Atomic Resolution Structural and Chemical Imaging Revealing the

May 9, 2017 - Citation data is made available by participants in Crossref's Cited-by ... Iron migration and oxygen oxidation during sodium extraction ...
0 downloads 0 Views 6MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Atomic Resolution Structural and Chemical Imaging Revealing the Sequential Migration of Ni, Co, and Mn upon the Battery Cycling of Layered Cathode Pengfei Yan,† Jianming Zheng,‡ Ji-Guang Zhang,‡ and Chongmin Wang*,† †

Environmental Molecular Sciences Laboratory, ‡Energy and Environment Directorate, Pacific Northwest National Laboratory, 902 Battelle Boulevard, Richland, Washington 99354, United States S Supporting Information *

ABSTRACT: Layered lithium transition metal oxides (LTMO) are promising candidate cathode materials for next-generation high-energy density lithium ion battery. The challenge for using this category of cathode is the capacity and voltage fading, which is believed to be associated with the layered structure disordering, a process that is initiated from the surface or solid-electrolyte interface and facilitated by transition metal (TM) reduction and oxygen vacancy formation. However, the atomic level dynamic mechanism of such a layered structure disordering is still not fully clear. In this work, utilizing atomic resolution electron energy loss spectroscopy (EELS), we map, for the first time at atomic scale, the spatial evolution of Ni, Co and Mn in a cycled LiNi1/3Mn1/3Co1/3O2 layered cathode. In combination with atomic level structural imaging, we discovered the direct correlation of TM ions migration behavior with lattice disordering, featuring the residing of TM ions in the tetrahedral site and a sequential migration of Ni, Co, and Mn upon the increased lattice disordering of the layered structure. This work highlights that Ni ions, though acting as the dominant redox species in many LTMO, are labile to migrate to cause lattice disordering upon battery cycling, while the Mn ions are more stable as compared with Ni and Co and can act as pillar to stabilize layered structure. Direct visualization of the behavior of TM ions during the battery cycling provides insight for designing of cathode with high structural stability and correspondingly a superior performance. KEYWORDS: Layered cathode, lithium ion battery, structural degradation, EELS, annular bright field imaging

F

aggregation, lithium depletion, and oxygen loss. Lattice transformation, from layered to either spinel-like structure or rock-salt structure, is realized through TM migration, noting that the oxygen sublattice adopts the same stacking sequence during the phase transformation. Therefore, suppression of TM migration is the key to stabilize the layered structure during cycling. The mobility of TM is mainly determined by its valence state and ionic radius. When TM is at a high valence state, the activation energy for migration is high, and simultaneously the large difference of the ionic radius between TM and Li+ can significantly suppress a potential Li/TM interlayer mixing. When TM is reduced to a low valence state, the energy barrier for migration is decreased and therefore gains a high mobility, and at the same time, its ionic radius is increased to close to that of Li+, facilitating the migration of TM into the Li layer and correspondingly causing lattice transformation. One of the key questions that needs to be answered is which species migrates preferentially to the Li layers. Based on simulation and indirect deduction,18−20 it has been proposed that, among Ni, Co, and Mn, Ni ions are prone

or rechargeable batteries, pushing toward high energy density and stable cycling, among other performance factors, appear to be an insatiable goal. One way to push toward this goal is to reversibly extract a high fraction of lithium from a cathode lattice. Layered lithium transition metal oxides (LTMO), as a category of cathode for lithium ion battery, appear to be feasible for pushing toward high energy density and cycling stability due to their high theoretical capacities, rich redox species, and diversified composition versatility.1−4 Typical trigonal LTMO (LiMO2, M = Co, Mn, Ni, etc.), such as LiCoO2, LiNi1/3Mn1/3Co1/3O2, and LiNi0.5Mn0.5O2, have been commercialized. As another category of layered cathode, Li-rich LTMO cathode can provide an even higher energy density.5−8 However, the practical capacities of both trigonal and Li-rich LTMO are much lower than their theoretical capacities, indicating that the fraction of Li utilization can be further increased to boost energy density. Unfortunately, many previous studies have indicated that increasing Li utilization will aggravate the degradation of the cell, especially on the cathode side.9−12 One of the well-known degradation mechanisms occurred on LTMO is the phase transformation which has been widely observed at the particle surface.13−17 Intensive studies reveal that such a phase transformation is initiated by a local chemistry change, including transition metal (TM) reduction and © 2017 American Chemical Society

Received: April 12, 2017 Revised: May 5, 2017 Published: May 9, 2017 3946

DOI: 10.1021/acs.nanolett.7b01546 Nano Lett. 2017, 17, 3946−3951

Letter

Nano Letters

Figure 1. Electrochemical data of NMC333 using half-cell configuration. (a) Cycling performance and (b, c) the corresponding charge/discharge voltage profile evolutions of NMC333 during cycling at C/10 in the voltage range of (b) 2.7−4.2 V and (c) 2.7−4.8 V.

Figure 2. STEM-HAADF images of the surface layer of NMC333 cathode. (a) Pristine sample without cycling. (b) After 100 cycles with a high cutoff voltage of 4.2 V. (c) After 100 cycles with a high cutoff voltage of 4.8 V. (d) Enlarged area from the region marked with red box in panel c. The insets in panel d are the structural models illustrating lattice transformation from surface into bulk, where transition metal ions, Li ions, and oxygen ions are denoted by blue, green, and red balls, respectively. The yellow dashed lines denote the surface reconstruction layer. Fast Fourier transformation from the three different regions in panel d are shown at the bottom, indicating structure transformation. Blue arrows indicate extra spots, and dashed black frames highlight unit cell changes due to lattice disordering.

3947

DOI: 10.1021/acs.nanolett.7b01546 Nano Lett. 2017, 17, 3946−3951

Letter

Nano Letters

Figure 3. STEM-HAADF and STEM-ABF images simultaneously captured on NMC333 cathode following 100 cycles with a high cutoff voltage of 4.8 V. (a, b) From a region with the well-preserved layered structure. (e, f) From disordered region. (i, j) From heavily disordered region. Enlarged images from panels b, f, and j are shown in panels c, g, and k, respectively. Panels d and h are the corresponding crystal model for panels c and g. (l) Illustrates tetrahedral site TM ion and lattice distortion in panel k. The scale bars are 1 nm.

to migrate to initialize the lattice transformation. However, direct evidence for this claim is still lacking. Based on high energy electron irradiation, Lu et al. noticed that Ni ions, as contrasted with Mn ion, are prone to migrate into Li-layer in the Li-rich LTMO.21 Essentially, migration behavior of Ni, Co, and Mn during the electrochemical cycling is still not experimentally addressed. In this work, we captured atomic resolution elemental maps on a well-known LiNi1/3Mn1/3Co1/3O2 (NMC333) layered cathode by virtue of electron energy loss spectroscopy (EELS) in scanning transmission electron microscopy (STEM), indicating a sequential migration of Ni, Co, and Mn to Li layer, which correspondingly contributes to the lattice transformation. The electrochemical properties of the battery with the NMC333 cathodes were tested using a coin cell configuration with Li metal as an anode and cycled at the rate of 0.1 C at room temperature with the high-cutoff voltages of 4.2 and 4.8 V. Figure 1 shows the electrochemical performance of the cells at different high-cutoff voltages after 100 cycles demonstrating that NMC333 charged to 4.8 V shows much faster capacity decay and voltage fading during cycling, while the cell charged to 4.2 V shows negligible performance decay. Although the initial capacity is higher when charged to 4.8 V (∼220 mA h/g), after 100 cycles, both the discharge capacity and the averaged discharge voltage are lower than that of charged to 4.2 V, indicating a high voltage cycling leads to poor cyclability in terms of both capacity retention and voltage fading. Structurally, our previous investigation has indicated that intragranular cracking is one of the major features for the

high voltage cycled sample, which directly contributes to the fast degradation of the cathode.10 In this work, we focus on probing the atomic level process that contributes to the degradation mechanism of cathode particles when cycled at different voltages. The cathode foils were disassembled from Li/NMC333 halfcells, and cross-sectional TEM specimens were prepared using focused ion beam (FIB) lift-out technique (Figure S1) as detailed in a previous publication.10 STEM high-angle annular dark field (HAADF) imaging reveals that the surface of the NMC333 particle was modified after 100 cycles. As shown in Figure 2, cycling introduces a surface layer, which shows a bright contrast under STEM-HAADF imaging due to the TM aggregation and formation of solid electrolyte interphase.13−15,22 Figure S2 and Figure 2d show the lattice images from the surface layer of both pristine and cycled samples. Figure 2 clearly demonstrates that cycling of the battery at 4.8 V can introduce a thicker surface reconstruction layer of ∼25 nm as compared with that of ∼2 nm for cycling at 4.2 V. In Figure 2c and d, we can identify three zones from surface into bulk, which are highlighted by purple and yellow dashed lines. Beyond the purple dashed line is the outmost surface layer, which is heavily disordered, and the layered structure features can be hardly seen. In between the purple and yellow dashed lines, the layered structure is mildly disordered, still maintaining a layered feature. Beyond the yellow dashed line toward the bulk lattice, the layered structure is well-preserved. Such a gradual structure evolution from bulk lattice to surface that featuring a layered to disordered layer and then to rock salt is also evidenced by the fast Fourier transformation images 3948

DOI: 10.1021/acs.nanolett.7b01546 Nano Lett. 2017, 17, 3946−3951

Letter

Nano Letters

Figure 4. Atomic resolution STEM-EELS mapping. (a) From the pristine NMC333 without cycling. (b−d) From NMC333 after 100 cycles with a high cutoff voltage of 4.8 V. (b) From the well-preserved layered region. (c) From the disordered region. (d) From the heavily disordered region. The dashed blue frame at bottom right highlights TM maps with disordered structure. The white arrows in row b indicates a correspondence between the STEM-HAADF image and EELS map for identifying the Ni that migrates to the Li layer. The scale bars in HAADF images are 0.5 nm.

(shown in Figure 2d). It is interesting to note that, for the case of Li-rich LTMO, such a surface reconstruction layer is more likely to be the spinel structure.15 The structural details of the surface layer were further revealed by using a combination of HAADF and annular bright field (ABF) imaging. As shown in Figure 3, the HAADF and ABF images are collected simultaneously from the wellpreserved layered region (Figure 3a and b), the moderately disordered region (Figure 3e and f), and the heavily disordered region (Figure 3i and j). Both HAADF and ADF series indicate the TM ions in the Li layer are gradually increased, causing disordering of the layered structure. As shown in Figure 3k, the magnified image from Figure 3j clearly shows TM-ions sitting at tetrahedral sites (highlighted by a pink arrow) as well as severe lattice distortion (highlighted by a blue arrow). The lattice distortion is associated with the reduction of the TM cation and formation of oxygen vacancies for which the former can rise Jahn−Teller distortion, while the later can cause TM− O debonding. This observation for the first time provides direct evidence that TM-ions reside at tetrahedral sites in NMC layered cathode, which indicates that the degradation mechanism of trigonal LTMO is similar to that of Li-rich LTMO for which trapping of TM ions at tetrahedral ions are believed to be the origin of voltage decay.9

To address the key question as which TM ions initiates the disordering of layered structure, we use atomic resolution STEM-EELS mapping to directly correlate the lattice structure with the spatial distribution of Ni, Co, and Mn. Figure 4a was taken from the pristine NMC333 (collected under 200 kV imaging electron beam). For the cycled samples, we used 80 kV imaging electron beam to minimize the beam damage on the cycled samples (Figure 4b−d). As evidenced by Figures S3 and S4, following the EEL mapping, the lattice shows no appreciable change, indicating that the sample damage during the STEM-EELS mapping is well-mitigated with the 80 keV. Details on EELS acquisition and data processing are provided in the Supporting Information. Four typical mapping results are shown in Figure 4 for which the oxygen maps are presented and set as references to ensure our mapping results with high spatial resolution (the oxygen columns can be well-resolved in all series, indicating the spatial resolution is better than 2 Å). For the pristine sample (Figure 4a), the maps of Mn, Co, and Ni clearly reveal that Ni, Co, and Mn reside at the TM layer and the lattice maintains a wellordered layered structure. However, following the battery cycling, elemental distribution exhibits disordering, and such a disordering is different for Ni, Co, and Mn, which in turn depends on the degree of the overall structural disordering of the lattice as shown in Figure 4b−d. Moreover, in Figure 4b, for 3949

DOI: 10.1021/acs.nanolett.7b01546 Nano Lett. 2017, 17, 3946−3951

Letter

Nano Letters the slightly structural disordered region, although the layered structure is presented as evidenced by the corresponding HAADF image, the Ni map shows the Ni distribution is deviated from a layered feature, indicating substantial Ni has migrated from the TM layer into the Li layer. Clear evidence is highlighted by the arrows in Figure 4b, where Ni is sitting in the Li layer and leads to the bright contrast under HAADF image. In contrast, Co and Mn maps show a well-preserved layered structure as indicated in Figure 4b. This set of maps captured on the slightly structural disordered region indicate that it is the Ni ions, rather than the Mn or Co ions, that preferentially migrate into the Li layer to contribute the lattice disordering. In other words, Ni ions disorder first upon battery cycling. For the moderate structural disordered region as judged from the HAADF image shown in Figure 4c that is captured from a region that is adjacent to the region of Figure 4b, the elemental maps clearly indicate at this region both Ni and Co distributions are significantly disordered, while the Mn distribution is only slightly disordered. For the heavily structural disordered region as evidenced by the HAADF image shown in Figure 4d, the corresponding maps of Ni, Co, and Mn indicate all of the TM elements are heavily mixed and lost their original layered arrangement. The map series shown in Figure 4 indicates that, among Ni, Co, and Mn, the Mn ions are the most stable ions in keeping the layered structure. Though the regions we chose in Figure 4b−d are spatially different, the increasing degree of lattice structural disordering actually reflect the temporal evolution of layered structure disordering with progression of the battery cycling. Therefore, what we have observed on the dependence of the spatial distribution of Ni, Co, and Mn on the degree of the lattice disorder actually reflects different propensity of migrating from TM layer to Li layer for Ni, Co, and Mn, featuring a sequential migration behaviors of Ni, Co, and Mn in accordance with the lattice disordering of layered structure. The structural and chemical evolution as described above is also consistently supported by the correspondingly electronic structure evolution as indicated by the EELS spectra shown in Figure 5, which are captured from the four mapping regions in Figure 4. In Figure 5, spectra i−iv correspond to regions a−d in Figure 4. In Figure 5a, these EELS spectra show two significant features with the progression of the structural lattice disordering. One is the oxygen prepeak, which is gradually decreased from i to iv as denoted by the arrows. The oxygen prepeak is associated with the transition of electrons from the 1s core state to unoccupied 2p states hybridized with 3d states in transition metals.23 Therefore, a gradual suppression of oxygen prepeak indicates the formation of oxygen vacancy and the reduction of TM that coordinates with the oxygen.24 The other trend is the evolution of the Mn L3/L2 ratio, which gradually increases as shown in Figure 5a, indicating Mn reduction is more significant in the heavily disordered regions. The EELS L-edges of Co and Ni are shown in Figure 5b. Clearly, the chemical shift on the L edge of Co can be noticed as marked in iii and iv of Figure 5b, indicating the reduction of Co, which is consistent with the disordering of Co as shown in Figure 4c and d. The chemical shift for the L-edge of Ni can also be noticed as indicated in ii, iii, and iv in Figure 5b. Combined the elemental mapping (Figure 4) with EELS valence state analysis (Figure 5), it is clearly indicated that lattice disordering (TM migration) always associates with the reduction of TM ions. Therefore, our EELS analysis corroborates the argument that the surface phase transformation is facilitated by oxygen loss and TM reduction.

Figure 5. EELS spectra acquired from the regions in Figure 4. (a) O K-edges and Mn L-edges; and (b) Co L-edges and Ni L-edges. Spectra labeled as i−iv correspond to the regions a−d in Figure 4, respectively. As marked in panel a, with the gradual increase of the lattice disordering, the prepeak on the oxygen K-edge gradually decrease, indicating formation of oxygen vacancies; while s and the L3/L2 ratio on Mn-L edge gradually increases, indicating the reduction of Mn. As indicated by the black dashed line in panel b, the chemical shift of of Ledge for both Ni and Co indicates the reduction of Ni and Co.

In pristine LTMO, previous studies have shown that Ni ions, due to a similar ionic radius to Li+, can have slight interlayer mixing with Li ions, while Li/Co and Li/Mn interlayer mixing is negligible.25−28 After high voltage cycling, oxygen vacancies are introduced into the surface layer either through oxygen gas evolution and/or cathode/electrolyte side reactions. Oxygen vacancy on one hand can lead to the reduction of the adjacent TM to a lower valence states, and on the other hand it may facilitate the migration of the reduced TM.24 The present atomic resolution structural and chemical imaging reveal that, upon battery cycling, the disordering of layered structure is not a process of random hopping of all TM ions to the Li layer; rather, it is a process with strong preference of one species over the others. For the case of coexistence of Ni, Co, and Mn, the Ni ion is observed to be the most labile one to migrate during the battery cycling, followed by Co and Mn. Therefore, in terms of lattice stability, among Ni, Co, and Mn, the Co ion shows better stability than Ni in countering layered structure disordering, and the Mn ion is the most stable one among all of these three, indicating Mn ions play a key role in stabilizing layered structure. On the other hand, suppressing oxygen anion loss is also very crucial for stabilizing the layered structure,6,29,30 3950

DOI: 10.1021/acs.nanolett.7b01546 Nano Lett. 2017, 17, 3946−3951

Letter

Nano Letters

(4) Rozier, P.; Tarascon, J. M. J. Electrochem. Soc. 2015, 162, A2490− A2499. (5) Qiu, B.; Zhang, M.; Xia, Y.; Liu, Z.; Meng, Y. S. Chem. Mater. 2017, 29, 908−915. (6) Qiu, B.; Zhang, M.; Wu, L.; Wang, J.; Xia, Y.; Qian, D.; Liu, H.; Hy, S.; Chen, Y.; An, K.; Zhu, Y.; Liu, Z.; Meng, Y. S. Nat. Commun. 2016, 7, 12108. (7) Ceder, G.; Chiang, Y. M.; Sadoway, D. R.; Aydinol, M. K.; Jang, Y. I.; Huang, B. Nature 1998, 392, 694−696. (8) Sathiya, M.; Rousse, G.; Ramesha, K.; Laisa, C. P.; Vezin, H.; Sougrati, M. T.; Doublet, M. L.; Foix, D.; Gonbeau, D.; Walker, W.; Prakash, A. S.; Ben Hassine, M.; Dupont, L.; Tarascon, J. M. Nat. Mater. 2013, 12, 827−835. (9) Sathiya, M.; Abakumov, A. M.; Foix, D.; Rousse, G.; Ramesha, K.; Saubanere, M.; Doublet, M. L.; Vezin, H.; Laisa, C. P.; Prakash, A. S.; Gonbeau, D.; VanTendeloo, G.; Tarascon, J. M. Nat. Mater. 2014, 14, 230−8. (10) Yan, P.; Zheng, J.; Gu, M.; Xiao, J.; Zhang, J.-G.; Wang, C.-M. Nat. Commun. 2017, 8, 14101. (11) Kim, H.; Kim, M. G.; Jeong, H. Y.; Nam, H.; Cho, J. Nano Lett. 2015, 15, 2111−2119. (12) Lee, E.-J.; Chen, Z.; Noh, H.-J.; Nam, S. C.; Kang, S.; Kim, D. H.; Amine, K.; Sun, Y.-K. Nano Lett. 2014, 14, 4873−4880. (13) Lin, F.; Markus, I. M.; Nordlund, D.; Weng, T. C.; Asta, M. D.; Xin, H. L.; Doeff, M. M. Nat. Commun. 2014, 5, 3529. (14) Boulineau, A.; Simonin, L.; Colin, J. F.; Bourbon, C.; Patoux, S. Nano Lett. 2013, 13, 3857−63. (15) Yan, P.; Nie, A.; Zheng, J.; Zhou, Y.; Lu, D.; Zhang, X.; Xu, R.; Belharouak, I.; Zu, X.; Xiao, J.; Amine, K.; Liu, J.; Gao, F.; ShahbazianYassar, R.; Zhang, J. G.; Wang, C. M. Nano Lett. 2015, 15, 514−22. (16) Xu, B.; Fell, C. R.; Chi, M. F.; Meng, Y. S. Energy Environ. Sci. 2011, 4, 2223−2233. (17) Yan, P.; Xiao, L.; Zheng, J.; Zhou, Y.; He, Y.; Zu, X.; Mao, S. X.; Xiao, J.; Gao, F.; Zhang, J.-G.; Wang, C.-M. Chem. Mater. 2015, 27, 975−982. (18) Carroll, K. J.; Qian, D.; Fell, C.; Calvin, S.; Veith, G. M.; Chi, M.; Baggetto, L.; Meng, Y. S. Phys. Chem. Chem. Phys. 2013, 15, 11128−38. (19) Nam, K. W.; Bak, S. M.; Hu, E. Y.; Yu, X. Q.; Zhou, Y. N.; Wang, X. J.; Wu, L. J.; Zhu, Y. M.; Chung, K. Y.; Yang, X. Q. Adv. Funct. Mater. 2013, 23, 1047−1063. (20) Breger, J.; Meng, Y. S.; Hinuma, Y.; Kumar, S.; Kang, K.; ShaoHorn, Y.; Ceder, G.; Grey, C. P. Chem. Mater. 2006, 18, 4768−4781. (21) Lu, P.; Yuan, R. L.; Ihlefeld, J. F.; Spoerke, E. D.; Pan, W.; Zuo, J. M. Nano Lett. 2016, 16, 2728−33. (22) Yang, P.; Zheng, J.; Kuppan, S.; Li, Q.; Lv, D.; Xiao, J.; Chen, G.; Zhang, J.-G.; Wang, C.-M. Chem. Mater. 2015, 27, 7447−7451. (23) Hwang, S.; Chang, W.; Kim, S. M.; Su, D.; Kim, D. H.; Lee, J. Y.; Chung, K. Y.; Stach, E. A. Chem. Mater. 2014, 26, 1084−1092. (24) Qian, D.; Xu, B.; Chi, M.; Meng, Y. S. Phys. Chem. Chem. Phys. 2014, 16, 14665−8. (25) Yan, P. F.; Zheng, J. M.; Lv, D. P.; Wei, Y.; Zheng, J. X.; Wang, Z. G.; Kuppan, S.; Yu, J. G.; Luo, L. L.; Edwards, D.; Olszta, M.; Amine, K.; Liu, J.; Xiao, J.; Pan, F.; Chen, G. Y.; Zhang, J. G.; Wang, C. M. Chem. Mater. 2015, 27, 5393−5401. (26) Fell, C. R.; Carroll, K. J.; Chi, M.; Meng, Y. S. J. Electrochem. Soc. 2010, 157, A1202−A1211. (27) Lu, Z. H.; Beaulieu, L. Y.; Donaberger, R. A.; Thomas, C. L.; Dahn, J. R. J. Electrochem. Soc. 2002, 149, A778−A791. (28) Kobayashi, H.; Sakaebe, H.; Kageyama, H.; Tatsumi, K.; Arachi, Y.; Kamiyama, T. J. Mater. Chem. 2003, 13, 590−595. (29) Li, J.; Zhan, C.; Lu, J.; Yuan, Y.; Shahbazian-Yassar, R.; Qiu, X.; Amine, K. ACS Appl. Mater. Interfaces 2015, 7, 16040−16045. (30) Long, B. R.; Croy, J. R.; Park, J. S.; Wen, J.; Miller, D. J.; Thackeray, M. M. J. Electrochem. Soc. 2014, 161, A2160−A2167.

because the absence of oxygen vacancy can maintain a high valence state for the TM cations and thus immobile. In summary, we investigated the surface layer structural and chemical evolution of NMC333 following the high-cutoff voltages cycling at 4.2 and 4.8 V and found 4.8 V cycling can introduce a much thicker surface phase transformation layer. STEM-ABF imaging directly reveals the residing of TM-ions at the tetrahedral site in the heavily disordered region. Atomicresolution STEM-EELS mapping captured at the regions with different levels of structural disordering in the cycled sample indicates a sequential migration behaviors of Ni, Co, and Mn in facilitating the disordered layered structure, suggesting that Ni ion is the most labile one and Mn-ion is the most stable one in terms of the tendency of migrating to Li layers. The present observations provide unprecedented insights on the lattice degradation mechanism of layered LTMO at atomic level, shedding new light on the material selection and composition optimization for NMC-based LTMO in countering lattice degradation and correspondingly mitigating capacity and voltage fading.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.7b01546. Snapshot images during FIB lift-out, STEM-HAADF images, and EELS experimental details (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Ji-Guang Zhang: 0000-0001-7343-4609 Chongmin Wang: 0000-0003-3327-0958 Author Contributions

P.Y. and J.M.Z. contributed equally to this work. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work is supported by the Assistant Secretary for Energy Efficiency and Renewable Energy, Office of Vehicle Technologies of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231, Subcontract No. 18769 and No. 6951379 under the Advanced Battery Materials Research (BMR) program. The microscopic analysis in this work was conducted in the William R. Wiley Environmental Molecular Sciences Laboratory (EMSL), a national scientific user facility sponsored by DOE’s Office of Biological and Environmental Research and located at PNNL. PNNL is operated by Battelle for the Department of Energy under Contract DE-AC0576RLO1830.

■

REFERENCES

(1) Wang, J.; He, X.; Paillard, E.; Laszczynski, N.; Li, J.; Passerini, S. Adv. Energy Mater. 2016, 6, 1600906. (2) Manthiram, A.; Knight, J. C.; Myung, S. T.; Oh, S. M.; Sun, Y. K. Adv. Energy Mater. 2016, 6, 1501010. (3) Hy, S.; Liu, H.; Zhang, M.; Qian, D.; Hwang, B.-J.; Meng, Y. S. Energy Environ. Sci. 2016, 9, 1931. 3951

DOI: 10.1021/acs.nanolett.7b01546 Nano Lett. 2017, 17, 3946−3951