In operando X-shaped cell online electrochemical mass spectrometry (OEMS): New online analysis enables insight into lab scale lithium ion batteries during operation

https://doi.org/10.1016/j.jelechem.2016.04.023Get rights and content

Highlights

  • New method for the in operando analysis of evolving gases during cyclic aging of lithium ion batteries

  • Application of the new system to LiNi1/3Co1/3Mn1/3O2 (NCM)/Li half cells

  • Identification of decomposition products with a new method in commercially electrolytes and self-assembled cells

Abstract

A new method for the in operando analysis of evolving gases during cyclic aging of lithium ion batteries (LIBs) was developed to better assess safety concerning cell processes, especially those arising from the electrochemical degradation of the lithium hexafluorophosphate LiPF6/organic carbonate solvent based electrolyte. For electrochemical characterization at lab-scale, a cell in the shape of T-connector (“T-cell”) is usually used, offering connections to working, counter and reference electrode. To maintain comparability to this established system, an in operando X-shaped cell, i.e., a T-cell (“X-cell”), which varies only by an additional connector from the original setup, was designed. The new OEMS cell based on DEMS cell designs was linked to a modified GC–MS System and a potentiostat for in operando analysis of the evolving gases and the voltammetry experiments, respectively. This work comprises the evaluation of this new OEMS method in potentiostatic aging experiments of the conventional electrolyte 1M LiPF6 in EC:EMC (1:1, by wt.) in LiNi1/3Co1/3Mn1/3O2 (NCM)/Li half cells as a function of the applied cut-off potential. Mainly CO2 release at onset potentials > 4.6 V vs. Li/Li+ could be identified. At a potential of > 5.4 V vs. Li/Li+, the evolution of silicon tetrafluoride (SiF4) was observed mainly stemming from the HF induced degradation of the used glass fiber separator. Furthermore, triethyl phosphate (TEP) evolved from the LiPF6 decomposition at > 5.5 V vs. Li/Li+. Oxygen evolution either coming from the oxidative decomposition of the electrolyte or degradation of the NCM cathode material was not detected at even 5.5 V vs. Li/Li+ and at 20 °C.

Introduction

Advantages like, high cell voltage, low self-discharge rate and very high energy density have led to the extended proliferation of lithium ion batteries (LIBs) in portable consumer electronics and in the emerging market of E-mobility [1], [2], [3]. However, to be implemented as large-scale traction batteries, life and safety requirements are raised [4], [5]. R&D efforts are focused on the development of new positive electrode materials owing to higher operation voltages and larger specific charges, in order to increase the resulting energy and power of the LIB. Safety issues in LIBs arise to large extent from flammable electrolyte components, which release potential toxic degradation products for example during thermal runaway [6], [7], [8], [9], [10], [11], [12], [13], [14], as well as possible gas evolution of either the electrolyte, or the electrodes under abuse conditions [15], [16], [17], [18]. Previous publications deal with the post-mortem analysis of the harvested electrolyte storage tests (calendar life), abuse tests (after thermal aging) and electrochemical processing (cycle life) [6], [10], [19], [20], [21], [22], [23]. However, to fulfill the need for safety in large-scale applications one necessary aspect is to characterize toxic compounds and detect the buildup of pressure within the LIB cell due to evolution of gaseous products during operation of the cell.

The use of the DEMS method for online analyses is well-known in literature [24], also with regard to investigation of lithium ion battery components [25], [26]. DEMS instruments basically consist of three main parts: 1. Electrochemical cell, 2. Membrane interface (PTFE or other fluorinated polymer membrane, frit) and 3. Vacuum system with the quadrupole mass spectrometer [27]. A porous membrane is used to separate the electrochemical cell, especially the electrolyte, from the vacuum system. Until now there is no commercial DEMS system available [27]. Self-made solutions of the research groups [7], [25], [28], [29], [30], [31], [32], [33], [34], [35] have to deal with several challenging aspects. For example minor comparability of results between the research groups, high material consumption, partial inert cell housings and design intrinsic difficult reproducibility.

The main limitation is given by the fluorinated membranes in the DEMS systems [36]. Compounds have to pass through the membrane and according to that only gaseous or volatile and relatively non-polar substances with a matching size are able to evaporate through the membrane. In addition to the porosity of the membrane, an additional phase transition came into the system and thus absorbance of organic compounds and other reactions is possible. Furthermore, DEMS systems have to deal with the problem of shifting the reaction equilibrium during the measurement [24]. Finally, cold traps, pressure reducing valves or similar devices may hold molecules back and thus make true online measurements impossible, but increase the limit of detection [37], [38].

The developed in operando X-shaped cell online electrochemical mass spectrometry (OEMS) technique, oriented on DEMS systems, enables the detection of volatile gaseous substances during operation of the LIB. Hereby, OEMS monitors the time dependent intensities of selected mass/charge signals almost direct to the applied electrochemical experiment, which is comparable to a continuously performed headspace measurement. Multiple headspace extraction technique based on dynamic gas extraction carried out repetitively, is the method of choice for headspace quantification and the theoretical foundation for our new designed system [39]. Many challenges of existing DEMS methods could be overcome by avoiding an excessive amount of electrolyte compared to the active electrode materials. Instead of > 2 mL [40], 600 μL [25] or 400 μL [8] or flow through solutions [41] in our setup 120 μL is sufficient. Moreover massive dilution of the parasitic reaction products with carrier gas was prevented. Thus both effects leading to poor validity of the obtained results are excluded. Handling and sample preparation are simple and the often criticized fluorinated membranes or porous frits are not used in the newly developed method [24]. Furthermore, glassy components in the setup can be avoided.

The main challenge of our new setup was to overcome pressure differences between the cell and the applied vacuum in the MS device which could be solved with a capillary connection to the MS. As a consequence of the missing membrane, direct detection particularly of polar substances is enhanced. It remains as difficulty that electrochemical reactants which pass in the capillary are sensitive to the inlet position of the capillary at the top of the cell setup, which varies slightly between measurements. This leads to a challenging quantification because of the convection in the cavity, which results in a longer response time, as in other DEMS systems [24], [27]. The focus of this work is the on the detection of the evolution of O2, CO2 and the decrease of the EMC vapor signal as an indicator for electrolyte decomposition [42], [43], [44].

Section snippets

Sample Preparation

The preparation of the X-shaped cell was carried out in an argon-filled (argon 4.6, Westfalen AG) glove box (H2O < 0.1 ppm; O2 < 0.1 ppm). 1M LiPF6 dissolved in ethylene carbonate/ethyl methyl carbonate (EC/EMC 1:1, by wt. LP50, Selectilyte™, BASF) was used as the electrolyte in all experiments. LiNi1/3Co1/3Mn1/3O2 (NCM) electrodes with an average active mass loading of 14 mg cm 2 were provided by the BMW group and used as working electrode. As counter and reference electrode metallic lithium

Results and Discussion

First the cell was flushed with helium and continuously monitored for 11 h. Constant values of the air residue signals (impurity background, e.g. O2 and N2) in the used helium gas (purity 6.0) were reached after 60 min and represent the lowest possible contaminations (Fig. 2). Therefore, all upcoming measurements included a 60 min rest step at the beginning, which is much better than 120 min [8]. Furthermore, the measured background intensities were used for background substraction in the following

Conclusion

Within this work, a proof of principle setup of a newly developed OEMS method was demonstrated. A T-shaped three-electrode cell with one additional connector for hyphenation possibilities was designed to measure evolving gases during LIB full cell operation in real time. Advantages of the system, like high comparability to electrochemically generated data, in particular with regard to the cell setup, no additional phase transition (e.g. at fluorinated membranes), inert cell housing and no

Acknowledgments

The authors gratefully acknowledge the financial support by the German Federal Ministry of Education and Research (BMBF) for funding this work in the projects ‘Electrolyte-Lab 4e’ (03X4632) and ‘SafeBatt’ (03X4631N). The material support by BMW Group and BASF is especially acknowledged.

References (54)

  • L. Terborg et al.

    Investigation of thermal aging and hydrolysis mechanisms in commercial lithium ion battery electrolyte

    J. Power Sources

    (2013)
  • H. Baltruschat

    Differential electrochemical mass spectrometry

    J. Am. Soc. Mass Spectrom.

    (2004)
  • M. Winter et al.

    FTIR and DEMS investigations on the electroreduction of chloroethylene carbonate-based electrolyte solutions for lithium-ion cells

    J. Power Sources

    (1999)
  • M. Lanz et al.

    DEMS study of gas evolution at thick graphite electrodes for lithium-ion batteries: the effect of γ-butyrolactone

    J. Power Sources

    (2001)
  • Y. Gao et al.

    New on-line mass spectrometer system designed for platinum-single crystal electrode and electroreduction of acetylene

    J. Electroanal. Chem.

    (1994)
  • C. Niether et al.

    Development of a novel experimental DEMS set-up for electrocatalyst characterization under working conditions of high temperature polymer electrolyte fuel cells

    J. Electroanal. Chem.

    (2015)
  • H. Buqa et al.

    SEI film formation on highly crystalline graphitic materials in lithium-ion batteries

    J. Power Sources

    (2006)
  • B.B. Berkes et al.

    Simultaneous acquisition of differential electrochemical mass spectrometry and infrared spectroscopy data for in situ characterization of gas evolution reactions in lithium-ion batteries

    Electrochem. Commun.

    (2015)
  • M. Holzapfel et al.

    Oxygen, hydrogen, ethylene and CO2 development in lithium-ion batteries

    J. Power Sources

    (2007)
  • S.E. Sloop et al.

    The role of Li-ion battery electrolyte reactivity in performance decline and self-discharge

    J. Power Sources

    (2003)
  • D.P. Abraham et al.

    Diagnostic examination of thermally abused high-power lithium-ion cells

    J. Power Sources

    (2006)
  • P. Lanz et al.

    Oxygen release from high-energy xLi2MnO3·(1 − x)LiMO2 (M = Mn, Ni, Co): Electrochemical, differential electrochemical mass spectrometric, in situ pressure, and in situ temperature characterization

    Electrochim. Acta

    (2013)
  • J.-M. Kim et al.

    Role of transition metals in layered Li[Ni,Co,Mn]O2 under electrochemical operation

    Electrochim. Acta

    (2004)
  • D.R. Gallus et al.

    The influence of different conducting salts on the metal dissolution and capacity fading of NCM cathode material

    Electrochim. Acta

    (2014)
  • D.R. Gallus et al.

    New insights into the structure–property relationship of high-voltage electrolyte components for lithium-ion batteries using the pKa value

    Electrochim. Acta

    (2015)
  • S.F. Lux et al.

    The mechanism of HF formation in LiPF6 based organic carbonate electrolytes

    Electrochem. Commun.

    (2012)
  • J. Vetter et al.

    In situ study on CO2 evolution at lithium-ion battery cathodes

    J. Power Sources

    (2006)
  • Cited by (22)

    • Charactering and optimizing cathode electrolytes interface for advanced rechargeable batteries: Promises and challenges

      2022, Green Energy and Environment
      Citation Excerpt :

      Analyzed by the online electrochemical MS, Fig. 11d shows the change in the release of O2 and CO2 at different potentials through the electrolyte (1 mol L−1 LiPF6 in EC + EMC) decomposition and the degradation of NCM cathode. It is inferred that the growth of the CEI causes additional diffusion barrier and electron transfer, thereby deactivating the reaction on the electrode surface [134]. Fig. 11e introduces the application of operando differential electrochemical MS, which evaluates the gas released during the first cycle of the charge–discharge process.

    • A novel DEMS approach for studying gas evolution at battery-type electrode|electrolyte interfaces: High-voltage LiNi<inf>0.5</inf>Mn<inf>1.5</inf>O<inf>4</inf> cathode in ethylene and dimethyl carbonate electrolytes

      2019, Electrochimica Acta
      Citation Excerpt :

      Also these components can evaporate and enter the analysis chamber of the mass spectrometer to some extent through the non-porous membrane. The latter problem becomes even more severe when using a porous membrane interface [43,44], or for the head space approach [24,25,53,56,66–69]. In the former case, this limits the choice of the organic electrolytes to those having a low vapor pressure [42,45–47].

    • Lithium-secondary cell: Sources of risks and their effects

      2018, Electrochemical Power Sources: Fundamentals, Systems, and Applications Li-Battery Safety
    • Capacity fading behavior of Ni-rich layered cathode materials in Li-ion full cells

      2016, Journal of Electroanalytical Chemistry
      Citation Excerpt :

      Some important factors that can be evaluated in full-cell tests include active material loading level, electrode density, heat produced in a large-scale cell, mechanical degradation by electrode swelling, and supply limitation of Li to the cathode by irreversible consumption in the graphite anode. From the postmortem analysis of electrodes cycled in full cells, we can observe the actual fading behavior of batteries in various applications [21–26]. In this study, we prepared pouch-type full cells by employing commercial graphite as a negative electrode (anode) material and high-Ni layered oxide (LiNi0.8Co0.1Mn0.1O2) as a positive electrode (cathode) material.

    View all citing articles on Scopus
    View full text