Angewandte A Journal of the Gesellschaft Deutscher Chemiker
International Edition
Chemie www.angewandte.org
Accepted Article Title: Lithium azide as a novel electrolyte additive for all-solid-state Li-S batteries Authors: Gebrekidan Gebreslase Eshetu, Xabier Judez, Chunmei Li, Alex Bondarchuk, Lide Mercedes Rodriguez-Martinez, Heng Zhang, and Michel Armand This manuscript has been accepted after peer review and appears as an Accepted Article online prior to editing, proofing, and formal publication of the final Version of Record (VoR). This work is currently citable by using the Digital Object Identifier (DOI) given below. The VoR will be published online in Early View as soon as possible and may be different to this Accepted Article as a result of editing. Readers should obtain the VoR from the journal website shown below when it is published to ensure accuracy of information. The authors are responsible for the content of this Accepted Article. To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201709305 Angew. Chem. 10.1002/ange.201709305 Link to VoR: http://dx.doi.org/10.1002/anie.201709305 http://dx.doi.org/10.1002/ange.201709305
10.1002/anie.201709305
Angewandte Chemie International Edition
COMMUNICATION Lithium azide as a novel electrolyte additive for all-solid-state Li-S batteries
[*] Dr. G. G. Eshetu, X. Judez, Dr. C. Li, Dr. A. Bondarchuk, Dr. L. M. Rodriguez-Martinez, Dr. H. Zhang,* and Prof. M. Armand* CIC Energigune, Parque Tecnológico de Álava, Albert Einstein 48, 01510 Miñano, Álava, Spain. E-mail: [email protected]; [email protected]. Supporting information for this article is can be found under: http://dx.doi.org/10.1002/anie.2015xxxxx.
LiN3
LiN3 -e
+e
Cathode
Among post lithium metal batteries (LMBs), Li-S batteries appear to be the most appealing and viable energy storage technologies. This is attributed to their overwhelming advantages in terms of energy density, with a theoretical value of about 2600 Wh kg−1 computed on the basis of Li° anode and S cathode, abundance of sulphur resources, low cost, and environmental benignancy.[1] However, the practical deployment is still hampered due to the presence of several inherent challenges, including the electronically insulating nature of elemental sulphur (S8) and lithium sulphide (Li2S), polysulphide (PS) shuttling effect and most importantly, the formation of lithium dendrites upon continuous charging/discharging.[2] The stability of Li° anode in Li-S batteries has become one of the top urgent challenges that need to be addressed so as to meet the long-term stability requirement of such technologies. Among various strategies to overcoming issues related to Li° anode and polysulfide shuttling, it is well acknowledged that the use of electrolyte additives (usually no more than 5-10% either by weight or volume) is one of the most feasible, economical and effective approaches.[3] To date, LiNO3 has been investigated as state-of-the-art additive for Li-S batteries containing liquid electrolytes (LEs), owing to its ability to form a good passivation layer on Li° surface and also catalyse the conversion of highly soluble polysulfide to slightly soluble S8. However, side effects such as uncontrollable passivation layer thickness, side reactions with polysulfide and possibly with the cathode, electrolyte and electrode additive materials etc. are among the limitations of LiNO3, thus encouraging the search for other more robust additives. [4] One of the basic requirements for Li° anode passivating species includes high ionic conductivity, good and homogeneous
surface coverage on Li° surface, mechanical stability etc. In this regard, lithium nitride (Li3N) is known to be a good candidate due to its high ionic conductivity (σ = 6 × 10–3 S cm–1 at 25 ºC for the single crystal structure[5]) and superior stability against Li metal compared to other species such as LiF (σ = 10–31 S cm–1),[6] Li2S (σ = 10–13 S cm–1)[7] and Li2CO3 (σ = 10–8 S cm–1)[6]. Thus, it could serve as a robust Li° solid electrolyte interphase (SEI) building material. In the present work, lithium azide (LiN3), for the first time, has been conceived as a novel Li3N precursor. A compact and conductive passivation layer, comprising of Li3N as principal SEI component can be formed by adding LiN3 in the electrolyte. Oxidation of LiN3 on the cathode leads to the formation of molecular nitrogen (N2), which after could migrate to the anode side and further react with Li° resulting in additional Li3N (Scheme 1). With all above-mentioned merits, the use of small amount of LiN3 (2 wt.%) in the Li-S cells effectively inhibits the PS shuttling by forming a vigorous passivation layer on Li° anode, thus significantly improving the cycling performances and sulphur utilization.
Li metal anode
Abstract: Of the various beyond lithium-ion battery technologies, lithium-sulphur (Li-S) batteries have an appealing theoretical energy density and are being intensely investigated as next generation rechargeable lithium metal batteries. However, the stability of the Li° anode is among the most urgent challenges that need to be addressed to ensure the long-term stability of Li-S batteries. In this work, we report lithium azide (LiN3) as a novel electrolyte additive for all-solid-state Li-S batteries (ASSLSBs). It results in the formation of a thin, compact and highly conductive passivation layer on Li° anode, thereby avoiding dendrite formation, and polysulfide shuttling. It superbly enhances the cycling performance, Coulombic and energy efficiencies of ASSLSBs, outperforming the state-of-the-art additive lithium nitrate (LiNO3).
Li3N +e
Li+
Diffusion N2
N2
Scheme 1. Electrochemical reactions of LiN3 in lithium metal batteries.
The role of LiN3 as a novel electrolyte additive to form superior SEI on Li° anode using common poly(ethylene oxide) (PEO)-based polymer electrolyte (PE) are comprehensively investigated.[8] The physicochemical and electrochemical properties of the LiN3-added electrolytes are comparatively studied with the LiNO3-added ones, aiming at shedding light on the structure-activity relationship of these two additives. In order to find out how additives generate and enhance the SEI layer, lithium bis(trifluoromethansulfonyl)imide (Li[N(SO 2CF3)2], LiTFSI, not a good SEI builder[9]) (PE1) and lithium bis(fluorosulfonyl)imide (Li[N(SO2F)2], LiFSI, a well-known SEI former[10]) (PE0) were chosen as blank electrolytes. Due to the difficulties of access to post-mortem surface analysis from polymer cells, LiTFSI (LE1) and LiFSI (LE0) in 1,2dimethoxyethane (DME), were selected as archetype electrolytes for XPS and SEM studies, enabling to mimic the working phenomena in PEO-based electrolytes due to the analogous chemical structure of DME vs. PEO.[11] Figure 1a shows the photographs of the membranes prepared from PE1 with 2 wt.% LiN3 and LiNO3 as electrolyte additives (hereafter referred without specifying the amount of additives). It can be clearly seen that self-standing and translucent membranes are obtained, with good ductility and mechanical strengths. The X-ray diffraction (XRD) patterns and
This article is protected by copyright. All rights reserved.
Accepted Manuscript
Gebrekidan Gebresilassie Eshetu, Xabier Judez, Chunmei Li, Alex Bondarchuk, Lide M. RodriguezMartinez, Heng Zhang* and Michel Armand*
10.1002/anie.201709305
Angewandte Chemie International Edition
differential scanning calorimetric (DSC) traces of PE1 and PE0 with LiN3 or LiNO3 additives are presented in Figure 1b and c, respectively and are compared to the corresponding blank electrolytes. XRD patterns of neat salts, additives and PEO membrane are given in Figure S1. Firstly, the XRD pattern does not show any strong peak of either the salts or additives, suggesting their complete dissolution in the PEO matrix via complexation. For all the electrolytes, the typical diffraction peaks of neat PEO are observed (i.e., 2θ = 17.5 and 22.2º), suggesting the presence of highly crystalline phase in the polymer electrolyte at 25 oC. This is further confirmed by the melting transition of PEO at ca. 60 ºC in the DSC profiles (Figure 1c). Upon addition of the additives, the peak intensities of PEO in XRD and melting enthalpy in DSC slightly increase, indicating relatively higher degree of crystallinity of the polymer (see Table S1 and Figure S2 for DSC data). This would be ascribed to the increased Coulombic interactions upon the addition of LiN3 or LiNO3, since PEO itself has a strong solvating ability towards various lithium salts due to the relatively high donicity of the ethylene oxide unit (EO, donor number = 22) and favourable configurational entropy term.[8a] Thus, the LiN3 and LiNO3 additives, behaving like normal lithium salts, have non negligible interactions with PEO matrix. Possibly, the higher melting P(EO)3LiN3 and P(EO)3LiNO3 stoichiometric complexes can be formed. Figure 1d shows the thermogravimetric analysis (TGA) enlisting the influence of LiN3 and LiNO3 additives on the weight loss of the PEs. For the pristine additives, while LiNO3 is stable in the tested temperature range; LiN3 presented a very sharp mass loss at 236 ºC, which is related to the release of N2 as evidenced from mass spectra measurement (Figure S3). Surprisingly, LiN3 improves the thermal stability of PE0 while it does for LiTFSI marginally. The noticeable enhancement with LiFSI-based electrolyte could be linked to the formation of a more thermally stable [N(SO2F)(SO2N3)] anion via the reaction of N3– on the S─F bond, chemically as well as electrochemically labile linkage) compared to the C─F of TFSI–, as shown in Scheme S1b. Overall, the addition of LiN3 has a positive impact on the thermal stability, as well as lesser influence on the glass transition temperatures (Tg) of PEs (see Figure S2 for the DSC test with various LiN3 concentrations). These are beneficial requirements while hunting for functional electrolyte additives. The temperature dependence of the ionic conductivities of the as-prepared PEs is given in Figure 1e. Near room temperature range (< 50 ºC), the conductivity is found to get slightly decreased with the incorporation of both additives and this could be attributed to the increase in the crystallinity of the polymer as also evidenced from the XRD and DSC results. However, the ionic conductivities for all the electrolytes show small differences at the temperature ranging from 70 to 90 ºC, a suitable temperature domain for the operation of Li-S polymer cells. The anodic stabilities of the PEs with and without LiN3 and LiNO3 are depicted in Figure 1f. Both PE1 and PE0 employing LiN3 additive presented a mildly peak around 3.7 V vs. Li+/Li, and is associated to the oxidation of LiN3 as shown in Scheme 1. In general, the anodic stabilities are not affected by the presence of either LiN3 or LiNO3, signalling its favourability with respect to the electrochemical stability of the parent PEs. The quality of the SEI films formed between Li° electrode and the electrolyte is one of the most important elements in
rechargeable LMBs. It dictates the overall electrochemical performance enlisting cycle life, rate capability, Coulombic/energy efficiency as well as safety of the batteries. Thus, the effect of additives on the electrochemical stability towards Li° electrode is shown in Figure 2. The addition of LiN3 or LiNO3 to PE1 and PE0 proved to increase the cycling life of the stripping/plating tests, and at the same time reducing the voltage (V) hysteresis compared to blank electrolytes (Figure 2a). This indicates that both LiN3 and LiNO3 result in much more stable SEI layers. More interestingly, LiN3 provided much lower voltage compared to LiNO3, e.g., ca. 5 mV for PE1 + LiN3 and 15 mV for PE1 + LiNO3 electrolytes, linked to the high ionic conductivity of the principal passivating species (i.e., Li 3N).
Figure 1. Physicochemical and electrochemical properties of polymer electrolytes. a) Appearances of PE1 with LiN3 and LiNO3 additives. b) XRD patterns and c) DSC traces of the electrolytes. d) TGA traces for the electrolytes, and the neat components, e) Arrhenius plots of ionic conductivity and f) anodic stabilities at 70 oC for the electrolytes.
To ascertain the mechanism behind the super enhancement in the Li/PEs interfacial stability upon the addition of additives, Li plating/stripping tests, post-mortem analysis using scanning electron microscopic (SEM) and X-ray photoelectron spectroscopy (XPS), were conducted in liquid Li-S cell configuration. As seen in Figure 2b, the Li symmetric cells with LiN3 and LiNO3 in LE1 performed the same trend as shown in Figure 2a (more than 2000 h continuous cycling for LiN3-added one with a low voltage of ca. 6 mV). Figure 2c shows SEM images of Li° deposited onto the bare Cu substrate in various liquid electrolytes. Both additives have an effect on the Li° surface morphology; however, uniform and smooth pancake-like morphologies with much lower surface area were formed in LiN 3added electrolytes, instead of those fibre and needle like Li dendrites in the respective LiNO3-added and blank electrolytes. This is in agreement with the improved cycling stability of Li symmetrical cells using LiN3-added electrolytes (Figure 2a and b).
This article is protected by copyright. All rights reserved.
Accepted Manuscript
COMMUNICATION
10.1002/anie.201709305
Angewandte Chemie International Edition
COMMUNICATION LiTFSI and LiN3 seem to be stable at this potential. However, they get reduced with 2e- extraction from the biphenyl radical dianion, which is assumed to have a reduction potential very close to that of Li° (Table S3). To vividly comprehend the working mechanism of LiN3 and LiNO3 as electrolyte additives, a schematic illustration is provided in Figure 3. We hereby proposed 8e- reduction of LiN3 leading to Li3N as the only product, while LiNO3 leading to Li3N and Li2O (Figure 3b). In literature, it is also inferred that the reduction of LiNO3 could result in the formation of species such as with nitro R-NO2, and/or nitrite (NO2–) groups, Li2O,[15] LiNO2,[16] etc. However, Li2O is much less conductive as compared to Li3N.[5b] The reduction mechanisms of LiTFSI, LiFSI and DME on Li electrode are summarized in Scheme S2 for comparison. Shortly, the SEI layer obtained in the presence of LiN3 additive is richer in Li3N, more conductive as shown by lower overpotential, of high surface coverage (due to the higher molar volume of Li3N vs. Li°) and discrete in comparison with LiNO3 containing and the blank electrolytes. Li
(a) N
N
N
Conductive and compact SEI layer
LiN3 + 8 Li + 8 e
3 Li3N
Resistive and compact SEI layer
Li metal electrode O
(b)
Li
LiNO3 + 2 Li + 2 e
LiNO2 + 3 Li2O
LiNO3 + 8 Li + 8 e
Li3N + 3 Li2O
N O
resistive and discrete SEI layer
O
resistive and discrete SEI layer
Li metal electrode
Figure 3. Schematic illustration of the reaction mechanisms for the electrolytes containing LiN3 and LiNO3 additives.
Figure 2. Electrochemical behaviour of Li° electrode in the prepared electrolytes. Galvanostatic cycling of Li symmetric cells at 0.1 mA cm−2 (half cycle time 2 h) using a) polymer electrolytes at 70 oC and b) liquid electrolytes at 25 oC. c) Typical SEM images and d) XPS spectra of Li deposited onto Cu substrates at 0.1 mA cm−2 (plating time 20 h) in TFSI-based liquid electrolytes at 25 oC.
Aiming at reinforcing the above XPS results, chemical simulation of the neat LiN3, LiNO3, LiTFSI and LiFSI was carried following the procedures mentioned in literature. [12] It is found that while LiNO3 and LiFSI easily get reduced with biphenyl radical anion (ca. 0.4 V vs. Li+/Li°) via 8e- and 4e- respectively,
As a proof of concept, the feasibility of LiN 3 as an additive for ASSLSBs is manifested by cyclability test of Li-S cells at 70 ºC (Figure 4 and S8). Noteworthy, the Li-S cell using PE1 with LiN3 can deliver high reversible discharge capacity ca. 800 mAh g–1 with superior capacity retention after 30 cycles (Figure 4a), while the one with the blank electrolyte shows PS shuttling during the first charge process (Figure 4e). The addition of LiN3 into both TFSI- and FSI-based electrolytes results in an enhancement of the Coulombic efficiency compared to that of LiNO3 (Figure S8a vs. S8c; S8b vs. S8d). These impressive results suggest that stable and robust SEI layers on Li° anode can be not only formed but also maintained during continuous cycling by adding LiN 3 additive, as supported by the results on symmetric Li° cells (Figure 2). In conclusion, lithium azide (LiN3) has been identified for the first time as a novel electrolyte additive for all-solid-state Li-S batteries (ASSLSBs). The SEI layer formed on Li° electrode in the presence of LiN3 is uniform, dendrite-free, and rich in Li3N which is a highly conductive and high molar volume SEI-building material. Such superior SEI layer is found to boost the Li/electrolyte interfacial stability, leading to the enhancement of the cyclability, Coulombic/energy efficiencies and discharge
This article is protected by copyright. All rights reserved.
Accepted Manuscript
Figure 2d presents the most relevant spectra, F1s and N1s for LE1 + LiN3, LE1 + LiNO3 and blank electrolyte systems. The survey and C1s spectra are given in Figures S4 and S5, respectively. For the sake of comparison, the survey, F1s and N1s spectra of LE0-based electrolytes are provided in Figures S6 and S7, respectively. The F 1s spectra for LE1 displays two peaks at 685.2 eV and 688.7 eV, corresponding to LiF and –CF3 (from residual LiTFSI and/or its reduction products, such as Li2NSO2CF3),[12] respectively. From the intensity of the peak at 685.7 eV, it is evident that the passivation film from LE1 + LiN3 contains less LiF compared to LE1 + LiNO3. Interestingly, the most distinguishing feature comes from the N1s spectra where it strongly evidences the effect of the additives compared to the blank electrolyte. The N 1s binding energy for LE1 shows a singlet peak at ~ 399.9 eV, a characteristic binding energy of residual LiTFSI and/or its reduction derivatives. In the presence of LiN3 and LiNO3 additives, there exist additional binding energies at 398.2 eV and 404.2 eV for LE1 + LiN3 and at 398.5 eV and 402.7 eV for LE1 + LiNO3. The peak in the range of 398.2−398.5 eV is assigned to Li3N, [13] and is found to be the most predominant SEI species in the case of LE1 + LiN3. This is supported by the higher atomic percentage of N1s belonging to Li3N for the LiN3-added electrolyte, which is again corroborated by decrement in the atomic percentage of F1s (Table S2). Moreover, one of the basic requirement for good SEI-building materials is to have higher equivalent volume compared to Li° anode (i.e., 13.0 cm3 mol−1), [14] and in this regard, Li3N (27.4 cm3 mol−1) rich SEI is preferable compared to the one rich in LiF (8.9 cm3 mol−1). This makes LiN3 a relatively novel and desired electrolyte additive.
10.1002/anie.201709305
Angewandte Chemie International Edition
capacity in the Li-S cells. We hope, this finding will help to facilitate the development of practical Li-S battery systems and other related technologies in the future.
[8] [9]
[10]
[11] [12] [13]
[14] [15] [16]
a) M. Armand, Annu. Rev. Mater. Res. 1986, 16, 245; b) H. Zhang, C. Li, M. Piszcz, E. Coya, T. Rojo, L. M. Rodriguez-Martinez, M. Armand, Z. Zhou, Chem. Soc. Rev. 2017, 46, 797. a) X. Judez, H. Zhang, C. Li, J. A. González-Marcos, Z. Zhou, M. Armand, L. M. Rodriguez-Martinez, J. Phys. Chem. Lett. 2017, 8, 1956; b) X. Judez, H. Zhang, C. Li, G. G. Eshetu, Y. Zhang, J. A. González-Marcos, M. Armand, L. M. Rodriguez-Martinez, J. Phys. Chem. Lett. 2017, 8, 3473. a) H. Zhang, W. Feng, J. Nie, Z. Zhou, J. Fluorine Chem. 2015, 174, 49; b) H. Zhang, C. Liu, L. Zheng, F. Xu, W. Feng, H. Li, X. Huang, M. Armand, J. Nie, Z. Zhou, Electrochim. Acta 2014, 133, 529. D. Aurbach, J. Power Sources 2000, 89, 206. G. G. Eshetu, T. Diemant, S. Grugeon, R. J. Behm, S. Laruelle, M. Armand, S. Passerini, ACS Appl. Mater. Interfaces 2016, 8, 16087. a) D. Aurbach, E. Pollak, R. Elazari, G. Salitra, C. S. Kelley, J. Affinito, J. Electrochem. Soc. 2009, 156, A694; b) J. Yang, R. C. de Guzman, S. O. Salley, K. Y. S. Ng, B.-H. Chen, M. M.-C. Cheng, J. Power Sources 2014, 269, 520. E. Peled, S. Menkin, J. Electrochem. Soc. 2017, 164, A1703. T. Jaumann, J. Balach, M. Klose, S. Oswald, J. Eckert, L. Giebeler, J. Electrochem. Soc. 2016, 163, A557. V. Etacheri, U. Geiger, Y. Gofer, G. A. Roberts, I. C. Stefan, R. Fasching, D. Aurbach, Langmuir 2012, 28, 6175.
Figure 4. Discharge/charge profiles of the Li-S cells using polymer electrolytes at 70 oC.
Acknowledgements This work was supported by GV-ELKARTEK-2016 from the Basque Government and MINECO RETOS (ref: ENE201564907-C2-1-R) from the Spanish Government. X.J. thanks Basque Government for PhD funding and C.L. thanks the Spanish Government for the Juan de la Cierva scholarship (Ref: FJCI-2015-23898). Keywords: electrolytes; lithium metal electrode; batteries; lithium sulphur battery; lithium azide. [1] [2]
[3] [4] [5] [6] [7]
a) J.-M. Tarascon, M. Armand, Nature 2001, 414, 359; b) M. Armand, J.-M. Tarascon, Nature 2008, 451, 652; c) Z. W. Seh, Y. Sun, Q. Zhang, Y. Cui, Chem. Soc. Rev. 2016,45, 5605. a) N.-S. Choi, Z. Chen, S. A. Freunberger, X. Ji, Y.-K. Sun, K. Amine, G. Yushin, L. F. Nazar, J. Cho, P. G. Bruce, Angew. Chem. Int. Ed. 2012, 51, 9994; b) Y.-X. Yin, S. Xin, Y.-G. Guo, L.-J. Wan, Angew. Chem. Int. Ed 2013, 52, 13186; c) A. Manthiram, Y. Fu, S. H. Chung, C. Zu, Y. S. Su, Chem. Rev. 2014, 114, 11751; d) Q. Pang, X. Liang, C. Y. Kwok, L. F. Nazar, Nat. Energy 2016, 1, 16132; e) H. J. Peng, J. Q. Huang, X. B. Cheng, Q. Zhang, Adv. Energy Mater. 2017, 1700260. a) W. Xu, J. Wang, F. Ding, X. Chen, E. Nasybulin, Y. Zhang, J.-G. Zhang, Energy Environ. Sci. 2014, 7, 513; b) S. Zhang, K. Ueno, K. Dokko, M. Watanabe, Adv. Energy Mater. 2015, 5, 1500117. a) S. S. Zhang, J. Power Sources 2013, 231, 153; b) W. Li, H. Yao, K. Yan, G. Zheng, Z. Liang, Y.-M. Chiang, Y. Cui, Nat. Commun. 2015, 6, 7436. a) U. v. Alpen, A. Rabenau, G. H. Talat, Appl. Phys. Lett. 1977, 30, 621; b) Y. Zhu, X. He, Y. Mo, Adv. Sci. 2017, 1600517. Q. Zhang, J. Pan, P. Lu, Z. Liu, M. W. Verbrugge, B. W. Sheldon, Y. T. Cheng, Y. Qi, X. Xiao, Nano Lett. 2016, 16, 2011. Z. Lin, Z. Liu, N. J. Dudney, C. Liang, ACS Nano 2013, 7, 2829.
This article is protected by copyright. All rights reserved.
Accepted Manuscript
COMMUNICATION
10.1002/anie.201709305
Angewandte Chemie International Edition
COMMUNICATION COMMUNICATION G. G. Eshetu, X. Judez, C. Li, A. Bondarchuk, L. M. Rodriguez-Martinez, H. Zhang, * and M. Armand* Page No. – Page No. Lithium azide as a robust electrolyte additive for all-solid-state Li-S batteries
Accepted Manuscript
Lithium azide effectively favours the formation of dendrite-free and highly ionic conductive solid electrolyte interphase on Li electrode, and thereby improving the cycling performances and sulphur utilization of Li-S cells.
This article is protected by copyright. All rights reserved.
International Edition
Chemie www.angewandte.org
Accepted Article Title: Lithium azide as a novel electrolyte additive for all-solid-state Li-S batteries Authors: Gebrekidan Gebreslase Eshetu, Xabier Judez, Chunmei Li, Alex Bondarchuk, Lide Mercedes Rodriguez-Martinez, Heng Zhang, and Michel Armand This manuscript has been accepted after peer review and appears as an Accepted Article online prior to editing, proofing, and formal publication of the final Version of Record (VoR). This work is currently citable by using the Digital Object Identifier (DOI) given below. The VoR will be published online in Early View as soon as possible and may be different to this Accepted Article as a result of editing. Readers should obtain the VoR from the journal website shown below when it is published to ensure accuracy of information. The authors are responsible for the content of this Accepted Article. To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201709305 Angew. Chem. 10.1002/ange.201709305 Link to VoR: http://dx.doi.org/10.1002/anie.201709305 http://dx.doi.org/10.1002/ange.201709305
10.1002/anie.201709305
Angewandte Chemie International Edition
COMMUNICATION Lithium azide as a novel electrolyte additive for all-solid-state Li-S batteries
[*] Dr. G. G. Eshetu, X. Judez, Dr. C. Li, Dr. A. Bondarchuk, Dr. L. M. Rodriguez-Martinez, Dr. H. Zhang,* and Prof. M. Armand* CIC Energigune, Parque Tecnológico de Álava, Albert Einstein 48, 01510 Miñano, Álava, Spain. E-mail: [email protected]; [email protected]. Supporting information for this article is can be found under: http://dx.doi.org/10.1002/anie.2015xxxxx.
LiN3
LiN3 -e
+e
Cathode
Among post lithium metal batteries (LMBs), Li-S batteries appear to be the most appealing and viable energy storage technologies. This is attributed to their overwhelming advantages in terms of energy density, with a theoretical value of about 2600 Wh kg−1 computed on the basis of Li° anode and S cathode, abundance of sulphur resources, low cost, and environmental benignancy.[1] However, the practical deployment is still hampered due to the presence of several inherent challenges, including the electronically insulating nature of elemental sulphur (S8) and lithium sulphide (Li2S), polysulphide (PS) shuttling effect and most importantly, the formation of lithium dendrites upon continuous charging/discharging.[2] The stability of Li° anode in Li-S batteries has become one of the top urgent challenges that need to be addressed so as to meet the long-term stability requirement of such technologies. Among various strategies to overcoming issues related to Li° anode and polysulfide shuttling, it is well acknowledged that the use of electrolyte additives (usually no more than 5-10% either by weight or volume) is one of the most feasible, economical and effective approaches.[3] To date, LiNO3 has been investigated as state-of-the-art additive for Li-S batteries containing liquid electrolytes (LEs), owing to its ability to form a good passivation layer on Li° surface and also catalyse the conversion of highly soluble polysulfide to slightly soluble S8. However, side effects such as uncontrollable passivation layer thickness, side reactions with polysulfide and possibly with the cathode, electrolyte and electrode additive materials etc. are among the limitations of LiNO3, thus encouraging the search for other more robust additives. [4] One of the basic requirements for Li° anode passivating species includes high ionic conductivity, good and homogeneous
surface coverage on Li° surface, mechanical stability etc. In this regard, lithium nitride (Li3N) is known to be a good candidate due to its high ionic conductivity (σ = 6 × 10–3 S cm–1 at 25 ºC for the single crystal structure[5]) and superior stability against Li metal compared to other species such as LiF (σ = 10–31 S cm–1),[6] Li2S (σ = 10–13 S cm–1)[7] and Li2CO3 (σ = 10–8 S cm–1)[6]. Thus, it could serve as a robust Li° solid electrolyte interphase (SEI) building material. In the present work, lithium azide (LiN3), for the first time, has been conceived as a novel Li3N precursor. A compact and conductive passivation layer, comprising of Li3N as principal SEI component can be formed by adding LiN3 in the electrolyte. Oxidation of LiN3 on the cathode leads to the formation of molecular nitrogen (N2), which after could migrate to the anode side and further react with Li° resulting in additional Li3N (Scheme 1). With all above-mentioned merits, the use of small amount of LiN3 (2 wt.%) in the Li-S cells effectively inhibits the PS shuttling by forming a vigorous passivation layer on Li° anode, thus significantly improving the cycling performances and sulphur utilization.
Li metal anode
Abstract: Of the various beyond lithium-ion battery technologies, lithium-sulphur (Li-S) batteries have an appealing theoretical energy density and are being intensely investigated as next generation rechargeable lithium metal batteries. However, the stability of the Li° anode is among the most urgent challenges that need to be addressed to ensure the long-term stability of Li-S batteries. In this work, we report lithium azide (LiN3) as a novel electrolyte additive for all-solid-state Li-S batteries (ASSLSBs). It results in the formation of a thin, compact and highly conductive passivation layer on Li° anode, thereby avoiding dendrite formation, and polysulfide shuttling. It superbly enhances the cycling performance, Coulombic and energy efficiencies of ASSLSBs, outperforming the state-of-the-art additive lithium nitrate (LiNO3).
Li3N +e
Li+
Diffusion N2
N2
Scheme 1. Electrochemical reactions of LiN3 in lithium metal batteries.
The role of LiN3 as a novel electrolyte additive to form superior SEI on Li° anode using common poly(ethylene oxide) (PEO)-based polymer electrolyte (PE) are comprehensively investigated.[8] The physicochemical and electrochemical properties of the LiN3-added electrolytes are comparatively studied with the LiNO3-added ones, aiming at shedding light on the structure-activity relationship of these two additives. In order to find out how additives generate and enhance the SEI layer, lithium bis(trifluoromethansulfonyl)imide (Li[N(SO 2CF3)2], LiTFSI, not a good SEI builder[9]) (PE1) and lithium bis(fluorosulfonyl)imide (Li[N(SO2F)2], LiFSI, a well-known SEI former[10]) (PE0) were chosen as blank electrolytes. Due to the difficulties of access to post-mortem surface analysis from polymer cells, LiTFSI (LE1) and LiFSI (LE0) in 1,2dimethoxyethane (DME), were selected as archetype electrolytes for XPS and SEM studies, enabling to mimic the working phenomena in PEO-based electrolytes due to the analogous chemical structure of DME vs. PEO.[11] Figure 1a shows the photographs of the membranes prepared from PE1 with 2 wt.% LiN3 and LiNO3 as electrolyte additives (hereafter referred without specifying the amount of additives). It can be clearly seen that self-standing and translucent membranes are obtained, with good ductility and mechanical strengths. The X-ray diffraction (XRD) patterns and
This article is protected by copyright. All rights reserved.
Accepted Manuscript
Gebrekidan Gebresilassie Eshetu, Xabier Judez, Chunmei Li, Alex Bondarchuk, Lide M. RodriguezMartinez, Heng Zhang* and Michel Armand*
10.1002/anie.201709305
Angewandte Chemie International Edition
differential scanning calorimetric (DSC) traces of PE1 and PE0 with LiN3 or LiNO3 additives are presented in Figure 1b and c, respectively and are compared to the corresponding blank electrolytes. XRD patterns of neat salts, additives and PEO membrane are given in Figure S1. Firstly, the XRD pattern does not show any strong peak of either the salts or additives, suggesting their complete dissolution in the PEO matrix via complexation. For all the electrolytes, the typical diffraction peaks of neat PEO are observed (i.e., 2θ = 17.5 and 22.2º), suggesting the presence of highly crystalline phase in the polymer electrolyte at 25 oC. This is further confirmed by the melting transition of PEO at ca. 60 ºC in the DSC profiles (Figure 1c). Upon addition of the additives, the peak intensities of PEO in XRD and melting enthalpy in DSC slightly increase, indicating relatively higher degree of crystallinity of the polymer (see Table S1 and Figure S2 for DSC data). This would be ascribed to the increased Coulombic interactions upon the addition of LiN3 or LiNO3, since PEO itself has a strong solvating ability towards various lithium salts due to the relatively high donicity of the ethylene oxide unit (EO, donor number = 22) and favourable configurational entropy term.[8a] Thus, the LiN3 and LiNO3 additives, behaving like normal lithium salts, have non negligible interactions with PEO matrix. Possibly, the higher melting P(EO)3LiN3 and P(EO)3LiNO3 stoichiometric complexes can be formed. Figure 1d shows the thermogravimetric analysis (TGA) enlisting the influence of LiN3 and LiNO3 additives on the weight loss of the PEs. For the pristine additives, while LiNO3 is stable in the tested temperature range; LiN3 presented a very sharp mass loss at 236 ºC, which is related to the release of N2 as evidenced from mass spectra measurement (Figure S3). Surprisingly, LiN3 improves the thermal stability of PE0 while it does for LiTFSI marginally. The noticeable enhancement with LiFSI-based electrolyte could be linked to the formation of a more thermally stable [N(SO2F)(SO2N3)] anion via the reaction of N3– on the S─F bond, chemically as well as electrochemically labile linkage) compared to the C─F of TFSI–, as shown in Scheme S1b. Overall, the addition of LiN3 has a positive impact on the thermal stability, as well as lesser influence on the glass transition temperatures (Tg) of PEs (see Figure S2 for the DSC test with various LiN3 concentrations). These are beneficial requirements while hunting for functional electrolyte additives. The temperature dependence of the ionic conductivities of the as-prepared PEs is given in Figure 1e. Near room temperature range (< 50 ºC), the conductivity is found to get slightly decreased with the incorporation of both additives and this could be attributed to the increase in the crystallinity of the polymer as also evidenced from the XRD and DSC results. However, the ionic conductivities for all the electrolytes show small differences at the temperature ranging from 70 to 90 ºC, a suitable temperature domain for the operation of Li-S polymer cells. The anodic stabilities of the PEs with and without LiN3 and LiNO3 are depicted in Figure 1f. Both PE1 and PE0 employing LiN3 additive presented a mildly peak around 3.7 V vs. Li+/Li, and is associated to the oxidation of LiN3 as shown in Scheme 1. In general, the anodic stabilities are not affected by the presence of either LiN3 or LiNO3, signalling its favourability with respect to the electrochemical stability of the parent PEs. The quality of the SEI films formed between Li° electrode and the electrolyte is one of the most important elements in
rechargeable LMBs. It dictates the overall electrochemical performance enlisting cycle life, rate capability, Coulombic/energy efficiency as well as safety of the batteries. Thus, the effect of additives on the electrochemical stability towards Li° electrode is shown in Figure 2. The addition of LiN3 or LiNO3 to PE1 and PE0 proved to increase the cycling life of the stripping/plating tests, and at the same time reducing the voltage (V) hysteresis compared to blank electrolytes (Figure 2a). This indicates that both LiN3 and LiNO3 result in much more stable SEI layers. More interestingly, LiN3 provided much lower voltage compared to LiNO3, e.g., ca. 5 mV for PE1 + LiN3 and 15 mV for PE1 + LiNO3 electrolytes, linked to the high ionic conductivity of the principal passivating species (i.e., Li 3N).
Figure 1. Physicochemical and electrochemical properties of polymer electrolytes. a) Appearances of PE1 with LiN3 and LiNO3 additives. b) XRD patterns and c) DSC traces of the electrolytes. d) TGA traces for the electrolytes, and the neat components, e) Arrhenius plots of ionic conductivity and f) anodic stabilities at 70 oC for the electrolytes.
To ascertain the mechanism behind the super enhancement in the Li/PEs interfacial stability upon the addition of additives, Li plating/stripping tests, post-mortem analysis using scanning electron microscopic (SEM) and X-ray photoelectron spectroscopy (XPS), were conducted in liquid Li-S cell configuration. As seen in Figure 2b, the Li symmetric cells with LiN3 and LiNO3 in LE1 performed the same trend as shown in Figure 2a (more than 2000 h continuous cycling for LiN3-added one with a low voltage of ca. 6 mV). Figure 2c shows SEM images of Li° deposited onto the bare Cu substrate in various liquid electrolytes. Both additives have an effect on the Li° surface morphology; however, uniform and smooth pancake-like morphologies with much lower surface area were formed in LiN 3added electrolytes, instead of those fibre and needle like Li dendrites in the respective LiNO3-added and blank electrolytes. This is in agreement with the improved cycling stability of Li symmetrical cells using LiN3-added electrolytes (Figure 2a and b).
This article is protected by copyright. All rights reserved.
Accepted Manuscript
COMMUNICATION
10.1002/anie.201709305
Angewandte Chemie International Edition
COMMUNICATION LiTFSI and LiN3 seem to be stable at this potential. However, they get reduced with 2e- extraction from the biphenyl radical dianion, which is assumed to have a reduction potential very close to that of Li° (Table S3). To vividly comprehend the working mechanism of LiN3 and LiNO3 as electrolyte additives, a schematic illustration is provided in Figure 3. We hereby proposed 8e- reduction of LiN3 leading to Li3N as the only product, while LiNO3 leading to Li3N and Li2O (Figure 3b). In literature, it is also inferred that the reduction of LiNO3 could result in the formation of species such as with nitro R-NO2, and/or nitrite (NO2–) groups, Li2O,[15] LiNO2,[16] etc. However, Li2O is much less conductive as compared to Li3N.[5b] The reduction mechanisms of LiTFSI, LiFSI and DME on Li electrode are summarized in Scheme S2 for comparison. Shortly, the SEI layer obtained in the presence of LiN3 additive is richer in Li3N, more conductive as shown by lower overpotential, of high surface coverage (due to the higher molar volume of Li3N vs. Li°) and discrete in comparison with LiNO3 containing and the blank electrolytes. Li
(a) N
N
N
Conductive and compact SEI layer
LiN3 + 8 Li + 8 e
3 Li3N
Resistive and compact SEI layer
Li metal electrode O
(b)
Li
LiNO3 + 2 Li + 2 e
LiNO2 + 3 Li2O
LiNO3 + 8 Li + 8 e
Li3N + 3 Li2O
N O
resistive and discrete SEI layer
O
resistive and discrete SEI layer
Li metal electrode
Figure 3. Schematic illustration of the reaction mechanisms for the electrolytes containing LiN3 and LiNO3 additives.
Figure 2. Electrochemical behaviour of Li° electrode in the prepared electrolytes. Galvanostatic cycling of Li symmetric cells at 0.1 mA cm−2 (half cycle time 2 h) using a) polymer electrolytes at 70 oC and b) liquid electrolytes at 25 oC. c) Typical SEM images and d) XPS spectra of Li deposited onto Cu substrates at 0.1 mA cm−2 (plating time 20 h) in TFSI-based liquid electrolytes at 25 oC.
Aiming at reinforcing the above XPS results, chemical simulation of the neat LiN3, LiNO3, LiTFSI and LiFSI was carried following the procedures mentioned in literature. [12] It is found that while LiNO3 and LiFSI easily get reduced with biphenyl radical anion (ca. 0.4 V vs. Li+/Li°) via 8e- and 4e- respectively,
As a proof of concept, the feasibility of LiN 3 as an additive for ASSLSBs is manifested by cyclability test of Li-S cells at 70 ºC (Figure 4 and S8). Noteworthy, the Li-S cell using PE1 with LiN3 can deliver high reversible discharge capacity ca. 800 mAh g–1 with superior capacity retention after 30 cycles (Figure 4a), while the one with the blank electrolyte shows PS shuttling during the first charge process (Figure 4e). The addition of LiN3 into both TFSI- and FSI-based electrolytes results in an enhancement of the Coulombic efficiency compared to that of LiNO3 (Figure S8a vs. S8c; S8b vs. S8d). These impressive results suggest that stable and robust SEI layers on Li° anode can be not only formed but also maintained during continuous cycling by adding LiN 3 additive, as supported by the results on symmetric Li° cells (Figure 2). In conclusion, lithium azide (LiN3) has been identified for the first time as a novel electrolyte additive for all-solid-state Li-S batteries (ASSLSBs). The SEI layer formed on Li° electrode in the presence of LiN3 is uniform, dendrite-free, and rich in Li3N which is a highly conductive and high molar volume SEI-building material. Such superior SEI layer is found to boost the Li/electrolyte interfacial stability, leading to the enhancement of the cyclability, Coulombic/energy efficiencies and discharge
This article is protected by copyright. All rights reserved.
Accepted Manuscript
Figure 2d presents the most relevant spectra, F1s and N1s for LE1 + LiN3, LE1 + LiNO3 and blank electrolyte systems. The survey and C1s spectra are given in Figures S4 and S5, respectively. For the sake of comparison, the survey, F1s and N1s spectra of LE0-based electrolytes are provided in Figures S6 and S7, respectively. The F 1s spectra for LE1 displays two peaks at 685.2 eV and 688.7 eV, corresponding to LiF and –CF3 (from residual LiTFSI and/or its reduction products, such as Li2NSO2CF3),[12] respectively. From the intensity of the peak at 685.7 eV, it is evident that the passivation film from LE1 + LiN3 contains less LiF compared to LE1 + LiNO3. Interestingly, the most distinguishing feature comes from the N1s spectra where it strongly evidences the effect of the additives compared to the blank electrolyte. The N 1s binding energy for LE1 shows a singlet peak at ~ 399.9 eV, a characteristic binding energy of residual LiTFSI and/or its reduction derivatives. In the presence of LiN3 and LiNO3 additives, there exist additional binding energies at 398.2 eV and 404.2 eV for LE1 + LiN3 and at 398.5 eV and 402.7 eV for LE1 + LiNO3. The peak in the range of 398.2−398.5 eV is assigned to Li3N, [13] and is found to be the most predominant SEI species in the case of LE1 + LiN3. This is supported by the higher atomic percentage of N1s belonging to Li3N for the LiN3-added electrolyte, which is again corroborated by decrement in the atomic percentage of F1s (Table S2). Moreover, one of the basic requirement for good SEI-building materials is to have higher equivalent volume compared to Li° anode (i.e., 13.0 cm3 mol−1), [14] and in this regard, Li3N (27.4 cm3 mol−1) rich SEI is preferable compared to the one rich in LiF (8.9 cm3 mol−1). This makes LiN3 a relatively novel and desired electrolyte additive.
10.1002/anie.201709305
Angewandte Chemie International Edition
capacity in the Li-S cells. We hope, this finding will help to facilitate the development of practical Li-S battery systems and other related technologies in the future.
[8] [9]
[10]
[11] [12] [13]
[14] [15] [16]
a) M. Armand, Annu. Rev. Mater. Res. 1986, 16, 245; b) H. Zhang, C. Li, M. Piszcz, E. Coya, T. Rojo, L. M. Rodriguez-Martinez, M. Armand, Z. Zhou, Chem. Soc. Rev. 2017, 46, 797. a) X. Judez, H. Zhang, C. Li, J. A. González-Marcos, Z. Zhou, M. Armand, L. M. Rodriguez-Martinez, J. Phys. Chem. Lett. 2017, 8, 1956; b) X. Judez, H. Zhang, C. Li, G. G. Eshetu, Y. Zhang, J. A. González-Marcos, M. Armand, L. M. Rodriguez-Martinez, J. Phys. Chem. Lett. 2017, 8, 3473. a) H. Zhang, W. Feng, J. Nie, Z. Zhou, J. Fluorine Chem. 2015, 174, 49; b) H. Zhang, C. Liu, L. Zheng, F. Xu, W. Feng, H. Li, X. Huang, M. Armand, J. Nie, Z. Zhou, Electrochim. Acta 2014, 133, 529. D. Aurbach, J. Power Sources 2000, 89, 206. G. G. Eshetu, T. Diemant, S. Grugeon, R. J. Behm, S. Laruelle, M. Armand, S. Passerini, ACS Appl. Mater. Interfaces 2016, 8, 16087. a) D. Aurbach, E. Pollak, R. Elazari, G. Salitra, C. S. Kelley, J. Affinito, J. Electrochem. Soc. 2009, 156, A694; b) J. Yang, R. C. de Guzman, S. O. Salley, K. Y. S. Ng, B.-H. Chen, M. M.-C. Cheng, J. Power Sources 2014, 269, 520. E. Peled, S. Menkin, J. Electrochem. Soc. 2017, 164, A1703. T. Jaumann, J. Balach, M. Klose, S. Oswald, J. Eckert, L. Giebeler, J. Electrochem. Soc. 2016, 163, A557. V. Etacheri, U. Geiger, Y. Gofer, G. A. Roberts, I. C. Stefan, R. Fasching, D. Aurbach, Langmuir 2012, 28, 6175.
Figure 4. Discharge/charge profiles of the Li-S cells using polymer electrolytes at 70 oC.
Acknowledgements This work was supported by GV-ELKARTEK-2016 from the Basque Government and MINECO RETOS (ref: ENE201564907-C2-1-R) from the Spanish Government. X.J. thanks Basque Government for PhD funding and C.L. thanks the Spanish Government for the Juan de la Cierva scholarship (Ref: FJCI-2015-23898). Keywords: electrolytes; lithium metal electrode; batteries; lithium sulphur battery; lithium azide. [1] [2]
[3] [4] [5] [6] [7]
a) J.-M. Tarascon, M. Armand, Nature 2001, 414, 359; b) M. Armand, J.-M. Tarascon, Nature 2008, 451, 652; c) Z. W. Seh, Y. Sun, Q. Zhang, Y. Cui, Chem. Soc. Rev. 2016,45, 5605. a) N.-S. Choi, Z. Chen, S. A. Freunberger, X. Ji, Y.-K. Sun, K. Amine, G. Yushin, L. F. Nazar, J. Cho, P. G. Bruce, Angew. Chem. Int. Ed. 2012, 51, 9994; b) Y.-X. Yin, S. Xin, Y.-G. Guo, L.-J. Wan, Angew. Chem. Int. Ed 2013, 52, 13186; c) A. Manthiram, Y. Fu, S. H. Chung, C. Zu, Y. S. Su, Chem. Rev. 2014, 114, 11751; d) Q. Pang, X. Liang, C. Y. Kwok, L. F. Nazar, Nat. Energy 2016, 1, 16132; e) H. J. Peng, J. Q. Huang, X. B. Cheng, Q. Zhang, Adv. Energy Mater. 2017, 1700260. a) W. Xu, J. Wang, F. Ding, X. Chen, E. Nasybulin, Y. Zhang, J.-G. Zhang, Energy Environ. Sci. 2014, 7, 513; b) S. Zhang, K. Ueno, K. Dokko, M. Watanabe, Adv. Energy Mater. 2015, 5, 1500117. a) S. S. Zhang, J. Power Sources 2013, 231, 153; b) W. Li, H. Yao, K. Yan, G. Zheng, Z. Liang, Y.-M. Chiang, Y. Cui, Nat. Commun. 2015, 6, 7436. a) U. v. Alpen, A. Rabenau, G. H. Talat, Appl. Phys. Lett. 1977, 30, 621; b) Y. Zhu, X. He, Y. Mo, Adv. Sci. 2017, 1600517. Q. Zhang, J. Pan, P. Lu, Z. Liu, M. W. Verbrugge, B. W. Sheldon, Y. T. Cheng, Y. Qi, X. Xiao, Nano Lett. 2016, 16, 2011. Z. Lin, Z. Liu, N. J. Dudney, C. Liang, ACS Nano 2013, 7, 2829.
This article is protected by copyright. All rights reserved.
Accepted Manuscript
COMMUNICATION
10.1002/anie.201709305
Angewandte Chemie International Edition
COMMUNICATION COMMUNICATION G. G. Eshetu, X. Judez, C. Li, A. Bondarchuk, L. M. Rodriguez-Martinez, H. Zhang, * and M. Armand* Page No. – Page No. Lithium azide as a robust electrolyte additive for all-solid-state Li-S batteries
Accepted Manuscript
Lithium azide effectively favours the formation of dendrite-free and highly ionic conductive solid electrolyte interphase on Li electrode, and thereby improving the cycling performances and sulphur utilization of Li-S cells.
This article is protected by copyright. All rights reserved.
