Na Solid-State NMR Reveals the Local Na-Ion ... - ACS Publications

Article pubs.acs.org/cm

Ex Situ 23Na Solid-State NMR Reveals the Local Na-Ion Distribution in Carbon-Coated Na2FePO4F during Electrochemical Cycling Danielle L. Smiley and Gillian R. Goward* Department of Chemistry & Chemical Biology and Brockhouse Institute for Materials Research, McMaster University, 1280 Main Street West, Hamilton, Ontario L8S 4M1, Canada S Supporting Information *

ABSTRACT: The potential Na-ion cathode material Na2FePO4F is investigated here by ex situ 23Na solid-state nuclear magnetic resonance (ssNMR) in order to characterize the structure and ion mobility as a function of electrochemical cycling. The use of fast magic angle spinning (MAS) speeds of 65 kHz allows for the collection of high-resolution 23Na NMR spectra that reveal two unique peaks at +450 and −175 ppm, corresponding to the two crystallographically unique Na sites in the material of interest. Two-dimensional NMR exchange spectroscopy results reveal that chemical exchange between the Na ions residing in distinct environments has a maximum hopping rate of ∼200 Hz. The collection of one-dimensional NMR spectra as a function of electrochemical cycling reveals the reproducible formation of a new peak at +320 ppm in the 23Na NMR spectrum at all intermediate states of charge. The appearance of this resonance at +320 ppm is attributed to the fully oxidized (NaFePO4F) phase that is present even upon initial electrochemical oxidation. The simultaneous existence of both the pristine and oxidized phases suggest formation of two distinct phases upon charging, consistent with a two-phase desodiation mechanism. This two-phase arrangement of Na ions persists for multiple charge/discharge cycles and is congruent with high reversibility of Na (de)intercalation in Na2FePO4F cathodes. These findings imply that the Na2FePO4F framework is incredibly structurally stable with a robust intercalation process, despite a lack of ideal sodium-ion kinetics.

1. INTRODUCTION The discovery and development of commercial lithium-ion batteries (LIBs) has revolutionized the energy storage market with their successful implementation in portable electronics and, more recently, in commercially available electric vehicles. However, this overwhelming success has provoked concerns regarding limited global lithium supply and the consequently rising cost of lithium materials.1,2 Sodium-ion batteries (NIBs) were initially considered alongside Li before the breakthrough of commercial LIBs and have garnered attention of late as a viable alternative to lithium due to the relatively high global abundance of sodium resources coupled with the low cost of raw materials. While the larger, heavier Na ion ultimately decreases overall energy density for a given material vs Li, the cost and resource availability benefits largely outweigh this disadvantage.1,3−5 Moreover, depending on choice of Na insertion compound for the NIB cathode, electrochemical performance on par with Li has been observed in some cases, thus reinforcing the viability of this technology for commercial applications.2−5 The search for practical NIB electrode materials is ongoing, with the focus on cathode materials encompassing both layered oxide materials similar to the already commercially realized Li counterparts, and the similarly Li-analogous polyanionic frameworks including phosphates, fluorophosphates, and © 2016 American Chemical Society

sulfates. In particular, NaFeO2 has garnered attention as it has not only shown promising electrochemical behavior but has also taken advantage of the accessible Fe2+/Fe3+ redox couple that is highly sought after due to the low toxicity and low cost relative to other transition metals. Nevertheless, the discovery of a high-performance Fe-based polyanionic framework remains an attractive target, as these materials would be relatively inexpensive in addition to being structurally robust owing to their thermally and electrochemically stable three-dimensional network. Possible candidates include materials such as NaFePO 4 , 6 , 7 Na 3 Fe 2 (PO 4 ) 3 , 8 , 9 Na 2 FeP 2 O 7 , 1 0 , 1 1 and Na2FePO4F,12,13 with the latter three demonstrating reasonable electrochemical performance in NIBs. The layered fluorophosphate-type intercalation material A2FePO4F (A = Li, Na) has great potential as a cathode material for both LIBs and NIBs.12,13 The Na variant crystallizes in the orthorhombic Pbcn space group where facesharing FeO4F2 octahedra are connected via bridging F atoms to form bioctahedral units that are joined by PO4 anions to make up the layered structure (see Figure 1).12 Na+ ions are housed in two unique crystallographic positions, Na1 and Na2, Received: June 22, 2016 Revised: October 6, 2016 Published: October 6, 2016 7645

DOI: 10.1021/acs.chemmater.6b02539 Chem. Mater. 2016, 28, 7645−7656

Article

Chemistry of Materials

Figure 1. 23Na MAS NMR spectrum of Na2FePO4F at 65 kHz. The two crystallographic sites Na1 (teal) and Na2 (pink) depicted in the crystal structure are resolved in the NMR spectrum as peak A and peak B. The deconvolution of the isotropic resonances is shown in blue, with the full fit of the spectrum shown in red. Residual impurities from the synthetic process are visible as the gray peaks in the deconvoluted spectrum. All spinning sidebands are marked with an asterisk (*). In the inset the Na2FePO4F crystal structure is shown with FeO4F2 and PO4 polyhedra depicted as orange and yellow polyhedra respectively.

chemical behavior has prompted the modeling of diffusion pathways in an effort to identify the mechanism of Na-ion migration in this material.14 In 2013, Tripathi et al.14 were able to demonstrate the possibility for Na+ diffusion along the a- and c-axes, explaining the improvement in Na intercalation properties relative to the poorly performing NaFePO4 material and confirming the earlier predictions of Ellis et al.13 Their results suggest a network of Na ions that include Na+ hops between connecting Na1−Na1 and Na1−Na2 sites that form an extensive diffusion pathway throughout the material with sufficiently low activation energy barriers for alkali-ion hopping.14 Solid-state nuclear magnetic resonance (ssNMR) has been used to study a variety of cathode candidates for Li-ion batteries, and its unique ability to probe local properties of both the mobile and framework atoms has allowed it to be used to elucidate important mobility and structural information in a number of materials.24 Recently, the application of 23Na ssNMR to study electrode materials for NIBs has been implemented in some cases,8,25−35 although its use for characterizing Na-ion diffusion pathways in cathode materials of interest has been relatively unexplored. Unlike their LIB counterparts, NIB cathode materials have not yet been thoroughly investigated, and as such there exists a disparity between materials development and characterization of emerging materials. Na-23 ssNMR has shown promise as the

both of which exhibit a [6 + 1]-type coordination to four O atoms and two F atoms with each having an additional longrange contribution to the bond valence sum from a neighboring oxygen atom.12,13 The Na2 environment is found to be slightly more compact, having shorter Na−O and Na−F distances on average.13 As a result of this layered structural arrangement, the Na atoms are expected to diffuse readily within the layers with relatively limited motion between layers, predictions that were later corroborated by modeling data.14 Favorable electrochemical performance of this material was demonstrated by Nazar and co-workers when cycled vs Li metal, forming a novel Na−Li-ion battery hybrid, exhibited an average charge potential of 3.3 V vs Li and a reputable 115 mA h g−1 capacity.12,13 Chemical oxidation of the parent Na2FePO4F phase yielded a near-identical NaFePO4F structure with only the Na1 positions occupied, implying that the Na2 site houses the more mobile of the two ions.12,13 Following the initial proposal and synthesis of this material, several studies have expanded upon the work of Nazar and coworkers. A variety of modified synthetic methods for preparation of Na2FePO4F with improved electrochemical performance have been proposed, including new synthetic pathways for nanosizing the material, as well as an assortment of carbon-containing methods that both act as a particle growth inhibitor and deposit a conductive carbon coating for improved electronic conductivity.15−23 Moreover, this promising electro7646


Article


spinning sidebands were fit individually using a Gaussian/Lorentzian model. Details of the parameters from the deconvolution are provided in the Supporting Information. The quadrupolar parameters for Na2MgPO4F were fit using the same SOLA software fitting only the central quadrupole line shape. Powder X-ray diffraction data for the as-synthesized powders were collected on a PANalytical diffractometer using Cu Kα radiation from 2θ = 10° to 60°. Ex situ 2D diffraction data were collected at the MAX Diffraction Facility at McMaster University, with a Bruker Smart6000 CCD area detector and a Bruker three-circle D8 goniometer using a Rigaku RU200 Cu Kα rotating anode to collect a range of 5−98° (2θ). Air sensitive samples were packed into capillaries and data collected in the transmission geometry.

method of choice for providing useful information to explain ion dynamics and structural stability that lead to the observed electrochemical performance. The paramagnetic nature of the transition metal center in many of the Na cathode materials complicates their analysis by NMR, as the unpaired-electron spins that align in the external magnetic field can interact strongly with the NMR observed nucleus.36−38 This interaction gives rise to a series of spinning sidebands in typical 6,7Li and 23Na spectra of paramagnetic electrode materials, where the breadth of the spectrum is indicative of the strength of this through-space interaction and therefore the number of unpaired electrons at the paramagnetic metal center. In addition to anisotropic broadening, the transfer of unpaired-electron spin density to the atom of interest through a Fermi-contact interaction typically dictates chemical shifts in the NMR spectrum which acts to dramatically increase the otherwise narrow chemical shift range for Li or Na, an effect that can ultimately be exploited to resolve signals corresponding to unique crystallographic environments. In an attempt to minimize the effect of severe broadening due to strong electron−nuclear interactions, fast magic angle spinning (MAS) and low external magnetic fields are typically utilized. Na-23 MAS ssNMR is used here to probe Na+-ion dynamics and examine local structural changes as a function of electrochemical cycling with the goal of elucidating the Naion (de)insertion mechanism as well as assess the reversibility of the electrochemical process of Na2FePO4F in a Na-ion cell.

3. RESULTS AND DISCUSSION 3.1. 23Na NMR Characterization of Pristine Na2FePO4F Powders. Sodium-23 ssNMR with fast MAS is implemented here in an effort to characterize the structure and ion mobility in the layered fluorophosphate cathode material, Na2FePO4F. The 1D 23Na NMR spectrum is depicted in Figure 1, where two distinct chemical shifts dominate the spectrum at +440 and −175 ppm in a 0.85:1 integration ratio of the isotropic shifts, labeled sites A and B, respectively. Deviation of this ratio from 1:1 is rationalized by a marked difference in the CSA of the two sites as will be discussed below, making the ability to infer the relative populations from the isotropic resonances alone difficult. Evidence of impurities at −127 and −230 ppm are present in small quantities of 8 and 2% of the total 23Na signal, respectively, the latter being attributed to residual NaFePO4 from the solid-state synthesis. The more significant of the two impurities (8%) at −127 ppm is potentially the result of antisite disorder in Na2FePO4F, whereby Na and Fe atoms switch lattice positions, a phenomenon predicted for this material elsewhere.14,21 Finally, small amounts (∼2%) of diamagnetic NaF and Na2CO3 starting material remaining from the synthetic process are present around 0 ppm. None of the aforementioned impurities demonstrate any observable electrochemical activity, and thus their presence does not affect the results presented herein. Peaks A and B in the spectrum are assigned to the two crystallographically unique Na sites in the orthorhombic Na2FePO4F structure, with A corresponding to the more mobile Na2 site and B to the Na1 site in the crystal lattice. The robust assignment of paramagnetic shifts in related Li fluorophosphate cathode materials is notoriously difficult, owing to the lack of distinct 90° or 180° Li−O−Fe interactions that give rise to predictable 7Li chemical shifts in the layered transition metal oxide structures. While the mechanism of unpaired-electron spin density transfer from the transition metal to Na ion is expected to be the same as in analogous Li structures, prediction of the strength of that paramagnetic contribution due to incomplete orbital overlap can prove difficult. As such, absolute certainty of paramagnetic shift assignments is elusive in materials of this type without the help of complex DFT calculations.24,37,38 Nevertheless, correlating differences in the NMR spectrum to differences in the crystallographic sites housing the Na ions in question allows for the prediction of a reasonable site assignment in some cases. Examination of the spinning sidebands arising from each of the two unique Na sites in the 23Na NMR spectrum of Na2FePO4F/C reveals an observable difference in the size of the spinning sideband manifold for each site as revealed in the deconvolution of the spectrum in Figure 2. Site B has spinning sideband intensity spanning ∼500 kHz, while site A’s spinning sidebands extend well beyond a breadth of 1 MHz, providing

2. EXPERIMENTAL DETAILS 2.1. Preparation of Na2FePO4F and Na2MgPO4F. A two step solid-state synthetic route initially proposed by Kosova et al.18 was employed here to prepare pristine carbon-coated Na2FePO4F powders. Using this method, Fe(C2O4)·2H2O, Na2CO3, and NH4H2PO4 were initially mixed and ball-milled for 1 h in acetone before annealing at 575 °C for 2 h with a pause at 350 °C for 1 h. The resulting NaFePO4 powder was then ball-milled with a stoichiometric quantity of NaF and 3 wt % carbon black, followed by heating to 600 °C for 4 h under flowing Ar gas. The resulting Na2FePO4F/C samples were characterized by PXRD (see Supporting Information Figure S1). Preparation of Na2MgPO4F powder was carried out via the standard solid-state method reported in the literature,13 whereby Na2PO3F and MgCO3 were mixed stoichiometrically, ball-milled for 12 h, and heated initially to 350 °C for 6 h. The powder was then further heated to 625 °C for 6 h following an intermittent grinding. 2.2. Electrochemical Measurements. Cathodes were prepared by grinding the as-synthesized Na2FePO4F/C in a 75:15:10 (wt %) ratio with carbon black (CB) and poly(vinylidene fluoride) (PVDF) in NMP solvent to create a thick slurry. This slurry was then cast onto Al foil and dried under vacuum for 12 h at 120 °C. The dried cast was then punched into discs 1.27 cm in diameter that were incorporated into coin cells. Sodium-ion cells were prepared with 0.6 M NaPF6 in 30:70 EC/DMC (by volume) solution mixed in house and a Na-metal counter electrode. These cells were cycled on a multichannel potentiostat at a rate of C/20 using a standard galvanostatic cycling procedure. Cells extracted for ex situ NMR measurements were washed with acetonitrile in an Ar-filled glovebox before preparation for NMR. 2.3. Solid-State NMR and Powder X-ray Diffraction Measurements. Sodium-23 ssNMR spectra were acquired at a Larmor frequency of 79.39 MHz (1.5 μs, solid, π/2 pulse) on a Bruker wide bore 300 MHz spectrometer. A Bruker 1.3 mm double resonance probe was employed for fast MAS rates of 50−65 kHz. All spectra were referenced to 1 M NaCl at 0 ppm. For experiments with 25 kHz MAS spinning rates, a custom-built double resonance probe housing 1.8 mm diameter rotors was employed. Deconvolutions of the 23Na spectra were carried out in Bruker’s Topspin (version 3.1) solids lineshape analysis (SOLA) tool. Each peak and corresponding 7647


Article


Table 1. Na−F/Na−O Bond Lengths Used for Chemical Shift Assignment in Na2FePO4F13 atoms

distance (Å)

atoms

distance (Å)

Na1−F1 Na1−O2 Na1−O3 Na1−O1 Na1−O2 Na1−F2 Na1−O2

2.3525(11) 2.3641(13) 2.4109(13) 2.4149(13) 2.4643(13) 2.5171(12) 2.9467(14)

Na2−O2 Na2−F1 Na2−O3 Na2−F2 Na2−O4 Na2−O1 Na2−O1

2.2882(13) 2.3247(7) 2.3828(13) 2.4083(7) 2.4088(13) 2.5307(14) 2.8039(14)

av

2.4958

av

2.4496

chemical shift assignment in Na2FePO4F is ongoing and will be discussed elsewhere. In addition to the consideration of the effect of this strong paramagnetic interaction, 23Na is known to have a measurable quadrupole moment, particularly for nonspherically symmetric Na environments. Based on the known crystal structure for this material, the distortion of the octahedral environment around both Na1 and Na2 sites in Na2FePO4F is therefore expected to produce an observable effect due to a nonzero quadrupole coupling interaction. In order to investigate this effect in the absence of the line broadening introduced by interactions with paramagnetic transition metal atoms, the diamagnetic structural analogue, Na2MgPO4F, was synthesized and investigated by 23 Na NMR spectroscopy. Comparison of the 1D MAS NMR spectra of Na2MgPO4F and Na2FePO4F illustrates the extreme effect of the addition of a paramagnetic transition metal to the structure. The massive broadening of the isotropic resonances is coupled with a significant change in the spinning sideband intensities of the paramagnetic material relative to the diamagnetic equivalent (Figure 4a). In addition, the excellent chemical shift resolution of the two crystallographic sites is almost completely lost in the diamagnetic example, necessitating the use of multiple-quantum magic angle spinning (MQMAS) and/or deconvolution methods as shown in Figure 4 to identify the two unique Na line shapes in the 1D spectrum of Na2MgPO4F. A fit of the line shapes reveals two distinct Na environments with comparable quadrupole coupling constants

Figure 2. Deconvolution of the spinning sideband manifolds of (a) site A, (b) site B, and (c) the sum of all sites including impurities, in the 23 Na MAS spectrum of Na2FePO4F/C out to approximately ±3000 ppm.

evidence of a stronger paramagnetic contribution at site A. Analysis of the proximity of the Na crystallographic sites to nearby Fe atoms (shown in Figure 3) exhibits closer Na2−Fe contact relative to Na1−Fe interactions, an effect that is reflected in the difference of the spinning sideband intensities of sites A and B in the 23Na NMR spectrum. Furthermore, the larger magnitude of site A’s chemical shift can be understood by examining the Na−F and Na−O contacts for the Na1 and Na2 sites (Table 1), again revealing a more compact Na2 crystallographic environment arising from shorter Na−F and Na−O contacts on average. A more robust analysis of the

Figure 3. Depiction of the two Na environments in Na2FePO4F highlighting the closest Na−Fe contacts in Na2FePO4F out to the second coordination sphere for (a) Na1 (teal) and (b) Na2 (pink). The Fe atoms are labeled and depicted in orange with the phosphorus, oxygen, and fluorine atoms shown as yellow, red, and gray spheres, respectively. 7648


Article


Figure 4. (a) Comparison of the 1D 23Na MAS spectra of Na2FePO4F (bottom) to Na2MgPO4F (top) spinning at 25 kHz at 7.05 T. (b) Comparison of the isolated isotropic resonances for the Mg (red) vs the Fe (black) analogues, depicting the profound effect the addition of paramagnetism imparts on the peak widths. (c) Deconvolution of the 1D 23Na MAS spectrum of Na2MgPO4F depicting the two Na sites corresponding to the two unique crystallographic sites.

Alkali-ion cathode materials of this type typically exhibit fast nuclear relaxation of the site of interest (7Li, 6Li, or 23Na) owing primarily to the paramagnetic nature of the samples.24 The spin−lattice relaxation rates (T1) were measured for both sites in Na2FePO4F and were found to be 7.5 and 1.5 ms for sites A and B, respectively. As a result, the 2D EXSY mixing time, or period during which 23Na nuclei are allowed to exchange with one another, is in practice limited to a maximum of 5 ms before signal is completely lost to relaxation. This limits the measurable time scale for Na ion exchange via this method as compared to similar studies of analogous lithium polyanionic structures, where T1’s were more favorable, allowing for detection of comparatively slow ion exchange processes.39−42,44,46 Several methods were employed in an attempt to observe and quantify Na1−Na2 exchange in Na2FePO4F, including variable temperature experiments, 23Na 2D EXSY, and selective inversion NMR technique. Variable temperature 1D spectra show no evidence of coalescence between the unique crystallographic sites. Additionally, 2D EXSY experiments over a temperature range of 295−340 K exhibit no visible off-diagonal (cross-peak) intensity. In all cases, the NMR studies reveal that fast (>200 Hz) Na1−Na2 exchange does not occur in Na2FePO4F at room temperature. Select 2D EXSY spectra are depicted in Figure 5, where the lack of cross-peaks at the longest achievable mixing time is evident. All selective inversion studies only further corroborate the results of the 2D EXSY experiment, with no Na ion exchange on the observable time scale. Atomistic modeling data reported by Tripathi et al.14 in 2013 suggested that the available pathways for Na-ion

of 1.5 and 1.6 kHz for sites A and B, respectively, a result that might be expected due to the similar coordination motifs of the Na1 and Na2 crystallographic environments. When overlaying the 23Na NMR spectra of the Mg and Fe analogues, it becomes clear that the chemical shifts, line widths, and spinning sideband manifolds are largely independent of the quadrupolar interaction in this material, and thus the interpretation of the spectra can be done by primarily considering the effect of the paramagnetic Fe ion on the Na environments. 3.2. Investigation of Na−Na Chemical Exchange in Pristine Na2FePO4F/C by 23Na NMR. Ion exchange rates are typically a good predictor of cathode success, because they provide insight into local ion dynamics in the structure independent of particle size and thus grain boundary effects. Quantification of these exchange rates in Li polyanionic materials by6,7 Li NMR spectroscopy has been very successful for a number of materials via the use of both one- and twodimensional NMR methods such as selective inversion (SI)39−43 and 2D exchange spectroscopy (2D EXSY).44,45 In theory, these methods can be extended to the study of similar Na cathode materials, where 23Na ssNMR can be utilized to probe the ion exchange between crystallographically unique Na sites. For both 1D and 2D NMR exchange experiments, a simple requirement is that the lifetime of spin polarization is long enough to capture changes in resonance frequency resulting from chemical exchange. In short, the exchange rates must be faster than or equal to the inherent nuclear spin relaxation times for the ions of interest. In many cases, this significantly limits the time scale of exchange that can be probed by these methods and is largely sample dependent. 7649


Article


cross-peaks in the 2D EXSY spectrum up to mixing times of 5 ms. It is noted that there is evidence of very slow alkali ion exchange in a similar Li fluorophosphate cathode material, Li2VPO4F,39 where the Li−Li exchange time scale was determined to be on the order of 20 Hz, a rate too slow to be captured by the analogous 23Na 2D EXSY experiment in the material of interest here. While this ion exchange rate limit is slow relative to the fast Li-ion diffusion in other polyanionic lithium phosphate materials such as Li3V2(PO4)347 and Li3Fe2(PO4)3,42,45 it is in good agreement with results for other lithium fluorophosphates structures, such as LiVPO4F,39 where the addition of the fluoride anion appears to impose an attractive hold on the Na atom, thus impeding alkali-ion mobility. The comparatively sluggish ion kinetics in Na2FePO4F at typical operational temperatures would certainly contribute to its lack of commercial viability for high-rate applications. 3.3. Ex Situ 23Na NMR and PXRD of Electrochemically Cycled Na2FePO4F/C. The success of a cathode is contingent not only on the ion diffusion rates but also the ability of the structure to withstand the removal of Na ions during electrochemical cycling. As such, efforts to understand changes to the local Na2FePO4F structure during charge and discharge are undertaken here utilizing ex situ ssNMR techniques. In particular, Na2FePO4F is considered to be an attractive cathode candidate for Na-ion batteries as its lack of significant structural change during cycling ultimately yields high-capacity retention over many cycles.16,17,19,22,23 As a result, a clear understanding of the (re)arrangement of Na ions within the layered structure during the charge/discharge process is beneficial. Na-23 ssNMR is an ideal tool for elucidating the Na+ (de)intercalation mechanism throughout the cycling process as we have demonstrated the sensitivity of the Na nucleus to the local environment in Na2FePO4F, thus allowing for small structural and/or electronic changes to be readily captured by this method. NIBs containing Na2FePO4F/C cathodes were assembled and cycled to various states of charge/discharge, demonstrating voltage profiles with two distinct regions centered at 2.9 and 3.1 V vs Na upon charging, in excellent agreement with literature electrochemical tests.12,13 Cycled coin cells were carefully deconstructed and the recovered cathodes handled exclusively in a dry Ar atmosphere with no exposure to moisture or air during testing. The 1D 23Na NMR spectrum of each cycled cathode (Figure 6) was collected at a MAS speed of 65 kHz to ensure resolution of the distinct Na resonances. Upon partially charging the cathode a new Na site centered at +350 ppm (labeled here as site C) is evident in the 23Na NMR spectrum coupled with the partial loss in intensity of the original A and B sites that correspond to Na ions in the original Na2FePO4F phase. This dramatically different chemical shift is not thought to arise from Na ions in an entirely new crystallographic environment but rather is indicative of Na ions within an environment now surrounded by Fe3+ ions rather than Fe2+ as a result of the electrochemical oxidation process. Further cycling to a fully charged state yields a 23Na NMR spectrum consisting of only the new site C and a small amount of electrochemically inactive impurities at −127 and −230 ppm remaining from the solid-state synthesis, as discussed above. Upon subsequently discharging the cell, site C is completely diminished accompanied by the reappearance of sites A and B including a full recovery of the initial relative intensities. Over the course of the full charge/discharge cycle, the ratio of site A to site B remains relatively constant with no

Figure 5. 23Na 2D EXSY spectra of Na2FePO4F at 335 K with mixing times of (a) 0 ms and (b) 5 ms. As evidenced by the lack of offdiagonal cross-peak intensity, no exchange between sites Na1 (peak B) and Na2 (peak A) is occurring on the time scale probed here.

diffusion in Na2FePO4F include a combination of Na1−Na1 and Na1−Na2 site hops. The results of the 2D EXSY experiment presented here indicate that either ion motion occurs via an alternate pathway that does not involve Na1−Na2 exchange or that the ion diffusion is slow in this material overall. In the case of the latter, where the ion mobility is inherently limited, the results obtained from 2D EXSY NMR method allow for the determination of an ion hopping limit, whereby Na1−Na2 exchange is expected to occur at a rate of less than 200 Hz at 340 K based on the lack of observable 7650


Article


Figure 6. Na-ion cells cycled to various points along the first charge/discharge cycle, extracted and analyzed by 23Na MAS NMR spectroscopy. Colored boxes on the electrochemical curve indicate positions at which samples were extracted for NMR analysis. At the midpoint of the first charge cycle (green) both the original A and B sites are evident in addition to a novel site C. The original A and B peaks in the spectrum of pristine Na2FePO4F fully disappear upon charging (blue) and are fully recovered by the end of discharge (red). The peaks surrounding 0 ppm that appear in electrochemically cycled samples correspond to a combination of starting materials from the synthesis and residual electrolyte. Their varying relative intensities are a result of variations in the washing and drying procedure of the cathodes.

NaFe3+FePO4F phases. Site assignment is relatively straightforward in the end members of the redox process, with sites A and B corresponding to ions in positions Na2 and Na1, respectively, in Na2FePO4F, influenced exclusively by Fe2+ TM ions, and site C consistent with Na+ occupying the Na1 position (the sole

more than a 15% variance attributable to the poor signal-tonoise of spectra at intermediate states of charge. This chemical shift behavior can be explained simply by consideration of possible Na environments in the pristine Na2Fe2+FePO4F, partially oxidized Na2−xFe2+/3+PO4F, and fully oxidized 7651


Article


Figure 7. Proposed structural changes that occur in Na2FePO4F upon a charge/discharge cycle in a Na-ion cell. Na2 (pink) ions leave the structure via a two-phase mechanism, leaving the Na1 (teal) site fully occupied, whereby at the midpoint (2) of the charging process there are two distinct regions: Na2FePO4F and fully oxidized NaFePO4F. Upon fully charging the cell only NaFePO4F remains with subsequent full recovery of the original Na2FePO4F phase after discharge (4).

remaining Na site as per the literature13) in NaFePO4F, with the chemical shift dominated by paramagnetic interactions with only the Fe3+ ion. The most informative results arise from partially charged/discharged samples corresponding to mixed Fe2+/Fe3+ species, where the apparent coexistence of both the original Na2FePO4F phase and the fully oxidized NaFePO4F phase is indicative of a so-called two-phase desodiation mechanism occurring at all states of charge. There is no evidence of Na ions in a mixed TM environment resulting from a combination of partial Fe2+ and partial Fe3+ contributions that change with the state of charge, as might be expected for a solid solution-like mechanism. Visual representation of the proposed mechanism is depicted in Figure 7 where the Na occupancies in the crystal structure at varying states of charge as determined by 23 Na NMR are correlated to the electrochemical cycling curve. This proposal of a two-phase extraction process of the Na ions from Na2FePO4F is in contrast to the original assumption that this material desodiates via a solid-solution mechanism.13 The previous interpretation was based in part on X-ray diffraction results reported by Nazar and co-workers13 for Na1.5FePO4F and NaFePO4F phases generated by chemical oxidation, which demonstrated the presence of a single phase in each case, with the latter exhibiting the full removal of Na ions from the Na2 site and otherwise full retention of the parent structure. Moreover, the apparent sloping voltage profile in the electrochemical curves is often suggestive of solid-solution formation. The possibility for this solid-solution behavior was thus carefully considered in light of the NMR results obtained; however, there is little justification for a true solid-solution based on the NMR data shown here. In the event of solidsolution formation, both Fe2+ and Fe3+ would surround the Na ions at all intermediate states of charge, resulting in a 23Na chemical shift that varies with relative Fe2+/Fe3+ content. For instance, the chemical shift of Na1 in the pristine material would be expected to gradually migrate to the final Fe3+-derived chemical shift, experiencing all intermediate possibilities along the way. This process would be easily traceable via 23Na NMR spectroscopy, where the state of charge could be estimated

simply by measuring the chemical shift of the Na1 ion in a partially oxidized environment. Rather than such a chemical shift evolution, we observe a single new site, at 350 ppm, that changes in intensity but remains at a constant chemical shift irrespective of the state of charge. Therefore, it is concluded that Fe2+ and Fe3+ domains are wholly isolated from each other, resulting in 23Na chemical shifts dominated either by interactions with Fe2+ OR Fe3+, but not both simultaneously. This finding is in good agreement with Mössbauer results for partially charged/discharged samples, suggesting Fe2+/Fe3+ ions are isolated from one another and do not form a mixed valence Fe2.5+-type state.13,48 In addition to ssNMR measurements, ex situ powder X-ray diffraction (PXRD) at various states of charge was performed for electrochemically cycled Na2FePO4F cathodes. The diffraction results for an uncycled cathode as well as three samples cycled to 3.16, 3.4, and 3.5 V are depicted in Figure 8. A full interpretation including a refinement of the PXRD data was not possible owing to the nature of the samples; however the analysis of some key peaks in the diffraction pattern provide some insight into the composition of the partially charged samples. As was observed via 23Na NMR, samples at partial states of charge (3.16, 3.4, and 3.56 V) along the initial charging curve appear to be comprised of a mixture of distinct phases, nominally Na2FePO4F, NaFePO4F, and an NaFePO4 impurity present in the uncycled cathodes. Closer analysis of two regions (Figure 1b,c) reveal the changing ratio of Na2FePO4F to fully oxidized NaFePO4F as the material is desodiated electrochemically. At intermediate states of charge, two unique sets of reflections are observed, one for Na2FePO4F and the other for the fully desodiated NaFePO4F phase, rather than a gradual change in 2θ value as the material is cycled. This coexistence of two unique phases during charging is in excellent agreement with the results observed by ssNMR technique and, in combination with the high-resolution 23Na ssNMR method, provides further evidence for the two-phase desodiation mechanism proposed for Na2FePO4F here. 7652


Article


Figure 8. (a) Ex situ powder X-ray diffraction for cathode materials at various compositions; an uncycled Na2FePO4F cathode and three cathodes charged to 3.16, 3.4, and 3.56 V. The latter three samples are expected to be partially desodiated and thus contain both Fe3+ and Fe2+ ions. (b) Closer view of two particularly informative regions where it is evident that at partial states of charge there are two distinct components changing in intensity relative to each other.

measurable time scale via this experiment. As such, it is deduced that despite excellent structural stability as demonstrated by the robust nature of the Na chemical shift upon multiple electrochemical cycles, the material ultimately suffers from relatively limited intrinsic ion motion, preventing it from competing with current state-of-the-art materials. This is not to say that attempts to enhance ionic conductivity are unwarranted, and strategies such as carefully controlling particle size may help to maximize the diffusion through a real electrode. Indeed, the structural stability of the fluorophosphates is highly attractive, and therefore efforts to obtain commercially viable materials based on this structure should continue to be pursued.

Further NMR measurements of cathodes cycled beyond the initial charge/discharge cycle reveal that this two-phase extraction mechanism does indeed persist over multiple cycles (see Supporting Information Figure S5), appearing to be highly reversible. The robust nature of the fluorophosphate framework is evident, with the local Na environment remaining unchanged even after more than 40 full charge/discharge cycles. Finally, in order to confirm that the appearance of a new site in the NMR spectra of electrochemically cycled cathodes is in fact a new Na environment and not a result of chemical exchange between sites A and B, 1D variable temperature NMR and 2D EXSY experiments of partially charged Na1+xFePO4F/C samples were undertaken and are depicted in Figure 9. No evidence of coalescence or cross-peak intensity is observed, confirming that site C does not arise from partial A−B exchange, and is indeed a unique Na environment. In addition, these experiments allowed for the probing of ion exchange rates at intermediate states of charge, where the results are analogous to that of the pristine phase, wherein any exchange between unique sites is limited to motion slower than that of the

4. CONCLUSIONS The layered fluorophosphate cathode material Na2FePO4F was synthesized and investigated here by ssNMR and ex situ X-ray diffraction techniques. Variable temperature 2D EXSY studies confirmed that chemical exchange between Na+ atoms residing 7653



■

Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemmater.6b02539. Powder X-ray diffraction data of as-prepared Na2FePO4F, deconvolution of the spinning sideband manifolds in the one-dimensional 23Na NMR spectrum, and ex situ NMR spectra acquired following extended cycling (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Phone: (905)-525-9140 x24176. Fax: (905) 522-2509. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We are grateful to Victoria Jarvis at the MAX Diffraction Facility at McMaster University for her assistance with the powder X-ray diffraction measurements and helpful discussions. G.R.G. thanks the Natural Science and Engineering Research Council of Canada (NSERC) for financial support through the Discovery Grant program. D.L.S. is grateful to the Ontario Graduate Scholarship program for financial support.

■

REFERENCES

(1) Ellis, B. L.; Nazar, L. F. Current Opinion in Solid State and Materials Science. Curr. Opin. Solid State Mater. Sci. 2012, 16, 168− 177. (2) Kim, S.-W.; Seo, D.-H.; Ma, X.; Ceder, G.; Kang, K. Electrode Materials for Rechargeable Sodium-Ion Batteries: Potential Alternatives to Current Lithium-Ion Batteries. Adv. Energy Mater. 2012, 2, 710−721. (3) Palomares, V.; Serras, P.; Villaluenga, I.; Hueso, K. B.; CarreteroGonzález, J.; Rojo, T. Na-Ion Batteries, Recent Advances and Present Challenges to Become Low Cost Energy Storage Systems. Energy Environ. Sci. 2012, 5, 5884. (4) Yabuuchi, N.; Kubota, K.; Dahbi, M.; Komaba, S. Research Development on Sodium-Ion Batteries. Chem. Rev. 2014, 114, 11636− 11682. (5) Kundu, D.; Talaie, E.; Duffort, V.; Nazar, L. F. The Emerging Chemistry of Sodium Ion Batteries for Electrochemical Energy Storage. Angew. Chem., Int. Ed. 2015, 54, 3431−3448. (6) Prosini, P. P.; Cento, C.; Masci, A.; Carewska, M. Sodium Extraction From Sodium Iron Phosphate with a Maricite Structure. Solid State Ionics 2014, 263, 1−8. (7) Zhu, Y.; Xu, Y.; Liu, Y.; Luo, C.; Wang, C. Comparison of Electrochemical Performances of Olivine NaFePO4 In Sodium-Ion Batteries and Olivine LiFePO4 In Lithium-Ion Batteries. Nanoscale 2013, 5, 780−787. (8) Trad, K.; Carlier, D.; Croguennec, L.; Wattiaux, A.; Lajmi, B.; Ben Amara, M.; Delmas, C. A Layered Iron(III) Phosphate Phase, Na 3Fe 3(PO 4) 4: Synthesis, Structure, and Electrochemical Properties as Positive Electrode in Sodium Batteries. J. Phys. Chem. C 2010, 114, 10034−10044. (9) Masquelier, C.; Wurm, C.; Rodríguez-Carvajal, J.; Gaubicher, J.; Nazar, L. A Powder Neutron Diffraction Investigation of the Two Rhombohedral NASICON Analogues: Γ-Na 3 Fe 2 (Po 4 ) 3 And Li3Fe2(Po4)3. Chem. Mater. 2000, 12, 525−532. (10) Barpanda, P.; Ye, T.; Nishimura, S.-I.; Chung, S. C.; Yamada, Y.; Okubo, M.; Zhou, H.; Yamada, A. Sodium Iron Pyrophosphate: a Novel 3.0V Iron-Based Cathode for Sodium-Ion Batteries. Electrochem. Commun. 2012, 24, 116−119.

Figure 9. (a) 23Na 2D EXSY at 335 K with a 2 ms mixing time and (b) variable temperature 1D NMR spectra of partially charged NaxFePO4F. Despite resolution of both pristine A and B sites as well as the oxidized C site, no Na−Na exchange is observed between any of the sites. The peak at ∼0 ppm corresponds to a combination of residual starting materials and electrolyte not completely removed from the cathode washing procedure.

in unique crystallographic sites is not rapid, and if it does indeed occur as predicted, the ion hopping rate does not exceed 200 Hz in either the pristine or partially desodiated materials. Ex situ NMR studies of electrochemically cycled cathodes yielded novel information regarding the deintercalation mechanism in Na2FePO4F, where the results are wholly indicative of a two-phase desodiation mechanism in this material. The results from ex situ NMR experiments were corroborated by PXRD data of partially charged cathodes that suggested two distinct phases present in varying quantities at several states of charge. All results reported here indicate that while the two-phase mechanism of Na intercalation is highly reversible and the framework is structurally very robust, less than ideal ion dynamics is likely what ultimately limits this cathode material, particularly at fast cycling rates. Efforts to optimize electrodes derived from this framework should therefore be focused on improvement of Na ion diffusion properties. 7654


Article

Chemistry of Materials (11) Barpanda, P.; Liu, G.; Ling, C. D.; Tamaru, M.; Avdeev, M.; Chung, S. C.; Yamada, Y.; Yamada, A. a2FeP2O7: A Safe Cathode for Rechargeable Sodium-Ion Batteries. Chem. Mater. 2013, 25, 3480− 3487. (12) Ellis, B. L.; Makahnouk, W. R. M.; Makimura, Y.; Toghill, K.; Nazar, L. F. A Multifunctional 3.5 v Iron-Based Phosphate Cathode for Rechargeable Batteries. Nat. Mater. 2007, 6, 749−753. (13) Ellis, B. L.; Makahnouk, W. R. M.; Rowan-Weetaluktuk, W. N.; Ryan, D. H.; Nazar, L. F. Crystal Structure and Electrochemical Properties of A2MPO4F Fluorophosphates (a = Na, Li; M = Fe, Mn, Co, Ni). Chem. Mater. 2010, 22, 1059−1070. (14) Tripathi, R.; Wood, S. M.; Islam, M. S.; Nazar, L. F. Na-Ion Mobility in Layered Na2FePO4F and Olivine Na[Fe,Mn]PO4. Energy Environ. Sci. 2013, 6, 2257. (15) Kawabe, Y.; Yabuuchi, N.; Kajiyama, M.; Fukuhara, N.; Inamasu, T.; Okuyama, R.; Nakai, I.; Komaba, S. Synthesis and Electrode Performance of Carbon Coated Na2FePO4F for Rechargeable Na Batteries. Electrochem. Commun. 2011, 13, 1225−1228. (16) Langrock, A.; Xu, Y.; Liu, Y.; Ehrman, S.; Manivannan, A.; Wang, C. Carbon Coated Hollow Na2FePO4F Spheres for Na-Ion Battery Cathodes. J. Power Sources 2013, 223, 62−67. (17) Song, W.; Ji, X.; Wu, Z.; Zhu, Y.; Yao, Y.; Huangfu, K.; Chen, Q.; Banks, C. E. Na2FePO4F Cathode Utilized in Hybrid-Ion Batteries: a Mechanistic Exploration of Ion Migration and Diffusion Capability. J. Mater. Chem. A 2014, 2, 2571−2577. (18) Kosova, N. V.; Podugolnikov, V. R.; Devyatkina, E. T.; Slobodyuk, A. B. Structure and Electrochemistry of NaFePO4 And Na2FePO4F Cathode Materials Prepared via Mechanochemical Route. Mater. Res. Bull. 2014, 60, 849−857. (19) Brisbois, M.; Krins, N.; Hermann, R. P.; Schrijnemakers, A.; Cloots, R.; Vertruyen, B.; Boschini, F. Spray-Drying Synthesis of Na2FePO4F/Carbon Powders for Lithium-Ion Batteries. Mater. Lett. 2014, 130, 263−266. (20) Brisbois, M.; Caes, S.; Sougrati, M. T.; Vertruyen, B.; Schrijnemakers, A.; Cloots, R.; Eshraghi, N.; Hermann, R. P.; Mahmoud, A.; Boschini, F. Na2FePO4F/Multi-Walled Carbon Nanotubes for Lithium-Ion Batteries_ Operando Mössbauer Study of SprayDried Composites. Sol. Energy Mater. Sol. Cells 2016, 148, 67−72. (21) Law, M.; Ramar, V.; Balaya, P. Synthesis, Characterisation and Enhanced Electrochemical Performance of Nanostructured Na2FePO4F for Sodium Batteries. RSC Adv. 2015, 5, 50155−50164. (22) Yan, J.; Liu, X.; Li, B. Nano-Assembled Na2FePO4F/Carbon Nanotube Multi-Layered Cathodes for Na-Ion Batteries. Electrochem. Commun. 2015, 56, 46−50. (23) Cui, D.; Chen, S.; Han, C.; Ai, C.; Yuan, L. Carbothermal Reduction Synthesis of Carbon Coated Na2FePO4F for Lithium Ion Batteries. J. Power Sources 2016, 301, 87−92. (24) Grey, C. P.; Dupré, N. NMR Studies of Cathode Materials for Lithium-Ion Rechargeable Batteries. Chem. Rev. 2004, 104, 4493− 4512. (25) Carlier, D.; Blangero, M.; Ménétrier, M.; Pollet, M.; Doumerc, J.-P.; Delmas, C. Sodium Ion Mobility in NaXCoO2 (0.6 < X < 0.75) Cobaltites Studied by 23Na MAS NMR. Inorg. Chem. 2009, 48, 7018− 7025. (26) Stallworth, P.; Greenbaum, S. G.; Ma, Y.; Ding, L.; Doeff, M. M.; Visco, S. 23Na-NMR Studies of NaxCoO2 Cathode Materials. Solid State Ionics 1996, 86−88, 797−803. (27) Ma, J.; Bo, S.-H.; Wu, L.; Zhu, Y.; Grey, C. P.; Khalifah, P. G. Ordered and Disordered Polymorphs of Na(Ni2/3Sb1/3)O2: Honeycomb-Ordered Cathodes for Na-Ion Batteries. Chem. Mater. 2015, 27, 2387−2399. (28) Liu, Z.; Hu, Y.-Y.; Dunstan, M. T.; Huo, H.; Hao, X.; Zou, H.; Zhong, G.; Yang, Y.; Grey, C. P. Local Structure and Dynamics in the Na Ion Battery Positive Electrode Material Na3V2(Po4)2F3. Chem. Mater. 2014, 26, 2513−2521. (29) Serras, P.; Palomares, V.; Alonso, J.; Sharma, N.; López del Amo, J. M.; Kubiak, P.; Fdez-Gubieda, M. L.; Rojo, T. Electrochemical Na Extraction/Insertion of Na 3V 2O 2 X(PO 4) 2F 3−2 X. Chem. Mater. 2013, 25, 4917−4925.

(30) Carlier, D.; Cheng, J. H.; Berthelot, R.; Guignard, M.; Yoncheva, M.; Stoyanova, R.; Hwang, B. J.; Delmas, C. The P2Na2/3Co2/3Mn1/3O2 Phase: Structure, Physical Properties and Electrochemical Behavior as Positive Electrode in Sodium Battery. Dalton Trans. 2011, 40, 9306−9307. (31) Cabana, J.; Chernova, N. A.; Xiao, J.; Roppolo, M.; Aldi, K. A.; Whittingham, M. S.; Grey, C. P. Study of the Transition Metal Ordering in Layered NaxNiX/2Mn 1−X/2O 2(2/3 ≤ X≤ 1) and Consequences of Na/Li Exchange. Inorg. Chem. 2013, 52, 8540−8550. (32) Chen, H.; Hao, Q.; Zivkovic, O.; Hautier, G.; Du, L.-S.; Tang, Y.; Hu, Y.-Y.; Ma, X.; Grey, C. P.; Ceder, G. Sidorenkite (Na3MnPO4CO3): a New Intercalation Cathode Material for Na-Ion Batteries. Chem. Mater. 2013, 25, 2777−2786. (33) Billaud, J.; Clément, R. J.; Armstrong, A. R.; Canales-Vázquez, J.; Rozier, P.; Grey, C. P.; Bruce, P. G. B-NaMnO2: A High-Performance Cathode for Sodium-Ion Batteries. J. Am. Chem. Soc. 2014, 136, 17243−17248. (34) Singh, G.; Lôpez del Amo, J. M.; Galceran, M.; Pérez-Villar, S.; Rojo, T. Structural Evolution During Sodium Deintercalation/ Intercalation in Na2/3[Fe1/2Mn1/2]O2. J. Mater. Chem. A 2015, 3, 6954−6961. (35) Clément, R. J.; Billaud, J.; Robert Armstrong, A.; Singh, G.; Rojo, T.; Bruce, P. G.; Grey, C. P. Structurally Stable Mg-Doped P2Na2/3Mn1−YMgYO2 Sodium-Ion Battery Cathodes with High Rate Performance: Insights From Electrochemical, NMR and Diffraction Studies. Energy Environ. Sci. 2016, 9, 3240−3251. (36) Castets, A.; Carlier, D.; Trad, K.; Delmas, C.; Ménétrier, M. Analysis of the 7Li NMR Signals in the Monoclinic Li3Fe2(Po4)3 And Li3V2(Po4)3 Phases. J. Phys. Chem. C 2010, 114, 19141−19150. (37) Kim, J.; Middlemiss, D. S.; Chernova, N. A.; Zhu, B. Y. X.; Masquelier, C.; Grey, C. P. Linking Local Environments and Hyperfine Shifts: a Combined Experimental and Theoretical 31P and 7 Li Solid-State NMR Study of Paramagnetic Fe(III) Phosphates. J. Am. Chem. Soc. 2010, 132, 16825−16840. (38) Middlemiss, D. S.; Ilott, A. J.; Clément, R. J.; Strobridge, F. C.; Grey, C. P. Density Functional Theory-Based Bond Pathway Decompositions of Hyperfine Shifts: Equipping Solid-State NMR to Characterize Atomic Environments in Paramagnetic Materials. Chem. Mater. 2013, 25, 1723−1734. (39) Davis, L. J. M.; Ellis, B. L.; Ramesh, T. N.; Nazar, L. F.; Bain, A. D.; Goward, G. R. 6Li 1D EXSY NMR Spectroscopy: a New Tool for Studying Lithium Dynamics in Paramagnetic Materials Applied to Monoclinic Li2VPO4F. J. Phys. Chem. C 2011, 115, 22603−22608. (40) Davis, L. J. M.; Goward, G. R. Differentiating Lithium Ion Hopping Rates in Vanadium Phosphate Versus Vanadium Fluorophosphate Structures Using 1D 6Li Selective Inversion NMR. J. Phys. Chem. C 2013, 117, 7981−7992. (41) Smiley, D. L.; Tessaro, M. Z.; He, X.; Goward, G. R. Correlation of Electrochemical Performance with Lithium Environments and Cation Dynamics in Li2(Mn1−YFeY)P2O7Using 6Li Solid-State NMR. J. Phys. Chem. C 2015, 119, 16468−16474. (42) Smiley, D. L.; Davis, L. J. M.; Goward, G. R. An Improved Understanding of Li+ Hopping Pathways and Rates in Li3Fe2(Po4)3 Using Selective Inversion 6Li NMR Spectroscopy. J. Phys. Chem. C 2013, 117, 24181−24188. (43) Langer, J.; Smiley, D. L.; Bain, A. D.; Goward, G. R.; Wilkening, M. An Unexpected Pathway: 6Li-Exchange NMR Spectroscopy Points to Vacancy-Driven Out-of-Plane Li-Ion Hopping in Crystalline Li 2SnO 3. J. Phys. Chem. C 2016, 120, 3130−3138. (44) Cahill, L. S.; Chapman, R. P.; Britten, J. F.; Goward, G. R. 7Li NMR and Two-Dimensional Exchange Study of Lithium Dynamics in Monoclinic Li3V2(Po4)3. J. Phys. Chem. B 2006, 110, 7171−7177. (45) Davis, L. J. M.; Heinmaa, I.; Goward, G. R. Study of Lithium Dynamics in Monoclinic Li3Fe2(Po4)3Using 6Li VT and 2D Exchange MAS NMR Spectroscopy. Chem. Mater. 2010, 22, 769−775. (46) Davis, L. J. M.; He, X. J.; Bain, A. D.; Goward, G. R. Studies of Lithium Ion Dynamics in Paramagnetic Cathode Materials Using 6Li 1D Selective Inversion Methods. Solid State Nucl. Magn. Reson. 2012, 42, 26−32. 7655


Article

Chemistry of Materials (47) Cahill, L. S.; Kirby, C. W.; Goward, G. R. 6Li{31P} RotationalEcho, Double-Resonance Studies of Lithium Ion Site Dynamics in Li3V2(Po4)3. J. Phys. Chem. C 2008, 112, 2215−2221. (48) Lee, I. K.; Shim, I.-B.; Kim, C. S. Phase Transition Studies of Sodium Deintercalated Na2−XFePO4F (0 ≤ X ≤ 1) by Mössbauer Spectroscopy. J. Appl. Phys. 2011, 109, 07E136−4.

7656


Na Solid-State NMR Reveals the Local Na-Ion ... - ACS Publications

Recommend Documents