Subscriber access provided by UNIV OF LOUISIANA
A: Kinetics, Dynamics, Photochemistry, and Excited States
Singlet Oxygen Reactivity with Carbonate Solvents Used for Li-Ion Battery Electrolytes Anna T. S. Freiberg, Matthias K. Roos, Johannes Wandt, Regina de Vivie-Riedle, and Hubert A. Gasteiger J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.8b08079 • Publication Date (Web): 24 Oct 2018 Downloaded from http://pubs.acs.org on October 26, 2018
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 19 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
The Journal of Physical Chemistry
Singlet Oxygen Reactivity with Carbonate Solvents used for Li-Ion Battery Electrolytes Anna T.S. Freiberg,‡a* Matthias K. Roos,‡b Johannes Wandt,a Regina de Vivie-Riedle,b** Hubert A. Gasteigera aChair
of Technical Electrochemistry, Department of Chemistry and Catalysis Research Center, Technische Universität München, Garching, Germany bDepartment
of Chemistry, Ludwig-Maximilians-Universität München, München, Germany
Singlet Oxygen, ab initio calculations, anodic oxidation of H2O2, chemical oxidation of ethylene carbonate
ABSTRACT: High degrees of delithiation of layered transition metal oxide cathode active materials (NCMs and HE-NCM) for lithium-ion batteries (LIBs) was shown to lead to the release of singlet oxygen, which is accompanied by enhanced electrolyte decomposition. Here we study the reactivity of chemically produced singlet oxygen with the commonly used cyclic and linear carbonate solvents for LIB electrolytes. On-line gassing analysis of the decomposition of ethylene carbonate (EC) and dimethyl carbonate (DMC) reveals different stability toward the chemical attack of singlet oxygen, which is produced in-situ by photo-excitation of the Rose bengal dye. Ab initio calculations and on-the-fly simulations reveal a possible reaction mechanism, confirming the experimental findings. In the case of EC, hydrogen peroxide and vinylene carbonate (VC) are found to be the products of the first reaction step of EC with singlet oxygen in the reaction cascade of the EC chemical decomposition. In contrast to EC, simulations suggested DMC to be stable in the presence of singlet oxygen, which was also confirmed experimentally. Hydrogen peroxide is detrimental for cycling of a battery. For all known cathode active materials, the potential where singlet oxygen is released is found to be already high enough to electrochemically oxidize hydrogen peroxide. The formed protons and/or water both react with the typically used LiPF6 salt to HF that then leads to transition metal dissolution from the cathode active materials. This study shows how important the chemical stability toward singlet oxygen is for today’s battery systems and that a trade-off will have to be found between chemical and electrochemical stability of the solvent to be used.
INTRODUCTION Lithium-ion batteries (LIBs) power today’s consumer electronics such as laptops and mobile phones.1-2 Their use in battery electric vehicles (BEVs) promises to substantially reduce the (local) CO2 emissions in the transportation sector.3 This, however, still requires significant increases in the gravimetric as well as volumetric energy density of LIBs in order to reduce battery weight/volume and to enable BEV driving ranges competitive with conventional combustion engine powered vehicles.4-5 Most state-of-the-art LIBs rely on a layered transition metal oxide cathode active material6 (CAM), currently paired with a graphite anode or, in the mid- to near-term with a higher energy density silicon anode.5, 7 To further increase the energy density, two types of mixed transition meal layered oxides are being investigated as high energy density CAMs, namely so-called Ni-rich NCMs (Li1(NixCoyMnz)O2, with x+y+z=1)8 and Li-rich9 NCMs also referred to as HE-NCMs (Li1+w(NixCoyMnz)1-wO2, with x+y+z=1 and w typically ≤0.2). While NCMs can
theoretically deliver discharge capacities of up to 280 mAh/g upon complete delithiation, the long-term charge/discharge cycling stability of NCMs can only be met by limiting the degree of delithiation to ~70% state of charge (SOC),10-12 i.e., by limiting the discharge capacity to ~190 mAh/g. Delithiation beyond ~70% SOC leads to higher charge cut-off potentials, which are reported to promote electrochemical electrolyte oxidation,13 whereby protic species proposed to result from electrolyte oxidation14 are the likely reason for the observed transition metal leaching from the CAM10-11, 15-18 and the cell impedance growth.10-11 Furthermore, the reaction of electrolyte oxidation products and dissolved transition metal ions with the graphite anode can chemically delithiate the graphite anode, thereby leading to an irreversible active lithium loss.15, 17 Oxygen release is another issue and has been known to occur for NCMs and HE-NCMS at high SOC,12, 19-23 indicating changes in the active material composition. The surface-near layers change from a layered to a spinel-like structure and/or a rock-salt structure.24-27 Simultaneous to
ACS Paragon Plus Environment
The Journal of Physical Chemistry 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
the O2 release from the near-surface region, the formation of electrolyte decomposition products such as CO2 and CO was detected by on-line electrochemical mass spectrometry (OEMS) by Jung et al.12 and Guéguen et al.23 Jung et al.12 emphasized that the point at which O2 is evolved depends on the SOC rather than on the cathode potential. The simultaneous evolution of O2, CO2 and CO was therefore assigned to a chemical reaction of the electrolyte with the released O2 rather than to the electrochemical decomposition of the electrolyte, as the latter would depend on cathode potential and not on the SOC. The reactive O2 released from the cathode was postulated to be singlet oxygen (excited state 1∆𝑔, furtheron referred to as 1O2). Recently, we were indeed able to prove the SOC dependent release of 1O2 from NCMs and HE-NCM at high SOC using an operando photomultiplier set-up to record the 633 nm photon emission of the collision-induced bimolecular radiative decay of 1O2 to its triplet state.28 As will be shown in this work, the following seemingly unrelated recent finding may also be linked to the release of singlet oxygen: It was demonstrated by the Dahn group13, 29-31 that electrolytes free of ethylene carbonate (EC) show superior capacity retention for NCM based CAMs operating at high cathode potentials (i.e., at high SOCs), whereby a lower overall impedance growth and a lower extent of gassing was observed.13, 30 This finding is somewhat puzzling, as it could not be explained by solely considering the electrochemical anodic stability of the solvents, as the cyclic ethylene carbonate has superior stability at high anodic potentials compared to the experimentally observed superior linear carbonate solvent (ethyl methyl carbonate (EMC)).32-34 A possible hypothesis to explain the inferior performance of EC solvent may be that it will react with 1O2 released at high SOC, forming reaction products, which are detrimental to battery durability. Therefore, we will examine in this study the reactivity and the reaction products of the commonly used cyclic ethylene carbonate solvent with singlet oxygen and then compare it with the reactivity of a typical linear carbonate (dimethyl carbonate (DMC)) solvent. To this purpose, the reaction of in-situ formed 1O2 with EC and DMC was followed by analyzing the gaseous decomposition products and the consumption rate of oxygen by on-line mass spectrometry. To gain further insights into the reaction pathways and the nature of the reaction products, ab initio calculations are performed. The simulations identify the key characteristics, which govern the stability of aprotic electrolyte solvents toward 1O2. As H2O2 is predicted and found to be formed upon reaction of EC with 1O2, its decomposition pathway in the battery environment is further evaluated by on-line electrochemical mass spectrometry (OEMS) measurements. Finally, we will show that the here presented ab initio simulations and experimental findings on the reaction and reaction products of singlet oxygen with carbonate based electrolyte solvents are able to rationalize some of the observed aging mechanisms observed for LIBs employing
Page 2 of 19
NCM or HE-NCM cathodes. We therefore believe that our analysis will be able to guide the design of improved electrolyte solvents for use with oxygen releasing cathode active materials.
MATERIALS AND METHODS On-line gassing analysis during the reaction of 1O2 with EC and DMC The reaction of singlet oxygen with EC and DMC (both battery grade, < 10 ppm H2O, BASF, Germany) is studied in a set-up adapted from our cell design for on-line electrochemical mass spectrometry (OEMS).35 Here, singlet oxygen (1O2) is produced in triplet oxygen (3O2) saturated solutions of EC or DMC using Rose bengal (RB, disodiumsalt, >95%, Sigma Aldrich) as photosensitizer to form 1O2 in solution from 3O2. Exciting RB with light at 525 nm ultimately leads to the formation of unstable triplet state RB with an excitation energy of 42 kcal/mol,36 which transfers its excitation energy to a nearby 3O2 by promoting it to the first singlet state. RB was chosen due to: i) its sufficient solubility in carbonate solvents; ii) its high quantum yield in solution and efficiency to convert 3O2 to 1O 37 and, iii) its high chemical stability.38 The RB powder 2; was dried for 72 h at 130 °C under vacuum in a glass-oven (Büchi, Switzerland), and solutions of 100 µM RB in EC and DMC were prepared in an Ar-filled glovebox (MBraun, < 0.1 ppm H2O, < 0.1 ppm O2) by stirring on a hot-plate for several days at 50°C (i.e., above the melting temperature of EC). The set-up for the on-line mass spectrometry (MS) measurements was assembled inside the glovebox to avoid contamination from ambient air. For on-line MS analysis, the electrochemical cell of our OEMS set-up was replaced by a standard UV-Vis cuvette (square 10 mm bottom, 2 transparent sides, 4 ml volume, Spectrosil glass, Starna, Germany) that is connected via a GL-tube-fitting (PPS, Bola, Germany) to a metal fitting, which in turn connects directly to the MS via a flowrestricting capillary (at a standard flow rate of ~1 µl/min). The cuvette was filled with either 4 or 1 ml of the carbonate solutions with RB dye, leaving a head-space volume of either 2.7 or 5.7 ml, respectively, which can be purged with argon or oxygen. A 6 mm magnetic stirrer was added to induce some convective mixing. The cuvette is located in a non-transparent plastic encasing that prevents unintended excitation of the photosensitizer and that holds 4 LEDs (two on either side of the cuvette) with a characteristic wave-length of 525 nm at a maximum power of 3 W each (at an operating voltage of 3 V). The LEDs are connected in series and placed onto cooling fins to remove most of the heat generated during illumination. The LED power supply cables, the metal tube connecting the cuvette with the MS capillary inlet and the gas flushing valves are fed through the encasing. For details of the assembled cell and its location in the encasing see Fig. S4. After assembly, the setup is placed onto a magnetic stirrer inside a temperature chamber at 45°C (chosen to be above the EC melting point), the capillary is connected to the mass spectrometer, and the set-up is flushed with pure argon and left to equilibrate for 4 h before starting an experiment. During
2 ACS Paragon Plus Environment
Page 3 of 19 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
The Journal of Physical Chemistry
these 4 h, the leak tightness of the assembly is assured prior to each measurement by following the MS signal for nitrogen; the measured nitrogen leaks were also negligible (
