Reference electrodes with ionic liquid salt bridge: when will these

Feb 14, 2019 - In this paper, we raise questions that researchers have to ask if they intend to replace a conventional reference electrode with an ion...
0 downloads 0 Views 523KB Size
Subscriber access provided by TULANE UNIVERSITY

Review

Reference electrodes with ionic liquid salt bridge: when will these innovative novel reference electrodes gain broad acceptance? Erno Lindner, Marcin Guzinski, Taskia Ahammad Khan, and Bradford D. Pendley ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.8b01651 • Publication Date (Web): 14 Feb 2019 Downloaded from http://pubs.acs.org on February 17, 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 34 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

ACS Sensors

Reference electrodes with ionic liquid salt bridge: when will these innovative novel reference electrodes gain broad acceptance? Ernő Lindner, Marcin Guzinski, Taskia A. Khan, Bradford D. Pendley Department of Biomedical Engineering, The University of Memphis, Memphis, TN, 38152 KEYWORDS: reference electrodes, ionic liquid, flow-rate dependence, ionic liquid purity, ionic liquid membrane/gelling material ABSTRACT In this paper, we raise questions that researchers have to ask if they intend to replace a conventional reference electrode with an ionic liquid-based reference electrode and try to answer these questions based on our experiences and literature data. Among these questions, the most important is which ionic liquid should be used. However, beyond the chemical composition of the ionic liquid, to realize all the potential benefits of ionic-liquid based reference electrodes, there are additional, equally important considerations. Through examples we will show the importance of the (i) purity of the ionic liquid and the consequences of deviations from its stoichiometric salt composition, (ii) form of implementation of the ionic liquid-based reference electrode membrane (free flowing salt bridge, or ionic liquid embedded in a membrane), (iii) membrane/gelling agent material and its composition, (iv) experimental conditions (steady state or flowing conditions) under which it will be used. Finally, we recommend methods to test the performance criteria of the ionic liquid-based reference electrodes.



Corresponding author E-mail: [email protected] 1 ACS Paragon Plus Environment

ACS Sensors 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 34

In the preamble of their famous book on reference electrodes, Ives and Janz1 expressed their intention to provide useful information for the experimentalist, some of which is difficult to track down even if the researcher traces back successively earlier references. Our goal with this contribution is similar but much narrower in its focus; we hope to provide useful information on the critical aspects of ionic liquid-based reference electrodes without liquid junction. In these reference electrodes, room-temperature molten salts, also referred to as ionic liquids (ILs) or lipophilic electrolytes, comprised of lipophilic cations and anions, are used as salt bridges both in free flowing or immobilized forms, i.e., as reference electrode membranes. These new reference electrodes, after more than 100 years, offer a viable alternative to conventional reference electrodes with concentrated KCl solutions as salt bridge electrolyte.2-5 We try to unravel reasons why these innovative novel reference electrodes, with the unique advantage of diffusion potential free potentiometric measurement, still did not make a commercial breakthrough among the available reference electrodes despite the numerous disadvantages of conventional reference electrodes6 and their alternatives with solidified reference electrolytes.7, 8 We will evaluate the performance characteristics of the liquid junction free reference electrodes (LJFREs) with focus on their utility in potentiometric measurements with uniquely stringent requirements, i.e., the measurement of blood electrolytes. The trend in potentiometric analysis targets the development of simple calibration free analytical systems, e.g., paper-based microfluidic manifolds,9-17 using solid contact ion-selective electrodes (ISEs) in combination with solid state reference electrodes,18-26 with reproducible sensor-to-sensor standard potentials,23, 27 short equilibration time28, 29 and minimal drift19, 21 that matches the performance of commercial blood electrolyte analyzers. A LJFRE, which can be simply miniaturized, could be useful in many areas of electroanalytical chemistry but it is expected to have a significant impact in multi analyte flow analytical systems (e.g., blood electrolyte analyzers,30 microfluidic manifolds9-17, 31-34 and single use devices and) where conventional reference electrodes have serious limitations. Conventional reference electrodes most commonly are filled with a large volume of electrolyte solution (inner filling solution, IFS), which is in contact with the sample through a flow-restricting sleeve junction, diaphragm or frit. For a good reference electrode, the potential drop across this liquid junction (liquid junction or 2 ACS Paragon Plus Environment

Page 3 of 34 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

ACS Sensors

diffusion potential) is expected to be independent from the sample composition and stable in time. With concentrated filling solutions in the reference electrodes (e.g., 3.5 M KCl) these expectations are well approached as long the measurements are performed in high ionic strength solutions (I > 0.01 M). But with reference electrodes manufactured with nanoporous frits, especially in low ionic strength solutions, significant systematic errors were reported.35, 36 Using frits with a few μm pore sizes, the errors could be reduced but it increased the outflow of electrolyte from the reference electrodes. Minute outflow of electrolyte is considered beneficial for reproducible liquid junction potentials (LJP),37 but large outflow can contaminate the sample and lead to potential drifts. To eliminate the liquid filling solution of conventional reference electrodes38 reference electrodes with a variety of solidified reference electrolytes (melted, gelled, embedded in a polymer matrix etc.) were developed.7, 8 In these arrangements the slow leaching salt, most commonly KCl, stabilizes the potential on the reference electrode/sample solution phase boundary. Although reference electrodes with solidified reference electrolyte can be implemented in microfluidic devices,38 due to the geometrical constraints, the lifetime of these miniature reference electrodes is often limited. If the reference electrolyte is implemented in highly crosslinked polymers for prolonged life time, the reference electrodes had high resistances and long equilibration times.7, 8 Liquid Junction Free Reference Electrodes (LJFRE) Conventional reference electrodes have several inherent drawbacks: (i) leaching related sample contamination, and need for frequent renewal of the filling solution, (ii) clogging of the liquid junction in certain samples, and (iii) differences in the LJPs with different types of liquid junctions. To overcome these problems, Horvai39 envisioned a possible reference electrode design without liquid junction using a lipophilic salt loaded plasticized PVC membrane electrode. In this system, the reference electrode membrane is sandwiched between two aqueous electrolyte solutions, the sample and the IFS of the reference electrode. Such a membrane is expected to have a sample solution independent potential if the anion and cation of the lipophilic salt have moderate and comparable lipophilicities. Under these conditions, 3 ACS Paragon Plus Environment

ACS Sensors 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 4 of 34

both the anion and the cation will partition into the sample and become potential determining ions (Eq.1). The validity of Eq. 1 assumes that the hydrophilic ions of the aqueous phase do not partition into the membrane. 𝑅𝑇

𝐸𝑃𝐵 = 𝑧𝐼𝐹𝑙𝑛

𝑘𝐼𝑎𝐼 𝑧𝐼 +

𝛾𝐼[𝐼

]

Eq. 1

where 𝑎𝐼 is the activity of the potential determining ion in the solution, [𝐼𝑧𝐼 + ] is the concentration of the same ion in hydrophobic membrane, 𝑧𝐼 is the charge of the ion and 𝛾𝐼 is its activity coefficient in the membrane R, T, and F are the universal gas constant, the absolute temperature, and the Faraday number. Finally, the constant 𝑘𝐼 incorporates the free energy of transfer of the ion 𝐼𝑧𝐼 + : 𝑜

𝑜

𝑘𝐼 = 𝑒(𝜇𝑎𝑞 ― 𝜇𝑜𝑟𝑔)/𝑅𝑇

Eq. 2

where 𝜇𝑜 is the standard chemical potential of the ion in the indicated phase. Since equation 1 is valid for both cations and anions (with 𝑧𝐼 having a positive or negative charge sign, respectively) the membrane may act as a reference electrode membrane because its potential is determined by both the cations and anions of the membrane partitioned into the aqueous sample.40 Instead of a lipophilic salt-loaded PVC membrane, Kakiuchi used room-temperature molten salts (lately referred to as ionic liquids, ILs) as separation membrane between two aqueous phases.3 He considers the IL layered between two aqueous solutions as water immiscible salt bridge41 with “liquid-junction potential” at the interface of two immiscible electrolyte solutions (ITIES).2, 3, 42-44 To prepare LJFREs, the ionic liquids were used as separation membranes between two aqueous electrolyte solutions in a variety of forms, i.e., (i) as free flowing IL layer,2, 3, 42-44 (ii) loaded in plasticized39 or plasticizer free PVC membranes45, (iii) incorporated in silicone rubber and polyurethane membranes46, 47, (iv) embedded in poly(vinylidene fluoride-co-hexafluoropropene) [P(VdFHFP)] as gelling material2 or (v) incorporated into an IL-miscible poly(methyl methacrylate) composite membrane.48 Instead of incorporating a lipophilic salt (ionic liquid) into a membrane, Nagy et al. recommended of coupling two electrodes with cation and anion exchanger membranes in parallel49 while Mattinen50 incorporated stoichiometric quantities of two salts into the membrane, a cation 4 ACS Paragon Plus Environment

Page 5 of 34 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

ACS Sensors

exchanger and an anion exchanger, that in situ formed the intended lipophilic electrolyte in the membrane. Almost all the papers on LJFREs report nearly ideal reference electrode behaviors, i.e., sample composition and concentration independent potentials and excellent long-term stability etc. Despite these encouraging data and all the claimed advantages of LJFREs, their spread in commercially available reference electrodes is minimal.51 We assume that the main reason impeding the widespread commercial success of these innovative devices is that from the reported data it is often difficult to deduce how far can these reference electrodes considered non-polarizable or what were the required performance criteria with respect of the (i) sample composition and concentration independent potentials, (ii) reproducibility of the potential measurements, (iii) the influence of interfering compounds on the reference electrode potential, (vi) sensitivity of the reference electrode potential to flow rate, (v) the rate of equilibration following the first exposure of the reference electrode to an aqueous solution (especially important with single use devices), (vi) the rate of equilibration following a change in the sample concentration or flow rate, (vii) and the lifetime of miniaturized LJFREs. In addition, since these LJFREs are prepared from ILs with different chemical compositions, in a variety of formulations (e.g., as pure liquids, or liquids incorporated into a variety plastics or gelling materials), and sensor constructions (with liquid or solid contacts), it is almost impossible to infer what are the critical issues for selecting an IL-based reference electrode that would meet specific criteria in certain samples. The advantage of conventional reference electrodes is that the potential errors related to the LJP can be estimated and a salt bridge electrolyte with minimal LJP can be selected for specific samples. For those who are considering the replacement of a conventional reference electrode with a LJFRE it would be essential to know which ionic liquid to choose, in which formulation and sensor construction and to understand quantitatively the advantages of these novel reference electrodes in solving their own analytical problems in potentiometric, voltammetric, or coulometric analysis, impedance spectroscopy, etc. Characterizing the existing and new LJFREs according to a standardized protocol, similar to the IUPAC recommendations for ion-selective electrodes (ISEs),52 could have a positive impact on their broader acceptance. 5 ACS Paragon Plus Environment

ACS Sensors 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 6 of 34

In this paper, we summarize our experiences with a variety ILs and lipophilic electrolytes as the key components of LJFREs from the end user’s practical points of view. In addition, we raise questions that researchers have to ask if they intend to replace a conventional reference electrode with an IL-based reference electrode. We also try to answer these questions based on our experiences and literature data. The ILs were tested as water immiscible liquid membranes like in the works of Kakiuchi53 and by incorporating them into PVC membranes with and without plasticizers21, 45, 54 as well as in P(VdFHFP) membranes.2 Detailed experimental data are provided for the ILs listed in Table 1. However, simultaneously to the ILs listed in Table 1, we evaluated the performance of a variety of additional ILs. The chemical structures of these ILs are provided in Table S1 in the supplementary material. The ILs tested were commercial products (e.g., [C8mim+][C1C1N−], i.e., IL1 and [TBMOEP+][C1C1N−], i.e., IL2), synthesized in our laboratory (e.g., TDDA-TPFPhB, i.e., IL3) or received as a generous gift ([TBMOEP+][C2C2N−], i.e., IL13) from Profs. Kakiuchi and Yamamoto. The ILs-based reference electrode responses were tested both in steady state and under variable flow conditions in a broad range of electrolyte solutions, e.g., NaCl, KCl, and in electrolyte solutions with multiple cations and anions, like blood serum like electrolytes, and in the presence of potential interferences, e.g., salicylate and tetraalkyl ammonium ions. We will show that depending on the applications, a variety of ILs might be appropriate. However, beyond the chemical composition of an IL, there are additional issues that could significantly influence the performance characteristics of an IL-based reference electrode: (i) the purity of the IL and/or potential deviations from its stoichiometric salt composition, (ii) the form of implementation of the IL-based reference electrode membrane (ITIES like liquid salt bridge, IL embedded in a membrane, etc.), and (iii) the experimental conditions (steady state or flowing conditions).

6 ACS Paragon Plus Environment

Page 7 of 34

Table 1. Chemical structure of ILs and lipophilic electrolytes with the logP values of the cationic and anionic constituents calculated with four different programs: (1) Molinspiration,24 2) Chemaxon,25 3) Chemdraw,55 4) ChemSketch (ACDLabs)56

IL2 [TBMOEP+][C1C1N−]

IL1 [C8mim+][C1C1N−]

Name

Structure of the cation N+

(CH2)7CH3

logP†

Structure of the anion

1) 0.0 2) -0.4

N

3) -0.15

CH3

4) 3.2

1-Methyl-3-octylimidazolium+ (CH2)3CH3

F3C

2) 4.2

(CH2)3CH3 Tributyl (2-methoxyethyl)phosphonium+

3) 2.2 4) nd

O

S

N- S

O

O

F3C

O

O

S

N- S

IL3 TDDA-TPFPhB

CH2(CH2)10CH3 CH3(CH2)10CH2

N

+

CH2(CH2)10CH3

CH2(CH2)10CH3

CF3

2) 1.94 3) -0.35 4) 1.5 1) 2.97

CF3

O O Bis(trifluoromethanesulfonyl)imide−

F

logP 1) 2.97

Bis(trifluoromethanesulfonyl)imide−

1) 2.62

CH3OCH2CH2 P+ (CH2)3-CH3

O

1) 10.4

2) 1.94 3) -0.35 4) 1.5

1) 4.6

F

2) 15.0

2) 8.9 -

B

F

3) 18.0 4) 11.4

F

3) 9.6 4) nd

F 4

Tetradodecylammonium+ IL13* [TBMOEP+][C2C2 N−]

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

ACS Sensors

1) 2.62 2) 4.2

tetrakis(pentafluorophenyl)borate− O O

CF3CF2

3) 2.2 4) nd Tributyl(2-methoxyethyl)phosphonium+

S

N- S

O

O

CF2CF3

bis(pentafluoroethanesulfonyl)imide−

1) -1.0 2) nd 3) 0.9 4) 4.6

* Generous gift from Professors Yamamoto and T.Kakiuchi; nd: not determined † Calculated

value of the logarithm of the partition coefficient between octanol and water

EXPERIMENTAL Materials ILs and reference electrode membranes: 1-Methyl-3-octylimidazolium bis(trifluoromethanesulfonyl)imide (IL1 or [C8mim+][C2C2N−], 99%) was purchased from IoLiTec Inc., (Tuscalosa, AL) while Tributyl(2-methoxyethyl)phosphonium 7 ACS Paragon Plus Environment

ACS Sensors 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 8 of 34

bis(trifluoromethanesulfonyl)imide (IL2 or [TBMOEP+][C1C1N−], 95%) from TCI America. Tetradodecylammonium tetrakis (pentafluorophenyl) borate (IL3 or TDDA-TPFPhB), was prepared by metathesis as described previously57, 58 from tetradodecylammonium chloride (TDDACl, >97%, Sigma-Aldrich) and potassium tetrakis (pentafluorophenyl) borate (KTPFPhB ≥ 97%, Boulder Scientific). Finally, Tributyl(2-methoxyethyl)phosphonium bis(pentafluoroethanesulfonyl)amide (IL13 or [TBMOEP+][C2C2N−]) was a generous gift from Profs. Yamamoto and Kakiuchi. For the preparation of the LJFRE membrane cocktails, poly(vinylchloride) (PVC, high molecular weight), poly(vinylidene fluoride-co-hexafluoropropene) (P(VdFHFP), average MW 400000), 2-Nitrophenyl octyl ether (oNPOE) and tetrahydrofuran (THF), were purchased from Sigma Aldrich. The PVC-based reference electrode membranes were cast in three different PVC/Plasticizer/IL ratio. 1) Membranes without any plasticizer with 1:2 and 1:1 PVC/IL ratio; 2) Membranes with 1:2 PVC/plasticizer ratio containing 7 w/w% IL, 3) Membranes with 1:1 PVC/plasticizer ratio and 20 w/w% IL.21, 27 The membrane cocktails contained approximately 300 mg of the membrane ingredients dissolved in ~ 2mL tetrahydrofuran (THF). The membrane cocktails were cast into a 25.4 mm diameter glass ring fixed on a glass plate or over 0.24 mm diameter hydrogel coated Ag/AgCl sensing sites implemented on the bottom of a PVC-based microfluidic channel.19 After the evaporation of the THF an approximately 200 μm thick soft film remains on the glass plate from which ~ 7 mm diameter discs were cut and implemented into Philips electrode bodies. The P(VdFHFP)-based membranes were cast from a cocktail containing 200 mg P(VdFHFP) and 200 mg ionic liquid in ~3.5 mL in acetone. After the evaporation of acetone, the P(VdFHFP)-based membranes were tested the same way as the PVC-based membranes, in Philips electrode bodies. Synthesis and purification of ILs and lipophilic electrolytes is provided in the supplementary material. Aqueous solutions: The aqueous solutions were prepared with 18.2 MΩcm resistivity deionized water from Millipore Milli-Q A10 system. NaCl, KCl, Sodium salicylate and Benzalkonium chloride for the test solutions, were purchased from Thermo-Fisher or Sigma-Aldrich. 8 ACS Paragon Plus Environment

Page 9 of 34 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

ACS Sensors

Electrodes: The IL-loaded membranes were tested in Philips IS-560 liquid membrane electrode bodies (Möller Glasbläserei, Zurich Switzerland) or in a microfluidic channel after their deposition over hydrogel coated Ag/AgCl sensing sites. For testing the responses of the pure ILs in liquid form they were filled into a CHI111 reference electrode body with a porous glass frit on its lower end to hold the IL and separate it from the sample solution. The responses of the ILbased LJFREs were tested in combination of commercial reference electrodes in steady and stirred or flowing solutions. As commercial reference electrodes FisherbrandTM accumetTM (single junction calomel and double junction Ag/AgCl), Thermo Scientific Sure-flow and Oesch Sensor Technology Inc. OSTECH double junction Ag/AgCl reference electrodes were used with saturated KCl as salt bridge electrolyte. Instruments Potentiometric measurements: For the potentiometric data acquisition, a Lawson Lab (Malvern, PA) 16-channel high input impedance data acquisition system was used. The data acquisition system has been connected to a computer equipped with the EMF Suite version 2.0.0.2 program. The potentiometric measurements with the macro electrodes were performed at room temperature. The responses of the miniature reference electrodes implemented in the PVCbased flow channel were tested at 37 oC. Procedures Measurements in steady state solution: Following the complete equilibration of the reference electrodes they were exposed to a variety of test solutions, which were replaced manually. Between the exposures to the different solutions the electrodes were wiped dry with Kimwipes. The individual electrode potentials were recorded for 5 to 10 minutes after the electrodes were exposed to a new solution. In these experiments, the responses (response slopes, potential reproducibility/hysteresis) of the LJFREs were tested in NaCl, KCl, and homemade blood serum like electrolytes. To test the influence of lipophilic anions (salicylate, logP~2.0) and cations (benzalkonium, logP~4.0) on the responses of the LJFREs 0.14 M NaCl solution with 1, 2 and 4 mM Na salicylate concentration, or with 2.5 mg/L or 5.0 mg/L benzalkonium chloride concentration was used. 9 ACS Paragon Plus Environment

ACS Sensors 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 10 of 34

Determination of the of the flow rate dependence of the LJFREs in stirred solutions, a wall jet type flow cell and in a miniature flow channel: For testing the influence of intense stirring on the potential of the LJFRE membranes, the electrode potentials were recorded upon periodically switching on and off a magnetic stirrer. For certain electrodes, the flow rate dependence was also tested in a wall-jet type cell.59 The design of the wall jet cell has been published in several of our earlier papers59, 60 and it is shown schematically as an inset in Figure 2. The sample solution was pumped perpendicularly to the center of the tested electrode surface with a Gilson® Minipulse 3 peristaltic pump with adjustable flow rate. An advantage of the wall-jet arrangement is that only the tested electrode is exposed the solution flow. Changes in the sample solution composition were induced with the help of a manual multi-position valve. For assessing the flow sensitivity of the LJFREs implemented in a flow channel on a PVC-based planar sensor platform the outlet of the flow channel was inserted a ~ 500 mL volume beaker, partially filled with saturated KCl solution. The electrode potentials of the planar LJFREs in the flow channel were measured against conventional reference electrodes placed into this large volume KCl solution, i.e., the conventional reference electrodes with liquid junction were not exposed to the fluctuating flow rates. In the experiments the electrode potentials were recorded at 0.05 mL/min flow rate continuously, then the flow rate was increased to 10 mL/min for 15 s and the flow rate was reduced back to 0.05 mL/min for at least 15 minutes. In other experiments after the 15 s 10 mL/min pumping period the flow was stopped for at least 5 minutes. The electrode potentials during these experiments were continuously recorded.

RESULTS AND DISCUSSION The theoretical and experimental challenges of assessing the performance characteristics of the IL- based reference electrodes are similar to the difficulties of pH measurement: (i) immeasurable nature of single ion activity coefficients, (ii) the differences in the determination of the pH of primary ion standards (in cells without transference, Harned cell) and real samples with glass electrodes in electrochemical cells with liquid junction.61 Consequently, the most reliable data on the performance of the IL-based reference electrodes can be obtained by measuring the reference electrode membrane potential between two 10 ACS Paragon Plus Environment

Page 11 of 34 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

ACS Sensors

hydrogen gas electrodes in an electrochemical cell without transference. This method has been used to show the utility of IL-based reference electrodes for pH measurements in low ionic strength solutions.62 In pure electrolyte solutions, Kakiuchi et al. measured the responses of the IL-based reference electrodes against Ag/AgCl,53 Ag/AgBr63 or Ag/AgI64 reference electrodes of the second kind without liquid junction and plotted the measured potentials against the mean activity of the electrolytes. A Nernstian response of the silver halide electrodes in these experiments was considered as a proof for stable, concentration independent potential of the IL-based reference electrode. Although these experiments clearly prove the validity of the assumptions related the working mechanism of the IL-based reference electrodes, they might not provide the necessary information about the utility of these electrodes for solving specific tasks. Beyond the slopes of the silver halide electrodes, the reproducibility of the measured potentials in certain solutions and/or the residual mean standard deviations of the potential values around the fitted line are equally important. In complex solutions however, testing the response of a LJFRE against a silver halide electrode53, 63, 64 may not be feasible. Although, instead of a silver halide electrode, one might record the response of an ion-selective electrode against the IL-based reference electrode for the assessment of the IL-based reference electrode performance, the uncertainties related to (i) the determination of ionic activities, and (ii) deviations from the assumed Nernstian response of the ISE, especially in very dilute and very concentrated solutions, cannot be eliminated. Alternatively, similar to this work, the IL-based reference electrode responses can be compared to commercial reference electrodes with liquid junction.45,54 In such experiments the uncertainties in the LJP limits the attainable precision and accuracy of the information on the IL-based reference electrode. Which Ionic Liquid Should Be Used? To the best of our knowledge, the only IL that is used as a bridge electrolyte in a commercial reference electrode is IL13 ([TBMOEP+][C2C2N−]).51, 65, 66 It is implemented in gelled form in the reference electrode half-cell of a Horiba combination glass electrode. As a gelling 11 ACS Paragon Plus Environment

ACS Sensors 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 12 of 34

agent, P(VdFHFP),2 is used. The gelled IL separates the sample solution and the 0.1M KCl inner filling solution (saturated with IL13 and AgCl) of the Ag/AgCl reference electrode. This HORIBA pH glass combination electrode (PUREIL) is specifically recommended for solutions with low conductivity (κ