# Fileset

[Batteries   Supercaps - 2025 - Matsuda - Multichannel Electrochemical Cell and Liquid‐Handling Dispenser for (2).pdf](https://mdr.nims.go.jp/filesets/11eaac31-047c-4df0-acaf-857cfc506f81/download)

## Creator

[Shoichi Matsuda](https://orcid.org/0000-0002-0640-3404), Misato Takahashi

## Rights

[Creative Commons BY-NC-ND Attribution-NonCommercial-NoDerivs 4.0 International](https://creativecommons.org/licenses/by-nc-nd/4.0/)

## Other metadata

[Multichannel Electrochemical Cell and Liquid‐Handling Dispenser for High‐Throughput Combinatorial Screening of Multicomponent Electrolytes for Advanced Lithium‐Ion Batteries](https://mdr.nims.go.jp/datasets/36af5079-6fd2-49bd-adde-1a920776da85)

## Fulltext

Multichannel Electrochemical Cell and Liquid‐Handling Dispenser for High‐Throughput Combinatorial Screening of Multicomponent Electrolytes for Advanced Lithium‐Ion Batterieswww.batteries-supercaps.orgMultichannel Electrochemical Cell and Liquid-HandlingDispenser for High-Throughput Combinatorial Screeningof Multicomponent Electrolytes for Advanced Lithium-IonBatteriesShoichi Matsuda* and Misato TakahashiThe performance requirements of Li-ion batteries as energy stor-age devices are continuously increasing. To meet these demands,optimizing the electrolyte composition, especially for expandingthe operating temperature range of LiBs, remains a critical chal-lenge. High-throughput experimentation represents an effectiveapproach for accelerating the discovery of multicomponent elec-trolytes. However, most high-throughput experiment studies onbattery electrolytes are focused on evaluating the battery perfor-mance at room temperature owing to the challenges of integrat-ing temperature control systems. To address this limitation, thisstudy introduces a high-throughput experimental setup com-posed of a (i) closed-type 36-well multichannel electrochemicalcell module, (ii) noncontact liquid-handling dispenser, and (iii)multielectrochemical analyzer installed within a temperature-controlled chamber. This setup enables the preparation of mul-ticomponent electrolyte additives in a combinatorial manner andthe evaluation of battery performance with the prepared electro-lytes across a wide temperature range, achieving a throughput ofover 400 samples per week.1. IntroductionThe demand for enhancing the performance of Li-ion batteries(LiBs) as energy storage devices is growing. In particular,enhancements such as increased cell-level energy density, pro-longed cycle life, reduced charging time, and extended operatingtemperature range are highly desirable for the industrial applica-tion of LiBs. To meet these requirements, advanced electrolytematerials must be identified. Electrolytes not only facilitateLi-ion transport between the negative and positive electrodesbut also contribute to the formation of a stable electrode–electrolyte interface, which considerably affects the overallperformance of LiBs.[1,2] For instance, at the graphite negativeelectrode, the reductive decomposition of electrolyte compo-nents results in the formation of a solid–electrolyte interphase(SEI). Ideally, the SEI must be electronically insulating yet Li-ionconductive, preventing further electrolyte decomposition whileenabling the reversible intercalation and deintercalation of Li ionsinto the graphite electrode. In general, to optimize SEI formation,various compounds are incorporated into electrolytes as addi-tives. As these additives work synergistically to facilitate SEI for-mation, the appropriate combinations that can promote theformation of an ideal SEI must be identified to maximize LiB per-formance. However, predicting the optimal combination remainschallenging owing to the complexity of the SEI formation process.Despite extensive experimental and computational efforts, theprecise formation mechanism and structure of the SEI remain elu-sive. Thus, researchers often rely on conventional trial-and-errormethods to identify the optimal additive combinations, which aretime- and cost-intensive.High-throughput experimental techniques offer a promisingapproach for accelerating electrolyte development.[3–7] Previ-ously, we developed a 96-well microplate-based multichannelelectrochemical (MCE) cell system and used it to conduct auto-mated experiments for screening multicomponent electrolyteadditives for LiBs, achieving a throughput of over 1000 samplesper day.[8–10] This system, installed in an argon-filled glove box,consisted of a liquid-handling dispenser, multielectrochemicalanalyzer, plate stacker, and robotic arm. Using a data-drivensearch algorithm, we identified a combination of five chemicalsthat enhanced battery cycle life. Although this system provedeffective for electrolyte additive screening, several technical chal-lenges remain for further methodological improvements.One challenge is electrolyte volatilization. As the microplate-based electrochemical cell is not completely sealed,S. Matsuda, M. TakahashiCenter for Green Research on Energy and Environmental MaterialsNational Institute for Material Science1-1 Namiki, Tsukuba, Ibaraki 305-0044, JapanE-mail: MATSUDA.Shoichi@nims.go.jpS. MatsudaCenter for Advanced Battery CollaborationNational Institute for Material Science1-1 Namiki, Tsukuba, Ibaraki 305-0044, JapanS. MatsudaNIMS-SoftBank Advanced Technologies Development CenterNational Institute for Material Science1-1 Namiki, Tsukuba, Ibaraki 305-0044, JapanSupporting information for this article is available on the WWW under https://doi.org/10.1002/batt.202400777© 2025 The Author(s). Batteries & Supercaps published by Wiley-VCH GmbH.This is an open access article under the terms of the Creative CommonsAttribution-NonCommercial-NoDerivs License, which permits use anddistribution in any medium, provided the original work is properly cited,the use is non-commercial and no modifications or adaptations are made.Batteries & Supercaps 2025, 8, e202400777 (1 of 7) © 2025 The Author(s). Batteries & Supercaps published by Wiley-VCH GmbHBatteries & SupercapsResearch Articledoi.org/10.1002/batt.202400777http://www.batteries-supercaps.orghttps://orcid.org/0000-0002-0640-3404mailto:MATSUDA.Shoichi@nims.go.jphttp://creativecommons.org/licenses/by-nc-nd/4.0/http://creativecommons.org/licenses/by-nc-nd/4.0/http://doi.org/10.1002/batt.202400777http://crossmark.crossref.org/dialog/?doi=10.1002%2Fbatt.202400777&domain=pdf&date_stamp=2025-03-30high-vapor-pressure components gradually evaporate, leading toelectrolyte depletion. Another challenge is temperature control.From a practical standpoint, evaluating battery performanceacross a wide temperature range is desirable.[11–13] However, pre-cise temperature control within the Ar-filled glove box duringbattery cycling is difficult owing to space constraints.To address these limitations, in this study, we establish a high-throughput experimental setup consisting of a (i) closed-type36-well MCE cell module, (ii) a noncontact liquid-handling dis-penser, and (iii) multielectrochemical analyzer housed within atemperature-controlled chamber (Figure 1). The noncontactliquid-handling dispenser enables rapid electrolyte mixing andinjection into the MCE cell module, within a few minutes(36 wells). Additionally, each MCE cell is tightly sealed with boltedcaps to minimize electrolyte volatilization. Thus, this systemallows the preparation of multicomponent electrolyte additivesin a combinatorial manner and enables battery performance eval-uation across a wide temperature range, achieving a throughputof over 400 samples per week. Notably, this setup eliminates theneed for a robotic arm for MCE cell transport, further streamliningthe workflow. The proposed methodology offers an effectivemeans for accelerating the discovery of multicomponent electro-lytes for advanced LiBs.2. Results and DiscussionAutomated robotic experimental techniques represent an effec-tive approach for preparing a wide range of electrolyte solutionswith diverse compositions. Various automated liquid-handlingsystems are commercially available at present. Using thesetechnically established devices appropriately, the electrolyte-preparation process can be automated with high throughput.In standard laboratory procedures, electrolyte preparationtypically involves three key steps: (i) weigh solid compounds,(ii) add liquid compounds in appropriate volumes, and (iii) stirthe solution to ensure complete dissolution of solid compounds.Automated robotic systems are typically designed to performspecific tasks with high accuracy and throughput. Therefore, torealize the standard electrolyte-preparation process using auto-mated robotic systems, at least three types of robotic equipmentmust be available: (i) for weighing solid compounds, (ii) for inject-ing liquid compounds, and (iii) for stirring solutions. However, theintroduction of multiple automated robotic systems necessitatessignificant cost and space. These requirements hinder the imple-mentation of automated robotic systems in laboratory settings.Considering these aspects, this study focuses on an alterna-tive approach to electrolyte preparation, in which several types ofsolutions, with solid compounds pre-dissolved, are mixed. Thus,electrolyte solutions with different compositions can be effec-tively prepared by appropriately selecting the solutions and con-trolling the mixing ratio. Importantly, this method requires only asingle automated robotic system (liquid-handling dispenser) forelectrolyte mixing, which helps minimize investment cost andspace requirements. In this study, the noncontact liquid dis-penser, CERTUS, is selected, which can mix up to eight differentsolutions with accuracy at the nanoliter scale (Figure S1,Supporting Information). Using this dispenser, multicomponentelectrolytes with various compositions can be prepared in a com-binatorial manner within minutes.When preparing electrolytes by mixing several solutions, theexpression of the electrolyte composition must be carefullytreated. For example, consider a multicomponent electrolytecomposed of LiPF6, LiBF4, lithium difluoro(oxalato)borate (LiBOB),ethylene carbonate (EC), propylene carbonate (PC), vinylenecarbonate (VC), and fluoroethylene carbonate (FEC), which arecommonly used for preparing LiB electrolyte solutions. In general,the composition of electrolyte solutions is expressed in terms ofmolar concentration (mol/L), while the composition of additives isexpressed in terms of weight percentage (wt%). For instance, therepresentative expression for a LiB electrolyte LiB is 0.5 MLiPF6þ 0.5 M LiBF4 in EC/PC (30:70 vol%) with 1 wt% LiBOB, 5 wt%VC, and 5 wt% FEC. However, this expression presents severalproblems in the case of multicomponent electrolytes. First, theLi salt concentration is indicated in units of mol L�1. The denomi-nator of this unit represents the volume of the entire solution,Figure 1. Schematic of proposed approach for high-throughput combinatorial screening of multicomponent electrolytes for advanced lithium-ion batteries.Batteries & Supercaps 2025, 8, e202400777 (2 of 7) © 2025 The Author(s). Batteries & Supercaps published by Wiley-VCH GmbHBatteries & SupercapsResearch Articledoi.org/10.1002/batt.202400777 25666223, 2025, 8, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/batt.202400777 by National Institute For, Wiley Online Library on [08/09/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/batt.202400777which may change owing to solvent mixing and dissolution of Lisalt. In the case of a multicomponent electrolyte, the dissolutionof a large amount of Li salt may alter the solution volume, and thischange may be challenging to accurately predict. Consequently,when using the mol L�1 unit, the final volume of the electrolytesolution must be measured. Second, additive concentrations aretypically expressed in wt%. When the additive content exceeds10 wt%, the concentration of the principal components maydiminish by �10%. Consequently, expressing electrolyte compo-sition based on the weight percentage may not be an accuraterepresentation.One effective approach for addressing the aforementionedproblem is to express all chemical compositions in terms of theirmolar ratios. For example, the electrolyte composition can beexpressed as LiPF6 (0.466) LiBF4 (0.466) LiBOB (0.068) EC (4.20)PC (7.73) VC (0.76) FEC (0.62). By defining the following sixparameters, the electrolyte can be defined in a manner that rep-resents its physicochemical properties, such as concentrationand solvent ratio.x ¼ LiPF6LiPF6 þ LiBF4 þ LiBOB(1)y ¼ ECþ PCþ VCþ FECLiPF6 þ LiBF4 þ LiBOB(2)z ¼ ECECþ PCþ VCþ FEC(3)a ¼ LiBOBLiPF6 þ LiBF4 þ LiBOB(4)b ¼ VCECþ PCþ VCþ FEC(5)c ¼ FECECþ PCþ VCþ FEC(6)In these expressions, x and a represent the ratios of LiPF6 andLiBOB among the Li salts, respectively; and z, b, and c representthe EC, VC, or FEC ratios among the solvents. The electrolyte com-position can be consistently expressed using these six parame-ters, as the electrolyte is composed of six chemical compounds.Furthermore, the constituents of the electrolyte can be expressedin terms of the molar ratio using the aforementioned parametersthrough mutual conversion.LiPF6 ¼ x (7)LiBOB ¼ a (8)LiBF4 ¼ 1� ðx þ aÞ (9)EC ¼ yz (10)VC ¼ yb (11)FEC ¼ yc (12)PC ¼ y � ð1� ðz þ bþ cÞÞ (13)Consider the actual preparation of electrolytes using theliquid-handling dispenser. As the design principle for the mothersolution, first, solutions with a high concentration level of Li saltsmust be prepared, as they can be subsequently diluted withsolvents to achieve electrolytes with lower Li salt concentration.In the case of the sample electrolyte, EC and PC are the mainsolvents, while LiPF6 and LiBF4 are the main Li salts. Thus, thefollowing four mother solutions should be prepared:1: EC solution with a high concentration of LiPF6.2: PC solution with a high concentration of LiPF6.3: EC solution with a high concentration of LiBF4.4: PC solution with a high concentration of LiBF4.In addition, a solution with a high concentration of LiBOBshould be prepared, along with a pure PC solvent.5: PC solution with a high concentration of LiBOB.6: Pure PC solvent.Notably, an EC-based solution with LiBOB or pure EC solventis not included as a mother solution owing to the limited solu-bility of LiBOB into EC and the fact that EC is a solid at room tem-perature. Lastly, pure VC and FEC solvents are selected as mothersolutions.7: Pure VC solvent.8: Pure FEC solvent.The compositions of these eight mother solutions are summa-rized in Table S1, Supporting Information.To conduct electrolyte-preparation experiments using a liquid-handling dispenser, it is necessary to consider the number ofelectrolyte solutions that can be prepared by mixing the eightmother solutions. Figure S2, Supporting Information, illustratesthe mother solution mixing protocol. First, the required amountof LiPF6 or EC is introduced by injecting mother solution No. 1. Inthe case of 54 candidates, the required amount of EC can be sup-plied using only mother solution No. 1. For these candidates, thenext step is to add the required amount of LiPF6 using mothersolution No. 2 (LiPF6 (1) PC (10)). For the other 189 candidates,the remaining amount of EC against the required level is intro-duced using mother solution No. 3 (LiBF4 (1) EC (7.5)). Therequired amount of LiPF6 or EC for all 234 candidates is intro-duced by adding appropriate quantities of mother solutions.The next step involves the addition of mother solutions No. 4(LiBF4 (1) PC (4)) and No. 5 (LiBOB (1) PC (200)) to meet therequired amounts of LiBF4 and LiBOB, respectively. However,for 44 candidates, the amount of PC exceeds the required levelupon adding these solutions. For the remaining 199 candidates,the required amounts of VC and FEC are introduced by addingmother solutions No. 7 and No. 8. As shown in Figure S2, Support-ing Information, 236 candidate electrolytes can be prepared bymixing eight mother solutions in the appropriate ratios.This procedure enables the realization of various electrolytecompositions using a liquid-handling dispenser by simply mixingseveral mother solutions in the appropriate ratio. Furthermore,electrolyte mixing and injection into the MCE cell modulecan be accomplished in a single step, thereby, improving thetotal throughput of the experiments. Although a specific liquid-handling dispenser is used as the model equipment in this study,Batteries & Supercaps 2025, 8, e202400777 (3 of 7) © 2025 The Author(s). Batteries & Supercaps published by Wiley-VCH GmbHBatteries & SupercapsResearch Articledoi.org/10.1002/batt.202400777 25666223, 2025, 8, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/batt.202400777 by National Institute For, Wiley Online Library on [08/09/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/batt.202400777the proposed methodology can be applied to other liquid-handling dispensers. One MCE cell module contains 36 indepen-dent electrochemical cells, arranged to be compatible with theautomatic dispensing equipment. Each electrochemical cell canbe assembled with a positive electrode, separator, and negativeelectrode using a semiautomated punching equipment (Figure S3,Supporting Information), and the electrolyte can be injecteddirectly from the liquid-handling dispenser. Following electro-lyte injection, electric connection parts are installed within eachMCE cell (Figure S4, Supporting Information). Subsequently, theMCE cells are tightly sealed by capping their tops with bolts,thereby, suppressing electrolyte volatilization (Figure S5,Supporting Information). The prepared MCE cell is then electro-chemically connected to a suitable multichannel battery tester,allowing for parallel assessment of the battery performance ofdifferent cells (Figure S6, Supporting Information). Details of theexperimental procedure for MCE cells are provided in Figure S7,Supporting Information.Figure 2a shows the representative charge/discharge profilesobtained from an MCE cell. These experiments involved a graph-ite negative electrode (5mg cm�2, 1.6 mAh cm�2), a LiCoO2 posi-tive electrode (10 mg cm�2, 1.5 mAh cm�2), and an electrolytesolution of 1 M LiPF6 in EC:PC (1:1 vol%). To mitigate concernsregarding electrolyte wettability on the separator, a glass fiberseparator was used. The results demonstrate that stablecharge/discharge reactions proceeded in all cells with a currentdensity of 0.1 mA cm�2. At 25°C condition, the 1st chargingcapacity exhibited �1.5 mAh cm�2. In contrast, in 1st dischargeprocess, capacity is 1.0 mAh cm�2. Such low capacity in the 1stdischarge process is due to the large irreversible capacity causedby SEI formation process because the electrolyte does not containsuitable SEI forming additives to minimize the irreversible capac-ity during 1st cycle. Finally, the coulombic efficiency (CE) of the5th cycle was 97.9%. After the measurement, the temperature wasincreased to 60°C and an additional five-cycle test was performed.In this condition, the capacity gradually decreased with cycling.Thus, the CE of the 5th cycle was 91.6% (Figure 2b). The experi-ments were repeated using three different cells, and the standarddeviation was recorded. The standard deviation of the 5th CE was0.17% at 25 °C and 0.61% at 60 °C, revealing the high reproduc-ibility in battery performance testing (Figure 2c).Next, we discuss the actual experimental throughput of theproposed system for preparing multicomponent electrolytes andassessing their battery performance. This setup involved six MCEcell modules. Using a liquid-handling dispenser, 216 electrolyteswere prepared and injected into the MCE cells (36 cells per mod-ule � 6 modules = 216 cells). This process could be completedwithin several hours. Subsequently, the prepared MCE cellmodules were subjected to battery performance evaluations in atemperature-controlled chamber. This process required one orFigure 2. a,b) Charge/discharge profiles obtained using an MCE cell with a graphite negative electrode, a LiCoO2 positive electrode, and an electrolyte of1 M LiPF6 in EC:PC (1:1 vol%) at a) 25 °C and b) 60 °C. c) 5th CE obtained in each temperature condition. d,e) Charge/discharge profiles of cell with electro-lyte composition of 0.09 M LiPF6þ 0.78 MLiBF4þ 0.02 M LiBOB in (EC:PC:VC:FEC= 45:45:9:1 vol%), obtained at d) 25 °C and e) 60 °C. f ) Relationship between5th CE obtained under 25 and 60 °C conditions.Batteries & Supercaps 2025, 8, e202400777 (4 of 7) © 2025 The Author(s). Batteries & Supercaps published by Wiley-VCH GmbHBatteries & SupercapsResearch Articledoi.org/10.1002/batt.202400777 25666223, 2025, 8, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/batt.202400777 by National Institute For, Wiley Online Library on [08/09/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/batt.202400777two days. Thus, experiments involving two series of six MCE cellmodules could be completed within one week, achieving anexperimental throughput of over 400 samples per week.To demonstrate the effectiveness of the proposed system, weevaluated the battery performance of a multicomponent electro-lyte solution consisting of LiPF6, LiBF4, LiBOB, EC, PC, VC, and FEC.We defined 243 potential electrolyte solutions by setting fiveparameters with three levels each. Thus, the parameter valueswere set as follows: x= 0.1, 0.3, 0.5; y= 15; z= 0.1, 0.3, 0.5;a= 0.005, 0.01, 0.02; b= 0.01, 0.05, 0.1; and c= 0.01, 0.05, 0.1(Table S2, Supporting Information). Note that these parameterswere selected as an example to demonstrate the effectivenessof the experimental setup using MCE cells and a liquid-handlingdispenser, and other parameter values could also be selected.Among the 243 candidates, 236 samples could be prepared bymixing the eight mother solutions. The battery performance testswere conducted using LiCoO2 as the positive electrode andgraphite as the negative electrode. Repeated charge/dischargetests were performed for five cycles at 25 °C and five cycles at60 °C with a cutoff voltage of 4.2/2.0 V.Figure 2f presents the results of the 5th CE of MCE cells con-taining 236 electrolytes, obtained at 60 °C. Each plot representsdata from a single MCE cell with a specific electrolyte composi-tion. At 25 °C, most cells exhibited a 5th CE exceeding 90%. Incontrast, at 60 °C, the 5th CE of several cells was below 90%.Research has shown that during charge/discharge cycling at ele-vated temperatures, the SEI on graphite becomes unstable, lead-ing to the decomposition of electrolyte components. In most ofthe electrolyte compositions investigated in this study, this sidereaction occurred at 60 °C. Consequently, the 5th CE was lowerthan that at 25 °C. The electrolyte compositions exhibiting thehighest 5th CE at 60 °C are listed in Table S3, SupportingInformation. Notably, in all top-30 samples, the value of parame-ter a was 0.02, corresponding to the highest concentration ofLiBOB. No clear trend was observed for the other parameters(parameter y was fixed in this experiment). These results clearlyindicate that LiBOB plays a pivotal role in achieving a high 5th CEat 60 °C.Among the 236 candidates, electrolyte No. 79 (x, y, z, a, b, c)=(0.1, 15, 0.5, 0.02, 0.1, 0.01) exhibited the highest 5th CE at 60 °C.This electrolyte, with a composition of 0.09 M LiPF6þ 0.78 MLiBF4þ 0.02 M LiBOB in (EC:PC:VC:FEC= 45:45:9:1 vol%), achieveda 5th CE of 97.8% at 60 °C. The charge/discharge profiles at 25 and60 °C are shown in Figure 2d and e, respectively. This compositionwas characterized by the lowest concentrations of LiPF6 and FECand the highest concentrations of LiBF4, LiBOB, EC, and VC. LiBOB,EC, VC, and FEC are well-known compounds that act as SEI-forming agents for graphite electrodes.[14–16] In addition, LiBF4demonstrates greater stability at higher temperatures comparedwith LiPF6.[17] Therefore, it is reasonable to conclude that anincrease in the concentration of these elements results in a higher5th CE, even at 60 °C.Table S4, Supporting Information, summarizes the batteryperformance results for the cell containing electrolytes similarto No. 79. The results indicate that varying x and z had minimalimpact on the battery performance. In contrast, the 5th CE sig-nificantly decreased when a, b, and c were varied. Therefore,we focused on the concentration dependence of VC and FEC.Additional experiments were conducted using three differentcells, and the standard deviation was assessed. Figure 3a showsthe 1st CE at 25 °C for five electrolytes with different VC and FECconcentrations. The electrolyte composition (VC: 0.1, FEC 0.01)exhibited the highest 5th CE at 60 °C. These results indicate thatincreasing VC and FEC concentrations enhanced the 1st CE.Figure S8, Supporting Information, shows the voltage profilesof cells with electrolytes containing different VC and FEC concen-trations. The shift in the profile at the beginning of the chargingprocess reveals that the increase in VC and FEC concentrationsresulted in a decrease in charge capacity required for SEI forma-tion on the graphite electrode. This result is consistent with theobserved increase in the 1st CE at this condition. Figure 3b showsthe 5th CE values at 60 °C. Although the increase in VC concen-tration resulted in enhanced CE, the increase in FEC concentra-tion significantly reduced the 5th CE. These results clearlydemonstrate that the increase in VC concentration and decreasein FEC concentration are crucial for maximizing the 5th CE at60 °C. One possible explanation is that the negative effect ofFEC on the positive electrode side is more pronounced at60 °C. We are currently exploring this phenomenon in greater detail.Next, we compared the electrochemical evaluation results ofour developed MCE cell and a standard coin-type cell. Two elec-trolytes were considered: 1 M LiPF6 in EC/PC (1:1 vol%) and0.09 M LiPF6þ 0.78 M LiBF4þ 0.02 M LiBOB in (EC:PC:VC:FEC=45:45:9:1 vol%). Three independent cells were fabricated foreach type of cell, and their battery performance was evaluatedFigure 3. a) Summary of 1st CE at 25 °C condition and b) 5th CE at 60 °Ccondition, obtained by MCE cell with the electrolyte containing differentVC and FEC concentrations level.Batteries & Supercaps 2025, 8, e202400777 (5 of 7) © 2025 The Author(s). Batteries & Supercaps published by Wiley-VCH GmbHBatteries & SupercapsResearch Articledoi.org/10.1002/batt.202400777 25666223, 2025, 8, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/batt.202400777 by National Institute For, Wiley Online Library on [08/09/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/batt.202400777at 25 and 60°C. At 25°C, both cells exhibited similar profiles,although the 1st CE of the MCE cell was lower than that ofthe coin-type cell (Figure 4a,c). After the five-cycle test at25°C, the cells were subjected to cycling tests at 60 °C for fiveadditional cycles. The results are summarized in Figure 4b–d,with magnified images shown in Figure S9, SupportingInformation. At 60 °C, the CE between the 2nd and 5th cyclesexceeded 95% for both electrolytes in the coin-type cell.However, for the MCE cell, the corresponding CE was below95%. In the coin-type cell, the areal size of the electrode was2.0 cm2 and electrolyte amount was 100 μL. Thus, the electrolyteamount per electrode was 50 μL cm�2. In contrast, for MCE cells(areal size of electrode: 0.237 cm2 and electrolyte amount:40 μL), this value increased to 168 μL cm�2. The excess electro-lyte in MCE cells likely resulted in prominent side reactions, suchas electrolyte volatilization and decomposition at the surface ofelectric connection parts. These differences, related to the mini-aturization of electrochemical cells, can likely explain the abso-lute CE variations observed between the MCE and coin-typecells. Our future work will be aimed at enhancing the designof MCE cells, for example, by minimizing the dead volume.In this study, battery performance evaluations were primarilyconducted at 60 °C. Therefore, we focused on electrolyte systemscomposed of high-boiling-point solvents, such as PC and EC.Nevertheless, the proposed high-throughput experimental meth-odology is highly versatile and can also be applied to electrolytesystems containing low-boiling-point solvents, such as dimethylcarbonate and diethyl carbonate. In fact, we are actively applyingthis methodology for the development of electrolyte materials fornext-generation rechargeable battery systems.Lastly, we examined the experimental throughputs of MCEcells and coin-type cells for electrolyte screening (Figure S10,Supporting Information). Assume that a single technical assistantworks for one week, and the maximum number of available bat-tery evaluation channels is 200. In the coin-type cell experiment,the electrolyte is prepared manually. As the electrolyte must beprepared in an Ar-filled glove box, at most 40 electrolyte solutionscan be prepared each day. Subsequently, 40 coin-type cells arefabricated, which takes one additional day. By repeating this pro-cess two times, it is possible to evaluate the battery performanceof 80 electrolyte compositions in one week. In comparison, in thecase of MCE cells, more than 200 electrolytes with different com-positions can be prepared within one day using a high-speedliquid-handling dispenser. The technical assistant is required toprepare only the mother solution and configure the programfor the liquid-handling dispenser. Electrolyte mixing and injectioninto MCE cells can be simultaneously performed by the liquid-handling dispenser within minutes in a high-throughput manner.The corresponding MCE cells can be fabricated within a singleday. By repeating this procedure twice, the battery performanceof more than 400 electrolytes can be accomplished withinone week.In this study, the battery performance evaluation was limitedto the first five cycles. However, longer cycling tests are requiredFigure 4. Battery performance evaluation for coin-type cell and MCE cell with electrolytes 1 M LiPF6 in EC:PC (1:1 vol%) (black curve) and 0.09 MLiPF6þ 0.78 M LiBF4þ 0.02 M LiBOB in (EC:PC:VC:FEC= 45:45:9:1 vol%) (blue curve). CE evaluation in a,b) coin-type cell and c,d) MCE cell at a,c) 25 °Cand b,d) 60 °C.Batteries & Supercaps 2025, 8, e202400777 (6 of 7) © 2025 The Author(s). Batteries & Supercaps published by Wiley-VCH GmbHBatteries & SupercapsResearch Articledoi.org/10.1002/batt.202400777 25666223, 2025, 8, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/batt.202400777 by National Institute For, Wiley Online Library on [08/09/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/batt.202400777to gain deeper insights about the battery performance. Strategies,such as accelerated evaluation protocols or lifetime predictionmodels, can be adopted to maintain high throughput whileensuring meaningful performance differentiation. Moreover, fur-ther advancements in electrolyte material screening methods arenecessary, leveraging the advantages of the liquid-handling dis-penser, which enables the rapid and precise preparation of a widevariety of electrolytes. Overall, the proposed methodology canaccelerate the development of multicomponent electrolytes formaximizing battery performance and help clarify the complexreaction mechanism in LiBs.3. ConclusionMCE cells, which are highly compatible with liquid-handlingdispensers, were used to construct an electrolyte screening sys-tem for LiBs, achieving a high throughput of over 400 samples perweek. This system enables the combinatorial preparation of dif-ferent electrolytes and their direct injection into MCE cells. Batteryperformance testing can then be performed in temperature-controlled conditions. To demonstrate the effectiveness of theproposed system for identifying multicomponent electrolytesthat can enhance LiB performance, we selected a model electro-lyte system composed of LiPF6, LiBF4, LiBOB, EC, PC, VC, and FECand investigated the LiCoO2/graphite battery performance at60 °C. We experimentally verified the electrolyte composition thatexhibited the highest 5th CE among 236 candidates. The system-atic investigation also revealed that a high concentration of LiBOBis crucial for achieving a high 5th CE at this temperature. Thisstudy offers a high-throughput experimentation system for accel-erating the discovery of multicomponent electrolytes to enhancethe performance of next-generation rechargeable batteries. Inparticular, by combining autonomous search algorithms,[10] theefficiency of complex optimization processes, conventionally per-formed through time- and labor-intensive trial-and-error basedapproaches, can be enhanced.AcknowledgementsThe authors thank Dr. Yuki Maruyama for providing technicalassistance in the experiments. This work was partially supportedby JST COI-NEXT (grant no. JPMJPF2016) and Ministry ofEducation, Culture, Sports, Science, and Technology (MEXT)Program: Data Creation and Utilization Type Materials Researchand Development Project (grant no. JPMXP1121467561).Additionally, this work was supported by the National Institutefor Materials Science (NIMS) Battery Research Platform.Conflict of InterestThe authors declare no conflict of interest.Keywords: experimental automation · high-throughputexperiment · lithium-ion batteries · multicomponentelectrolyte · solid electrolyte interfaces[1] K. Xu, Chem. Rev. 2014, 114, 11503.[2] E. Peled, S. Menkin, J. Electrochem. Soc. 2017, 164, A1703.[3] A. Dave, J. Mitchell, K. Kandasamy, H. Wang, S. Burke, B. Paria, B. Póczos,J. Whitacre, V. Viswanathan, Cell Rep. Phys. Sci. 2020, 1, 100264.[4] J. T. Yik, L. Zhang, J. Sjölund, X. Hou, P. H. Svensson, K. Edström, E. J. Berg,Digital Discovery 2023, 2, 799.[5] A. Dave, J. Mitchell, S. Burke, H. Lin, J. Whitacre, V. Viswanathan, Nat.Commun. 2022, 13, 5454.[6] J. Noh, H. A. Doan, H. Job, L. A. Robertson, L. Zhang, R. S. Assary, K. Mueller,V. Murugesan, Y. Liang, Nat. Commun. 2024, 15, 2757.[7] A. Suzumura, H. Ohno, N. Kikkawa, K. Takechi, J. Power Sources 2022, 541,231698.[8] S. Matsuda, K. Nishioka, S. Nakanishi, Sci. Rep. 2019, 9, 6211.[9] S. Matsuda, G. Lambard, K. Sodeyama, Cell Rep. Phys. Sci. 2022, 3, 100832.[10] R. Tamura, K. Tsuda, S. Matsuda, Adv. Mater. Methods 2023, 3, 2232297.[11] T. Taskovic, A. Eldesoky, C. P. Aiken, J. R. Dahn, J. Electrochem. Soc. 2022,169, 100547.[12] E. Logan, A. Eldesoky, E. Eastwood, H. Hebecker, C. Aiken, M. Metzger,J. Dahn, J. Electrochem. Soc. 2022, 169, 040560.[13] T. Taskovic, A. Eldesoky, W. Song, M. Bauer, J. R. Dahn, J. Electrochem. Soc.2022, 169, 40538.[14] R. Fong, U. von Sacken, J. R. Dahn, J. Electrochem. Soc. 1990, 137, 2009.[15] K. Xu, U. Lee, S. S. Zhang, T. R. Jow, J. Electrochem. Soc. 2004, 151,A2106.[16] C. Täubert, M. Fleischhammer, M. Wohlfahrt-Mehrens, U. Wietelmann,T. Buhrmester, J. Electrochem. Soc. 2010, 157, A721.[17] S. S. Zhang, K. Xu, T. R. Jow, J. Electrochem. Soc. 2002, 149, A586.Manuscript received: December 10, 2024Revised manuscript received: March 19, 2025Version of record online: March 21, 2025Batteries & Supercaps 2025, 8, e202400777 (7 of 7) © 2025 The Author(s). Batteries & Supercaps published by Wiley-VCH GmbHBatteries & SupercapsResearch Articledoi.org/10.1002/batt.202400777 25666223, 2025, 8, Downloaded from https://chemistry-europe.onlinelibrary.wiley.com/doi/10.1002/batt.202400777 by National Institute For, Wiley Online Library on [08/09/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://doi.org/10.1002/batt.202400777 Multichannel Electrochemical Cell and Liquid-Handling Dispenser for High-Throughput Combinatorial Screening of Multicomponent Electrolytes for Advanced Lithium-Ion Batteries 1. Introduction 2. Results and Discussion 3. Conclusion