74 75 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 1 9 / 2 0 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 1 9 / 2 0 Spectroscopy Group Beamline B18 Nanoengineering a fully reversiblemultielectron cathode for lithium-ion batteries Related publication: Rana J., ShiY., ZubaM. J., Wiaderek K. M., Feng J., Zhou H., Ding J.,WuT., Cibin G., BalasubramanianM., Omenya F., Chernova N. A., Chapman K.W.,WhittinghamM. S. & Piper L. F. J. Role of disorder in limiting the truemulti-electron redox in ε-LiVOPO 4 . J. Mater. Chem. A 6 , 20669 (2018). DOI: 10.1039/C8TA06469E Publication keywords: Photocatalyst; Metal oxide; X-ray Photoelectron Spectroscopy; Electronic structure E lectric vehicles (EV) are expected to replace the internal combustion engine by 2050. Despite impressive progress over the last decade, several technological hurdles still need to be overcome to make EVs practical and economical. One of the major challenges to the EV industry is the driving range, which is dictated by the energy density of the battery pack. This, in turn, is governed by the number of electrons that can be exchanged per transition-metal (TM) cation in the cathodematerials. Currently, the state-of-the-art cathodematerials used in EV, such as the Tesla Model 3 battery pack, exchange less than one electron per TM cation. One strategy being explored to improve energy density is to use intercalated (‘doped’) compounds capable of multiple electron transfer per TM cation. Vanadyl phosphates (VOPO 4 ) are an appealing class of compounds, since they offer two-electron exchange within the voltage window safe and suitable for EV applications. Researchers used core X-ray Absorption Spectroscopy (XAS) at beamline B18 to directlymonitor the vanadiumoxidation state of ε-VOPO 4 . This technique was vital for identifying side reactions promoted by high energy ball milling-induced disorder and defects and for identifying a path to realising fully reversible two lithium ion intercalation. The results demonstrate that realising full two lithium intercalation in ε-VOPO 4 requires the consideration of different kinetics and reaction pathways associated with multielectron reactions. In addition, parasitic side reactions associated with electrolyte degradation can complicate the analysis of the electrochemistry. Realising full two lithium intercalation in ε-VOPO 4 requires the consideration of different kinetics and reaction pathways associated with multielectron reactions 1 . The two-electron exchange in vanadyl phosphates (VOPO 4 ) involveV 5+ /V 4+ and V 4+ /V 3+ redox reactions (Fig. 1). Various polymorphs of VOPO 4 exist but the epsilon (ε) phase is considered to be the most promising for full reversible two lithium ion intercalation. The insertion of Li + into ε-LiVOPO 4 (V 4+ /V 3+ redox at ~2.5V vs. Li/Li + ) during discharge is kinetically favored over the extraction of Li + from ε-LiVOPO 4 (V 5+ /V 4+ redox at ~4.0V vs. Li/Li + ) during charge 2 . As a result, overcoming the kinetic limitations of the high-voltage region was critical for realising full two lithium ion intercalation in ε-VOPO 4 . Reducing the Li-ion diffusion pathways can reduce the kinetic barriers of high-voltage region. High-energy ball milling is an essential step in the synthesis to obtain materials with finer particles. However, increased ball milling deteriorated the electrochemical performance of our cathodes 3 . Our operando XAS measurements revealed the lack of evolution of vanadium oxidation state despite significant capacity during charge, which clearly indicated that side reactions were promoted by the high-energy ball milling 4 . To investigate how the ball milling was further hindering the already sluggish kinetics and promoting side reactions, rate dependent XAS studies were performed at beamline B18 of Diamond Light Source. Tracking of vanadium oxidation state in the ε-LiVOPO 4 electrodes lithiated/delithiated at different rates revealed the expected reduction of V 4+ to V 3+ upon inserting the second lithium ion into ε-LiVOPO 4 and subsequent oxidation of V 3+ back to V 4+ upon its extraction, irrespective of rate 4 . In contrast, the extraction of Li + from ε-LiVOPO 4 was found to be sluggish as the oxidation of V 4+ to V 5+ increased at slower rates (Fig. 2). Yet, even at the slowest cycling rate of C/100 (which exceeded the theoretical capacity for two Li + extraction) only half of the vanadium was oxidised to the expected V 5+ . Quantitative analysis of the XAS data further revealed a significant portion of the capacity arose from side reactions during charge in the high-voltage region at all rates. These data demonstrate the kinetic limitations imposed by ball milling-induced disorder and/or defects that ultimately trigger detrimental side reactions and hinder full V 4+/5+ redox. Prompted by this finding our recent work employed hydrothermally synthesised (i.e. ball milling-free approach) ε-VOPO 4 with highly crystalline nanosized particles, which demonstrated reversible full 2Li intercalation without any indication of side reactions 5 . This research was awarded a Ten at Ten Scientific Ideas award by the U.S. Department of Energy. Future efforts are dedicated to further improve rate-capabilities and minimise the voltage gap between the high-voltage and low-voltage plateaus through chemical substitutions of nanoengineered ε-VOPO 4 . In summary, the present research reiterates the need for synchrotron studies in battery research to identify detrimental side reactions in multielectron cathodes, which would otherwise remain elusive based on the electrochemical characterisation alone. References: 1. Lin Y.-C. et al. Thermodynamics, Kinetics and Structural Evolution of ε-LiVOPO 4 over Multiple Lithium Intercalation. Chem. Mater. 28 , 1794- 1805 (2016). DOI: 10.1021/acs.chemmater.5b04880 2. Wangoh L. W. et al . Uniform second Li ion intercalation in solid state ε -LiVOPO 4 , Appl. Phys. Lett. 109 , 053904 (2016). DOI: 10.1063/1.4960452 3. Shi Y. et al . Electrochemical Performance of Nanosized Disordered LiVOPO 4 . ACS Omega . 3 , 7310-7323 (2018). DOI: 10.1021/acsomega.8b00763 4. Rana J. et al . Role of disorder in limiting the true multi-electron redox in ε-LiVOPO 4 . J. Mater. Chem . A. 42 , 20669-20677 (2018). DOI: 10.1039/C8TA06469E 5. Siu C. et al . Enabling multi-electron reaction of ε-VOPO 4 to reach theoretical capacity for lithium-ion batteries. Chem. Commun . 54 , 7802- 7805 (2018). DOI: 10.1039/C8CC02386G Funding acknowledgement: This work was supported as part of the NorthEast Center for Chemical Energy Storage (NECCES), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences under Award# DE- SC0012583. Corresponding author: Dr Louis F. J. Piper, Binghamton University,
[email protected] Li + = 1 Li + = 2 Li + = 0 Capacity 4 3 2 Voltage Poor Kinetics? V 3+ V 4+ V 5+ Figure 1: Schematic illustration of the voltage profiles of ε-VOPO 4 indicating poor kinetics of the high-voltage region caused by high energy ball milling-induced disorder/defects. 5474 5472 5470 5468 5466 Photon energy (eV) Chg. 4.5V Obs. Exp. V valence 350 300 250 200 150 100 50 0 2Li 1Li C/10 C/20+10h CV C/50 C/100 Decreasing rate of Li extracton Specific charge capacity (mA h g -1 ) Facile (V 3+ V 4+ ) (1.6V - 3.5V) Sluggish (V 4+ V 5+ ) 3.5V - 4.5V Side rxns (3.5V - 4.5V) 1.6V - 3.5V 3.5V - 4.5V (based on obs. V 5+ ) 3.5V - 4.5V (based on Echem.) Dis. 1.6V Obs. Exp. V valence 5472 Photon energy (eV) 5469 5466 5463 @ C/100 4.5 4.0 3.5 3.0 2.5 2.0 1.5 -150 -100 -50 0 50 100 150 C/10 C/20 C/50 C/100 Voltage vs. Li/Li + (V) Specific capacity (mA h g -1 ) 1C=153 mA g -1 2Li 1Li Excess capacity (a) (b) Figure 2: (a) voltage profiles of ε-LiVOPO 4 at different cycling rates and (b) sluggish kinetics of the high-voltage region giving rise to side reaction contributions at all rates. Reproduced from 4 . Li + = 1 Li + = 2 Li + = 0 Capacity 4 3 2 Voltage V 3+ V 4+ V 5+ Figure 3: Schematic illustration of the voltage profiles of highly crystalline, nanoengineered (ball milling-free synthesis) ε-VOPO 4 showing reversible full 2Li intercalation.