Angewandte

Communications Chemie

International Edition: DOI: 10.1002/anie.201509299

Ion Conductors German Edition: DOI: 10.1002/ange.201509299

Single Lithium-Ion Conducting Polymer Electrolytes Based on a Super-

Delocalized Polyanion
Qiang Ma, Heng Zhang, Chongwang Zhou, Liping Zheng, Pengfei Cheng, Jin Nie,
Wenfang Feng, Yong-Sheng Hu,* Hong Li, Xuejie Huang, Liquan Chen, Michel Armand, and
Zhibin Zhou*
Abstract: A novel single lithium-ion (Li-ion) conducting gradients of the salt and cell polarization, which causes
polymer electrolyte is presented that is composed of the lithium premature battery failure.[15, 16]
salt of a polyanion, poly[(4-styrenesulfonyl)(trifluorometh- One of the best solutions to the above-mentioned prob-
yl(S-trifluoromethylsulfonylimino)sulfonyl)imide] lem is to prepare single Li-ion conductors (SLICs), in which
(PSsTFSI¢), and high-molecular-weight poly(ethylene oxide) the anions are immobilized by anion trapping agents or by
(PEO). The neat LiPSsTFSI ionomer displays a low glass- fixing the anions to the polymeric backbones,[17–20] so that the
transition temperature (44.3 8C; that is, strongly plasticizing transference number (tLi+) of Li+ cations is close to one. Until
effect). The complex of LiPSsTFSI/PEO exhibits a high Li-ion now, most of SLICs have been synthesized by grafting
transference number (tLi+ = 0.91) and is thermally stable up to common alkyl carboxylate (RCO2¢)[17] and/or sulfonate
300 8C. Meanwhile, it exhibits a Li-ion conductivity as high as (RSO3¢)[18] anions to the polymeric backbones. However,
1.35 × 10¢4 S cm¢1 at 90 8C, which is comparable to that for the the ion dissociation in polyether media is very limited, owing
classic ambipolar LiTFSI/PEO SPEs at the same temperature. to the low degrees of negative charge distribution in these
These outstanding properties of the LiPSsTFSI/PEO blended anions.
polymer electrolyte would make it promising as solid polymer Armand et al.[21] have reported a new type of SLICs by
electrolytes for Li batteries. blending lithium poly[(4-styrenesulfonyl)(trifluoromethane-
sulfonyl)imide] (LiPSTFSI) with PEO. Owing to the
Solid polymer electrolytes (SPEs) for solvent-free lithium enhanced flexibility and improved negative charge distribu-
(Li) rechargeable batteries have captured much attention tion of the ¢SO2¢N(¢)¢SO2¢CF3 vs. common ¢CO2¢ and ¢
owing to their potential advantages, including safety, ease of SO3¢ structures, the relatively high ionic conductivity of about
packaging, excellent flexibility and containment, and their 10¢5 S cm¢1 at 70 8C has been obtained for the blended
functioning as a separator compared with liquid or gel polymer electrolytes. Then, the conductivities were further
electrolytes.[1–8] To date, great progress has been made by improved by elegantly preparing random and triblock
utilizing LiTFSI (Li[(CF3SO2)2N])/poly(ethylene oxide) copolymers of LiPSTFSI through incorporating ethylene
(PEO) SPEs for Li/LiFePO4 batteries, which is being used oxide (EO) units in side and main chains, respectively.[22, 23]
as power source for an electric car, Autolib.[9] However, A prototype Li battery using this triblock polymer electrolyte
conventional SPEs formed by dissolving Li salt with a small exceeds a conventional battery based on a dual-ion polymer
anion in a polymer host (usually high-molecular-weight electrolyte, suggesting the importance of single Li-ion nature
PEO), are dual-ion conductors, in which both cations and in improving the cycling performances. Inspired by these
anions are mobile.[10–14] This would generate the concentration results, we envisaged that it would be beneficial to further
enhance ionic conductivities of such imide anion-based SLICs
by increasing its negative charge distribution.
Herein, we report a new type of single Li-ion conducting
[*] Q. Ma, H. Zhang, C. Zhou, L. Zheng, P. Cheng, Prof. J. Nie, polymer electrolytes, which are prepared by simply dissolving
Prof. W. Feng, Prof. Z. Zhou
lithium salt of a super-delocalized polyanion, namely poly[(4-
Key Laboratory for Large-Format Battery Materials and System,
Ministry of Education, School of Chemistry and Chemical Engineer- styrenesulfonyl)(trifluoromethyl(S-trifluooromethylsulfony-
ing, Huazhong University of Science and Technology limino)sulfonyl)imide] (PSsTFSI¢ ; Scheme 1) in PEO. We
1037 Luoyu Road, Wuhan 430074 (China) devised this kind of new anion structure on the base of
E-mail: [email protected] following consideration. Replacing a =O group in above ¢
Prof. Y.-S. Hu, Prof. H. Li, Prof. X. Huang, Prof. L. Chen SO2¢N(¢)¢SO2¢CF3 structure by a strong electron-withdraw-
Key Laboratory for Renewable Energy, Beijing Key Laboratory for New ing group, =NSO2CF3, can create much more delocalized
Energy Materials and Devices, Beijing National Laboratory for anion structure (¢SO2¢N(¢)¢SO(=NSO2CF3)¢CF3, sTFSI¢),
Condensed Matter Physics, Institute of Physics
in which the Li+ cations would be highly dissociated from this
Chinese Academy of Sciences, Beijing 100190 (China)
E-mail: [email protected] super-delocalized anion when mixed with a polymer media
Prof. M. Armand (that is, PEO), so that conductivities can be further
CIC energigune, Alava Technology Park, Albert Einstein improved.[24]
4801510 MIÄANO Þlava (Spain) The procedures for preparing the monomeric and poly-
Supporting information for this article is available on the WWW meric salts are detailed in the Supporting Information. The
under http://dx.doi.org/10.1002/anie.201509299. structure and composition of the monomer (KSsTFSI),

Angew. Chem. Int. Ed. 2016, 55, 2521 –2525 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 2521
 Angewandte
Communications Chemie

Scheme 1. The synthetic routes of a) monomer and b) polymers.

DMAP = dimethylaminopyridine, AIBN = azodiisobutyronitrile.

Figure 1. a) Photographs of membrane for the neat PEO and the LiX/
intermediate (KPSsTFSI), and target salt (LiPSsTFSI) were PEO (X = PSS, PSTFSI, and PSsTFSI) blended polymer electrolytes
characterized by 1H and 19F NMR (Supporting Information, (EO/Li+ = 20). b) DSC traces of neat PEO, LiPSsTFSI, and the LiX/PEO
Figure S1). The neat LiPSsTFSI salt has a number-average (X = PSS, PSTFSI, PSsTFSI) blended polymer electrolytes (EO/Li+ = 20)
at the first heating scan.
molecular weight (Mn) of about 2 × 105 g mol¢1 with a polydis-
persity index (PDI) of 2.21 (Supporting Information,
Table S1), and its complex of LiPSsTFSI/PEO exhibits an
oxidation potential at ca. 4.0 V vs. Li+/Li (Supporting measurement data are summarized in the Supporting Infor-
Information, Figure S2), and is thermally stable up to 300 8C mation, Table S2). The neat PEO shows a sharp melting
(Supporting Information, Figure S3). transition peak at 70.1 8C, which is consistent well with the
To understand the impact of degree of negative charge previous reports for neat PEO.[27, 28] Important note is that the
distribution in anion on the ionic conductivity of SPEs neat LiPSsTFSI salt displays a low glass transition temper-
reliably, the two related lithium polymer salts, lithium ature (Tg) at 44.3 8C (Figure 1 b; Supporting Information,
poly(4-styrenesulfonate) (LiPSS) and LiPSTFSI (Supporting Figure S5). This would be attributable to highly flexibility and
Information, Figure S4), were also synthesized using the same super-delocalized negative charge distribution of the ¢SO2¢
methods as described previously.[22, 25] The membranes of all of N(¢)¢SO(=NSO2CF3)¢CF3 structure in PSsTFSI¢ , which
the blended polymer electrolytes were prepared by a solu- impedes the motion of both Li+ cations and PSsTFSI¢
tion-casting method (the procedures are detailed in Support- anions to rearrange orderly in space (that is, slow kinetics of
ing Information). crystallization), thus making LiPSsTFSI glassformer at low
Figure 1 a shows the photographs of membrane for the temperature. In contrast, glass transitions below 50 8C have
neat PEO and the LiX/PEO (X = PSS, PSTFSI, and PSsTFSI) not observed for any neat Li ionomer based on lesser
blended polymer electrolytes at a molar ratio of EO/Li+ = 20, delocalized anions (for example, ¢CO2¢ , ¢SO3¢ , and ¢SO2¢
as the concentration of lithium salt to EO unit in the classic N(¢)¢SO2¢CF3) without PEO blending.[17, 21, 22, 25] It is interest-
LiTFSI/PEO SPEs around this ratio region has been found to ing to note that a glass transition at the low temperature of
afford relatively high ionic conductivities in medium high ¢15.1 8C is also observed for the LiPSsTFSI/PEO blended
temperatures, and its ionic conduction is little affected by its electrolyte (this glass transition at the low temperature of
heat history.[26] It can be seen that self-standing, translucent ¢15.3 8C is repeatedly observed at the second heating scan;
membranes are obtained. It is worthy to note that the Supporting Information, Figure S6), and is however not
membrane of LiPSsTFSI/PEO blended electrolyte has detected for both the LiPSS/PEO and LiPSTFSI/PEO
a better mechanical ductility than both the corresponding electrolytes. This result clearly indicates the stronger plasti-
LiPSTFSI/PEO and LiPSS/PEO electrolytes, which is per- cizing effect of the PSsTFSI¢ vs. PSS¢ and PSTFSI¢ anions,
ceptible from appreciable manual bending or stretching. This which is consistent with the glassformer nature of neat
would be attributed to the much better flexibility and greatest LiPSsTFSI, and is favorable for enhancing segmental motions
delocalized negative charge distribution of the ¢SO2¢N(¢)¢ of PEO in the blended electrolyte. The glass transition
SO(=NSO2CF3)¢CF3 structure (in PSsTFSI¢), among these temperature is a qualitative signature of ion mobility in SPEs,
three polyanions. and a low glass transition temperature for LiPSsTFSI/PEO is
Figure 1 b comparatively shows the DSC traces of the neat a sign of high ion mobility in the blended electrolyte.
PEO, LiPSsTFSI, and the three blended polymer electrolytes Moreover, the LiPSsTFSI/PEO blended electrolyte displays
of LiX/PEO (X = PSS, PSTFSI, PSsTFSI) with the same the lowest value of enthalpy of melting (that is, DHm =
molar ratio of EO/Li+ = 20 at the first heating scan (all the 58.3 J g¢1; Supporting Information, Table S2), which is obvi-

2522 www.angewandte.org Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2016, 55, 2521 –2525
 Angewandte
Communications Chemie

ously lower than that for the LiPSTFSI/PEO electrolyte (that PSsTFSI¢ . This is totally consistent with the degree of
is, DHm = 71.9 J g¢1; Supporting Information, Table S2), and is negative charge distribution and freedom (that is, structural
significantly lower than that for the LiPSS/PEO electrolyte flexibility) of the anion. This trend in XRD patterns is also
(that is, DHm = 101.2 J g¢1; Supporting Information, Table S2). well consistent with the DSC results in Figure 1 b (Supporting
This is a signature of a higher degree of amorphous phase in Information, Table S2), wherein the complex of LiPSsTFSI/
the former (LiPSsTFSI/PEO) than in the latter two (LiPSS/ PEO shows low glass transition before melting (that is, low
PEO and LiPSTFSI/PEO). All above results obtained from tendency to crystallize), while the latter two complexes of
DSC characterization suggest that the LiPSsTFSI/PEO LiX/PEO (X = PSS, PSTFSI) only show melting with larger
blended electrolyte has higher degrees of amorphous phases values of enthalpy (that is, a higher degree of crystalline
and larger segmental motions than both the LiPSS/PEO and phase) without glass transitions (Supporting Information,
LiPSTFSI/PEO electrolytes, which are beneficial for improv- Table S2).
ing its ionic conductivities. Figure 3 a compares the temperature dependence of ionic
Figure 2 comparatively displays the XRD patterns of the conductivities (s) for the LiX/PEO (X = PSS, PSTFSI,
LiX/PEO (X = PSS, PSTFSI, PSsTFSI) blended polymer PSsTFSI) blended polymer electrolytes with a molar ratio
electrolytes with a molar ratio of EO/Li+ = 20, as well as the of EO/Li+ = 20, as well as the classic LiTFSI/PEO blended
electrolyte for comparison, and some of the representative
data are presented in the Supporting Information, Table S3.
As seen from Figure 3 a, our conductivity values for the classic
LiTFSI/PEO blended electrolyte are consistent with the
previous report,[26] indicating that our measurements are
reliable. As a common feature for the PEO-based SPEs,[29] an
increase in the ionic conductivity to a variable extent is
observed with increasing the temperature up to about 70 8C
for all the LiX/PEO (X = PSS, PSTFSI, PSsTFSI, TFSI)
blended electrolytes (Figure 3 a), owing to a crystalline melt-
ing transition of the PEO host and/or the free-volume
activation. The ionic conductivities for all these four kinds
of blended electrolytes decrease in the order of LiTFSI/
PEO > LiPSsTFSI/PEO > LiPSTFSI/PEO > LiPSS/PEO in
the measured temperature range from 25 to 90 8C. It is
generally accepted that the ionic conductivity of SPEs is
mainly governed by the polymer crystallinity and/or glass
transition temperature, and the charge carrier concentra-
tion.[2, 22, 25, 30] The highest ionic conductivities for the blended
electrolyte containing LiTFSI are observed, which is expected
from that both the Li+ cations and relatively small TFSI¢
anions are mobile in the complex of LiTFSI/PEO. By
Figure 2. XRD patterns of the LiX/PEO (X = PSS, PSTFSI, PSsTFSI) contrast, the three blended electrolytes containing the
blended polymer electrolytes (EO/Li+ = 20), as well as the neat PEO
polymer salts show obviously lower ionic conductivities,
and LiPSsTFSI.
which is due to the huge volume of the polymeric anion.

neat PEO and LiPSsTFSI for

comparison. As shown in
Figure 2, two sharp diffraction
peaks are observed at 2q =
19.288 and 23.388 for the neat
PEO, suggesting the nature of
highly crystalline phase.[14] In
contrast, no characteristic dif-
fraction peak is observed for
the neat LiPSsTFSI salt, indi-
cating a mainly amorphous
phase. It is interesting to note
that the intensities of these
two representative XRD
peaks for neat PEO decrease Figure 3. a) The temperature dependence of ionic conductivities for the LiX/PEO (X = PSS, PSTFSI,
rapidly with addition of the PSsTFSI, TFSI) blended polymer electrolytes (EO/Li+ = 20). b) The temperature dependence of ionic
polymer salts in the anion conductivities of Li+ cations for the LiX/PEO (X = PSS, PSTFSI, PSsTFSI, TFSI) blended polymer electrolytes
order of PSS¢ < PSTFSI¢ < (EO/Li+ = 20).

Angew. Chem. Int. Ed. 2016, 55, 2521 –2525 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 2523
 Angewandte
Communications Chemie

Among these three blended electrolytes containing polymeric an oxidation potential at ca. 4.0 V vs. Li+/Li, and is thermally
anions, the complex of LiPSsTFSI/PEO displays the highest stable up to 300 8C. The ionic conductivities of the LiPSsTFSI/
conductivities, and is higher by about one order in magnitude PEO (EO/Li+ = 20) blended electrolyte are higher by about
than those of the previous state-of-the-art LiPSTFSI/PEO one order in magnitude than those of the LiPSTFSI/PEO
electrolyte, and higher by 2–3 orders in magnitude than those (EO/Li+ = 20) electrolyte, and by 2–3 orders in magnitude
of LiPSS/PEO electrolyte in the whole temperature range. than those of LiPSS/PEO (EO/Li+ = 20) electrolyte over the
This would be essentially attributed to more flexible and whole temperature range. This is essentially attributed to the
super-delocalized nature of the ¢SO2¢N(¢)¢SO(=NSO2CF3)¢ unique structure of ¢SO2¢N(¢)¢SO(=NSO2CF3)¢CF3 in the
CF3 structure (in PSsTFSI¢) compared with both the ¢SO2¢ PSsTFSI¢ anion with both super-delocalized negative charge
N(¢)¢SO2¢CF3 (in PSTFSI¢) and ¢SO3¢ structures (in PSS¢), distribution and highly flexible features, which not only makes
which can not only further improve polymeric segmental highly dissociation of Li+ ions from anions but also favorably
motions responsible for promoting Li-ion motion through the enhances the degrees of amorphous phases and the segmental
chains of PEO in amorphous phase, but also increase degree motions in SPEs required for ionic conduction. More
of dissociation of the Li salts. This anion-dependence trend in importantly, the ionic conductivities of individual Li+ cations
conductivity concurs with the results in DSC and XRD for the LiPSsTFSI/PEO blended electrolyte are comparable
measurements (Figures 1 b and 2), wherein a higher degree of to those for the classic ambipolar LiTFSI/PEO electrolyte
amorphous phase is observed in the complex of LiPSsTFSI/ above the melting point of PEO. All of these preliminary
PEO. results suggest that the LiPSsTFSI would be a promising salt
The lithium-ion transference number (tLi+) is a key of polymer electrolyte and would open the door to design new
parameter for evaluating the performance of SPEs. The highly conductive single lithium-ion conductors for lithium
value of tLi+ for the classic ambipolar LiTFSI/PEO blended batteries.
electrolyte is 0.22 at 60 8C, which is comparable to that of 0.29
reported previously.[30] The value of tLi+ for the complex of
LiPSsTFSI/PEO is as high as 0.91 at 60 8C, indicative of single Acknowledgements
Li-ion conducting behavior, which is expected from the huge
volume of the polyanion (Supporting Information, Tables S4 This work was supported by the National Natural Science
and Figure S7). Foundation of China (Nos. 51172083, 51222210, 51472268).
The ionic conductivities of individual Li+ cations in SPEs
play a paramount role in dictating the performance of Li Keywords: blended polymer electrolytes · ion conductors ·
batteries, as Li+ cations are only active species needed during lithium batteries · Li-ion conductivity
the cycling process of Li batteries; however, few work have
characterized the conduction behaviors of individual Li+ How to cite: Angew. Chem. Int. Ed. 2016, 55, 2521 – 2525
cations. Figure 3 b comparatively displays the temperature Angew. Chem. 2016, 128, 2567 – 2571
dependence of ionic conductivities of individual Li+ cations
(sLi+, sLi+ = s × tLi+) for the four LiX/PEO (X = PPS, PSTFSI, [1] M. B. Armand, Annu. Rev. Mater. Sci. 1986, 16, 245 – 261.
[2] J. M. Tarascon, M. Armand, Nature 2001, 414, 359 – 367.
PSsTFSI, TFSI) blended polymer electrolytes. It is important
[3] V. Di Noto, S. Lavina, G. A. Giffin, E. Negro, B. Scrosati,
to note that the values of sLi+ for the LiPSsTFSI/PEO blended Electrochim. Acta 2011, 57, 4 – 13.
electrolyte are comparable to those for the classic ambipolar [4] D. T. Hallinan, N. P. Balsara, Annu. Rev. Mater. Res. 2013, 43,
LiTFSI/PEO electrolyte above the melting of PEO (70 8C; for 503 – 525.
example, at 70 8C, sLi+ = 6.92 × 10¢5 S cm¢1 (LiPSsTFSI/PEO) [5] Q. W. Pan, D. M. Smith, H. Qi, S. J. Wang, C. Y. Li, Adv. Mater.
vs. sLi+ = 1.41 × 10¢4 S cm¢1 (LiTFSI/PEO); at 90 8C, sLi+ = 2015, 27, 5995 – 6001.
[6] D. Zhou, Y. B. He, R. L. Liu, M. Liu, H. D. Du, B. H. Li, Q. Cai,
1.35 × 10¢4 S cm¢1 (LiPSsTFSI/PEO) vs. sLi+ = 2.94 ×
¢4 ¢1 Q. H. Yang, F. Y. Kang, Adv. Energy Mater. 2015, 5, 1500353.
10 S cm (LiTFSI/PEO); Supporting Information, [7] J. J. Zhang, J. H. Zhao, L. P. Yue, Q. F. Wang, J. C. Chai, Z. H.
Table S5). This is inspiring as the PEO-based SPEs works Liu, X. H. Zhou, H. Li, Y. G. Guo, G. L. Cui, L. Q. Chen, Adv.
above its melting point in practical applications, and indicates Energy Mater. 2015, 5, 1501082.
that LiPSsTFSI/PEO blended electrolyte would have better [8] M. Balaish, E. Peled, D. Golodnitsky, Y. E. Eli, Angew. Chem.
electrochemical performance than LiTFSI/PEO electrolyte Int. Ed. 2015, 54, 436 – 440; Angew. Chem. 2015, 127, 446 – 450.
above its melting point, as the concentration gradients of the [9] http://www.autolib.eu/en/our-commitment/bluecar-menu-en/
100-electric/ (accessed 2014-2010-29).
salt and cell polarization, owing to the transference of small
[10] W. Wieczorek, D. Raducha, A. Zalewska, J. R. Stevens, J. Phys.
anions, could be removed in the SLICs of LiPSsTFSI/PEO. Chem. B 1998, 102, 8725 – 8731.
Among these SLICs, the highest the values of sLi+ for the [11] M. Kalita, M. Bukat, M. Ciosek, M. Siekierski, S. H. Chung, T.
complex of LiPSsTFSI/PEO are observed in the testing Rodr†guez, S. G. Greenbaum, R. Kovarsky, D. Golodnitsky, E.
temperature range. Peled, D. Zane, B. Scrosatif, W. Wieczorek, Electrochim. Acta
In summary, a new type of lithium polymer salt 2005, 50, 3942 – 3948.
(LiPSsTFSI) has been synthesized by free radical polymeri- [12] L. R. A. K. Bandara, M. A. K. L. Dissanayake, B. E. Mellander,
Electrochim. Acta 1998, 43, 1447 – 1451.
zation. The neat LiPSsTFSI ionomer displays a low glass
[13] Y. Zhao, R. Y. Tao, T. Fujinami, Electrochim. Acta 2006, 51,
transition temperature of 44.3 8C. Its blended polymer 6451 – 6455.
electrolyte of the LiPSsTFSI/PEO type exhibits single Li- [14] G. B. Appetecchi, W. Henderson, P. Villano, M. Berrettoni, S.
ion conducting behavior with a tLi+ value as high as 0.91 and Passerini, J. Electrochem. Soc. 2001, 148, A1171 – A1178.

2524 www.angewandte.org Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2016, 55, 2521 –2525
 Angewandte
Communications Chemie

[15] J. H. Shin, S. Passerini, Electrochim. Acta 2004, 49, 1605 – 1612. [25] C. H. Park, Y. K. Sun, D. W. Kim, Electrochim. Acta 2004, 50,
[16] M. Morita, T. Fukumasa, M. Motoda, H. Tsutsumi, Y. Matsuda, 375 – 378.
T. Takahashi, H. Ashitaka, J. Electrochem. Soc. 1990, 137, 3401 – [26] S. Lascaud, M. Perrier, A. Vall¦e, S. Besner, J. PrudÏhomme, M.
3404. Armand, Macromolecules 1994, 27, 7469 – 7477.
[17] T. Itoh, Y. Mitsuda, T. Ebina, T. Uno, M. Kubo, J. Power Sources [27] H. Zhang, C. Y. Liu, L. P. Zheng, F. Xu, W. F. Feng, H. Li, X. J.
2009, 189, 531 – 535. Huang, M. Armand, J. Nie, Z. B. Zhou, Electrochim. Acta 2014,
[18] S. S. Zhang, G. X. Wan, J. Appl. Polym. Sci. 1993, 48, 405 – 409. 133, 529 – 538.
[19] W. Xu, X. G. Sun, C. A. Angell, Electrochim. Acta 2003, 48, [28] M. Shaheer Akhtar, U. Y. Kim, D. J. Choi, O. B. Yang, Mater. Sci.
2255 – 2266. Forum 2010, 658, 161 – 164.
[20] M. A. Mehta, T. Fujinami, Chem. Lett. 1997, 915 – 916. [29] I. W. Cheung, K. B. Chin, E. R. Greene, M. C. Smart, S. Abbrent,
[21] R. Meziane, J. P. Bonnet, M. Courty, K. Djellab, M. Armand, S. G. Greenbaum, G. K. S. Prakash, S. Surampudi, Electrochim.
Electrochim. Acta 2011, 57, 14 – 19. Acta 2003, 48, 2149 – 2156.
[22] S. W. Feng, D. Y. Shi, F. Liu, L. P. Zheng, J. Nie, W. F. Feng, X. J. [30] W. Gorecki, M. Jeannin, E. Belorizky, C. Roux, M. Armand, J.
Huang, M. Armand, Z. B. Zhou, Electrochim. Acta 2013, 93, Phys. Condens. Matter 1995, 7, 6823 – 6832.
254 – 263.
[23] R. Bouchet, S. Maria, R. Meziane, A. Aboulaich, L. Lienafa, J. P.
Bonnet, T. N. T. Phan, D. Bertin, D. Gigmes, D. Devaux, R.
Denoye, M. Armand, Nat. Mater. 2013, 12, 452 – 457.
[24] H. Zhang, H. B. Han, X. R. Cheng, L. P. Zheng, P. F. Cheng, Received: October 19, 2015
W. F. Feng, J. Nie, M. Armand, X. J. Huang, Z. B. Zhou, J. Power Revised: November 11, 2015
Sources 2015, 296, 142 – 149. Published online: January 14, 2016

Angew. Chem. Int. Ed. 2016, 55, 2521 –2525 Ó 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.angewandte.org 2525