German Edition: DOI: 10.1002/ange.201607820
Lithium Ion Batteries
International Edition: DOI: 10.1002/anie.201607820Lithium n-Doped Polyaniline as a High-Performance Electroactive Material for Rechargeable Batteries
Pablo Jimnez,* Eric Levillain, Olivier AlvÞque, Dominique Guyomard, Bernard Lestriez, and Jol Gaubicher*
Abstract:The discovery of conducting lithium-doped polyan- iline with reversible redox chemistry allows simultaneous unprecedented capacity and stability in a non-aqueous Li battery. This compound (lithium emeraldinate) was synthe- sized by lithium–proton exchange on the emeraldine base in an anhydrous lithium-based electrolyte. A combination of UV/
Vis-NIR spectroelectrochemistry, XPS, FTIR, and EQCM characterization allowed a unified description of the chemical and electrochemical behavior, showing facile charge delocal- ization of the doped states and the reversibility of the redox processes in this form of polyaniline. From a practical point of view, lithium emeraldinate behaves as a high-capacity organic active material (230 mAh g 1) that enables preparation of relatively thick composite electrodes with a low amount of carbon additives and high energy density (460 Wh kg 1).
Concomitantly, at 1C rate, 400 cycles were achieved without significant capacity loss, while the coulombic efficiency is greater than 99 %.
I
ntrinsically conducting polymers (CPs) are organic, macro- molecular, redox-active materials for which electrochemical reactions are concomitant with the phenomena referred to as“doping” and “dedoping”.[1]Doping and dedoping, which can be also carried out by chemical reactions, result in the change of the net charge in the polymer backbone and lead to drastic changes in electrical, magnetic, and optical properties. The interest of CPs compared to other organic materials for electrochemical energy storage stems from their chemical stability (especially that of the charged states for which charges and unpaired electrons are stabilized by delocaliza- tion along the CP backbone), insolubility (macromolecules with strong non-covalent interactions), and their conducting/
semiconducting character (decreasing the need for conduc- tive additives).[2]
Polyaniline (PANI), the product from oxidative polymer- ization of aniline in aqueous acidic media,[3]stands out among other CPs due to its versatile doping chemistry;[4]not only does it possess a range of oxidation states with different levels of doping but also a range of protonation states that modify the net charge and hence the doping state of PANI owing to the presence of nitrogen atoms along the conjugated back- bone of the polymer. The most widely utilized forms of PANI are the water and air-stable, half-oxidized states[5](Figure 1):
the neutral, insulating, emeraldine base (EB) and the p- doped, conducting, emeraldine salt (ES), in which one can be converted into the other by simple acid–base reactions.
Nearly all of the electrochemistry studies performed on PANI have so far involved the participation of the p-doped states, generally either in water or protic media, so as to take advantage of the high electrical conductivity of ES.[6]During the 1980s, PANI was in fact one of the first organic materials tested as active material in electrochemical cells, and it showed an exceptional stability on cycling and excellent coulombic efficiency.[7]To put forward its possibilities, taking the fully protonated forms of PANI (right side of the square in Figure 1 a) the theoretical specific capacity going from the most reduced state (leucoemeraldine, LE) to the most oxidized state (pernigraniline salt, PNS) would be 294 mAh g 1 (corresponding to one electron per aniline unit), an exceptionally high value for a prospective positive active material.[2b]In practice, the capacity of PANI in a non- protic environment is usually limited to half of that value due to the irreversible character of the redox transition between ES and PNS; either PNS is not chemically stable (PNS is too strong as an acid or electrophile) and reacts with the electrolyte, or the overpotential to incorporate the charge compensating anions is too high and the completely oxidized state cannot be reached within the stability window of the electrolyte.[8] Even though some effort has been put into increasing the specific capacity of PANI by improving the reversibility of the ES-PNS redox transition in non-aqueous electrolytes, the most efficient strategy to date requires the use of high-molecular-weight anionic counterions, which limits the practical capacity of the PANI electrodes.[9]
Compared to covalent functionalization reactions of PANI, the deprotonation-induced n-doping of PANI has been studied far too low, as it implies the occurrence of formally negatively charged nitrogen atoms (nitrenes), the strong basicity of which requires strictly aprotic conditions.
The n-doping entails the presence of charge compensating cations; so far there have only been reports of n-doped PANI with Na+and K+countercations.[10]In the case of a complete deprotonation of PANI in the presence of Li+(the left side of [*] Dr. P. Jimnez, Prof. D. Guyomard, Prof. B. Lestriez,
Prof. J. Gaubicher
Institut des Matriaux Jean Rouxel (IMN), CNRS UMR 6502, Universit de Nantes
2 rue de la Houssinire BP 32229, 44322 Nantes Cedex 3 (France) E-mail: pablo.jimenez-manero@cnrs-imn.fr
joel.gaubicher@cnrs-imn.fr Prof. E. Levillain, Dr. O. AlvÞque
Laboratoire MOLTECH-Anjou, CNRS UMR 6200 Universit d’Angers
2 Bd Lavoisier, 49045 Angers Cedex 1 (France)
Supporting information and the ORCID identification number(s) for the author(s) of this article can be found under:
http://dx.doi.org/10.1002/anie.201607820.
the square in Figure 1 a), it would be possible to electro- chemically access to a one electron per aniline unit capacity going from the pernigraniline base (PNB) to a lithium pernigranilate state (PNLi) in a lithium electrolyte. Interest- ingly enough the EB form has been found to have a distinct interaction with respect to lithium cations, as opposed to other metals, which leads to a phenomenon of lithium doping in lithium electrolytes,[11]a form of p-doping induced by the coordination of Li+ on the imine-type nitrogen atoms (this state is depicted in Figure 1 b as LiEB, or lithium-doped EB).
It is known that nitrogen has a special affinity for lithium (both being hard atoms); when nitrogen has a formal negative charge, the interaction with lithium often presents an intermediate character between ionic and covalent bonding, which accounts for the particular stability and short intera- tomic distance found in compounds such as organic lithium amides, Li3N, Li2NH, or LiNH2.[12]
Our main discovery was the fact that the LiEB material, prepared by simple immersion of EB in a lithium electrolyte, could be effectively deprotonated by the addition of a lithium based, non-nucleophilic base (see the description and explan- ation of the deprotonation procedure in the Supporting Information).[13]The new form of PANI obtained following this deprotonation procedure has a yellowish-green color and was named lithium emeraldinate (ELi). ELi can only exist under inert atmosphere and aprotic environment: as evi- denced by FTIR and XPS analysis (see the Supporting
Information), ELi is converted back into the EB state when exposed to water, demonstrating that the chemical structure of the PANI backbone is not altered during the conversion of LiEB to ELi. The nature of this new species of PANI resulted to be intriguingly similar to that of ES and indeed different to LiEB state of PANI. First, cyclic voltammetry (Figure 2) showed two redox transitions: a fully reversible redox couple at 3.1 V vs. Li+/Li (one electron per two aniline units) and a second redox couple at 4 V, which is fully reversible if the upper potential limit is kept below 4.3 V (reaching to 0.75 electrons/aniline unit, see a further explanation in the Supporting Information). These transitions are analogous to the LE-ES and ES-PNS transitions in aqueous acidic media and different to the single one electron redox couple observed in LiEB in our experimental conditions (Figure 2). Thin films of this new material were made by deprotonation in water of an ES film (deposited by in situ polymerization[14]) to produce EB, followed by immersion in LiPF6salt solution to produce LiEB, and finishing with the deprotonation treatment to yield ELi (reaction pathway depicted by red arrows in Figure 1 b).
Conductivity measurements gave a value for ELi of 5 10 1S cm 1not far from the conductivity of ES 3 S cm 1and well above the values of LiEB 3 10 4S cm 1 and EB 7 10 10S cm 1; this supports the idea that as produced ELi is a truly doped state of PANI. Indeed UV/Vis-NIR spectroe- lectrochemistry of ELi thin films on platinum revealed a broad long-tail absorption extending into the infrared for Figure 1. a) Diagram showing the different possible protonation and oxidation degrees of polyaniline; the circles represent the canonical forms that can be represented with four aniline repeating unit formulae. b) Extension in 3D of the diagram “a” with an axis to include the possible lithium p-doped states. c) The formulae and reactions of some of the polyaniline states shown in the diagrams, including some of the new deprotonated–lithiated states (right).
the intermediate and higher oxidation states of ELi, which is characteristic of doped conducting polymers with highly delocalized charges along the backbone, whereas the com- pletely reduced state (PNLi) shows no absorption in the visible-NIR region as it would be expected of an insulating material, as is the case of LE.[15]This additional parallelism between the electronic spectra of our material and the LE- ES-PNS system allows for the elaboration of this analogy: as it is depicted on both sides of Figure 1 c, ELi can be seen as a lithium counterpart or doppelgnger of ES where all the hydrogen atoms attached to nitrogen have been replaced by lithium atoms. As expected, the spectroelectrochemistry features of LiEB films (Figure 3) show a much lesser degree of charge delocalization upon doping by electrochemical oxidation, which is typical of organic materials with a semi- conducting character.
The electrochemical quartz microbalance (EQCM, Figure 4) study of ELi films performed in the lithium electrolyte of a traditional battery (LP30: EC/DMC 1m LiPF6) shows that the oxidation of ELi to LiPNB follows a mechanism of anion insertion which is reversible on reduction (deinsertion). Upon the reduction of ELi to PNLi, two processes seem to happen concurrently: the deinsertion of PF6 anions and the incorporation of solvent molecules inside the polymer film. The transition from ELi to PNLi firstly creates a more densely charged or n-doped polymer backbone with a greater tendency to be solvated and secondly the release of PF6 anions leaves a vacancy that can be readily occupied by the carbonate molecules of solvent.[16]
The excellent reversibility of both redox and mass transfer processes in the studied potential window makes this material an especially promising candidate for use as an active material
in Li-ion batteries. The ELi material was thus incorporated in electrodes with 5 % of conducting additive (carbon fibers) and 5 % of binder (PTFE) and its performance was tested in half cells using Li as counterelectrode. These electrodes can be galvanostatically charged and discharged between 2.5 V Figure 2. a) Cyclic voltammetry (5 mVs 1) plot of a LiEB electrode in
EC/DMC 1mLiPF6before (red) and after (blue) the conversion to ELi by deprotonation. b) Galvanostatic charge and discharge curves of ELi electrode between 4.3 V and 2.5 V (see the Supporting Information, Figure S12 for further details).
Figure 3. UV/Vis-NIR spectroelectrochemistry plots of the difference absorption spectra (referenced to spectra at 2.5 V vs. Li+/Li) for a) LiEB film and b) ELi film.
Figure 4. EQCM plot showing the simultaneous CV curve (current vs.
potential, blue) and quartz frequency shift evolution (frequency vs.
potential, red) for a thin ELi film (a drop in frequency is caused by a gain in mass and an increment of frequency by a loss in mass).
and 4.25 V vs. Li /Li, that is, at a potential window superior to that of most organic materials for positive electrodes,[17]and they display a coulombic efficiency of greater than 99 % for more than 400 cycles (Supporting Information, Figure S9).
Moreover the specific capacity at charge–discharge rates currents aboveCrate (for example, 151 mAh g 1at 2 C with 5 wt % carbon content) is exceptionally high for an organic active material (Supporting Information, Figure S10) while at lower rates these electrodes yield a maximum value of 208 mAh g 1 (corresponding to a specific capacity of 230 mAh g 1 for PANI) and an energy density of 460 Wh kg 1. These values of stability and energy density are unprecedented for any CP studied to date for its use in Li- ion batteries.[2, 18] ELi also definitely surpasses in power performance most of the known organic active materials taking into account the low carbon content and the thickness of our electrodes;[19]this comes as a result of the fast kinetics for both electronic and ionic transport phenomena in ELi.
The new form of lithium-doped PANI presented herein definitely deserves further research due to its unique combi- nation of properties, particularly its intrinsic conductivity and redox stability, which are of paramount interest in several fields of application such as electrochemical storage and conversion of energy, thin film transistors, or electrooptic devices.
Keywords: conducting polymers · lithium ion batteries · organic active materials · polyaniline · spectroelectrochemistry
[1] a) T. A. Skotheim, J. Reynolds, Eds.Handbook of Conducting Polymers, 2 Volume Set. CRC press, Boca Raton, FL, 2007;
b) A. G. MacDiarmid,Angew. Chem. Int. Ed.2001,40, 2581 – 2590;Angew. Chem.2001,113, 2649 – 2659.
[2] a) Z. Song, H. Zhou,Energy Environ. Sci.2013,6, 2280 – 2301;
b) J. F. Mike, J. L. Lutkenhaus,J. Polym. Sci. Part B2013,51, 468 – 480.
[3] J. Stejskal, R. G. Gilbert,Pure Appl. Chem.2002,74, 857 – 867.
[4] A. G. MacDiarmid, J. C. Chiang, A. F. Richter, A. A. Epstein, Synth. Met.1987,18, 285 – 290.
[5] N. Gospodinova, L. Terlemezyan,Prog. Polym. Sci. 1998,23, 1443 – 1484.
[6] S. Bhadra, D. Khastgir, N. K. Singha, J. H. Lee,Prog. Polym. Sci.
2009,34, 783 – 810.
[7] a) E. M. Genies, M. Lapkowski, C. Santier, E. Vieil,Synth. Met.
1987,18, 631 – 636; b) Y. Kobayashi, T. Shishikura, H. Konuma, T. Sakai, H. Nakamura, M. Takeuchi, (Showa Denko Kabushiki Kaisha, Hitachi, Ltd., Japan), EP0208254,1986.
[8] O. A. Kozarenko, V. S. Dyadyun, M. S. Papakin, O. Y. Posudiev- sky, V. G. Koshechko, V. D. Pokhodenko, Electrochim. Acta 2015,184, 111 – 116.
[9] a) J.-W. Jeon, Y. Ma, J. F. Mike, L. Shao, P. B. Balbuena, J. L.
Lutkenhaus, Phys. Chem. Chem. Phys. 2013, 15, 9654 – 9662;
b) J.-W. Jeon, J. ONeal, L. Shao, J. L. Lutkenhaus,ACS Appl.
Mater. Interfaces2013,5, 10127 – 10136.
[10] a) M. Y. Hua, G. W. Hwang, Y. H. Chuang, S. A. Chen, R. Y.
Tsai, Macromolecules 2000, 33, 6235; b) I. Yamaguchi, T.
Nagano,Polymer2015,73, 79 – 85.
[11] a) K. S. Ryu, K. M. Kim, S. G. Kang, G. J. Lee, J. Joo, S. H.
Chang,Synth. Met.2000,110, 213 – 217; b) K. S. Ryu, J. H. Jung, J. Joo, S. H. Chang,J. Electrochem. Soc.2002,149, A478 – A482.
[12] a) C. Fressign, J. Maddaluno, C. Giessner-Prettre, B. Silvi, J.
Org. Chem.2001,66, 6476 – 6479; b) M. Lappert, A. Protchenko, P. Power, A. Seeber in Metal Amide Chemistry, Wiley, Chi- chester,2009, pp. 7 – 38; c) E. U. Wuerthwein, K. D. Sen, J. A.
Pople, P. v. R. Schleyer,Inorg. Chem.1983,22, 496 – 503.
[13] J. Gaubicher, P. Jimnez Manero, B. Lestriez, J-P. Bonnet, D.
Guyomard (CNRS, France), WO2015124757,2015.
[14] J. Stejskal, I. Sapurina, J. Prokes, J. Zemek,Synth. Met.1999,105, 195 – 202.
[15] a) E. M. Genis, M. Lapkowski,Synth. Met.1987,21, 117 – 121;
b) A. Neudeck, A. Petr, L. Dunsch,Synth. Met.1999,107, 143 – 158.
[16] H. Varela, R. M. Torresi,J. Electrochem. Soc.2000,147, 665 – 670.
[17] J. Xie, Q. Zhang,J. Mater. Chem. A2016,4, 7091 – 7106.
[18] a) N. Oyama, T. Tatsuma, T. Sato, T. Sotomura,Nature1995,373, 598 – 600; b) T. Tatsuma., T. Sotomura, T. Sato, D. A. Buttry, N.
Oyama,J. Electrochem. Soc.1995,142, L182 – L184; c) Z. Song, Y. Qian, M. L. Gordin, D. Tang, T. Xu, M. Otani, H. Zhan, H.
Zhou, D. Wang,Angew. Chem. Int. Ed.2015,54, 13947 – 13951;
Angew. Chem.2015,127, 14153 – 14157.
[19] As opposed to most previously reported examples that use thin films of CP (either electropolymerized or casted by coating) as electrodes, our electrodes consist of powder mixtures of CP and additives with a specific surface capacity of around 3 mAh cm 2, which is closer to the values of electrodes in real life Li-ion batteries.
Manuscript received: August 11, 2016 Revised: October 23, 2016
Final Article published:&& &&,&&&&
Communications
Lithium Ion Batteries
P. Jimnez,* E. Levillain, O. AlvÞque, D. Guyomard, B. Lestriez,
J. Gaubicher* &&&—&&&
Lithium n-Doped Polyaniline as a High- Performance Electroactive Material for Rechargeable Batteries
Staying positive: A new form of conduct- ing, n-doped polyaniline obtained by deprotonation in lithium electrolyte is able to yield an unprecedented combina- tion of capacity, stability, coulombic effi- ciency, and energy and power density as a positive electrode material for lithium ion batteries.
