Motif based Design of an Oxysulfide class of Lithium Superionic

11 hours ago - The novel concept of higher O-content with- out limiting the Li-ion conductivity could open up a new avenue of broad compositional spac...
0 downloads 0 Views 2MB Size
Subscriber access provided by Université de Strasbourg - Service Commun de la Documentation

Article

Motif based Design of an Oxysulfide class of Lithium Superionic Conductor: Towards Improved Stability and Record High Li-ion Conductivity Swastika Banerjee, Xiuwen Zhang, and Lin-Wang Wang Chem. Mater., Just Accepted Manuscript • DOI: 10.1021/acs.chemmater.9b01639 • Publication Date (Web): 20 Aug 2019 Downloaded from pubs.acs.org on August 20, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemistry of Materials

Motif based Design of an Oxysulfide class of Lithium Superionic Conductor: Towards Improved Stability and Record High Li-ion Conductivity



Swastika Banerjee,∗,‡,¶ Xiuwen Zhang,∗,§ and Lin-Wang Wang∗,¶ ‡Key Laboratory of Optoelectronic Devices and Systems of Ministry of Education and Guangdong Province, College of Physics and Optoelectronic Engineering, Shenzhen University, Shenzhen 518060, China ¶Materials Sciences Division, Lawrence Berkeley National Laboratory, 1 Cyclotron Road, Berkeley, CA 94720, USA. §Shenzhen Key Laboratory of Flexible Memory Materials and Devices, College of Physics and Optoelectronic Engineering, Shenzhen University, Shenzhen 518060, China. E-mail: [email protected]; [email protected]; [email protected]

Abstract Sulfur based solid-state-electrolyte with highest Li-ion conductivity shows promise towards next generation all-solid-state lithium-batteries. However, the moisture-driven (electro)chemical instabilities restrict them from a real battery setup. In contrast, moisture-resistant oxygen analogues exhibit poor Li-conductivity. To overcome these well known problems, we theoretically develop the chemistry of local structural motifs to build an unprecedented oxygen-sulfur mixed framework Li10 (MS4 )(PO4 )2 (M = †

Electronic Supplementary Information (ESI) available

1

ACS Paragon Plus Environment

Chemistry of Materials 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Ge and Sn) which combines the moisture stability with high Li-conductivity. Especially, Li10 (MS4 )(PO4 )2 oxysulfide framework exhibits an isotropic three-dimensional Li-diffusion associated with a lesser Li-migration barrier (0.10 ± 0.02 eV) compared to its sulfur-analogue Li10 GeP2 S12 (0.18 eV). Furthermore, oxysulfides exhibit wider electrochemical stability window compared to Li10 GeP2 S12 . As a reference for experimentalists we also tabulate the expected decomposition products at the interface while considering a number of high-performing cathodes in combination with Li10 (MS4 )(PO4 )2 electrolyte to discuss their compatibility. We emphasize that identification of specific moieties for oxygenation and sulfuration leads to a design principle for unique oxysulfide class of Li-superionic conductor (Li-SIC). The novel concept of higher O-content without limiting the Li-ion conductivity could open up a new avenue of broad compositional space for stable Li-SIC.

INTRODUCTION Solid-state Li-superionic conductor (Li-SIC) can revolutionize the all-solid-state batteries (ASSB) for the stationary applications as well as in electric automobiles. 1–3 In sharp contrast with organic liquid electrolyte (LE), inorganic solid-state electrolyte (SE) is intrinsically nonflammable and leakage-free, thus has the ultimate promise to resolve the safety issue of Li-ion batteries. 4,5 This could also overcome the unsolved problem with LE, such as, ‘shuttle effect’ and ‘transition metal dissolution effect’. 6 Among all existing SEs, sulfur based SEs (thiophosphate) are the most promising in terms of their high Li-ion conductivity (σLi ), favourable mechanical property, cost-effective synthesis as well as room-temperature processability. 7–11 It is identified that the Li-ion conductivity in SE can even surpass the value of σLi found in conventional liquid electrolytes (σLi = 12 mS cm−1 in Lithium germanium thiophosphate Li10 GeP2 S12 denoted as LGPS, 7,12 17 mS cm−1 in Li7 P3 S11 8 ). However, compared to the LE, the overall rate performance of SE in state-of-art ASSB set up is not satisfactory due to three main obstacles: (1) electrode-electrolyte interfacial incompatibility, (2) narrow elec-

2

ACS Paragon Plus Environment

Page 2 of 37

Page 3 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemistry of Materials

trochemical stability window, and (3) chemical reactivity in presence of unavoidable aerial O2 and H2 O. 13–18 In the current study, our main focus is to adopt an approach to solve the problem (3), which will also be beneficial to overcome the problems (1) and (2). A critical problem of poor rate performance and low cyclic efficiency for SE originates from the incompatibility and resistance at the electrode-electrolyte interface. Solving this issue is the main focus of recent battery research in order to combine sulfide based lithium superionic conductors with relatively high voltage cathodes and Li-metal anode. 16,19 However, all SEs are thermodynamically unstable with Li metal, forming either static or dynamic interfaces and increasing the cell impedance. 20 In addition, Lithium reduction of the electrolyte at the anode-electrolyte interface is detrimental, e.g. LGPS decomposes to form Lix Gey metallic alloy which propagates and results in cell shorting. 14 Besides, the interface with high voltage transition metal oxide cathode consists of ion-impeding layer resulting in poor rate performance of SE based batteries. 13,16,19 The necessity to minimize such interfacial reaction leads to different approaches including buffer layer coating at the cathode surface. 21,22 However, optimisation of external coating/buffer results in another dimension of complexity. In this regard, modification within electrolyte itself would be a smart approach to tune the interfacial compatibility. Together with interfacial compatibility, electrochemical stability window of an electrolyte determines the operating voltage of a battery. This window for S-based electrolytes are fundamentally limited by the low cathodic limit of S/S2− redox process (e.g. 1.9 to 2.3 V for LGPS). 16 On the other hand, every sulfur based SE undergoes anodic reduction (Li-metal alloy formation) at the bulk Li potential which results in the lowering of the anodic limit as well. Both of these electrolytic decomposition reactions result in the narrow operating voltage window and the formation of resistive layer leading to poor Coulombic efficiency (CE). 23,24 For example, the LGPS-based framework Li9.54 Si1.74 P1.44 S11.7 Cl0.3 attains the highest Li+ conductivities (25 mS cm−1 ), but, it performs with lesser CE compared to Li9.6 P3 S12 25 and oxide based electrolytes (Garnets), 26 although the later examples exhibit significantly lower

3

ACS Paragon Plus Environment

Chemistry of Materials 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 37

Li-conductivity (∼1 mS cm−1 ). 15,17,27 Despite the fact that the kinetic passivation effect increases the overpotential of SE-decomposition, a fine control over the decomposition to form the solid-electrolyte-interphase is still challenging. 16,28–31 In this regard, widening the electrochemical stability window of SE is necessary to reduce the extent of electrochemical decomposition. Future bulk scale solid state battery fabrication also necessitates the chemical stability of the components. However, all sulfur based SEs are reactive in presence of moisture at ambient condition due to the H2 O-induced degradation resulting in an irreversible loss after generation of H2 S. 15,17,27 The reaction equilibrium can be schematically described as follows,

M-S + H2 O* ) M-O + H2 S

(1)

where, M-S is the S-based SE. In a way, such moisture-instability is a serious issue because it involves relatively simple reaction path, thus is not protected by kinetic barrier. To enhance the air/moisture stability, chemical modification of sulfur based SE has been attempted through incorporation of dopants. 32,33 In particular, the oxygen analogues can avoid the above mentioned reaction equilibrium. However, the major issue with the oxide analogue is the poor Li-conductivity which is almost three orders of magnitude lesser than their sulfur counterpart. 34,35 For example, the Li-migration barrier for LGPO is 0.35±0.05 eV, while for LGPS, it is 0.20±0.04 eV. 36 From a recent experimental study, 34 the Li-ion conductivities of Li10 Ge(Si)P2 O12 is found to be 3.8 (3.5) x 10−3 mS cm−1 , which is three orders of magnitude lower than the σLi of LGPS. 7 Thus, the sulfide based SEs exhibit excellent Li-diffusivity at the cost of moisture-stability while the oxygenation results in lesser Li-diffusivity but superior moisture-stability at ambient condition, encouraging further studies on chemical modification of electrolytes through oxygen-sulfur mixing. However, it is a challenge till date to find an oxygen-rich framework, which can outperform the state-of-art S-based SEs in terms of Li-conductivity. This could be due to several practical difficulties, such as, (1) mixing oxygen within sulfur based framework would usually decrease the Li-conductivity at4

ACS Paragon Plus Environment

Page 5 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemistry of Materials

tributing to the common sense that the O-framework exhibits lesser polarisability and higher compactness. As a result, not sufficient effort has been tried to find O-rich SEs. However, relatively higher Li-conductivity were found in some of the pure oxygen based frameworks like Li1.3 Al0.3 Ti1.7 (PO4 )3 (LATP), 9 Li5 La3 Ta2 O12 (LLTO) 37 and Li7 La3 Zr2 O12 (LLZO). 10,11 Thus, the role of O-content to control the ionic diffusivity is still poorly understood. (2) Despite significant attention, the chemical stability of SEs and the fundamentals of lithium diffusion are not completely established. For example, although both the Li10 [MS4 ][PS4 ]2 (LMPS; M = Si, Ge, Sn, P, Al) and Li3 PS4 (LPS) exhibit very similar composition and structure, the σLi is 2-orders of magnitude lower in LPS. 7,38 Thus, it raises the question how does the local structural moiety namely MS4 plays a role to dictate the Li-ion conductivity. More importantly, it is worth to understand whether one can incorporate different moieties to introduce different features in a material, i.e., one for moisture-stability and another for high Li-ion conductivity. There have been efforts towards the understanding of Li-diffusion mechanism in existing SE library and it is found that the presence of polarisable S-anion sub-lattice helps in cationic migration and lower the grain boundary resistance. 39–45 However, the criteria for combination of desirable stability with ionic conductivity in oxysulfide-based SEs are still unexplored.

Figure 1: optimised structure of (a) Li10 GeP2 S12 (LGPS), (b) Li10 GeP2 O12 (LGPO), and (c) oxysulfide: Li10 GeP2 S4 O8 (LGPSO). Colour codes: PS4 -Td : yellow, PO4 -Td : red, GeS4 -Td : pink, GeO4 -Td : cyan, Li-ion : green sphere.

5

ACS Paragon Plus Environment

Chemistry of Materials 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In this work, we aim to provide a design principle for oxysulfide SEs based on Density Functional Theory and Molecular Dynamics simulations coupled with high throughput screening. We employ the well studied Li10 GeP2 S12 (LGPS) framework (see Figure 1a) to develop an insight into the influence of O-content within Li10 MP2 S12−x Ox (M = Si, Ge, Sn; x = 0-12). As a proof of concept, we derive an oxysulfide Li-SIC with a composition of Li10 MP2 S4 O8 which shifts the reaction equilibrium opposite to the H2 S generation (see eqn. 1), hence gains the moisture-stability. At the same time, it retains and even surpasses the Li-conductivity of its S-analogue. Nevertheless, our focus is not to numerically study the whole O-S compositional space, instead it is to point out the design principle of using GeS4 /PO4 combination. In contrast to the conventional understanding present study highlights that the choice of local tetrahedral moieties can circumvent the effect of polarisability which attributes to high Li-diffusivity even in O-rich oxusulfides. Moreover, there is an improved electrochemical stability window upon oxygen incorporation in sulfide framework. We also tabulate the expected decomposition products at the interface while considering a number of high-performing cathodes, revealing a distinct difference compared to its sulfide analogues. Finally, we summarize that tuning two different local structural motifs in a complementary way (through specific oxygenation/sulfuration) can combine the stability with high Li-conductivity which provides a new design approach to SE discovery.

METHODS First principles simulation details We have computed the bulk energies using first-principles density functional theory based calculations as implemented in Vienna Ab initio Simulation Package (VASP) 46,47 within the projector augmented-wave 48 approach. A cutoff energy of 520 eV is used for all calculations. k-point number multiplied with the number of atom is set to be greater than 1000. We have employed the DFT ground state structure of Li10 GeP2 S12 (LGPS) with P1 symmetry 6

ACS Paragon Plus Environment

Page 6 of 37

Page 7 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Chemistry of Materials

as the parent structure (see Figure 1a). 36 Notably, S-content in LGPS can be substituted by O-content (see Figure 1b) without major changes to the structural symmetry. All derived isovalent LMPSO structures are based on O substitution at S-sites of the DFT ground state LGPS structure, followed by structural relaxation (detailed in supporting information). For a particular value of x (0