Isotopically Labeled Carbon and Water to Deconvolute CO2 Evolution in Li-Batteries by on-Line Electrochemical Mass Spectrometry

Future Li-ion batteries are expected to reach high energy density by the use of high voltage cathode materials. At the high operating voltages (≥ 4.5 V vs. Li/Li⁺) most of the common alkyl carbonate electrolytes start to decompose followed by an evolution of gases upon cycling. In recent studies, Tsiouvaras et. al. and McCloskey et. al.  showed that CO₂ is the main gas evolved by the anodic oxidation of water free alkyl carbonate based electrolytes (e.g. PC) at high electrode potentials (≥ 4.5 V vs. Li/Li⁺) [1][2]. A recent study by our group [3], demonstrated that CO₂ is already evolved at lower potentials if the electrolyte contains traces of water. An accumulation of gas in a sealed battery cell after several cycles is a potential safety threat and its origins need to be understood.

In this work we employ on-line electrochemical mass spectrometry (OEMS) to analyze the gas evolution in battery test cells over a wide electrode potential window (2 - 6 V vs. Li/Li⁺). CO₂ is a known degradation product, but it can be either due to the electrochemical oxidation of the electrolyte or the carbon in the cathode through the reaction:

           C + 2H₂O    →    CO₂ + 4H⁺ + 4e^-            (1)

In order to deconvolute these two processes, isotopically labeled ¹³C carbon (99 atom% ¹³C, Sigma-Aldrich) is used to prepare a model electrode that consists exclusively of ¹³C carbon. This allows to assign the evolved CO₂ isotopes to either the degradation of the electrode or the electrolyte. In the context of Li-O₂ batteries, this strategy was used to analyze the formation of Li₂CO₃ in presence of Li₂O₂ [4].

Furthermore, the effect of water on the degradation of carbon and electrolyte is investigated. By adding a defined amount of water (4000 ppm) to the electrolyte one can mimic the effect of trace water that could unintentionally be introduced to the cell through the active materials. Joho et. al. studied the irreversible charge loss due to water reduction in graphite based Li-ion cells, however no correlation between water content and cycling stability could be found [5]. To clarify whether water promotes exclusively the degradation of either the carbon electrode or the electrolyte or the degradation of both components, isotopically labeled ¹⁸O-water (98 atom% ¹⁸O, Sercon) is added to the electrolyte.

Figure 1 features a linear potential sweep and the corresponding gas evolution in OEMS for a ¹³C carbon cathode and LP30 electrolyte containing 4000 ppm H₂¹⁶O. An evolution of the ¹³C¹⁶O₂ isotope of carbon dioxide is detected together with ¹²C¹⁶O₂ at potentials of 4.6 V vs. Li/Li⁺. The first isotope is likely to originate from the electrode corrosion, while the latter could result from the oxidation of the electrolyte. The signals at channels 48 and 49, corresponding to the ¹²C¹⁸O₂ and ¹³C¹⁸O₂ isotopes, respectively are not detected, since this system contains no labeled ¹⁸O-water.

References:

[1] N. Tsiouvaras, S. Meini, I. Buchberger, and H. A. Gasteiger, J. Electrochem. Soc., 160, A471 (2013).

[2] B. D. McCloskey, D. S. Bethune, R.M. Shelby, G. Girishkumar, and A. C. Luntz, J. Phys. Chem. Lett., 2, 1161 (2011).

[3] R. Bernhard, S. Meini, and H. A. Gasteiger, J. Electrochem. Soc., 161, A497 (2014).

[4] M. M. O. Thotiyl, S. A. Freunberger, P. Zhangquan, P. G. Bruce, J. Am. Chem. Soc., 135, 494 (2013).

[5] F. Joho, B. Rykart, R. Imhof, P. Novák, M. E. Spahr, A. Monnier, J. of Power Sources., 81-82, 243 (1999).

Acknowledgment:

Support of BASF SE in the framework of its scientific network on electrochemistry and batteries is acknowledged by TUM.