Potassium Metal as Reliable Reference Electrodes of Non-Aqueous

Subscriber access provided by Stockholm University Library

Energy Conversion and Storage; Plasmonics and Optoelectronics

Potassium Metal as Reliable Reference Electrodes of Non-Aqueous Potassium Cells Tomooki Hosaka, Shotaro Muratsubaki, Kei Kubota, Hiroo Onuma, and Shinichi Komaba J. Phys. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acs.jpclett.9b00711 • Publication Date (Web): 01 May 2019 Downloaded from http://pubs.acs.org on May 5, 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 13 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 Letters

Potassium Metal as Reliable Reference Electrodes of Non-Aqueous Potassium Cells Tomooki Hosaka,† Shotaro Muratsubaki, † Kei Kubota,†,‡ Hiroo Onuma,† and Shinichi Komaba*,†, ‡ AUTHOR ADDRESS. † Department of Applied Chemistry, Tokyo University of Science, Shinjuku, Tokyo 162-8601, Japan. ‡ Elements Strategy Initiative for Catalysts and Batteries (ESICB), Kyoto University, 1-30 Goryo-Ohara, Nishikyo-ku, Kyoto 615-8245, Japan AUTHOR INFORMATION Corresponding Author E-mail: [email protected]

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry Letters 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 2 of 13

ABSTRACT. Potassium metal electrochemical cells are widely utilized to examine potassium insertion materials for non-aqueous potassium-ion batteries. However, large polarization during K plating/stripping and unstable rest potential are found at potassium electrode, which leads to an underestimation of electrochemical performance of insertion materials.

In this study,

electrochemical behavior of K-metal electrodes is systematically investigated. Electrolyte salts, solvents, and additives influence the polarization of K metals. Although a highly concentrated electrolyte of 3.9 M KN(SO2F)2/1,2-dimethoxyethane realize the smallest polarization of 25 mV among the all electrolytes investigated in this study, the polarization of K metals is still larger than those of Li and Na metals. The issue of inaccurate rest potential is solved by pretreating the K electrodes with a plating/stripping process, which is essential in evaluating intrinsic electrode performance of potassium insertion materials.

TOC GRAPHICS


Page 3 of 13 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


Potassium-ion battery (KIB) has recently attracted much attention due to the abundance of potassium resource, lower standard electrode potential of K/K+,1 and weaker Lewis acidity of K+ ion compared to lithium.2 Since these features are advantageous for high-voltage/power operation of KIBs, researchers have shed light on development of new potassium insertion materials.3-5 To investigate their electrochemical potassium insertion, electrochemical cells equipped with metallic potassium as a counter electrode (CE) are widely utilized, similar to Li and Na cells.6 Recently, we reported polarization issue of potassium plating/stripping process unlike Li and Na metals.2 The undesired polarization results in inaccurate evaluation of the electrochemical properties of the electrode materials when K metal is used as a CE. In this manuscript, we investigate the polarization and rest potential of potassium electrode to confirm its applicability to counter and reference electrodes. Moreover, we find that a galvanostatic pretreatment to refresh K metal surface is highly effective to ensure an accurate equilibrium potential of K/K+ couple enabling us to demonstrate a reproducible and reliable evaluation of potassium insertion materials in K cells. Prior to evaluating polarization behavior of alkali-metal electrodes, open circuit voltage (OCV) of symmetric A//A cells (A = Li, Na, and K) was examined using APF6/ethylene carbonate (EC):diethyl carbonate (DEC) solution as an electrolyte. The OCV measurements were started after 20 min of the cell fabrication. The Li//Li and Na//Na cells showed quite small OCVs of less than 5 mV as shown in Fig. 1a. In contrast, the K//K cell represented a significantly large OCV of > 500 mV in an average of the 10 cells at the beginning and 50 mV even after 24 h of a relaxation period. Moreover, a large deviation of >100 mV from the average value (represented as error bars in Fig. 1a) raises a concern regarding the reproducibility (see detailed OCV data in Supporting Information, Fig. S1). Similar phenomena of unstable OCV were previously reported for Mg//Mg



Page 4 of 13

and Ca//Ca cells which deviated by ca. 40 mV from 0 V.7 From the OCV measurement, K metal is not suitable to be used as a reference electrode in KPF6/EC:DEC without any treatments.

Figure 1. (a) OCV curves of two-electrode Li//Li, Na//Na, and K//K cells filled with 1.0 M LiPF6/EC:DEC, 1.0 M NaPF6/EC:DEC, and 0.8 M KPF6 EC:DEC, respectively. Averaged OCV of 10 cells was plotted for the K//K cell. (b) Voltage profiles of the symmetric cells during continual plating/stripping process. Nyquist plots of (c) Li//Li, (d) Na//Na, and (e) K//K cells before and after 5 and 20 cycles of plating/stripping.

Figure 1b shows the voltage profiles of the symmetry cells during successive cycles of 10-h plating-stripping at ± 25 µA cm-2. The Li//Li and Na//Na cells showed small voltage-drop of less than 5 and 10 mV, respectively. These values are consistent with those in the literature.2, 8 In contrast , the K//K cell showed a significant polarization of 100 mV or larger. In addition, voltage spikes of nearly 1 V were frequently observed in initial few cycles. During the subsequent cycles, the polarization gradually decreased to 100－200 mV. A stark difference in their electrochemical impedance data was also found before and after the cycles as shown in Figs. 1c-e. Li//Li and Na//Na cells showed two semicircles in the high-frequency region of 0.1–0.9 kHz and middle-


Page 5 of 13 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


frequency region of 0.7–2 Hz (see Fig. S2) which can be ascribed to passivation and surface resistive species like alkaline oxide, respectively.9 The total interfacial resistance (Rint) ranges between 100–500 Ω cm2 in Li//Li and Na//Na cells which are consistent with a previous report.8 On the other hand, the K//K cell showed a huge semicircle in the frequency region including ca. 8 Hz, resulting in Rint = ca. 15 kΩ cm2 before plating/stripping. In addition, linear plots inclined at 45° corresponding to Warburg impedance appeared in low-frequency region, which should be attributed to formation of a resistive surface layer of K+ condition. Notably, the Warburg impedance disappeared after 5 and 20 cycles, and the Rint also decreased to less than 10 kΩ cm2 after 20 cycles, which agrees with the reduction of polarization after the cycles in Fig. 1b. As studied in Li- and Na-metal batteries,8 electrode resistance depends on the surface layer formed by electrolyte decomposition. Therefore, we further examined the polarization behavior of K//K cells with different solvents, salts, and additives. Figure 2a shows voltage profiles of K//K cells filled with different KPF6 solutions, 1 M KPF6/propylene carbonate (PC), 1 M KPF6/EC:PC, 0.8 M KPF6/EC:dimethyl carbonate (DMC), and 0.8 M KPF6/EC:DEC.

Similar to the

KPF6/EC:DEC, an use of KPF6/EC:PC and KPF6/EC:DMC leads to large polarization of >500 mV at initial plating/stripping cycles, and it converged to approximately 100 mV after several cycles. Oppositely, the KPF6/PC cell showed continuous increase in polarization during cycles, which is possibly due to continuous electrolyte decomposition and deposition of the decomposition products on K metal with simultaneous electrolyte degradation. Indeed, discoloration of KPF6/PC into yellow was observed when a piece of K metal was just soaked into the electrolyte for 14 d (Fig. 2d) while no visual change was confirmed in KPF6/EC:DEC (Fig. 2e). In contrast to KPF6/PC system, a potassium bis(fluorosulfonyl)amide (KFSA)/PC cell showed large but constant polarization as shown in Fig. 2b. Furthermore, no voltage spikes were observed with all of the



Page 6 of 13

KFSA solutions including 3.9 M KFSA/1,2-dimethoxyethane (DME) which is the highly concentrated and promising electrolyte for demonstrating KIBs

10

and K-metal batteries.11

Actually, the 3.9 M KFSA/DME cell undergoes the least polarization of ca. 50 mV among the all electrolytes used in this study. The suppressed polarization can be attributed to the lower resistive and durable SEI on the surface.11 Moreover, minor and no discolorations were observed in KFSA/PC and KFSA/EC:DEC in Figs. 2f and 2g, respectively, indicating better passivation of K metal in the KFSA-based electrolytes compared to in KPF6-based ones. The polarization is, however, at least 50 mV for a K//K cell, i.e. about 25 mV per K electrode at 25 µA cm-2 even in the highly concentrated electrolyte.

Figure 2. Voltage profiles of repeated K plating/stripping in K//K cells filled with (a) 1.0 M KPF6/PC, 1.0 M KPF6/EC:PC, 0.8 M KPF6/EC:DMC, 0.8 M KPF6/EC:DEC, (b) 1.0 M KFSA/PC, 1.0 M KFSA/EC:DEC, 3.9 M KFSA/DME, (c) 0.8 M KPF6/EC:DEC with or without 1 vol% of FEC, DFEC, VC, or ES. Photographs of K metal discs and electrolytes of the (d) KPF6/PC, (e)


Page 7 of 13 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


KPF6/EC:DEC, (f) KFSA/PC, (g) KFSA/EC:DEC, and KPF6/EC:DEC with (h) FEC, and (i) ES after 14 d.

Furthermore, influence of electrolyte additives was investigated with an aim to decrease the polarization. Electrolyte additives of fluoroethylene carbonate (FEC), difluoroethylene carbonate (DFEC), vinylene carbonate (VC), and ethylene sulfite (ES) which are known to be efficient in LIBs and SIBs 12-13 were added by 1 vol% into 0.8 M KPF6/EC:DEC solution. Figure 2c reveals that the addition of FEC, DFEC, and VC resulted in the increase in polarization compared to that of the additive-free electrolyte, and no remarkable difference was seen in the ES-added electrolyte. The polarization behavior also agrees with discoloration of the KPF6/EC:DEC with FEC, DFEC, and VC into yellow or brown, while no change in color was observed in additive-free and ESadded electrolytes (Figs. 2d-i and S3). Although these electrolyte additives might work effectively in the K-ion full cells such as a graphite//K2Mn[Fe(CN)6] cell, we observed no positive effect on K metal plating. All K based electrolyte components, such as solvents, salts, and additives, examined here are unsatisfactory for suppression of polarization as low as the one observed in Li and Na metal cells. Further developments of proper electrolytes and additives are urged to enable the application of K-metal cells. Since the polarization of K plating is problematic for testing K cells as discussed above, threeelectrode configuration is favorable to minimize the potential fluctuation of K/K+ electrode and bring out real performances of electrode materials in K cells. However, even in the three-electrode cell, the rest potential of K metal as a reference electrode (RE) is poorly reproducible and significantly different from 0 V as shown in Fig. 1a.

To solve this issue, an effect of

electrochemical pretreatment of K metal is applied to realize the reliable potential of K/K+.



Page 8 of 13

Because the reduction of polarization and Rint was successful as described in Figs. 1 and 2, the adapted pretreatment is plating/stripping of K metals between CE and RE inside three-electrode cells. Figure 3a displays a voltage curve of a K//K cell with 0.8 M KPF6/EC:DEC. As discussed above, an initial OCV was 0.87 V, far from 0 V. After the plating/stripping, OCV of the K//K cell is immediately equilibrated to

Potassium Metal as Reliable Reference Electrodes of Non-Aqueous

Recommend Documents