Beyond Oxygen Redox Strategy in Designing Cathode Material for

Subscriber access provided by Drexel University Libraries

C: Energy Conversion and Storage; Energy and Charge Transport

Beyond Oxygen Redox Strategy in Designing Cathode Material for Batteries: Dynamics of a Prussian Blue-like Cathode Revealed by Operando XRD and XAFS and by Theoretical Approach Angelo Mullaliu, Giuliana Aquilanti, Lorenzo Stievano, Paolo Conti, Jasper Rikkert Plaisier, Sylvain Cristol, and Marco Giorgetti J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.8b12116 • Publication Date (Web): 17 Mar 2019 Downloaded from http://pubs.acs.org on March 21, 2019

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

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

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

The Journal of Physical Chemistry

Beyond Oxygen Redox Strategy in Designing Cathode Material for Batteries: Dynamics of a Prussian Blue-like Cathode Revealed by Operando XRD and XAFS and by Theoretical Approach

Angelo Mullaliu(a), Giuliana Aquilanti(b), Lorenzo Stievano(c,d), Paolo Conti(e), Jasper R. Plaisier(b), Sylvain Cristol(f) and Marco Giorgetti(a)*

a

Department of Industrial Chemistry “Toso Montanari”, University of Bologna, Viale Risorgimento 4,

40136 Bologna, Italy. b

Elettra – Sincrotrone Trieste, ss 14, km 163.5, 34149 Basovizza, Trieste, Italy

c

Institut Charles Gerhardt Montpellier, CNRS UMR 5253, Université de Montpellier, Place Eugène

Bataillon, 34095 Montpellier Cedex 5, France d

Réseau sur le Stockage Electrochimique de l’Energie (RS2E), CNRS FR3459, 33 Rue Saint Leu,

80039 Amiens, Cedex, France e

Department of Chemistry, University of Camerino, Via S. Agostino 1, 62032 Camerino (MC), Italy

f

Univ. Lille, CNRS, Centrale Lille, ENSCL, Univ. Artois, UMR 8181, Unité de Catalyse et Chimie du

Solide, Lille 59000, France

*Corresponding author E-mail: [email protected], phone: + 39 051 209 3666; fax +39 051 209 3690

1 ACS Paragon Plus Environment

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

Page 2 of 34

Abstract The dynamics of the lithiation/delithiation process and the nitrosyl electroactivity in copper nitroprusside were studied by operando X-ray diffraction (XRD) and operando X-ray absorption fine structure (XAFS). Data were interpreted based on a joint study performed by means of density functional theory (DFT) calculations. This approach allows retrieving the relevant structural and electronic information from the measured scattering and absorption data and therefore to untangle the lithiation mechanism in copper nitroprusside, which occurs with the reduction of both metals generating a lattice basal plane contraction and an axial elongation. An increase of the Debye-Waller factors for Cu-N bonds and a decreasing trend for the Cu-NC-Fe linear chains along with lithium insertion reveal a general increase in the Cu local disorder, which is thought to be the main cause of the rapid capacity fading observed during cycling. The ligand electroactivity of the nitrogen atom, detected by following vibrational frequencies, delivers an extra capacity and represents an alternative path to cationic and oxygen redox.


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


1. Introduction Affordable and sustainable energy production is one of the fundamental ingredients for the economic growth.1 Within this framework, Li-ion batteries play a crucial role due to the ability to reversibly store electrochemical energy and release it when needed.2-4 Li-ion batteries generally rely on the rapid reversible insertion of lithium ions in electrode m1aterials characterized by a crystal structure with available insertion sites, channels and/or interlayer spacings. This is true especially for positive materials, which on the other side exhibit rather low specific capacities compared to the anodic counterpart.5-6 To address this issue nowadays, anionic oxygen activity has been coupled to common cationic redox in the Li-rich Li1+xMO3 class of materials to increase the capacity, thus the energy density.7-9 Prussian blue analogues (PBAs) belong to a class of insertion-type materials.10-15 PBAs, and in particular metal hexacyanoferrates (MHCFs), are bimetallic cyanides with a three-dimensional cubic lattice of repeating -Fe-CN-M-NC- units, where M denotes a transition metal, commonly Fe, Mn, Co, Ni, Cu, Zn.16 The cyanide acts as bridging ligand between Fe and M, linking them in a precise fashion, the carbon-end bound to Fe, while the nitrogen-end coordinates to M. PBAs have been widely adopted in several applications, for instance as analyte sensors,17-18 magnetic devices,19 electrochromic devices,20 charge storage devices,21 ion-exchange sieves,22 and antibacterial agents.23 Copper hexacyanoferrate (CuHCF) has shown considerable performance at high rate as cathodic material,24 with negligible lattice strain in the lithiation and delithiation processes.25 In a previous work,26 we suggested the substitution of one cyano ligand with the nitrosyl ligand (NO) to give copper nitroprusside (CuNP), Cu[Fe(CN)5(NO)]. Such strategy is expected to have several advantages: a) the NO does not bridge to the copper site, enhancing the porosity of the structure and favoring the intercalation process; b) the nitrogen constituting the NO+ ligand can be reduced during lithiation, becoming possibly the third redox centre in addition to Fe and Cu, giving rise to an extended specific capacity. The ligand electroactivity of the nitrogen represents a unique case in PBAs and is coupled to cationic redox. However, galvanostatic cycling of CuNP turned out to be only partially reversible, with a capacity loss during the first 20 cycles and a reversible behavior in the following cycles that was ascribed to a fraction of available electroactive species. To deeply investigate the redox reaction mechanism of the electrode material and clarify the observed capacity loss, a combined experimental-theoretical approach was applied: operando X-ray diffraction (XRD) and operando X-ray absorption fine structure (XAFS) experiments were performed to retrieve structural and electronic information, while the dynamic evolution of the system was thoroughly analyzed with the use of the multivariate curve resolution alternate least square (MCR-ALS) 3 ACS Paragon Plus Environment


Page 4 of 34

chemometric technique and further compared with ab initio simulations. Computational quantum mechanical modelling was accomplished by means of density functional theory (DFT) calculations. 2. Experimental As previously reported,26 CuNP was synthesized by bulk co-precipitation starting from 1000mL of a 20 mM CuSO4 5H2O solution and 1000mL of a 20mM Na2[Fe(CN)5(NO)] 2H2O solution. The stoichiometry of the obtained product, as determined by chemical analysis, is Cu0.8[Fe1.2(CN)5(NO)] 0.5 H2O.26 Suitable electrodes for electrochemical tests and operando analyses were obtained by thoroughly mixing the pure active material (70%), carbon black (CB) (10%), and vapor grown carbon fibers-high density (VGCF-H) (10%) in an agate mortar. Polytetrafluoroethylene (PTFE) (10%) was finally added and mixing continued until a homogenous paste was obtained. Disks were cut from the flattened paste using a puncher with an inner diameter of 8 mm. The mass loading of such disks was 6-8 mg/cm2 of active material. A suitable electrochemical cell for in situ experiments, described in detail elsewhere,27 was used for both XRD and XAFS measurement. The cell consisted of a large piece of lithium metal foil, used as negative electrode, while the CuNP electrode was used as positive electrode. A 1M solution of LiPF6 in a mixture of ethylene carbonate (EC), propylene carbonate (PC) and dimethyl carbonate (DMC) in volumetric ratio EC:PC:3DMC served as the electrolyte for XAFS measurements, whereas a 1M solution of LiPF6 in EC:DC was used for operando XRD. Positive electrode, Celgard separator soaked in the electrolyte, and negative electrode were stacked and assembled under inert atmosphere in an Arfilled glove-box. Galvanostatic cycling with potential limitation (GCPL) was conducted by considering that 1C rate equals the current needed to insert one equivalent of Li-ion per equivalent of Fe in one hour, corresponding to a theoretical specific capacity of 114 mAh g-1. Cycling started after a rest time at open circuit potential (OCP) conditions with a negative imposed current. Linear sweep voltammetry was performed in coin cell geometry by imposing a negative polarization from OCP value to 1.5 V vs. Li+/Li at 0.1 mV s-1 scan rate in 1M solution of LiPF6 in EC:PC:3DMC.

2.1 Operando XRD and XAFS experiments XRD and XAFS experiments were performed at Elettra - Sincrotrone Trieste (Italy), at the MCX28 and XAFS beamlines.29 The storage ring operated at 2.0 GeV in top up mode with a typical current of 310 mA. 4 ACS Paragon Plus Environment

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


XRD data were recorded on the electrochemical cell above described, using a monochromatic X-ray beam of 1 Å. Data were collected in the reflection mode consecutively from 10° to 30° 2θ range with a 0.01° step and 0.5 s/point acquisition time. Operando data were acquired during the first discharge to 1.8 V vs. Li+/Li and subsequent charge to 3.5 V vs. Li+/Li at C/22 current rate. XAFS data were recorded at Fe and Cu K-edges in transmission mode using ionization chambers filled with a mixture of Ar, N2, and He to have 10%, 70%, and 95% of absorption in the I0, I1, and I2 chambers. An internal reference of iron and copper foil was used for energy calibration in each scan. This allowed a continuous monitoring of the energy during consecutive scans. No energy drifts of the monochromator were observed during the experiments. Spectra at Fe and Cu K-edges were collected with a constant k-step of 0.03 Å-1 with 2 s/point acquisition time. Data were collected alternatingly from 6900 eV to 8000 eV and from 8750 eV to 9830 eV around the Fe and Cu K-edges, respectively. The energies were calibrated by assigning the first inflection point of the spectra of the metallic iron and copper to 7112 eV and 8979 eV, respectively. The white beam was monochromatized using a fixed exit monochromator equipped with a pair of Si(111) crystals. Harmonics were rejected by using the cutoff of the reflectivity of the platinum mirror placed at 3 mrad with respect to the beam upstream the monochromator for Cu K-edge data, and by detuning the second crystal of the monochromator by 30% of the maximum for Fe K-edge data. Data were recorded during discharge to 2.0 V vs. Li+/Li at C/31 current rate.

2.2 Operando XRD and XAFS data analysis Profile matching and Rietveld refinement were carried out on XRD patterns using FullProf Suite software 30 and assuming as structural model the one reported by Reguera et al..31 A pseudo-Voigt peak shape was used and spherical harmonics were included to take into consideration an anisotropic Lorentzian size broadening. Peaks corresponding to PTFE (contained in the electrode formulation) were not refined, excluding the corresponding regions, i.e., 11.60-11.95° and 16.78-17.31°.32 The 29.31-29.43° region was also not considered, due to the presence of a diffraction peak of beryllium arising from the in situ cell window. Further details are presented in the Supplementary Material. Graphical representation of structures was exploited by means of VESTA software.33 The Extended X-ray Absorption Fine Structure (EXAFS) analysis was performed using the GNXAS package34-35 which is based on the Multiple Scattering (MS) theory. The method uses the decomposition of the EXAFS signals into a sum of several contributions, namely the n-body terms. The theoretical 5 ACS Paragon Plus Environment


Page 6 of 34

signal is calculated ab-initio and contains the relevant two-body γ(2), three-body γ(3), and four-body γ(4) MS terms.36 The two-body terms are associated with pairs of atoms, and probe their distances and variances. The three-body terms are associated with triplets of atoms and probe angles, and bond-bond and bond-angle correlations. The four-body terms are associated to chains of four atoms, and probe distances and angles in-between, and bond-bond and bond-angle correlations. However, because of the linearity of the Fe-C-N-Cu chains, all the angles were set to 180°, hence the actual number of parameters used to define the γ(3) or the γ(4) peaks was reduced by symmetry. More details on the use of parameters correlation in the four-body term is out of the aim of the present work and can be found in the references.37-38 Data analysis was performed by minimizing a χ2-like residual function that compares the theoretical (model) signal, μmod(E), to the experimental one, μexp(E). The phase shifts for the photoabsorber and backscatterer atoms were calculated starting from the structure reported by Reguera et al.

31

according to the muffin-tin approximation and allowing 10% overlap between the muffin-tin

spheres. The Hedin-Lundqvist complex potential

39

was used for the exchange-correlation potential of

the excited state. The core-hole lifetime, Γc, was fixed to the tabulated value 40 and was included in the phase shift calculation. The experimental resolution used in the fitting analysis was around 1 eV, in agreement with the stated value for the beam line used. An independent analysis was performed by means of a chemometric approach. MCR-ALS can be used to analyze a full set of XAFS data during an electrochemical process in a battery41 to recover information on the system composition. In analytical terms, this means that the operando data matrix XS,W can be decomposed into two matrices CS,F and AW,F, as follows (Eq. 1): 𝑿𝑺,𝑾 = 𝑪𝑺,𝑭 ⋅ 𝑨𝑾,𝑭

(1)

where CS,F contains the pure concentration profiles and AW,F the pure spectral components. In this case, The MCR-ALS algorithm was applied to the whole series of operando Cu K-edge XANES spectra.

2.3 DFT Total energy calculations were performed within the DFT framework using the Vienna Ab initio Simulation Package (VASP).42-45 The electron wave functions were expanded on a plane-waves basis set with an energy cut-off of 850 eV. Pseudopotentials were used to describe the electron-ion interactions within the projector augmented waves (PAW) approach.46 The convergence criterion for the electronic self-consistent cycle was fixed at 0.1 meV per cell. The integrations in the Brillouin zone 6 ACS Paragon Plus Environment

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


were performed on a grid of 2×2×2. Geometry optimizations were carried out by means of the conjugate gradient technique using the exact Hellman-Feyman forces acting on the ions. The structure was considered as optimized when the forces acting on the atoms were smaller than 0.03 eVÅ-1 and the energies variation between successive geometries was below 1 meV. Lithiated structures during discharge were simulated by inserting consecutive atoms of lithium in the model structure: symmetry calculation was disabled not to take into account the multiplicity of lithium positions (cf. Supplementary Material). Different calculations have been performed by changing the spin multiplet for the structures containing different number of inserted lithium to probe the electronic structure of the inserted materials. Vibrational frequencies are derived by diagonalization of the dynamical matrix which is obtained by numerical differentiation of the forces acting on the atoms after displacement of 0.04 Å around equilibrium geometry.

2.4 XANES calculation The ab initio simulations of the X-ray Absorption Near Edge Structure (XANES) spectra were performed by using the FDMNES software

47

and the MS approximation. Cu and Fe K-edges were

calculated in the photoelectron energy range -1 < E < 120 eV with respect to the Fermi energy level. The Hedin-Lundqvist complex potential39 was used to calculate the excited states. The absorption crosssection was calculated within the dipolar approximation. Clusters of 7 Å built around each nonequivalent absorbing atom were considered. Space group symmetry (I4mm) was disregarded by using the P1 symmetry and inserting as input the positions obtained by the DFT calculations. Optimized convolution parameters were retrieved for the pristine material and kept fixed for all lithiated compounds. Finally, calculated and convoluted XANES spectra were compared with experimental data. FDMNES is ab initio in the sense that it generally does not need any auxiliary parameters. Only the atom positions are required (output crystallographic positions computed by DFT) and the code starts from the first-principles equations.

3. Results and discussion 3.1 Preface to the electrochemical behavior Figure 1 reports the linear sweep performed on CuNP. In our previous work,26 we ascribed the electroactivity in the 2.0 < E < 2.6 V vs. Li+/Li potential window to both copper and iron species, as



Page 8 of 34

corroborated by the XANES traces, while the nitrosyl showed activity at lower potentials, i.e., below 2.0 V, where the signal associated to its linear form (~1940 cm-1) disappeared from the IR spectra.

Figure 1. Linear sweep on CuNP. Imposed negative polarization starting from OCP value to 1.5 V vs. Li+/Li at 0.1 mV s-1

3.1 Operando XRD The Rietveld refinement on the XRD pattern of pristine CuNP electrode is presented in Figure 2a. The additional peaks of PTFE and Be are labeled, while the Miller planes are indicated in brackets. The resulting structure is tetragonal (space group I4mm), characterized by lattice parameters a = 7.089(4) Å and c = 10.888(3) Å, in agreement with the assumed model.31


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


Figure 2. (a) Rietveld refinement of pristine CuNP electrode. Inset shows the structure of CuNP. (b) (002) and (110) Miller planes at different lithiation steps (pristine, 0.4, 0.9 and 0.72 Li-inserted) A selection of operando XRD patterns is reported in Figure 2b. Pristine (blue line), partially discharged (incomplete lithiation, i.e., 0.40 Li-equivalents; red line), discharged (complete lithiation, i.e., 0.90 Liequivalents; yellow line), and partially charged (incomplete delithiation, i.e., 0.72 Li-equivalents; black line) states are here compared for the (002) and (110) Miller planes. During discharge (arrows direction in Figure 1), (00l) planes shift towards smaller 2θ angles, whilst the opposite occurs for the (hh0) planes. This indicates a gradual and simultaneous (ab) plane contraction and c-axis elongation (cf. Figure 3). This result can be explained as follows: due to the higher porosity in the c-direction created by the nonbridging nitrosyl ligand and the elongated structure of CuNP caused by the Jahn-Teller distorted CuII (d9 9 ACS Paragon Plus Environment


Page 10 of 34

electron configuration), lithium ions are preferentially inserted in the 8c Wyckoff positions (as graphically represented in Figure 3). Here they interact with the N-terminals of the equatorial cyanides, as suggested by DFT (vide infra). Moreover, peak intensities decrease during discharge, as the crystallinity is being reduced in response to lithium insertion, while an inversion of relative peak intensities for the (002) and (110) planes occurs at the end of the discharge. During charge, the opposite trends are observed, being the basal plane, the c-dimension and the peak intensities are restored, and so the crystallinity, although only to a certain degree (cf. Figure S3). A more quantitative analysis was carried out, further confirming the already observed trends. Figure 3 illustrates the variation of lattice parameters during the lithiation: the slight peak shifts presented in Figure 2 correspond to a variation of lattice parameters that never exceeds 0.5%. Such a small lattice strain during the lithiation, typical of socalled zero-strain materials, demonstrates that CuNP exhibits a relatively good long-range structural stability during the first cycle, despite the partial loss of crystallinity.

Figure 3. Evolution of the lattice parameters during discharge (left) and graphical representation of the lithiation process along the a and c axes (Fe: light- brown; Cu: blue; C: brown; N: light blue; O: red, Li: green)


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


Further details on XRD data treatment, including the goodness of fit, are available in the Supplementary Material.

3.2 Operando XAFS Figure 4 displays a selection of normalized XANES spectra recorded during the operando XAFS experiment at the Fe and Cu K-edge. The FeII+δ/ FeII reduction is appreciated not only as a slight shift of the edge position and the inflection point in the first derivative (cf. Supplementary Material), but also as a different occupancy of the t2g levels in the pre-edge region,25-26 in line with the reduction of Fe from a partial trivalent to a divalent state in a low-spin octahedral environment. Moreover, the peak centred at 7119 eV arises from normally forbidden dipole transitions to empty bound states and its evolution might reflect longer-range effects of the shells beyond the cyanides.48-50 The increase in intensity during discharge might mirror the disorder induced by lithium insertion. The evolution of Cu K-edge is consistent with the CuII/CuI reduction: the raise in intensity of the ~8981 eV peak (labeled as peak B, cf. Figure 5 and Figure S4) and the decrease of the ~8986 eV peak (labeled as peak C, cf. Figure 5) are attributed to the 1s-4p + shakedown transitions of CuI and CuII, respectively.25,26,51-53

Figure 4. Selected operando XAFS spectra at Fe and Cu K-edge during discharge. Pristine and discharged states (0.4 and 0.9 Li-inserted) are shown

A quantitative data treatment was carried out in the Cu pre-edge region and highlighted in Figure 5a, b. The analysis results are fully reported in the Supplementary Material. The pre-edge involves three contributions, labeled as A, B and C: the inset of Figure 5a illustrates a fit example on the pristine 11 ACS Paragon Plus Environment


Page 12 of 34

sample, where the signal was fitted with three Gaussian functions after background subtraction. As peak B is a signature of CuI species and already present in the pristine sample, we can deduce that the overall copper has a slight monovalent character so that we can write, on average, CuII-ε. Being peak B the most relevant dynamic feature in the pre-edge region, further data treatment presented in Figure 5b correlates the centroid shift of peak B to the inserted lithium equivalents. A drastic change in centroid position takes place in the 0.24 < Li < 0.60 range, assuming instead an almost constant value outside the range. If the slope of this graph could be used as reaction coordinate, we would then suppose that the highest rate for copper reduction occurs within this range.


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


Figure 5. (a) Pre-edge data analysis at the Cu K-edge: evolution of normalized area of three different features (A, B, C) during lithiation. Inset displays the fit performed for the pristine material. (b) Centroid shift for peak B during lithiation. (c) MCR-ALS analysis at the Cu K-edge: concentration profile plot and (d) pure spectral components (Cu2+ species is labeled as Species 1 and indicated in blue, while Cu+ species labeled as Species 2 in red) The MCR-ALS algorithm was applied to the whole series of operando Cu K-edge XANES spectra revealing a progressive transformation of the pristine species to a second one (cf. Figure 5c), which most likely corresponds to the reduction of CuII to CuI. This, in turn, suggests that only two species (represented in Figure 5d) are participating to the electrochemical reduction. A similar plot was obtained by looking at the Fe K-edge dataset (cf. SI), while the Cu data was also analyzed by using a threespecies model (cf. SI), confirming though the occurrence of two species only. Even though the copper reduction process continues until the end of the discharge, the slope of the concentration profile assumes qualitatively the highest value in the 0.12 – 0.60 Li range, suggesting an increase of the rate of the Cu sites reduction in good agreement with the pre-edge data analysis. Eventually, MCR-ALS analysis agrees to the XANES curve behavior of the Figure 4 (Cu K-edge) where the presence of a two well defined isosbestic points at about 8995 and 9005 eV might suggest an analytical system formed by only two species in equilibrium. Complementary structural information about both Cu and Fe sites can be gained by looking at the extended portion of their K-edge XAFS spectra. Analysis of the pristine electrode was first done to check the reliability of the structural model as well as to set the relevant parameters for the minimization. For an accurate extraction of the structural information, both Fe and Cu edges were analyzed by multiple edge approach,37 i.e., a simultaneous fitting procedure at both metal edges. This in turn means that the same CuNP structural parameters are probed using two independent measurements, hence the reliability of the fitting minimization results enhanced. Figure 6 displays the details of the EXAFS analysis for the pristine electrode at Fe (a) and Cu (b) K-edges, and respective Fourier transforms (FTs) (c and d). Briefly, (i) only few relevant single EXAFS contributions are necessary to simulate the overall EXAFS signal at each edge, as indicated in the Experimental section; (ii) not only the two-body signals, but also the three- and the four- body MS terms are relevant; (iii) the multiple edge refinement allows one to take into consideration the oscillation of the MS contributions at the Fe K-edge that overlay the Cu K-edge. The results of the fitting procedure, in terms of relevant bond length distances, corresponding EXAFS Debye Waller factors and coordination numbers are available in Table S5. The table also 13 ACS Paragon Plus Environment


Page 14 of 34

indicates the fitted values for S02(Fe) and S02(Cu) in the pristine sample, which were subsequently kept fixed at these values for the analysis of the lithiated samples. The fit was conducted on a selection of spectra, i.e., every three, by taking into account the result of the MCR-ALS analysis. Also, a more flexible and less rigid structural model was required respect to the pristine sample. As shown by the DFT calculations and XANES data, the lithiation does not only induce changes in local structural arrangements at the Cu site, but also affects the bending of the Fe-NO angle. The outcome of the fitting procedures for all samples, reported in Table S5, shows consistent structural modifications upon lithiation. Among the various structural parameters described in the fitting Table, the first shell distances and relative Debye-Waller factors deserve a close inspection. Their variation during the electrochemical reaction is plotted in Figures 6e and 6f. While bond lengths remain almost unchanged, a dramatic increase of the structural disorder, especially at the Cu site, is demonstrated by the corresponding increase in Debye-Waller factors. That corresponding to the C≡N bond, however, is less modified, and oscillates around the equilibrium position all through the discharge. Based on the results of these fits, two additional considerations can be done: (i) the number of Cu-NC-Fe linear atomic chains seen from the Cu site, labeled as CNchain in Table S5, acts as a key parameter, as it sets the degeneracy of the Cu-NC-Fe fragments and therefore can be considered an indicator of the linearity of the chain. The number is fixed to five as obtained from the XRD experiment in the pristine sample (also providing the excellent agreement between the theoretical and experimental EXAFS) but decreases upon lithiation, as revealed by Figure 6g. This in turn reveals that the accommodation of Li-ions induces a distortion of the lattice accompanied by a consistent structural disorder around copper. This agrees with DFT and XANES data. (ii) Even though the Fe-N-O triplet may allow one to estimate the evolution of the Fe-N-O angle, the little degeneracy of this MS path (1 over 6 with respect to the Fe center) did not permit a relevant and statistically significant evaluation of this angle while discharging the battery.


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




Page 16 of 34

Figure 6. Details of the EXAFS fitting for the pristine electrode: single EXAFS contributions for (a) Fe and (b) Cu K-edge, comparing in the bottom part the theoretical and experimental EXAFS signals; Fourier transforms at (c) Fe and (d) Cu K-edge. Relevant EXAFS fitting results: (e) First shell distances: Fe-C, C-N and Cu-N; (f) Pair EXAFS Debye-Waller factors; (g) Variation in Cu-N-C-Fe chains degeneracy during lithiation

3.3 DFT calculations Parameters optimization was carried out first on the pristine structure and the DFT-optimized geometry was compared to the CuNP structural model31 (cf. SI). The lowest energy value in the case of the pristine is obtained with two unpaired electrons, while the magnetization value, calculated as the integral over local spin densities in the atomic sphere, indicates that these two electrons are mainly located around copper atoms. Taking into consideration the electronic configuration of the atoms in the pristine lattice (Z = 2) and approximating the initial oxidation states of copper and iron to (+2), copper possesses a d9 electronic configuration, thus displaying one unpaired electron per copper, while iron has only paired electrons (d6 low-spin configuration). Cyanides and nitrosyl ligands in their linear form do not have unpaired electrons either. The obtained result is in line with the CuNP lattice model constituted by two formula units per cell. Optimized geometries for lithiated structures were obtained by inserting consecutive Li atoms in the lattice, corresponding to 0.5, 1.0 and 1.5 electrochemically inserted Li+ per formula unit, respectively. In fact, according to reaction (Eq. 2):

𝐶𝑢(𝐼𝐼)[𝐹𝑒(𝐼𝐼)(𝐶𝑁)5𝑁𝑂] + 𝐿𝑖 + + 𝑒 ―  𝐿𝑖𝐶𝑢(𝐼)[𝐹𝑒(𝐼𝐼)(𝐶𝑁)5𝑁𝑂]

(2)

the insertion of one lithium ion can be approximated to the insertion of a lithium atom (Li+ + e- = Li). Energy minimization for lithiated structures occurred for precise values of spin multiplet. Structures containing 0.5, 1.0 and 1.5 Li had a minimum of energy for total spin values equal to 1, 0 and 1, respectively. By analyzing the magnetization values, we can assume that the insertion of one Li atom/cell is responsible for the reduction of one Cu atom/cell (Eq. 2). Indeed, the CuII/CuI reduction modifies the copper electronic configuration from d9 to d10 resulting in a reduction of the number of unpaired electrons from 2 to 1 (1 Li/cell) and then to 0 (2 Li/cell). This condition matches the 16 ACS Paragon Plus Environment

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


experimental inserted Li during operando XRD and XAFS measurements. By adding extra Li atom/cell, that is 1.5 Li per formula unit, the structure is further reduced, although the metals are already in their low stable oxidation state. DFT calculations suggested a reduction of the nitrosyl ligand from a -NO+ oxidized state to the radical -NO, increasing the spin multiplicity due to the presence of one unpaired electron in the π* orbital.54 DFT optimized structures highlight a distortion in the lattice during discharge, especially in the copper environment (cf. SI). Individual nitrogen-copper bond lengths show a strong variation and the overall coordination number of Cu might be considered diminished (further details in SI). The relatively high amount of inserted Li and its significant interaction with the N-terminals of the equatorial cyanides may lead to the observed capacity fading, because the structure is not able to entirely extract the initially inserted amount. This may conduct to the loss of capacity experienced in the early stages of the cycling, after which a steady electrochemical performance is obtained, presumably due to the reversible insertion/extraction of a fraction of initially inserted lithium ions. When reduction of the NO occurs, a change in Fe-N-O bond angle is appreciated, which decreases from 178° to 142°, denoting a partial shift from a sp (-NO+) to a sp2 (-NO) hybridization of the nitrogen atom. Further calculations for higher number of lithium atoms (up to 3 Li equivalents) were carried out, even though this condition was not reached in the experiment, featuring a further reduction of the radical -NO to the anionic species -NO-. In this case, the angle of the Fe-N-O bond further decreases to 125°, which could be schematized by considering a planar sp2 hybridization of the nitrogen atom. The Fe-N-O bending suggested by DFT structure modelling is detectable also by computing vibrational frequencies (Figure 7a). While the cyanides frequencies are influenced by the reduction of copper sites, as already pointed out in reference,26 the wavenumber associated to the linearly coordinated NO drops of roughly 200 cm-1 for each reduction, i.e., -NO+/-NO and -NO/-NO- (cf. Figure 7a) as soon as the NO reduction and respective change in geometry take place. This outcome complements and agrees with the operando FT-IR experiment carried out in our previous study.26



Page 18 of 34

Figure 7. (a) DFT-calculated vibrational frequencies for pristine and lithiated structures: the left ordinate indicates the reduced species per unit cell (Z = 2), while the right ordinate the electrochemically inserted Li atoms. (b) Simulated XANES spectra at Fe K-edge (left) and Cu K-edge (right) during lithiation based on DFT retrieved atomic coordinates.

3.4 XANES calculations First, the accuracy of the CuNP structural model31 was verified with respect to the experimental CuNP data. A comparison between experimental and calculated spectra is reported in the Supplementary Material, evidencing that both calculated edges are representative of the experimental data in terms of main features in the pre-edge and edge regions. Concerning the post-edge region, the EXAFS analysis provides for a more reliable model.


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


XANES simulations during lithiation are displayed in Figure 7b: XANES spectra of DFT-derived lithiated structures containing 0.5, 1.0, and 1.5 Li are represented for both Fe and Cu K-edge. Both calculated edges present few striking features that fairly match the experimental data (cf. Figure 4). For instance, the 1s-4p transition in the Cu pre-edge increases in intensity during lithium insertion, lying 10.5 eV before the white line, in fair agreement with the experimental value (14.2 eV). In addition, the Fe K-edge is not much affected during lithiation and exhibits a decrease in intensity in the white line and following post-edge features, which further points out the agreement between calculated and experimental, thus making the DFT-derived geometries a good approximation of the real system.

4. Concluding Remarks A multi-technique approach was employed to thoroughly investigate the electrochemical lithiation process of copper nitroprusside. This PBA-based cathode material is characterized by the presence of the NO+ ligand that can be reduced during lithiation, becoming possibly the third redox centre in addition to the Fe and Cu metal site, giving rise to an extended specific capacity. A complementary operando experimental approach including XRD and XAFS was coupled to ab initio quantum mechanical modelling and presented in this work. This made possible to extract additional information on the system, which, in turn, can be considered an alternative approach to the well know oxygen redox strategy in designing new performance cathodes for rechargeable batteries. The results of this investigation indicate that both metals are reduced during discharge, simultaneously generating a contraction in the (ab) plane and an elongation in the c-direction along with the insertion of lithium ions in the lattice. The lithiation induces an increment in Debye-Waller factors for Cu-N bonds and a decreasing trend for the Cu-NC-Fe linear chains, indicating an augmented disorder in the system and a possible distortion provoked by lithium insertion. We have identified, by EXAFS, a key parameter related to the number of Cu-NC-Fe linear atomic chains, namely CNchain as it sets the degeneracy of the Cu-NC-Fe fragments and therefore can be considered an indicator of the linearity of the chain. According to DFT calculations, the structure is locally distorted upon lithiation around the copper site. After the CuII/CuI reduction, the nitrosyl ligand is also reduced, while the Fe-N-O angle is bent to 142°. The lattice deformation induced by the insertion of Li and the interaction with the N-terminals of the equatorial cyanides are thought to be the main cause of the observed capacity fading, while the



Page 20 of 34

reversible insertion/extraction of a fraction of initially inserted lithium ions is presumably the reason of the observed steady electrochemical performance after the early stages of cycling. In summary, a very good agreement with experimental data is reached throughout all calculations, making DFT-structures a good description of the system under study. The ligand electroactivity of the nitrogen, previously observed and supported by DFT calculation, represents a unique case in PBAs, contributing for an extra capacity. The new redox mechanism is unconventional, coupled to cationic activity, and alternative to oxygen reduction.

Acknowledgments XAFS and XRD measurements at Elettra – Sincrotrone Trieste have been done through the 20140337 and 20155185 projects, respectively (MG as PI). MG acknowledges the RFO funding of the University of Bologna. AM acknowledges Erasmus+ Mobility for rewarding a research period scholarship at Université Lille.

Appendix A. Supplementary data Supplementary data related to this article can be found at


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


References [1]

S. Chu, A. Majumdar, Opportunities and challenges for a sustainable energy future, Nature. 488 (2012) 294–303. doi:10.1038/nature11475.

[2]

J.M. Tarascon, M. Armand, Issues and challenges facing rechargeable lithium batteries, Nature. 414 (2001) 359–367. doi:10.1038/35104644.

[3]

J.B. Goodenough, A. Manthiram, A perspective on electrical energy storage, MRS Commun. 4 (2014) 135–142. doi:10.1557/mrc.2014.36.

[4]

J.B. Goodenough, Electrochemical energy storage in a sustainable modern society, Energy Environ. Sci. 7 (2014) 14–18. doi:10.1039/c3ee42613k.

[5]

A. Eftekhari, Lithium-Ion Batteries with High Rate Capabilities, ACS Sustain. Chem. Eng. 5 (2017) 2799–2816. doi:10.1021/acssuschemeng.7b00046.

[6]

A. Eftekhari, Low voltage anode materials for lithium-ion batteries, Energy Storage Mater. 7 (2017) 157–180. doi:10.1016/j.ensm.2017.01.009.

[7]

M. Saubanère, E. McCalla, J.M. Tarascon, M.L. Doublet, The intriguing question of anionic redox in high-energy density cathodes for Li-ion batteries, Energy Environ. Sci. 9 (2016) 984– 991. doi:10.1039/c5ee03048j.

[8]

W.E. Gent, K. Lim, Y. Liang, Q. Li, T. Barnes, S.-J. Ahn, K.H. Stone, M. McIntire, J. Hong, J.H. Song, Y. Li, A. Mehta, S. Ermon, T. Tyliszczak, D. Kilcoyne, D. Vine, J.-H. Park, S.-K. Doo, M.F. Toney, W. Yang, D. Prendergast, W.C. Chueh, Coupling between oxygen redox and cation migration explains unusual electrochemistry in lithium-rich layered oxides, Nat. Commun. 8 (2017) 2091. doi:10.1038/s41467-017-02041-x.

[9]

K. Luo, M.R. Roberts, R. Hao, N. Guerrini, D.M. Pickup, Y.-S. Liu, K. Edström, J. Guo, A. V. Chadwick, L.C. Duda, P.G. Bruce, Charge-compensation in 3d-transition-metal-oxide intercalation cathodes through the generation of localized electron holes on oxygen, Nat. Chem. 8 (2016) 684–691. doi:10.1038/nchem.2471.

[10]

Y. Lu, L. Wang, J. Cheng, J.B. Goodenough, Prussian blue: A new framework of electrode materials for sodium batteries, Chem. Commun. 48 (2012) 6544–6546. doi:10.1039/c2cc31777j.

[11]

L. Wang, Y. Lu, J. Liu, M. Xu, J. Cheng, D. Zhang, J.B. Goodenough, A superior low-cost cathode for a Na-Ion battery, Angew. Chemie - Int. Ed. 52 (2013) 1964–1967. 21 ACS Paragon Plus Environment


Page 22 of 34

doi:10.1002/anie.201206854. [12]

J. Song, L. Wang, Y. Lu, J. Liu, B. Guo, P. Xiao, J.J. Lee, X.Q. Yang, G. Henkelman, J.B. Goodenough, Removal of interstitial H2O in hexacyanometallates for a superior cathode of a sodium-ion battery, J. Am. Chem. Soc. 137 (2015) 2658–2664. doi:10.1021/ja512383b.

[13]

M. Pasta, C.D. Wessells, N. Liu, J. Nelson, M.T. McDowell, R.A. Huggins, M.F. Toney, Y. Cui, Full open-framework batteries for stationary energy storage, Nat. Commun. 5 (2014) 3007. doi:10.1038/ncomms4007.

[14]

C.D. Wessells, M.T. McDowell, S. V. Peddada, M. Pasta, R.A. Huggins, Y. Cui, Tunable reaction potentials in open framework nanoparticle battery electrodes for grid-scale energy storage, ACS Nano. 6 (2012) 1688–1694. doi:10.1021/nn204666v.

[15]

C.D. Wessells, S. V. Peddada, R.A. Huggins, Y. Cui, Nickel hexacyanoferrate nanoparticle electrodes for aqueous sodium and potassium ion batteries, Nano Lett. 11 (2011) 5421–5425. doi:10.1021/nl203193q.

[16]

M. Giorgetti, A Review on the Structural Studies of Batteries and Host Materials by X-Ray Absorption Spectroscopy, ISRN Mater. Sci. 2013 (2013) 1–22. doi:10.1155/2013/938625.

[17]

M. Giorgetti, E. Scavetta, M. Berrettoni, D. Tonelli, Nickel hexacyanoferrate membrane as a coated wire cation-selective electrode, Analyst. 126 (2001) 2168–2171. doi:10.1039/b107034g.

[18]

M. Giorgetti, D. Tonelli, M. Berrettoni, G. Aquilanti, M. Minicucci, Copper hexacyanoferrate modified electrodes for hydrogen peroxide detection as studied by X-ray absorption spectroscopy, J. Solid State Electrochem. 18 (2014) 965–973. doi:10.1007/s10008-013-2343-5.

[19]

O. Sato, T. Iyoda, A. Fujishima, K. Hashimoto, Photoinduced magnetization of a cobalt-iron cyanide, Science (80-. ). 272 (1996) 704–705. doi:10.1126/science.272.5262.704.

[20]

P.R. Bueno, D. Giménez-Romero, F.F. Ferreira, G.O. Setti, J.J. Garcia-Jareño, J. Agrisuelas, F. Vicente, Electrochromic switching mechanism of iron hexacyanoferrates molecular compounds: the role of fe2+(cn)6vacancies, J. Phys. Chem. C. 113 (2009) 9916–9920. doi:10.1021/jp901146w.

[21]

V.D. Neff, Some Performance Characteristics of a Prussian Blue Battery, J. Electrochem. Soc. 132 (1985) 1382. doi:10.1149/1.2114121. 22 ACS Paragon Plus Environment

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

[22]


R. Chen, H. Tanaka, T. Kawamoto, M. Asai, C. Fukushima, M. Kurihara, M. Watanabe, M. Arisaka, T. Nankawa, Preparation of a film of copper hexacyanoferrate nanoparticles for electrochemical removal of cesium from radioactive wastewater, Electrochem. Commun. 25 (2012) 23–25. doi:10.1016/j.elecom.2012.09.012.

[23]

M. Ciabocco, P. Cancemi, M.L. Saladino, E. Caponetti, R. Alduina, M. Berrettoni, Synthesis and antibacterial activity of iron-hexacyanocobaltate nanoparticles, J. Biol. Inorg. Chem. 23 (2018) 385–398. doi:10.1007/s00775-018-1544-x.

[24]

C.D. Wessells, R.A. Huggins, Y. Cui, Copper hexacyanoferrate battery electrodes with long cycle life and high power, Nat. Commun. 2 (2011) 550. doi:10.1038/ncomms1563.

[25]

A. Mullaliu, G. Aquilanti, P. Conti, J.R. Plaisier, M. Fehse, L. Stievano, M. Giorgetti, Copper Electroactivity in Prussian Blue-Based Cathode Disclosed by Operando XAS, J. Phys. Chem. C. 122 (2018) 15868–15877. doi:10.1021/acs.jpcc.8b03429.

[26]

A. Mullaliu, M.T. Sougrati, N. Louvain, G. Aquilanti, M.L. Doublet, L. Stievano, M. Giorgetti, The electrochemical activity of the nitrosyl ligand in copper nitroprusside: a new possible redox mechanism for lithium battery electrode materials?, Electrochim. Acta. 257 (2017) 364–371. doi:10.1016/j.electacta.2017.10.107.

[27]

J. Hannauer, J. Scheers, J. Fullenwarth, B. Fraisse, L. Stievano, P. Johansson, The Quest for Polysulfides in Lithium-Sulfur Battery Electrolytes: An Operando Confocal Raman Spectroscopy Study, ChemPhysChem. 16 (2015) 2755–2759. doi:10.1002/cphc.201500448.

[28]

L. Rebuffi, J.R. Plaisier, M. Abdellatief, A. Lausi, A.P. Scardi, Mcx: A synchrotron radiation beamline for X-ray diffraction line profile analysis, Zeitschrift Fur Anorg. Und Allg. Chemie. 640 (2014) 3100–3106. doi:10.1002/zaac.201400163.

[29]

G. Aquilanti, M. Giorgetti, R. Dominko, L. Stievano, I. Arčon, N. Novello, L. Olivi, Operando characterization of batteries using x-ray absorption spectroscopy: Advances at the beamline XAFS at synchrotron Elettra, J. Phys. D. Appl. Phys. 50 (2017) 074001. doi:10.1088/13616463/aa519a.

[30]

J. Rodríguez-Carvajal, Recent advances in magnetic structure determination by neutron powder diffraction, Phys. B Phys. Condens. Matter. 192 (1993) 55–69. doi:10.1016/09214526(93)90108-I.

[31]

A. Gómez, J. Rodríguez-Hernández, E. Reguera, Unique coordination in metal nitroprussides: 23 ACS Paragon Plus Environment


Page 24 of 34

The structure of Cu[Fe(CN)5NO]·2H2O and Cu[Fe(CN)5NO], J. Chem. Crystallogr. 34 (2004) 893–903. doi:10.1007/s10870-004-7724-2. [32]

M. Yao, K. Kuratani, T. Kojima, N. Takeichi, H. Senoh, T. Kiyobayashi, Indigo carmine: An organic crystal as a positive-electrode material for rechargeable sodium batteries, Sci. Rep. 4 (2014) 3650. doi:10.1038/srep03650.

[33]

K. Momma, F. Izumi, VESTA 3 for three-dimensional visualization of crystal, volumetric and morphology data, J. Appl. Crystallogr. 44 (2011) 1272–1276. doi:10.1107/S0021889811038970.

[34]

A. Filipponi, A. Di Cicco, C.R. Natoli, X-ray-absorption spectroscopy and n-body distribution functions in condensed matter. I. Theory, Phys. Rev. B. 52 (1995) 15122–15134. doi:10.1103/PhysRevB.52.15122.

[35]

A. Filipponi, A. Di Cicco, X-ray-absorption spectroscopy and n -body distribution functions in condensed matter. II. Data analysis and applications, Phys. Rev. B. 52 (1995) 15135–15149. doi:10.1103/PhysRevB.52.15135.

[36]

M. Giorgetti, M. Berrettoni, A. Filipponi, P.J. Kulesza, R. Marassi, Evidence of four-body contributions in the EXAFS spectrum of Na2Co[Fe(CN)6], Chem. Phys. Lett. 275 (1997) 108– 112. doi:10.1016/S0009-2614(97)00724-0.

[37]

M. Giorgetti, M. Berrettoni, Structure of Fe/Co/Ni hexacyanoferrate as probed by multiple edge X-ray absorption spectroscopy, Inorg. Chem. 47 (2008) 6001–6008. doi:10.1021/ic800289c.

[38]

M. Giorgetti, L. Guadagnini, D. Tonelli, M. Minicucci, G. Aquilanti, Structural characterization of electrodeposited copper hexacyanoferrate films by using a spectroscopic multi-technique approach, Phys. Chem. Chem. Phys. 14 (2012) 5527–5537. doi:10.1039/c2cp24109a.

[39]

L. Hedin, B. I. Lundqvist, S. Lundqvist, Local exchange-correlation potentials, Solid State Commun. 9 (1971) 537–541. doi:10.1016/0038-1098(71)90141-4.

[40]

M.O. Krause, J.H. Oliver, Natural widths of atomic K and L levels, K α X‐ray lines and several K L L Auger lines, J. Phys. Chem. Ref. Data. 8 (1979) 329–338. doi:10.1063/1.555595.

[41]

P. Conti, S. Zamponi, M. Giorgetti, M. Berrettoni, W.H. Smyrl, Multivariate Curve Resolution Analysis for Interpretation of Dynamic Cu K-Edge X-ray Absorption Spectroscopy Spectra for a Cu Doped V 2 O 5 Lithium Battery, Anal. Chem. 82 (2010) 3629–3635. doi:10.1021/ac902865h.

[42]

G. Kresse, J. Hafner, Ab initio molecular dynamics for liquid metals, Phys. Rev. B. 47 (1993) 24 ACS Paragon Plus Environment

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


558–561. doi:10.1103/PhysRevB.47.558. [43]

G. Kresse, J. Hafner, Ab initio molecular-dynamics simulation of the liquid-metalamorphoussemiconductor transition in germanium, Phys. Rev. B. 49 (1994) 14251–14269. doi:10.1103/PhysRevB.49.14251.

[44]

G. Kresse, J. Furthmüller, Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis set, Phys. Rev. B - Condens. Matter Mater. Phys. 54 (1996) 11169–11186. doi:10.1103/PhysRevB.54.11169.

[45]

G. Kresse, J. Furthmüller, Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set, Comput. Mater. Sci. 6 (1996) 15–50. doi:10.1016/0927-0256(96)00008-0.

[46]

D. Joubert, From ultrasoft pseudopotentials to the projector augmented-wave method, Phys. Rev. B - Condens. Matter Mater. Phys. 59 (1999) 1758–1775. doi:10.1103/PhysRevB.59.1758.

[47]

Y. Joly, X-ray absorption near-edge structure calculations beyond the muffin-tin approximation, Phys. Rev. B - Condens. Matter Mater. Phys. 63 (2001) 125120. doi:10.1103/PhysRevB.63.125120.

[48]

K. Hayakawa, K. Hatada, P. D’Angelo, S. Della Longa, C.R. Natoli, M. Benfatto, Full quantitative multiple-scattering analysis of X-ray absorption spectra: Application to potassium hexacyanoferrat(II) and -(III) complexes, J. Am. Chem. Soc. 126 (2004) 15618–15623. doi:10.1021/ja045561v.

[49]

A. Bianconi, M. Dell’Ariccia, P.J. Durham, J.B. Pendry, Multiple-scattering resonances and structural effects in the x-ray-absorption near-edge spectra of Fe II and Fe III hexacyanide complexes, Phys. Rev. B. 26 (1982) 6502–6508. doi:10.1103/PhysRevB.26.6502.

[50]

N. Kosugi, T. Yokoyama, H. Kuroda, Polarization dependence of XANES of square-planar Ni(CN)2−4 ion. A comparison with octahedral Fe(CN)4−6 and Fe(CN)3−6 ions, Chem. Phys. 104 (1986) 449–453. doi:10.1016/0301-0104(86)85034-0.

[51]

F. Tisato, C. Marzano, V. Peruzzo, M. Tegoni, M. Giorgetti, M. Damjanovic, A. Trapananti, A. Bagno, C. Santini, M. Pellei, M. Porchia, V. Gandin, Insights into the cytotoxic activity of the phosphane copper(I) complex [Cu(thp)4][PF6], J. Inorg. Biochem. 165 (2016) 80–91. doi:10.1016/j.jinorgbio.2016.07.007. 25 ACS Paragon Plus Environment


[52]

J. Chaboy, A. Muñoz-Páez, F. Carrera, P. Merkling, E.S. Marcos, Ab initio x-ray absorption study of copper K -edge XANES spectra in Cu(II) compounds, Phys. Rev. B - Condens. Matter Mater. Phys. 71 (2005) 134208. doi:10.1103/PhysRevB.71.134208.

[53]

N.C. Tomson, K.D. Williams, X. Dai, S. Sproules, S. Debeer, T.H. Warren, K. Wieghardt, Reevaluating the Cu K pre-edge XAS transition in complexes with covalent metal-ligand interactions, Chem. Sci. 6 (2015) 2474–2487. doi:10.1039/c4sc03294b.

[54]

D.M.P. Mingos, Nitrosyl Complexes in Inorganic Chemistry, Biochemistry and Medicine I, Springer Berlin Heidelberg, Berlin, Heidelberg, 2014. doi:10.1007/978-3-642-41187-8.


Page 26 of 34

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


TOC Graphic

ACS Paragon Plus Environment


Figure 1. Linear sweep on CuNP. Imposed negative polarization starting from OCP value to 1.5 V vs. Li+/Li at 0.1 mV s-1 361x153mm (96 x 96 DPI)


Page 28 of 34

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


Figure 2. (a) Rietveld refinement of pristine CuNP electrode. Inset shows the structure of CuNP. (b) (002) and (110) Miller planes at different lithiation steps (pristine, 0.4, 0.9 and 0.72 Li-inserted)



Figure 3. Evolution of the lattice parameters during discharge (left) and graphical representation of the lithiation process along the a and c axes (Fe: light- brown; Cu: blue; C: brown; N: light blue; O: red, Li: green)


Page 30 of 34

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


Figure 4. Selected operando XAFS spectra at Fe and Cu K-edge during discharge. Pristine and discharged states (0.4 and 0.9 Li-inserted) are shown 361x153mm (96 x 96 DPI)



Figure 5. (a) Pre-edge data analysis at the Cu K-edge: evolution of normalized area of three different features (A, B, C) during lithiation. Inset displays the fit performed for the pristine material. (b) Centroid shift for peak B during lithiation. (c) MCR-ALS analysis at the Cu K-edge: concentration profile plot and (d) pure spectral components (Cu2+ species is labeled as Species 1 and indicated in blue, while Cu+ species labeled as Species 2 in red)


Page 32 of 34

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


Figure 6. Details of the EXAFS fitting for the pristine electrode: single EXAFS contributions for (a) Fe and (b) Cu K-edge, comparing in the bottom part the theoretical and experimental EXAFS signals; Fourier transforms at (c) Fe and (d) Cu K-edge. Relevant EXAFS fitting results: (e) First shell distances: Fe-C, C-N and Cu-N; (f) Pair EXAFS Debye-Waller factors; (g) Variation in Cu-N-C-Fe chains degeneracy during lithiation



Figure 7. (a) DFT-calculated vibrational frequencies for pristine and lithiated structures: the left ordinate indicates the reduced species per unit cell (Z = 2), while the right ordinate the electrochemically inserted Li atoms. (b) Simulated XANES spectra at Fe K-edge (left) and Cu K-edge (right) during lithiation based on DFT retrieved atomic coordinates.


Page 34 of 34

Beyond Oxygen Redox Strategy in Designing Cathode Material for

Recommend Documents