Lowering Detection Limits Toward Target Ions Using Quasi-Symmetric

Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article

Lowering Detection Limits Toward Target Ions Using Quasi-Symmetric Polymeric Ion-Selective Membranes Combined with Amperometric Measurements Xénia Nagy, and Lajos Höfler Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.6b03043 • Publication Date (Web): 16 Sep 2016 Downloaded from http://pubs.acs.org on September 19, 2016

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 free 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 accessible to all readers and 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.

Analytical Chemistry 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 19

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

Analytical Chemistry

Lowering Detection Limits Toward Target Ions Using Quasi-Symmetric Polymeric Ion-Selective Membranes Combined with Amperometric Measurements Xénia Nagy1, Lajos Höfler1* 1

Department of Inorganic and Analytical Chemistry, Budapest University of Technology and Economics, Szt. Gellért tér 4, Budapest, 1111, Hungary

*

Corresponding Author: email: [email protected]

ABSTRACT An amperometric method is reported that compensates for the interference from marginally discriminated interfering ions when using traditional polymeric ionselective membrane (ISM) electrodes. The concept involves utilizing two ISMs in a three-compartment electrochemical cell configuration. The two ISMs are identical in composition except for the addition of an ionophore to one of the membranes. Initially, all three compartments contain the same concentration of interfering ion and the membrane does not contain primary ions. Reference electrodes are placed into each of the two outer compartments. At this point, there is no potential difference between the two reference electrodes. We show experimentally and theoretically that when the concentration of an interfering species is increased in the sample compartment, the phase-boundary potentials of both sample solution|ISMs change similarly. However, when the primary ion is added to the sample, an asymmetry will emerge, and the membrane with the ionophore will exhibit a larger phase-boundary potential change. At low concentrations the difference in membrane potentials can be too small for reliable potentiometric detection. Current, which can be routinely measured on pA levels, can be used instead to detect the small primary ion concentration changes with a significant lowering of detection limits. The theory of this method is described by Nernst-Planck-Poisson finite element simulations, and both amperometric and potentiometric experimental verification is demonstrated using ammonium ISM. It is shown that amperometric measurements enable 200 nM ammonium to be detected in the presence of 0.1 mM of potassium, detection capability that is not possible via conventional potentiometry. 1 ACS Paragon Plus Environment


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 19

INTRODUCTION In recent decades, many ionophores with high selectivity for their preferred ion have been reported for use in polymeric ion-selective membrane electrodes. With the utilization of these ionophores, selective, sub-nanomolar detection of many ions became possible with ion-selective electrodes (ISEs) with little or no sample preparation.1-3 For several biologically significant ions, such as chloride, fluoride, nitrate and ammonium, available ionophores often lack the required selectivity to provide adequate sensitivity when significant levels of interferent ions are present. For example, direct detection of elevated blood levels of ammonium can indicate liver dysfunction, such as cirrhosis or hepatitis.4 Unfortunately, the selectivity of the classical ammonium ionophores (e.g., nonactin) over potassium are less than two orders of magnitude because of the very similar radii of the ammonium and potassium ions.5 Indeed, the high levels of potassium in blood interfere with the potentiometric detection of ammonium using nonactin-based membrane electrodes. Further, there is a need for more selective detection of ammonium in environmental samples as well. Ammonium can be found in ground water due to anthropogenic sources, such as sewage and agricultural wastes, detecting elevated levels of ammonium is critical for identifying the source of the pollution and limiting the environmental impact.6

Currently, use of classical potentiometric measurements with available ammonium membrane electrodes would make the above biomedical applications impossible. However, application of current has been extensively researched to achieve enhanced analytical properties for ion-selective membranes. For instance, external current has been used – coupled with classical potentiometry – to neutralize the trans-membrane flux of primary ions, resulting in a lower detection limit.7-9 Monitoring the trans-membrane flux of ions lead to an important direction in potentiometry, the backside calibration potentiometry. Where the disappearance of concentration gradients across the membrane is used to determine the sample composition by the knowledge about the solution composition at the backside of the membrane.10-12 Direct amperometric13-17 and coulometric18-21 detection of ions with ion-selective membranes has been also reported in the literature. One major advantage of the amperometric methods over classical potentiometry is that current can be routinely 2 ACS Paragon Plus Environment

Page 3 of 19

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


measured on pA levels. According to Ohm’s law sufficiently large currents are expected when the resistance of the ISM is < 1 MΩ, which is generally the case for thin (e.g., < 200 µm) plasticized PVC membranes. Consequently, lower limit of detection (LOD) can be achieved.

The quasi-symmetric membrane/amperometric system proposed herein (see Figure 1) differs from the published methods (e.g., dynamic electrochemical methods22) in the fact that no external potential is applied between a working and reference/counter electrode. Thus, in theory, current flows between the electrodes only if there is an asymmetry in the phase-boundary potential change of the ISM with the ionophore and the ISM without the ionophore. In other words, in our system current emerges due to an internal process – the difference between the phase-boundary potentials – and not due to an externally applied potential. In principle, the species that gives rise to the largest asymmetry between the two ISMs is the preferred primary ion of the ionophore. In this work, the Nernst-Planck-Poisson (NPP) model23-25 is used to model the time-dependent amperometric behavior of this two ion-selective membrane, three-compartment measurement system. These theoretical simulations provided valuable insight to the experimentally measured current traces recorded for ammonium ISM prepared with nonactin. In the proposed setup, the two membrane compositions are similar (33% PVC, 66% plasticizer, 0.44% lipophilic ion-site additives) the only difference is the presence of the ionophore in the first membrane.

MATERIALS AND METHODS

Chemicals Poly(vinyl chloride), 2-nitrophenyl octyl ether (o-NPOE), potassium tetrakis[3,5-bis(trifluoromethyl)phenyl]borate (KTFPB), and tetrahydrofuran (THF), nonactin, and bovine serum albumin (BSA) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Sodium chloride, ammonium chloride, magnesium chloride, calcium chloride, and potassium chloride were products of Fluka AG (CH-9471 Buchs, Switzerland). All solutions were prepared with deionized water (>18 MΩ cm specific resistance) obtained with a Millipore water system (Millipore Corp., Billerica, MA, USA).

3 ACS Paragon Plus Environment


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 19

Membranes The ammonium selective membranes consisted of KTFPB (0.44 wt%, 5.0 mmol kg– 1

), PVC (33 wt%), and o-NPOE (66 wt%). In addition, the membrane with ionophore

(Mw/IP) contained 10 mM (0.8 wt%) nonactin. These components were dissolved in THF and poured into a glass ring (φ 21 mm) fixed on a glass plate. The membranes were mounted between the compartments of the cell (Fig. 1) after overnight evaporation of THF.

Experimental setup Measurements were carried out in a three compartment poly(methyl methacrylate) cell setup (Fig. 1 and Fig. S1). The left and the right compartments, each containing 3 mL of solution, were separated by membranes from the sample solution (3 mL). Thicknesses of the membranes were calculated from the disk weight to be 80 µm, and the diameter in direct contact with the solutions was 9 mm. Both outer compartments contained Ag/AgCl electrode.

Electrochemical measurements Potentials were monitored by a Lawson Labs 16 channel EMF meter (Pennsylvania, USA). Currents were registered by an Autolab PGSTAT10 amperometric workstation (ECO Chemie, Netherlands). The sampling frequency in both cases was 1 Hz. The galvanic cell was Ag / AgCl / left solution (0.1 mM KCl) / ISE membrane with ionophore / sample solution / ISE membrane without ionophore / right solution (0.1 mM KCl) / AgCl / Ag. Electrochemical impedance spectroscopy measurements (Gamry Instruments, Reference 600, Warminster, PA, USA) were used to assess the bulk resistances of the membranes. The impedance spectra were recorded in the frequency range 1 MHz - 1 Hz by using a sinusoidal excitation signal with amplitude of 10 mV and DC potential of 0 V. All compartments contained 10 mM KCl.

THEORY

The Nernst-Planck-Poisson (NPP) model can be applied to assess the timedependent

amperometric

behavior

of

the

proposed

dual

ion-selective 4

ACS Paragon Plus Environment

Page 5 of 19

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


membrane/three-compartment measurement system. The NPP model is the state-ofthe-art simulation technique to evaluate the dynamics of potential and concentration change across the membrane of ion-selective electrodes.23-25 It avoids the necessity to split the membrane potential into phase-boundary and diffusion potentials, which is essential for simulating current. Additionally, it is applicable to ions of any charge and it gives access to both time and space domains. Consequently, analyses of the transient membrane potential and its spatial distribution over the entire thickness of the membrane as well as the diffusion layers are possible.

The simulated system was one-dimensional: contacting reference solution (left) / ISM with ionophore / contacting sample solution (left) / contacting sample solution (right) / ISM without ionophore / contacting reference solution (right). The simulated membranes contained mobile lipophilic ion-exchangers (TFPB-), which were confined to the membranes, and primary and interfering ions could be extracted to and from the solutions in contact with the membranes. No association (e.g., no ion pairs) between extracted ions and lipophilic ion-exchangers was considered and convection in the diffusion layer was ignored. All activity coefficients were assumed to be unity. Dielectric permittivity and diffusion coefficients were constant within the phases (membrane and solution). The Nernst-Planck equation for a mobile ionic component is given by

 ∂c ( x, t ) F  J i ( x, t ) = − Di  i − zi ci ( x, t ) E ( x, t )  RT  ∂x 

(Eq. 1),

where J i ( x, t ) is the flux at given space (x) and time (t), ci is the concentration, Di is the diffusion coefficient, and zi is the valence of the i-th species. E is the electric field and F, R, T are the Faraday constant, the gas constant and the temperature, respectively. The total current density form of the Poisson equation is: I ( x, t ) = F ∑ zi J i ( x, t ) + ε i

∂E ( x, t ) ∂t

(Eq. 2),

where I ( x, t ) is the total current density and ε is the permittivity, due to the amperometric conditions: ∫ E ( x, t ) dx = 0 V . Ion extraction is described by first order heterogeneous rate constants. Consequently, the following boundary conditions are applied: 5 ACS Paragon Plus Environment


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 19

r p / p +1 s p / p +1 J ip / p +1 ( t ) = k i cip,L/ p +1 ( t ) − k i cip,R/ p +1 ( t ) where

ci ,L and

ci ,R

(Eq. 3),

are concentrations on the left and right side of the

interface p / p + 1 , respectively, and k is the forward and backward heterogeneous rate constants of transitions across the membrane/solution interface. The heterogeneous rate constants were chosen to obtain a selectivity of 10 for ammonium over potassium based on the principles laid out by Sokalski et al.24 The above equations are converted to totally implicit finite difference form with a space grid containing closely spaced points near the interfaces and a wider spacing inside the bulk membrane. The resulting system of differential equations can be solved by the modified Powell hybrid method.26 The program code was written in Python and C programming languages. The simulated system consisted of two diffusion layers in the sample compartment and two membranes. The concentration of the left cell, right cell, and the bulk of the sample cell was assumed to be constant and isotropic. The thicknesses of the diffusion layers and the membranes were considered to be 100 µm. The right membrane contained no ionophore, while the left membrane contained ammonium ionophore that had a selectivity of 10 fold over potassium (e.g., similar to that of nonactin). Relative permittivities of water and ISM were considered to be 80.1 and 10, respectively. Other simulation parameters are presented in Table 1. Table 1: Initial concentrations (M), diffusion coefficients (cm2s-1) and heterogeneous rate constants (ms-1). RL, M, RR, Mw/oIP, Mw/IP indexes indicate reference solution (left), middle solution, reference solution (right), membrane without ionophore, and membrane with ionophore, respectively. Lipophilic ion-exchanger was tetrakis[3,5bis(trifluoromethyl)phenyl]borate ( TFPB− ). Ion

ci ,RL

ci ,Mw/IP

ci ,M

ci ,Mw/oIP

ci ,RR

Di ,M

Di ,S

r k i ,Mw/IP

s k i ,Mw/IP

r s k i ,Mw/oIP k i ,Mw/oIP

NH +4

0

0

10−7 − 10−3

0

0

10−9

10−5

10−2

10−6

10−2

2 ⋅10−5

K+

10−4

5 ⋅10−3

10−4

5 ⋅10−3

10−4

10−9

10−5

10−2

10−5

10−2

2 ⋅10−5

TFPB−

-

5 ⋅10−3

-

5 ⋅10−3

-

10−9

-

0

0

0

0

Cl−

10−4

-

10−4 − 1.1 ⋅10−3

-

10−4

-

10−5

0

0

0

0


Page 7 of 19

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


RESULTS AND DISCUSSION

Theoretical modeling Simulated response curves (see Fig. 2) show the chronoamperometric response for 100 nM – 10 mM NH +4 in the presence of 0.1 mM K + for an ionophore system with selectivity for ammonium over potassium of 10 fold. The model predicts a negative slope at higher concentrations of ammonium after the peak. The phenomenon cannot be described in terms of the steady-state phase-boundary potential model,27 since it originates from the time-dependent migration and diffusion of the ionic species in the membrane (e.g. lipophilic anion, primary ion, and interfering ion). The emerging current causes non-linear potential profiles within the two membranes (Fig. 3b). The simulations predict sigmoidal calibration curve. At low ammonium concentrations neither of the membranes respond. For intermediate levels the response of the membrane with ionophore is larger than the response of the membrane without ionophore. At high enough concentrations when both membranes respond in Nernstian manner the plateau of the sigmoid is reached, since both of the phaseboundary potentials change similarly. In the amperometric measurement mode, the overall potential-drop across the cell is 0 V. Current flows through the dual-ISM-based cell only if phase-boundary potentials differ between the two membrane|sample solution interfaces. Since the only difference between the two membranes is the presence of the ammonium selective ionophore, in theory, the largest variance between the two phase-boundary potentials is expected to be caused by ammonium ions. When the primary ion is present in the middle solution, the phase-boundary potential is larger at the Mw/IP|sample solution interface than at the Mw/oIP|sample solution interface (Fig. 3b). Consequently, current flows through the cell. The emerging current results in a gradient in the counter ion concentration profile (Fig. 3a). Sodium or any other ion that the ionophore is not especially selective over (i.e., weak binding), will induce similar phaseboundary changes in both membranes. The potential profile in that case is a symmetrical rectangular function with steps at the interfaces. When phase-boundary potentials are the same they cancel each other out, so the overall current is 0 A.



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 19

Experimental results

Using electrochemical impedance spectroscopy the resistance of the membrane with and without nonactin was measured to be 45.9±2.1 and 32.4±0.4 kΩ, respectively. A disadvantage of the proposed amperometric setup is that the measured signal – unlike chronopotentiometry – depends on the membrane resistances, thus calibration is necessary.

In order to show the advantage of the proposed setup, classical two-compartment and

the

novel

three-compartment

potentiometric,

and

three-compartment

amperometric calibration curves for ammonium, potassium, sodium, magnesium and calcium can be seen in Figure 4. A signal for 200 nM ammonium in the presence of 0.1 mM KCl is detectable based on the 3σ-criterion with the amperometric setup. Calculating with the same 3σ-criterion,28 the potentiometric detection limit is worse by a factor of five, due to the observed lower signal-to-noise ratio of the background in the amperometric setup. An additional advantage of the three-compartment cell is that even when the interfering ions (e.g., potassium, calcium) show a Nernstian response at higher concentrations in the conventional cell, in the proposed cell their electrochemical signal is significantly lower, due to the above mentioned principles.

It is important to note that the quasi-symmetric amperometric setup might have a short term current response for ions that are not preferred, such as sodium or potassium. The reason for this behavior is the difference in the response time of the two membranes. This kinetic difference results in a pulse shaped amperometric curve (Fig. 5), and after a few minutes the current converges back to the baseline. On the other hand when the primary ion is injected in the sample compartment, a staircase amperometric response is observed (Fig. 6) with a theoretically predicted negative drift at high concentrations. It is this property of the system that is the most important in terms of the analytical selectivity. In a conventional potentiometric setup, it is not possible to differentiate between the addition of 1 mM potassium and 0.1 mM ammonium (assuming the potentiometric selectivity coefficient is 0.1). However, with quasi-symmetric dual ISM amperometry, potassium does not give any long term current response. Thus it can be more easily discriminated from the response to the primary ion, ammonium. The observed low signal to the interfering ions is critical 8 ACS Paragon Plus Environment

Page 9 of 19

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


when considering practical applications. The two most important consequences in this regard are a) the background composition of the sample does not need to be rigorously optimized, as long as the concentration of the interfering ion is high enough to support a stable phase-boundary potential and ionic conductivity; and b) the assumption that the ionophore responds only to the primary ion is unnecessary, as long as there is a ca. tenfold selectivity between the two most preferred ions. In case of the proposed nonactin based setup the latter was true for the tested ions. According to the unbiased selectivity experiments (Table S1), the presence of nonactin decreased the selectivity coefficients of the interfering ions by at least one order of magnitude.

To investigate the applicability of the proposed method for potential blood ammonium ion measurements, experiments were carried out at physiological serum albumin, hydrogen, sodium and potassium ion levels (Fig. 7). The cell was equilibrated for an hour in 44 g/L bovine serum albumin (BSA), 137 mM NaCl, 2,7 mM KCl, 10 mM phosphate buffer (pH=7.4) in the sample compartment. At high electrolyte concentrations, cations co-extract with their counter anions into the membrane. As shown in Figure 7, with the proposed amperometric method, 10 µM ammonium ion levels

are

detectable,

which

makes

it

potentially

suitable

for

biological

measurements, since the typical blood levels of ammonium is between 10 and 50 µM.29

CONCLUSIONS

These experiments demonstrate that the two most important analytical properties of a sensor, namely the selectivity and the detection limit, can be enhanced by using the dual quasi-symmetric ISMs with amperometric detection over that possible with conventional potentiometry. We showed that ammonium can be detected at 10 µM levels in the presence of physiological levels of sodium, potassium, and serum albumin. With improved selectivity and a lower LOD, this setup is an excellent choice for such measurements.

Utilizing the enhanced selectivity of the proposed differential membrane setup, the interference from blood components, e.g. proteins and other macromolecules can be 9 ACS Paragon Plus Environment


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 19

intrinsically compensated for. Further, It may also be possible to eliminate the two outer compartments by the application of all-solid-state ion-selective sensors based on

conducting

polymers

(such

as

polypyrrole

and

poly(3,4-

ethylenedioxythiophene)).30, 31 In the future, this approach can decrease the size of the sensors to a few mm2, and potentially allow arrays of sensors of this type to be employed.

ACKNOWLEDGEMENT L.H. gratefully acknowledges the support of the Bolyai János Fellowship. The support of the Lendület program of the Hungarian Academy of Sciences (LP2013-63/2013) is gratefully acknowledged.

ASSOCIATED CONTENT Potentiometric selectivity coefficients and picture of the three-compartment setup.


Page 11 of 19

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


FIGURES Figure 1. Scheme of the three-compartment experimental setup. Compartments are separated by two ionselective membranes, which are identical except for the presence of ionophore.



Figure 2. Simulated response curves for different primary ion concentrations of middle solution. The upper curve is 10-3 M [NH +4 ] , the bottom curve is 10-7 M [NH +4 ] . Simulation parameters can be found in Table 1.

-2

100

j / nAcm

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 19

50

0 0

2

4

6

8

10

12

14

t / min


Page 13 of 19

Figure 3. a) Concentration profiles through the cell at 10-5 M level of NH +4 . b) Potential profiles through the cell, NH +4 concentration is increased from 10-7 M (blue) to 10-3 M (red) by one order of magnitude step size. Membrane with ammonium ionophore is plotted from 0 to 100 µm, the diffusion layer in the sample solution contacting it is plotted from 100 to 200 µm, the diffusion layer contacting the membrane in the sample solution without ionophore is plotted from 200 to 300 µm, and from 300 to 400 µm is the membrane without ionophore.

a

b

+

NH4

100

+

K TFPB

6

50 4

U / mV

C / mM

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


2

0 -50

-100

0 0

100

200

300

400

0

100

d / µm

200

300

400

d / µm



Figure 4. a) Classical two-compartment potentiometric, b) three-compartment potentiometric, and c) three compartment amperometric calibration curves for ammonium (black), sodium (red), and potassium (blue), calcium (orange), and magnesium (green) ions in the background of 0.1 mM KCl. The error bars represent one standard deviation. 80

a

150

b

c

60 100

EMF / mV

100

I / nA

150

EMF / mV

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 14 of 19

40

50

50

20

0

0

0 -7

-6

-5

log c

-4

-3

-7

-6

-5

-4

-3

log c

-7

-6

-5

-4

-3

log c


Page 15 of 19

Figure 5. Amperometric response curve when increasing the concentration of potassium from 0.1 mM to 1 mM in the sample solution. 15

10

I / nA

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


5

0 110

115

120

t / min



Figure 6. Typical amperometric response curve for ammonium in 0.1 mM KCl background.

10-3 M

140 120

10-4 M

100

I / nA

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 16 of 19

80 60 40

10-5 M 10-6 M

20 0 500

600

700

800

900

t/s


Page 17 of 19

Figure 7. a) Amperometric calibration curve and b) typical response curve of ammonium ion in 44 g/L BSA, 0.137 M NaCl, 2.7 mM KCl, pH=7.4 background solution. 250

a

200

10-2 M

b

10-3 M

200 150

150

I / nA

I / nA

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


100 50

100 50

10-4 M 10-5 M

0 -7

-6

-5

-4 +

-3

log[NH4]

-2

0 200

400

600

t/s



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 18 of 19

REFERENCES (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)

Sokalski, T.; Ceresa, A.; Zwickl, T.; Pretsch, E. J Am Chem Soc 1997, 119, 11347−11348. Szigeti, Z.; Malon, A.; Vigassy, T.; Csokai, V.; Grun, A.; Wygladacz, K.; Ye, N.; Xu, C.; Chebny, V.; Bitter, I.; Rathore, R.; Bakker, E.; Pretsch, E. Anal Chim Acta 2006, 572, 1−10. Malon, A.; Vigassy, T.; Bakker, E.; Pretsch, E. J Am Chem Soc 2006, 128, 8154−8155. Traeger, H.; Gabuzda, G.; Ballou, A.; Davidson, C. Metabolism 1954, 3, 99−109. Suzuki, K.; Siswanta, D.; Otsuka, T.; Amano, T.; Ikeda, T.; Hisamoto, H.; Yoshihara, R.; Ohba, S. Anal Chem 2000, 72, 2200−2205. Jiao, J. J.; Wang, Y.; Cherry, J. A.; Wang, X.; Zhi, B.; Du, H.; Wen, D., Environ Sci Technol 2010, 44, 7470−7475. Pergel, E.; Gyurcsanyi, R.; Toth, K.; Lindner, E. Anal Chem 2001, 73, 4249−4253. Höfler, L.; Bedlechowicz, I.; Vigassy, T.; Gyurcsányi, R. E.; Bakker, E.; Pretsch, E. Anal Chem 2009, 81, 3592−3599. Lindner, E.; Gyurcsanyi, R.; Buck, R. Electroanal 1999, 11, 695−702. Malon, A.; Bakker, E.; Pretsch, E. Anal Chem 2007, 79, 632−638. Ngeontae, W.; Xu, Y.; Xu, C.; Aeungmaitrepirom, W.; Tuntulani, T.; Pretsch, E.; Bakker, E. Anal Chem 2007, 79, 8705−8711. Xu, Y.; Ngeontae, W.; Pretsch, E.; Bakker, E. Anal Chem 2008, 80, 7516−7523. Sutter, J.; Morf, W.; De Rooij, N.; Pretsch, E. J Electroanal Chem 2004, 571, 27−35. Morf, W.; Wuhrmann, P.; Simon, W. Anal Chem 1976, 48, 1031−1039. Horvath, V.; Horvai, G.; Pungor, E. Mikrochim Acta 1990, 1, 217−224. Horvai, G.; Pungor, E. Anal Chim Acta 1991, 243, 55−59. Hupa, E.; Vanamo, U.; Bobacka, J Electroanal Chem 2015, 27, 591-594 Grygolowicz-Pawlak, E.; Bakker, E. Anal Chem 2010, 82, 4537–4542. Shvarev, A.; Neel, B.; Bakker, E. Anal Chem 2012, 84, 8038–8044. Yoshida, Y.; Nakamura, S.; Uchida, J.; Hemmi, A.; Maeda, K. J Electroanal Chem 2013, 707, 95−101. Afshar, M. G.; Crespo, G. A.; Bakker, E. Angew Chem 2015, 127, 8228−8231. Crespo, G. A.; Bakker, E. RSC Adv 2013, 3, 25461−25474. Sokalski, T.; Lewenstam, A. Electrochem Commun 2001, 3, 107−112. Sokalski, T.; Lingenfelter, P.; Lewenstam, A. J Phys Chem B 2003, 107, 2443−2452. Jasielec, J. J.; Sokalski, T.; Filipek, R.; Lewenstam, A. Anal Chem 2015, 87, 8665−8672. Powell, M. J. D.; Rabinowitz, P. In Numerical methods for nonlinear algebraic equations; Gordon & Breach, 1970, pp 87. Bakker, E.; Buhlmann, P.; Pretsch, E. Talanta 2004, 63, 3−20. Bakker, E.; Pretsch, E. TRAC 2005, 24, 199−207. Dawson, A.M. Gut, 1978, 19, 504−509. Cadogan, A.; Gao, Z.; Lewenstam, A.; Ivaska, A.; Diamond, D. Anal Chem 1992, 64, 2496−2501. Bobacka, J.; Lahtinen, T.; Nordman, J.; Haggstrom, S.; Rissanen, K.; Lewenstam, A.; Ivaska, A. Electroanal 2001, 13, 723−726.


Page 19 of 19

For TOC only: +

NH4 Utilizing the phaseboundary potential difference

50 0

to create current

-50 -100

Na + K

100

I / nA

100

U / mV

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


+

50

0

0

100

200 300 d / µm

400

-7

-6

-5

-4

-3

log c


Lowering Detection Limits Toward Target Ions Using Quasi-Symmetric

Recommend Documents