrated in b) in the plane spanning the [001][001] and [110][110] directions. The energy minimum, [001][001], and maximum direction, [111][111], display similar values. d) Illustration of a semi-coherent interface with a misfit edge dislocation – one extra half-plane inserted from the top – that relieves the misfit strain in the direction of the Burgers vector perpendicular to the half-plane. e) Isosurface plot of elastic strain energy for a semi-coherent interface with coherency loss in the [110][110] direction. The blue arrow indicates the energy minimum direction along [1ˉ10][\bar{1}10]. f) 2D parametric plot in the plane spanning [100][100] and [010][010], comparing strain energy between a coherent (red) and a semi-coherent (blue) interface. The semi-coherent interface exhibits significant anisotropy in its strain energy.

To distinguish between the coherent and semi-coherent heterointerface, we adopt the microelasticity model within the framework of continuum mechanics33, which was adapted to phase transformation in LixFePO4\mathrm{Li}_x\mathrm{FePO}_4 (0<x<10<x<1), another technically important phase separating cathode material34,35. Specifically, we approximate the elastic strain energy, denoted as B\mathbf{B}, for an inclusion of the Li-rich phase inside the Li-poor phase (using Einstein notation for summation)

B(n⃗)=λijklεij0εkl0−niσij0Ωjl(n⃗)σlm0nm(1)\mathbf{B}(\vec{n}) = \lambda_{ijkl}\varepsilon_{ij}^{0}\varepsilon_{kl}^{0} - n_i\sigma_{ij}^{0}\Omega_{jl}(\vec{n})\sigma_{lm}^{0}n_m \tag*{(1)}

where n⃗\vec{n} is the interface normal, εij0\varepsilon_{ij}^{0} is the strain tensor, σij0\sigma_{ij}^{0} is the stress tensor, Ωij\Omega_{ij} is related to the elastic Green’s tensor and defined as Ωij−1=λikljn⃗kn⃗l\Omega_{ij}^{-1} = \lambda_{iklj}\vec{n}_k\vec{n}_l, and λijkl\lambda_{ijkl} is the elastic stiffness tensor (Figure S13). We first consider a coherent interface separating the two phases with an isotropic transformation strain of 1%. The calculated elastic strain energy from equation (1) reveals an energy minimum direction along the family of <100><100> directions (Figure 3b-Figure 3c). Notably, the strain energy along the maximum direction, <111><111>, differs from the minimum direction, <100><100>, by a mere 8%. Given this weak dependence of the elastic energy on the interface orientation, the modeling is consistent with an overall spherical shape of the interface, as shown in Figure 2e-Figure 2h.

Figure S13. Geometry of the dislocations used in microelasticity modelling. During discharge, LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} transforms from a cubic-spinel to another cubic-spinel phase with a lattice mismatch of 1% (consistent with literature and estimated directly from the diffraction data through peak splitting). The coherency strain is the same along all three principal axes of the ⟨100⟩\langle100\rangle family. We find the stress tensor by multiplying the strain tensor with the elastic stiffness tensor. The elastic stiffness tensor was adopted from a similar spinel material LiTi2O4\mathrm{LiTi}_{2}\mathrm{O}_{4} [5] and assumed equal for both phases. For the semi-coherent interface to reflect the coherency loss along [110][110], we rotate the strain matrix along the zz direction by 45 degrees to align [110][110] on [100][100] and set it to be 0, then transformed it back to the previous coordinates.

As the structural transformation advances and the interfacial area expands, significant coherency strain arises due to the lattice mismatch between the two phases. To alleviate this strain, the interface is expected to introduce misfit dislocations and form a semi-coherent interface (Figure 3d). In this configuration, the interface comprises regions that maintain lattice continuity but is interspersed with dislocation cores that release interfacial strain and disrupt this continuity31. These misfit dislocations introduce an anisotropy to the otherwise isotropic transformation strain, consequently altering the orientations of the low-energy phase boundaries34,35. The transformation from coherent to semi-coherent interface is reminiscent of the coherency loss in metals36. For the spinel structure, the slip directions and planes are <110><110> and {111}\{111\}37. We assume that the dislocations possess a Burgers vector along [110][110] and lie within the (1ˉ11)(\bar{1}11) plane (Figure S13). As a result, these dislocations lead to a loss of coherency along [110][110] by providing strain relief in this direction. Setting ε110=0\varepsilon_{110}=0 results in the minimum elastic strain energy, B(n⃗)\mathbf{B}(\vec{n}), along [1ˉ10][\bar{1}10] (Figure 3e), and the energy difference between the lowest and highest energy directions has increased to 197% (Figure 3f). This anisotropy results in a preferential interface orientation along [1ˉ10][\bar{1}10], and correspondingly, the microelasticity model predicts a transition of the interface geometry from a curved to a planar configuration, as we observe in Figure 2h-Figure 2i.

To validate the transition from a coherent to a semi-coherent interface, we examine if an array of misfit dislocations is visible at the interface after it transforms into a planar geometry. Dislocations can be directly observed in the displacement field from BCDI data as singularities20,21,27. At the fully charged state (DoD = 1.2%), the particle exhibits uniform negative strain, and the displacement field is continuous without singularities (Figure 4a). As the discharge progresses, the Li-rich phase with positive strain forms, leading to changes in the displacement field (Figure 3e). At the interface, the continuity of the displacement field remains uninterrupted, showing no indications of structural defects such as dislocations. Yet, for the subsequent discharge, when the planar interface forms, the displacement map in Figure 4c reveals an array of misfit dislocations. In this diffraction geometry, we measure the projection of the Burgers vector along [111]. Thus, a phase discontinuity of 2π2\pi from a dislocation with Burgers vector along [110] is expressed as a phase discontinuity of 5.1 radians along [111] (Figure S14). Given a 1% lattice mismatch, we anticipate observing one dislocation per 100 unit cells, which translates to approximately one dislocation per 65 nm, on the same order of magnitude as the observed dislocation density of about one per 95 nm in Figure 4c. The presence of these dislocations aligns with our earlier hypothesis regarding the introduction of a semi-coherent interface for strain relaxation in the microelasticity model. The preferential [1ˉ10][\bar{1}10] energy minimum direction is also evidenced in Figure 4c as the strain map shows the interface orientation almost normal to the scattering vector [111][111].

Cross-sections of the reconstructed strain and displacement fields in the xz plane

Figure 4. The cross-sections of the reconstructed strain (top) and displacement (bottom) field in the xzxz plane at different depths of discharge. a) Three-dimensional representation of the LNMO particle aligned at the same angle as in Figure 2, with the two slices shown as strain and displacement field in b) – e). b) Prior to the phase transformation, the strain map corresponds to a fully charged Li-poor phase. The false color in the phase maps reflects the displacement from the ideal lattice at each position. c) At the early stage of the phase transformation, the Li-rich phase (enclosed by the rectangle) nucleates at the lower right corner, leading to an intensified color gradient within the rectangle. Both xzxz slices in b) and c) are located at y1y_1 in a). d) During the phase transformation, the nucleated Li-rich phase forms a semi-coherent interface with the Li-poor phase (outlined by the rectangle), which contains an array of dislocations each denoted by ⊥\perp in the phase map. The dislocations run into the page and propagate along the negative xx-direction with the interface. e) At the end of phase transformation, the strain map shows the Li-rich phase with no observable signs of dislocations in the phase map. The xzxz slices in d) and e) are located at y2y_2 in a). The scattering vector, qq, points along the zz-axis.

Identifying singularities and the azimuthal scans around it indicating dislocations

Figure S14. Identifying singularities and the azimuthal scans around it indicating dislocations. a) The enlarged phase map at 58.8% DoD same as Figure 4c. b) - e) The azimuthal scans around the singularities as circled in a). All scans show an average phase jump around 5 radians.

Discussion

Various models have been proposed to explain intercalation-driven phase transformations in rechargeable battery electrodes. The ‘shrinking core’ model applied to LixFePO4\mathrm{Li}_{x}\mathrm{FePO}_{4} assumes uniform Li insertion in isotropic spherical particles38–40. Built on the anisotropic Li-ion diffusion in LixFePO4\mathrm{Li}_{x}\mathrm{FePO}_{4}, the intercalation wave mechanism (or “domino-cascade”) considers an interfacial zone sweeping through the particle41,42. Incorporating both models, Singh et al. develops a general continuum theory and categorizes intercalation dynamics based on crystal anisotropy and ion insertion kinetics across electrode/electrolyte interface (namely isotropic bulk transport limited, anisotropic bulk transport limited, isotropic surface reaction limited, and anisotropic surface reaction limited)30. In LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4}, the lithium diffusion is isotropic43. Thus, at (dis)charge rates we measured here (C/10), the phase transition falls under isotropic surface reaction limited regime. Our experiments suggest a nucleation and growth mechanism for the transformation, consistent with findings by Kuppan et al. using spectroscopic ex-situ imaging in chemically oxidated ordered LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} single crystals44. Within this framework, the nucleus grows from a single localized point within the particle and extends through interface propagation.

In addition to evaluating different models for phase growth, access to crystal structure in our operando data provides insights into the nanomechanics at the propagating interface. The direction of dislocation motion has a significant component perpendicular to the Burgers vector. This motion is reminiscent of dislocation climb, a process requiring diffusion of host species (Ni, Mn, and O). Yet, given the reversibility of hundreds of cycles in LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4}45, a diffusional flux of host species during every cycle seems unlikely. Drawing inspiration from the well-established theory of phase transformations in metals31, we propose a model involving a glissile motion of interfacial dislocations. In this framework, the interface is generated through an invariant shear deformation to accommodate the lattice mismatch between the two phases. As the interface propagates, dislocations enter the crystal at the surface along the slip planes created by the shear (Figure S15)32, maintaining a conservative glissile interface motion with no host species diffusion, which allows fast kinetics31. Operando imaging of a LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} nanoparticle measured at C/2 discharge rate shows that the two-phase reaction completes within 40 minutes (Figure S16), five times faster than the transformation at C/10. Despite the rapid reaction, phase separation is still evident in the strain maps (Figure S17), revealing that the two-phase transformation kinetics can be fast. Our results are consistent with recent phase-field modeling that suggests interfacial coherency loss can substantially improve reaction kinetics for high-rate cathode materials when the two-phase reaction is unavoidable35. When the phase transformation completes, the interfacial dislocations move along with the interface and exit the nanoparticle (Figure 4d).

Image not extracted

Figure S15. 2D illustration of the creation and propagation of a glissile interface. The example shows a growth of a square structural phase with a larger lattice parameter, similar to growth of Li-rich LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} that has a cubic structure with a larger lattice parameter. The operation to minimize coherency strain at the interface includes lattice deformation BB, lattice-invariant shear deformation SS, and rotation RR.[6] The lattice-invariant deformation introduces slip at the interface and misfit dislocations at the interface, which move along slip planes during interface propagation. Consider snapshots during interface propagation, each snapshots shows an array of dislocations. During subsequent snapshots the dislocation array (each consisting of a different set of dislocations) moves perpendicular to the Burgers vector. This occurs via conservative motion where new dislocations enter the interface at the surface and exit the interface at the opposite surface.

Electrochemical data and diffraction data of a single particle

Figure S16. Electrochemical data of the cell (top) and diffraction data of a single LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} particle within the cell (bottom) at C/2. The two-phase coexistence is still present at a much higher discharge rate. This supports our hypothesis that the phase separation in LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} does not necessarily limit its kinetics as dislocations play an important role in reducing energy barriers for the reaction.

Operando imaging of the structural phase transformation during discharge

Figure S17. Operando imaging of the structural phase transformation during discharge at C/2C/2. Phase retrieval of the data shown in S16 and collected at a discharge rate of C/2C/2. Akin to a slower discharge rate of C/10C/10, nucleation and growth are visible.

In conclusion, we have successfully imaged a discontinuous phase transformation inside a single battery electrode nanoparticle in a fully operational device. The 3D snapshots of the crystal structure with a spatial resolution in the tens of nanometers and a time resolution in a few minutes reveal a nucleation and growth transformation mechanism. At C/10 and C/2 discharge rate, we see no kinetic limitations due to loss of interface coherency from misfit dislocations. Although our current time resolution with BCDI limits the study for faster charge rates, ongoing developments of diffraction-limited synchrotron sources will boost BCDI to seconds46. The advancement will enable imaging of rapid, discontinuous phase transformations and investigating effects like interface pinning that we observed in Figure 2l. Operando observations of interface dynamics warrant further exploration into the atomic arrangement of interface structures. While BCDI measures the long-range displacement field at the interface, a multimodal approach, combining operando BCDI with atomistic simulations and high-resolution electron microscopy, is imperative. Eventually, imaging phase transformations in nanoparticulate functional materials embedded in multicomponent devices will establish a feedback loop between stimuli and characterization. This approach is critical for optimizing properties like durability and ionic transport in materials exhibiting discontinuous phase transformations.

Methods

Synthesis and coin cell assembly

Liδ_{\delta}Ni0.5_{0.5}Mn1.5_{1.5}O4−δ_{4-\delta} disordered spinel was synthesized using the sol-gel method39. The coin cell was assembled using LNMO as the cathode and graphite as the anode (Figure S2). A 3 mm opening was created around the center of the base and top shells and sealed with a Kapton film to allow for X-ray transmission.

Expanded view of the operando coin cell, showing base shell, Kapton tape, washer, anode, separator, cathode, top shell, incident X-rays, and diffracted X-rays

Figure S2. Expanded view of the operando coin cell. The cell top and base are from the standard CR2032 cells, which have a diameter of 20mm and a height of 3.2mm. Both sides have a hole drilled at the center of size around 3 mm in diameter, which is sealed by Kapton tape. The cell is placed so that the material of interest, the LixNi0.5Mn1.5O4\mathrm{Li}_{x}\mathrm{Ni}_{0.5}\mathrm{Mn}_{1.5}\mathrm{O}_{4} cathode nanoparticles, is located downstream from the incident X-rays. The anode is lithium metal, and the separator (Celgard C480) contains electrolyte of 1 M solution of lithium hexafluorophosphate (LiPF6\mathrm{LiPF}_{6}) in a 1:1 volume mixture of ethylene carbonate (EC) and dimethyl carbonate (DMC).

Bragg Coherent Diffractive Imaging experiment

The operando BCDI experiment was conducted at Sector 34-ID-C in the Advanced Photon Source at Argonne National Laboratory. A double crystal monochromator was used to select x-rays with energy E=9E = 9 keV. The coherent X-rays with a focus size of 800 nm were incident on a fully operational half-cell. The rocking curve around the 111 Bragg peak was collected by a 2D detector (Timepix, 256×256256 \times256 pixels, each pixel 55 μm×55 μm55\,\mu\mathrm{m} \times55\,\mu\mathrm{m}) around 2θ=172\theta= 17 degrees (Δθ=±0.3∘\Delta\theta= \pm0.3^\circ). The detector was placed 1.1 meter away from the sample and an evacuated flight tube was inserted between the sample and the detector. A total of 76 diffraction patterns were collected for a single 3D rocking scan with 1 second exposure time for each image. The 3D diffraction pattern of the same particle was continuously captured under operando conditions, while the coin cell battery was discharging at C/10 or C/2. The low discharge rate was chosen to ensure that the particle remained in its discharge state throughout the scan. Each scan had a duration of about 2 minutes. Two rounds of alignment scans in the labx, labz (sample position relative to the incident beam) and theta directions were taken between every three 3D rocking scans to ensure that the particle did not move out of the beam or rotate away from the diffraction condition.

Inversion of x-ray data using phase retrieval

The details of the correlated data inversion algorithm were reported elsewhere23. Here, for brevity we summarize the main aspects of the algorithm we used to invert the operando data. 10 diffraction scans that describe the entire two-phase reaction were inverted simultaneously. Every diffraction scan was aligned such that the center of each diffraction had the same scattering vector (during experiment, the scattering angle was shifted to follow the peak evolution). The data inversion started with each scan running 30 reconstructions individually, each initiated with a random phase. For every 10 iterations, the error matrix of the correlation within the 30 reconstructions was calculated for each scan. Then we averaged the support of the 5 best correlated reconstructions in each scan across all 10 scans (in total 50 reconstructions) while leaving the displacement fields unchanged. Then the averaged support was multiplied by the individual support and became the input support for the next set of the reconstructions. The reconstruction consisted of a total of 610 iterations with alternating 10 iterations of the ER algorithm and followed by 50 iterations of the RAAR algorithm. For the primary particle in the main text, five datasets that contain mutually different scans were reconstructed and then stitched back together into a single sequence. The series shown in Figure 2 and Figure S8 is a collage of multiple independent reconstruction procedures, i.e., some have no overlap of the diffraction data taken for the reconstructions. We take the continuous evolution of the nucleation and growth as a testament for the robustness of the algorithm. For the supplementary particle, one dataset that consists of 13 scans and one dataset that consists of 15 scans were reconstructed. The algorithm was performed 10 times on each dataset and the final imaging is the result of 5 ×\times 10 reconstructions of each scan.

Acknowledgements

The work at Cornell was supported by the National Science Foundation under award number CAREER DMR-1944907. The work at UC San Diego was supported by the Sustainable Power and Energy Center (SPEC) and US Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract No. DE-SC0001805. This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science user facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357.

Operando real-space imaging of a structural phase transformation in a high-voltage electrode: Supplementary Information

Yifei Sun; Sunny Hy; Nelson Hua; James Wingert; Ross Harder; Ying Shirley Meng; Oleg Shpyrko; Andrej Singer

Supplementary Information

2D diffraction slices and corresponding inverted strain maps

Figure S12. 2D diffraction slices and the corresponding inverted strain maps at three chosen planes for the supplementary particle during discharge at C/10. Similar to the particle discussed above, this supplementary particle also shows a nucleation and growth regime where the nucleated Li-rich phase grows through interface propagation. The images are less smooth as they represent averages over less reconstruction runs.

Microelasticity Modeling

Eight lattice diagrams labelled a) through h), showing lattice deformation, lattice-invariant deformation, and rotation

Figure 5 (PDF p. 31).

References

  1. [1]Williams, G. J., Pfeifer, M. A., Vartanyants, I. A. & Robinson, I. K. Three-Dimensional Imaging of Microstructure in Au Nanocrystals. Phys. Rev. Lett. 90, 4 (2003).pubmed.ncbi.nlm.nih.gov/12786079
  2. [2]Fienup, J. R. Phase retrieval algorithms: a comparison. Appl. Opt. 21, 2758 (1982).DOI
  3. [3]Luke, D. R. Relaxed averaged alternating reflections for diffraction imaging. Inverse Probl. 21, 37 (2004).
  4. [4]Kunduraci, M. & Amatucci, G. G. The effect of particle size and morphology on the rate capability of 4.7V LiMn1.5+δNi0.5−δO4 spinel lithium-ion battery cathodes. Electrochimica Acta 53, 4193–4199 (2008).DOI
  5. [5]Qi, Y., Hector, L. G., James, C. & Kim, K. J. Lithium Concentration Dependent Elastic Properties of Battery Electrode Materials from First Principles Calculations. J. Electrochem. Soc. 161, F3010 (2014).DOI
  6. [6]Balluffi, R. W., Allen, S. M. & Carter, W. C. Kinetics of Materials. (John Wiley & Sons, 2005).DOI

Paper details

vlnc:W_z6qaa4fy
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents