Research Article

Cite This: ACS Appl. Mater. Interfaces 2018, 10, 19248−19257 www.acsami.org

Conductive Metal−Organic Frameworks as Ion-to-Electron

Transducers in Potentiometric Sensors
Lukasz Mendecki and Katherine A. Mirica*
Burke Laboratory, Dartmouth College, 41 College Street, Hanover, New Hampshire 03755, United States
*
S Supporting Information

ABSTRACT: This paper describes an unexplored property of

conductive metal−organic frameworks (MOFs) as ion-to-
electron transducers in the context of potentiometric
detection. Several conductive two-dimensional MOF ana-
logues were drop-cast onto a glassy carbon electrode and then
covered with an ion-selective membrane to form a
potentiometric sensor. The resulting devices exhibited
excellent sensing properties toward anions and cations,
characterized by a near-Nernstian response and over 4 orders
of magnitude linear range. Impedance and chronopotentio-
metric measurements revealed the presence of large bulk
capacitance (204 ± 2 μF) and good potential stability (drift of
11.1 ± 0.5 μA/h). Potentiometric water test and contact angle
measurements showed that this class of materials exhibited hydrophobicity and inhibited the formation of water layer at the
electrode/membrane interface, resulting in a highly stable sensing response with a potential drift as low as 11.1 μA/h. The
property of ion-to-electron transduction of conductive MOFs may form the basis for the development of this class of materials as
promising components within ion-selective electrodes.
KEYWORDS: conductive metal−organic frameworks, potentiometric sensors, ion-selective electrode, electrochemical sensors,
ion-to-electron transduction

■ INTRODUCTION
Potentiometric sensors are an important class of analytical
electrode interface.15−17 Efforts to fulfill these combined criteria
through the implementation of conductive polymers and
devices with broad utility in clinical and environmental porous carbon-based materials as ion-to-electron transducers
analysis.1−5 Driven by the progress at the interfaces of have enabled major advancements in potentiometry, giving rise
analytical, supramolecular, and materials chemistry, modern to the most stable and robust potentiometric devices to
potentiometric devices have evolved from mechanically fragile date.8,18−21 Nonetheless, the existing classes of materials exhibit
liquid-filled ion-selective electrodes (ISEs) into solid-contact trade-offs and limitations in the context of SCEs.19,22−27
electrodes (SCEs) with excellent sensitivity, exceptional Conductive polymerssuch as poly(octylthiophene) (POT),
selectivity, and fast response time.6,7 This new generation of poly(3,4-ethylenedioxythiophene) (PEDOT), and polypyrrole
SCEs, based on a layered device architecture comprising an (PPy)can provide hydrophobicity (contact angle of water of
electrical contact coated with an ion-to-electron transducer and 50° and 100° measured on glassy carbon electrode (GCE)
further protected by an ion-selective membrane (ISM), seeks to surface modified with either PEDOT−polystyrene sulfonate
achieve outstanding analytical performance with stable and (PSS)28 or Ppy−perfluorooctanesulfonate,29 respectively) and
robust device design capable of long-term analysis without high redox capacitance (10−200 μF).30 However, their
deterioration in performance.8−12 Although recent advance- performance can also be diminished by: (i) variations in the
ments in the development of ISMs and ion-to-electron crystallinity that can alter charge transport;31 (ii) dopant-
transducing materials have led to significant progress in dependent changes in glass-transition temperature that may
potentiometric detection, overcoming the existing chemical affect the mechanical stability of the transducer layer;32 (iii)
limitations of available materials capable of forming well- sensitivity to O2, CO2, pH, and light that can lead to potential
defined interfaces between the electrode and the ISM may drift;33 and (iv) dependence of conductivity upon conforma-
enhance broad implementation of SCEs in chemical sensing.13 tional changes.34 Nanostructured carbon-based materials, such
Achieving stable and reliable measurements using SCEs as carbon nanotubes (CNTs), graphene, fullerenes, and three-
hinges on four fundamental conditions: (i) reversible ion-to-
electron signal transduction; (ii) non-polarizable interface with Received: March 9, 2018
high exchange current density; (iii) absence of any side Accepted: May 2, 2018
reactions;14 and (iv) absence of a thin water layer at the ISM/ Published: May 24, 2018

© 2018 American Chemical Society 19248 DOI: 10.1021/acsami.8b03956

ACS Appl. Mater. Interfaces 2018, 10, 19248−19257
ACS Applied Materials & Interfaces Research Article

dimensional ordered mesoporous (3DOM) carbons have been First, the class of 2D MOFs used in this study exhibits good
recently shown to perform the function of ion-to-electron conductivity64 that rivals that of carbon-based materials, making
signal transduction via the formation of the electrical double them suitable candidates for electronically transduced signal in
layer at the membrane/electrode interface.18,35,36 The inherent analytical devices. Second, being highly porous, conductive
hydrophobicity of these materials, combined with excellent MOFs have high surface areas on par with those reported for
electrical conductivity and high capacitance (625 μF for 3DOM nanoporous carbons.43,54,65,66 This characteristic is important
carbon8 and 302 μF for single-walled CNTs),25 reinforces the for obtaining high capacitance necessary for facilitating ion-to-
advantageous use of nanostructured materials as components electron transduction and obtaining high potential stability.
for the development of SCEs. However, several limitations to Third, these materials can be synthesized from metal ions
the use of carbon materials in sensing devices can also limit interconnected with organic ligands through solution-phase
their implementation, including: (i) limited control over the self-assembly to produce highly ordered, crystalline struc-
structure and electronic properties of the nanostructured tures.42,47,64,67 This bottom-up synthetic approach provides
carbon materials that may influence the electrical conductivity structural control and compositional modularity, allowing for
of the material (e.g., CNTs);37 (ii) requirement for expensive the development of conductive materials with predictable and
and sophisticated instrumentation for fabrication;38 (iii) harsh tuneable properties.43,68 The synthetic control over structure
synthetic conditions during the manufacturing process, such as offers a potential advantage over pyrolytic carbons and
high pressure and temperature;39 and (iv) limited control over conductive polymers in terms of chemical control of batch-
the amount/type of impurities introduced during the to-batch structural reproducibility and application-specific
fabrication process.36,40 Despite recent improvements in tunability.43,67,69 The synthetic conditions for MOF preparation
fabrication methods of SCEs based on carbon materials and are significantly milder than those typically employed in the
conductive polymers,18,41 fundamental research focused on synthesis of porous carbons39,42 and may offer molecular level
integrating novel conductive materials that exhibit the required control of interfaces, material composition, and device
characteristics of ion-to-electron transduction, high capacitance, performance. On the basis of these unique characteristics, we
and substantial hydrophobicity can promote future advance- reasoned that conductive MOFs can be designed to exhibit ion-
ments in potentiometry. to-electron transduction, high capacitance, and substantial
This paper describes the first implementation of conductive hydrophobicity critical for advancing the field of potentiometric
two-dimensional (2D) metal−organic frameworks (MOFs) as detection.

■
ion-to-electron transducers in potentiometric sensing. The
MOFs comprising metallic nodes (Ni, Cu, and Co) EXPERIMENTAL SECTION
interconnected with triphenylene-based organic linkers
(2,3,6,7,10,11-hexahydroxytriphenyleneHHTP) arranged in Materials and Instruments. Bis[(benzo-15-crown-5)-15-
a Kagome lattice42 (Figure 1) were integrated into layered ylmethyl]pimelate (potassium ionophore II), valinomycin (potassium
ionophore I), sodium tetrakis[3,5-bis(trifluoromethyl)phenyl]borate
(NaTFPB), tridodecylmethylammonium chloride (TDMACl), high-
molecular weight polyvinyl chloride (PVC), and bis(2-ethylhexyl)
sebacate (DOS) were purchased from Sigma-Aldrich. HHTP was
purchased from TCI Chemicals. All chemicals were of analytical
reagent grade. Solutions of metal ions were prepared in ultrapure water
obtained with a PicoPure 3 water system (resistance 18 MΩ cm−1).
Working solutions of different metal ions were prepared by serial
dilutions of a 1 M stock solution. GCEs (3 mm in diameter) were
purchased from CH instruments, USA. Scanning electron microscopy
(SEM) and energy dispersive X-ray spectroscopy (EDX) were
performed using Hitachi TM3000 SEM (Tokyo, Japan) equipped
for X-ray microanalysis with a Bruker Edax light element Si(Li)
detector (Billerica, MA). An EmStat MUX16 potentiostat (Palm
Instruments BV, Netherlands) was used for electrochemistry. The z-
dimension (height) for all electrodes was collected using a Zygo
NewView 7300 light interferometer equipped with a 10× lens. X-ray
photoelectron spectroscopy (XPS) experiments were conducted using
a Physical Electronics Versaprobe II X-ray photoelectron spectrometer
under ultrahigh vacuum (base pressure 10−10 mbar). The measure-
Figure 1. Schematic representation of the layered device architecture
ment chamber was equipped with a monochromatic Al Kα X-ray
employed in this study. A thin film of an MOF is drop-casted directly
source. Both survey and high-resolution spectra were obtained using a
on the top of a GCE and then covered with an ISM to enable
beam diameter of 200 μm. The spectra were processed with CasaXPS
potentiometric ion sensing.
software. Powder X-ray diffraction (PXRD) measurements were
performed with a Bruker D8 diffractometer equipped with a Ge-
device architectures to produce functional potentiometric monochromated 2.2 kW (40 kV, 40 kA) Cu Kα (α = 1.54 Å) radiation
devices. This general class of 2D materials exhibits good source and an NaI scintillation counter detector (Billerica, MA).
electrical conductivity,42−46 chemiresistive behavior,47−52 en- Synthesis and Characterization of Conductive MOFs. The
synthesis of MOFs using the organic linker HHTP was adapted from
ergy storage capacity,53−55 and catalytic activity,56−63 but the
Yaghi and co-workers.42
ability to carry out analytical ion-to-electron transduction and Ni3HHTP2. To a 100 mL round-bottom flask, 500 mg of HHTP and
the implementation in potentiometric detection have not been 767 mg of Ni(OAc)2·4H2O were added. Deionized water (70 mL) was
previously demonstrated. then added to the round-bottom flask. The resulting suspension was
We believe that conductive MOFs possess at least three sonicated for 15 min. The reaction mixture was then heated for 24 h at
unique characteristics for promising utility in potentiometry. 85 °C under stirring and air. The flask was left to cool for 1 h at room

19249 DOI: 10.1021/acsami.8b03956

ACS Appl. Mater. Interfaces 2018, 10, 19248−19257
ACS Applied Materials & Interfaces Research Article

temperature and the reaction product was filtered with a ceramic Preparation of Coated Wire Electrodes (GCE/ISM). Coated
funnel and a filter paper. The solid precipitate was washed with wire electrodes were prepared by drop-casting (10 μL) of the ISM
ultrapure water (1000 mL) and acetone (300 mL). The resulting black suspension directly onto the top of a GCE. Each device was then dried
powder was transferred into a glass vial and dried overnight under at room temperature for 24 h and then immersed in 5 mL of the 1.0 ×
vacuum (20 mTorr) at 85 °C. 10−3 M KCl solution for an additional 24 h.
Cu3HHTP2. To a 50 mL round-bottom flask 100 mg of HHTP and Potentiometric Measurements. Potentiometric responses of all
154 mg of copper(II) trifluoroacetylacetonate were added. Deionized electrodes were recorded using the EmStat MUX16 potentiostat
water (20 mL) was then added to the round-bottom flask. The (Palm Instruments BV, Netherlands) in a stirred solution against a
resulting suspension was sonicated for 15 min. The reaction mixture double-junction Ag/AgCl reference electrode with a 1 M KCl bridge
was then heated for 24 h at 70 °C under stirring and air. The flask was electrolyte (Sigma-Aldrich) at room temperature. For all potentio-
left to cool for 1 h at room temperature and the reaction product was metric measurements, at least three independent NO3− and K+-ISEs
filtered with a ceramic funnel and a filter paper. The solid precipitate were used.
was washed with ultrapure water (1000 mL) and acetone (300 mL). Electrochemical Impedance Spectroscopy. All impedance
The resulting black powder was transferred into a glass vial and dried measurements were performed by using an Ivium Technologies
overnight under vacuum (20 mTorr) at 85 °C. CompactStat impedance analyzer (Ivium Technologies). The electro-
Co3HHTP2. To a 50 mL round-bottom flask 50 mg of HHTP and chemical impedance spectroscopy (EIS) measurements of the
74 mg of CoCl2·6H2O were added. Deionized water (20 mL) was then deposited MOF layer and ISMs were conducted using established
added to the round-bottom flask. The resulting suspension was techniques.19 Briefly, impedance spectra were collected using
sonicated for 15 min. The reaction mixture was then heated for 24 h at excitation amplitude of 0.01 V within the frequency range spanning
70 °C under stirring and air. The flask was left to cool for 1 h at room from 100 kHz to 0.01 Hz for the GCE/MOF electrodes. However,
temperature and the reaction product was filtered with a ceramic higher amplitude (0.1 V) was used for electrodes consisting of a
funnel and a filter paper. The solid precipitate was washed with selected MOF layer and an ISM to improve the overall signal-to-noise
ultrapure water (1000 mL) and acetone (300 mL). The resulting black ratio. A conventional three electrode set-up was used for all impedance
powder was transferred into a glass vial and dried overnight under measurements using a glassy carbon working electrode, a platinum
vacuum (20 mTorr) at 85 °C. auxiliary electrode, and a silver chloride electrode as the reference.
Characterization of M3HHTP2 MOFs (M = Co, Ni, and Cu). Each measurement was performed at open-circuit potential in 0.1 M
SEM revealed the presence of distinct rod-shaped crystallites in aqueous solution of potassium chloride at room temperature. All
Ni3HHTP2 and Cu3HHTP2 MOFs and non-uniform nanoscale impedance spectra were fitted to equivalent circuits using the
morphology in Co3HHTP2 MOF (Figure S1). The presence of IviumStat software version 2.0. All measurements were done at least
metals (Co, Ni, and Cu) and C, H, and O atoms in the bulk of each in triplicate.
MOF was confirmed by EDX spectroscopy (Figure S1). PXRD Cyclic Voltammetry. Cyclic voltammetry (CV) experiments were
revealed stacked layered structure consistent with previous reports carried out at room temperature in a conventional three-electrode set-
(Figure S2).42 XPS showed exclusive presence of Ni2+ in Ni3HHTP2 up using an Ag/AgCl (1 M) reference electrode and a platinum wire
and mixed valency in Co3HHTP2 (Co2+/Co3+) and Cu3HHTP2 (Cu+/ and glassy carbon as the auxiliary and working electrode, respectively.
Cu2+) MOFs with no additional counterions, suggesting that the KCl (0.1 M) was used as the background electrolyte for all
redox-active triphenylene ligand maintains charge neutrality in the voltammetric measurements that were run under a nitrogen
framework (Figures S3−S5). These results are consistent with other atmosphere.
reports of similar materials.42,44,47,64 Chronopotentiometric Characterization. A constant current of
Preparation of MOF Dispersion. The MOF dispersion was −1.0 nA was applied to the working electrode for 60 s followed by a
prepared by transferring 1.0 mg of a selected MOF into an Eppendorf reversed current of the same magnitude for the same length of time.
tube containing 1.0 mL of acetonitrile. The same tube was then The resulting electromotive force was recorded in a solution of 0.1 M
sonicated for 4 h at 25 °C to give a homogenous dispersion. KCl at room temperature. All measurements were carried out in a
Preparation of K+ and NO3− Sensing Membranes. Traditional three-electrode configuration using the ISEs under study as working
potassium-selective membranes contained 10 mmol kg−1 of electrodes. The reference electrode was Ag/AgCl in KCl (3 M) and
valinomycin (K+-ISM-I) or potassium ionophore II (K+-ISM-II), 5 the auxiliary electrode was a platinum wire.

■
mmol kg−1 of NaTFPB, 33.3 wt % PVC, and 66.6 wt % DOS. NO3−-
ISM contained 5.0 mmol kg−1 of TDMACl, PVC (33.2 wt %), and
RESULTS AND DISCUSSION
DOS (66.4 wt %). The abovementioned components were then
dissolved in 1.5 mL of tetrahydrofuran and the resulting mixture was Measuring the Charge-Transfer Resistance, Capaci-
vortexed for 30 min for complete dissolution. tance, and Diffusional Impedance of Conductive MOFs
Preparation of MOF-Coated Electrodes (GCE/MOF). The Using EIS. To study ion-to-electron transduction, we drop-
intermediate conductive layer consisting of M3HHTP2 MOFs was casted aqueous suspensions of M3HHTP2 MOFs on top of
prepared by drop-casting 5.0 μL aliquots of each conductive MOF
GCEs to obtain a layered device. In this configuration, EIS was
directly onto a solid contact. The electrodes were then dried under
nitrogen for 3 h before electrochemical measurements were carried used to achieve simultaneous measurement of three critical
out. parameters: (i) bulk capacitance, (ii) charge transfer resistance,
Preparation of ISEs (GCE/MOF/ISM). For potentiometric and (iii) diffusion-induced impedance. Figure 2A illustrates the
measurements, a dispersion of selected MOF (prepared as described Nyquist plot of the EIS spectrum recorded in 0.1 M KCl
in the “Preparation of MOF Dispersion”) was drop-cast onto the solution for three distinct devices (GCE/MOF) comprising
glassy carbon electrode and dried for 3 h at room temperature. An GCE electrodes with a drop-cast layer of Co3HHTP2 (blue),
aliquot (10 μL) of the ISM was drop-cast onto the previously Ni3HHTP2 (green), and Cu3HHTP2 (red) MOFs. This
deposited MOF, and the electrodes were left at room temperature to spectrum (Figure 2A) exhibits three discrete features: (i) a
dry overnight. The following day, the resulting ISEs were conditioned
straight line angled at almost 90° exhibiting only slight
in 1.0 × 10−3 M potassium chloride solution for 24 h prior to the
potentiometric and impedance experiments. Membrane thicknesses curvature at lower frequencies characteristic of excellent
were measured using a digital micrometer. For the long-term stability capacitance; (ii) the presence of a partial semicircle in the
test, each electrode was only used once (during initial measurement) high-to-intermediate frequency range characteristic of Ohmic
and then stored in dark, at room temperature under air atmosphere, resistance; and (iii) lack of Warburg impedance as evidenced by
for three weeks in distilled water and reused. the absence of a line inclined at 45° in the low-frequency range.
19250 DOI: 10.1021/acsami.8b03956
ACS Appl. Mater. Interfaces 2018, 10, 19248−19257
ACS Applied Materials & Interfaces Research Article

presence of interconnected nanopores within the MOF lattice

that are permeable to the electrolyte,19,21 and the presence of a
large number of conductive pathways for electron transport
within the drop-cast MOF film. These recorded values are
within the same order of magnitude as those for conductive
polymers, such as PEDOT−PSS (204 μF)23 or single-walled
CNTs (302 μF),25 and are approximately 10 times lower than
the one reported for colloidally imprinted mesoporous (CIM)
carbon, which was used to fabricate the most stable ISEs to this
date. The large magnitude of capacitance of CIM carbon could
be attributed to the presence of large interconnected
mesopores (diameter of ∼24 nm) that are readily accessible
for the electrolyte diffusion,18 in comparison to the micropores
found in 2D M3HXTP2 MOFs (M = Cu, Ni, and Co; diameter
of ∼1.8 nM), which define the available surface area and
consequently the capacitance of the MOF.
To examine the dependence of the low-frequency
capacitance on MOF film thickness, we prepared three
electrodes with different thicknesses of the conductive
Ni3HHTP2 MOF layer (20 ± 8, 40 ± 5, and 60 ± 5 μm)
and performed impedance measurements. Increasing the
thickness, and consequently the surface area, of the underlying
MOF layer intensified the magnitude of capacitance 153 ± 1,
155 ± 1, and 204 ± 2 μF for thicknesses of 20 ± 8, 40 ± 5, and
60 ± 5 μm, respectively, indicating that this parameter
represents a bulk capacitance of the studied Ni3HHTP2 MOF
as reported for (Figure 2B).19,21 The small increase in the
capacitance of 2.0 μF when the Ni3HHTP2 MOF layer
thickness was varied from 20 ± 8 to 40 ± 5 μm may be a
function of inhomogeneity of the MOF film deposited on the
electrode surface. For example, the drop-casting method can
result in the formation of “islands” of material, which may lead
to poor electrical contact between the MOF and electrode, thus
diminishing the magnitude of observed capacitance. The
increase in the capacitance of 50 μF with the MOF layer
Figure 2. (A) EIS spectra obtained for a drop-cast layer of either
Co3HHTP2, Ni3HHTP2, and Cu3HHTP2, MOFs deposited on GCE.
thickness from 40 ± 5 to 60 ± 5 μm may arise from the: (i)
(B) Impedance spectra obtained for the GCEs coated with different increased amount of MOF material being available for
thickness of drop-cast Ni3HHTP2 layer: 20 ± 8 μm (red triangles), 40 electrolyte wetting; (ii) enhanced electrical contact between
± 5 μm (blue circles), and 60 ± 5 μm (black squares). The inset the layers of drop-cast MOF and the electrode; and (iii)
shows a zoomed-in representation of the high-frequency impedance increase in the density of states in the MOF material.71
data (100 kHz to 3 Hz). In both cases (A,B), the proposed equivalent The presence of small impedance in the high-to-intermediate
circuit comprises the resistance of the solution (R1), charge-transfer frequency region (100 kHz to 3 Hz; Z′: 44.0−250.0 Ω)
resistance (R2), and two CPEs (CPE1 and CPE2). Frequency range: characterized by the partial semicircle in Figure 2Bindicated
100 kHz to 10 mHz; amplitude potential: 0.01 V; solution: 0.1 M KCl. Ohmic resistance25 rather than double-layer capacitance as
typically observed for ISEs based on conductive polymers.23 In
The straight line angled at almost 90° in the low-frequency this electrode configuration, the magnitude of this resistance
region (1 Hz to 10 mHz; Z′: 280.0 Ω to 1.2 kΩ) of the Nyquist varied with the thickness of deposited conductive MOF layer
plot (Figure 2A) represents the capacitance of the MOF film. (Figure 2B inset): increasing the thickness of the MOF layer
The solution resistance (R1) for 0.1 M KCl was estimated at from 20 ± 8, 40 ± 5, to 60 ± 5 μm diminished the magnitude
R1 = 140 Ω and was consistent with the literature.21 The of recorded impedance from 186, 95 to 35 Ω, respectively
capacitance of the MOF can be computed from the total (Figures 2B and S7). These findings indicate that the
impedance of the electrode at low-frequency data (1 Hz to 10 conductivity of a MOF is the major contribution to the
mHz) in Figure 2A. Fitting this data with the constant-phase Ohmic resistance because the MOF itself can act as the
element (CPE) in an equivalent circuit shown in Figure 2A electronic conductor.44 Increasing the thickness of the
gave a substantial capacitance of 204 ± 2 μF and phase value of conductive film would lead to the increase in the number of
0.93 for Ni3HHTP2 MOF (60 ± 5 μm thickness), 177 ± 3 μF possible conductive pathways in a layered device, and thus
and phase value of 0.84 for the Cu3HHTP2 MOF (60 ± 5 μm would diminish the magnitude of the observed Ohmic
thickness, Figure S6), and 157 ± 2 μF and phase value of 0.75 resistance. These observations are consistent with findings of
for Co3HHTP2 MOF (60 ± 5 μm thickness, Figure S6). Crespo and co-workers, demonstrating that the increase in the
Because the CPE phase values indicated the behavior of nearly thickness of the layer of CNTs correlates with diminished
perfect capacitors (phase value of 1), the units of capacitance charge-transfer resistance in potentiometric measurements.25
(farads) were used instead.18,36,70 We attribute the large values We did not observe the characteristic straight line inclined at
of recorded bulk capacitance for all studied MOFs to the 45° indicative of Warburg impedance associated with the
19251 DOI: 10.1021/acsami.8b03956
ACS Appl. Mater. Interfaces 2018, 10, 19248−19257
ACS Applied Materials & Interfaces Research Article

diffusion-limited transport of the electrolyte through the pores

of the material.19,21 This observation indicates that ions can
diffuse into/through the MOF network without the significant
contribution from the mass transport resistance.
Additional analysis using CV (Figure S8 and associated
discussion) revealed that the electrochemical behavior of
M3HHTP2 MOFs was influenced by the identity of the metal
centers embedded within the MOFs, as evidenced by the
characteristically different CV profiles for Ni 3 HHTP 2 ,
Cu3HHTP2, and Co3HHTP2 MOFs (Figure S8). The observed
voltammetric behavior may be explained by the redox activity of
HHTP ligand that can undergo reversible redox trans-
formations between catechol, semiquinone, and quinone
forms.72 Given that MOFs exhibit permanent porosity, changes
in the cumulative pore volume as a function of the MOF layer
thickness may be further manifested by the unique voltammetry
of these materials.73
We hypothesize that the electrochemical double-layer
charging, characterized by the rectangular shape of the
potential−current response, plays a role in the capacitive
behavior of of Ni3HHTP2 and Co3HHTP2 MOF-based ISEs.54
Interestingly, the presence of reversible redox peaks in the
cyclic voltammograms of Cu3HHTP2 MOF/GCE electrodes,
together with large capacitive behavior, as confirmed with EIS
measurements, may indicate that the redox pseudocapacitance
is the dominant charge-storage mechanism in Cu3HHTP2
MOFs.54,55 The presence of small redox peaks in the CV of
Ni3HHTP2 MOFs may also indicate partial contribution of
redox pseudocapacitance to the total capacitance in Ni3HHTP2
MOF, thus providing highest capacitance amongst all
M3HHTP2 MOFs in this study.74
To examine whether the presence of the Faradaic process
influences the impedance response of the MOFs, we performed
EIS measurements at three different potentials (−0.3, 0.0, and Figure 3. Impedance spectra of (A) K+-ISM-II deposited onto bare
0.3 V) for Ni3HHTP2. The change in magnitude of the applied GCE and (B) the same K+-ISM-II deposited on top of the Ni3HHTP2
potential did not significantly influence the outcome of the MOF/GCE substrate. The inset demonstrates the zoomed-in
impedance measurements, indicating the recorded capacitance impedance response at the high-frequency spectral region. The
is independent of the redox processes observed in CV (Figure equivalent circuits used for fitting impedance data are shown in the
S9). Taken together, these results confirm that conductive top right corner of each presented spectrum. Frequency range: 100
MOF-based materials possess substantial capacitance, Ohmic kHz to 10 mHz; amplitude potential: 0.1 V; solution: 0.1 M KCl.
resistance, and charge-transfer characteristics suitable for
serving the function of ion-to-electron transducers in ISEs semicircle at the high-frequency range followed by the partial
and satisfy the first major criteria in SCE development. arc at lower frequencies (1 Hz to 10 mHz; Z′: 1.1−12.3 MΩ).
Quantifying the Efficiency of Ion-to-Electron Trans- We attributed the presence of the high-frequency semicircle
duction by Integrating Conductive MOFs into Solid- (100 kHz to 1 Hz; Z′: 60.0 Ω to 1.0 MΩ) to the bulk resistance
State ISE Devices. We focused our subsequent experimental in parallel with geometric capacitance of the drop-casted K+-
efforts on the use of Ni3HHTP2 for the development of stable ISM-II. The bulk membrane resistance (1.0 ± 0.2 MΩ) was
ISEs because of its highest bulk capacitance that would facilitate extracted from the impedance model shown in Figure 3A. The
ion-to-electron transduction and provide substantial signal second “larger” partial semicircle originated from the large
stability during potentiometric measurements. To assess the charge-transfer resistance at the interface between the ISM and
function of conductive MOFs as ion-to-electron signal GCE.19 This second spectral feature indicated ineffectiveness of
transducers in potentiometry, we used impedance measure- ion-to-electron transduction in the simple two-layer GCE/K+-
ments to compare the electrochemical performance of two ISM-II device configuration.
device configurations fabricated by sequential drop-casting of In contrast to Figure 3A, the impedance spectrum recorded
distinct layers (Figure 3). The first device configuration for the GCE/MOF/K+-ISM-II devices (Figure 3B) showed
involved the direct deposition of K+-ISM-II on top of GCE only one high-frequency spectral component (100 kHz to 1
to create a two-layer device: GCE/K+-ISM-II (Figure 3A). The Hz; Z′: 59.0 Ω to 1.1 MΩ), which we attributed to the bulk
second device configuration employed an Ni3HHTP2 MOF properties (resistance in parallel with geometric capacitance) of
thin film sandwiched between the GCE and K+-ISM-II to create the ISM (1.0 ± 0.3 MΩ).19 Notably, the absence of a second
a three-layer device (GCE/MOF/K+-ISM-II, Figure 3B) semicircle in the low-frequency region of the spectrum for the
capable of enhanced signal transduction. GCE/MOF/K+-ISM-II devices was characteristic of the low
For the control GCE/K+-ISM-II configuration (Figure 3A), charge-transfer resistance at ISM/MOF interface. In addition,
impedance measurements revealed the presence of a small CV of both device architectures could not be studied because of
19252 DOI: 10.1021/acsami.8b03956
ACS Appl. Mater. Interfaces 2018, 10, 19248−19257
ACS Applied Materials & Interfaces Research Article

the high resistance of the PVC-based ISM. The stark contrast EIS measurements (1.0 MΩ for GCE/MOF/K+-ISM-II
between Figure 3A,B suggests that the ion-to-electron signal electrodes).
transduction at all interfaces, as well as the phase transfer of The polarizability of the electrodes was further evaluated
ions through the solution/membrane interface in GCE/MOF/ from the slope (potential over time) at longer times (Figure 4)
K+-ISM-II device configuration, proceeded in a reversible as proposed by Bobacka (1999).23 In the absence of the MOF
manner and was kinetically fast. Thus, EIS measurements layer (GCE/ISM), a substantial drift of 2615 ± 14 μV/s, under
confirmed that the presence of the conductive MOF between polarizing conditions (−1 nA applied), was observed, whereas
the solid contact and the ISM strongly facilitates ion-to-electron the layer of conductive MOF in the GCE/MOF/K+-ISM-II
signal transduction. configuration significantly stabilized the electrode resulting in
Potential Stability of MOF-Containing Potentiometric the drift of 15 ± 1 μV/s. The potential drift (15 ± 1 μV/s)
Sensors under Polarizing Conditions. We employed estimated from constant-current chronopotentiometry under
constant-current chronopotentiometry to evaluate the ability polarizing conditions may also be related to the low-frequency
of a representative 2D MOF to provide a non-polarizable capacitance of a solid contact. Using the E/t curves to calculate
interface with high exchange current density and thus to the capacitance of the electrode (E/t = i/C) gave a substantial
improve signal stability of potentiometric devices. High signal capacitance for GCE/MOF/K+-ISM-II devices (60.6 ± 1.4 μF),
stability, with ideally no potential drift, is particularly important which was 150 times larger than that of GCE/K+-ISM-II (0.4 ±
in applications where the continuous monitoring of an analyte 0.1 μF). These results indicate that the MOF conductive layer
over an extended time period is required. However, larger contributes significantly toward the potential stability of tested
potential drifts (∼1 mV/h) can be tolerated during the short- devices under polarizing conditions and that the stability of
term measurements if the electrodes undergo frequent fabricated electrodes is on par with other ISEs based on
calibrations.8 To make this assessment, we compared the graphene (12.6 μV drift with 1 nA applied)75 or single-walled
potential stability under polarizing conditions of two layered CNTs (17.0 μV drift with 1 nA applied).25
device configurations previously examined by EIS: a control Potentiometric Ion Sensing Using Conductive MOFs
GCE/K+-ISM-II and GCE/MOF/K+-ISM-II. Each device was as Ion-to-Electron Transducers. To demonstrate the
first subjected to +1 nA for 60 s, after which the direction of the applicability of conductive MOFs as the ion-to-electron
current was reversed for the same time interval (−1 nA, 60 s), transducers in the development of solid-state ISEs in this
and the potential response was recorded. The potential versus proof-of-concept study, we fabricated and evaluated the
performance of potassium (model cation) and nitrate (model
time curves (E/t) for the analyzed GCE/K+-ISM-II and GCE/
anion) potentiometric sensors. Routine monitoring of blood
MOF/K+-ISM-II devices are shown in Figure 4. The potential
potassium is one of the examples where potassium-selective
jump that occurs during the reversal of applied current can be
electrodes can make a significant impact,76 whereas the
used directly to estimate the total resistance of the electrode
detection of NO3− has biological and environmental signifi-
(eq 1):
cance.77 We employed the electrode configuration in which the
R = E/I (1) MOF layer is drop-casted on the GCE and then covered with
an ISM to produce a potentiometric sensor.
In this equation, R represents the resistance of the electrode, The potentiometric responses of fabricated NO3−-ISM
E is potential of the electrode, and I is the current. The bulk electrodes comprising Ni3HHTP2 MOF sandwiched between
membrane resistance calculated from chronopotentiometric GCE and NO3−-ISM (GCE/MOF/NO3−-ISM) are depicted in
experiments was in close agreement with data extracted from Figure 5. Near-Nernstian responses (56.3 ± 0.5 mV decade−1)
were observed for all tested electrodes with the detection limits
(6.31 ± 0.01 × 10−7 M) and selectivity coefficient values closely
matching those reported for other SCEs equipped with ISM for
nitrate detection (Figures 5 and S11 and Table S2). Moreover,
tested electrodes demonstrated good short-term (11.1 ± 0.5
μA/h) and long-term potential stability (15 mV change in
recorded standard potential, E°, after 25 days) and exhibited
minimal (<1 mV) photosensitivity (Figures S14−S16). In
particular, the potential drift estimated from continuous
potentiometric measurements for 8 h was 11.1 ± 0.5 μA/h,
thus confirming excellent potential stability of resulting GCE/
MOF/ISM electrodes, whose magnitude is comparable to ISEs
based on 3DOM carbon (11.7 μA/h)36 or graphene (12.6 μA/
h).75 The crystallinity of M3HHTP2 MOFs (M = Cu, Ni and
Co) was retained after potentiometric analysis with MOFs used
as ion-to-electron transducers (Figure S17).
To demonstrate the general applicability of conductive
Figure 4. Chronopotentiograms obtained during the analysis of K+- MOFs as ion-to-electron transducers in potentiometric
ISM-II-based ISEs under polarizing conditions. (top) K+-ISM-II
detection, we fabricated K+-ISM-II electrodes containing
applied directly onto a GCE contact without the MOF as the
undelaying conductive layer; (bottom) GCE/Ni3HHTP2 MOF/K+- Ni3HHTP2 MOF sandwiched between GCE and K+-ISM-II
ISM-II electrode. The inset demonstrates a close-up of response (GCE/MOF/K+-ISM-II). The calibration curve for potassium
obtained for GCE/Ni3HHTP2 MOF/K+-ISM-II. Experimental con- ions (Figure 5B, green triangles) displayed a near-Nernstian
ditions: applied current +1 nA for 60 s followed by −1 nA for 60 s. slope of 54.1 ± 1.0 mV decade−1 and a micromolar LOD of
The background solution was 0.1 M KCl. 6.76 ± 0.03 × 10−6 M. A similar linear response range was
19253 DOI: 10.1021/acsami.8b03956
ACS Appl. Mater. Interfaces 2018, 10, 19248−19257
ACS Applied Materials & Interfaces Research Article

M),25 graphene (6.31 × 10−7 M),75 or CIM carbon (3.98 ×

10−6 M) as ion-to-electron transducers.18
The selectivity of potentiometric K+ sensors (GCE/MOF/
K -ISM-II and GCE/MOF/K+-ISM-I) was characterized using
+

an unbiased separate solution method as proposed by Bakker

(1997).79 Each device was first conditioned in a solution of less
discriminated interfering ions prior to the performed
potentiometric measurements. Table S1 and Figure S12 show
the selectivity values and corresponding experimental slopes
obtained for Na+, NH4+, and Ca2+ (log KPOTCa2+: −3.28 ±
0.02, Na+: −0.86 ± 0.10, and NH4+: −0.82 ± 0.05). Additional
potentiometric analysis of GCE/MOF/K+-ISM-I electrodes
revealed good match with theoretically expected selectivity
levels (log KPOTCa2+: −5.46 ± 0.09, Na+: −4.10 ± 0.01, and
NH4+: −2.12 ± 0.05; Table S3 and Figure S13).21 Together,
these findings indicate that 2D conductive MOFs can be used
to construct sensitive and selective potentiometric sensors with
good signal stability.
Assessing the Potential Stability of GCE/MOF/K+-ISM-
II and GCE/MOF/NO3−-ISM Devices Using the Aqueous
Layer Test. The formation of the undesirable water layer at the
membrane/electrode interface can cause substantial signal
instability (stability condition iv) and electrode drift, which
hinders the application of SCEs for long-term sample analysis.
Such a water layer can act as a small reservoir that can undergo
ion exchange with ions from the sample solution contacting an
ISM. In that instance, the transmembrane ion flux from the
water layer into the sample solution can diminish the
performance of the electrode in terms of detection limits
through the contamination of membrane/solution phase
boundary. Recent experimental reports demonstrated that
Figure 5. Potentiometric response of GCE/MOF/ISM devices to ISEs containing either 3DOM36 carbon or PEDOT/PSS
model anions and cations. In this electrode configuration, the
conductive polymer19 as ion-to-electron transducers possessed
Ni3HHTP2 MOF was drop-cast directly onto the GCE and then the
resulting conductive layer was covered with either (A) NO3−-ISM (red increased signal stability during potentiometric measurements
squares), (B) K+-ISM-I (blue diamonds), and K+-ISM-II (green because of the large interfacial capacitance and high hydro-
triangles) selective polymeric membrane. For the K+-ISM-I electrodes, phobicity of the transducer layer. We hypothesized that the
a near-Nernstian slope of 58.2 ± 1.0 mV decade−1 and a nanomolar hydrophobicity of 2D MOFs employed in this study would
limit of detection (LOD) of 5.01 ± 0.01 × 10−7 M was observed. K+- inhibit the formation of water layer at the membrane/electrode
ISM-II exhibited a near-Nernstian slope of 54.1 ± 1.0 mV decade−1 interface and thus minimize the effect of transmembrane ion
with LOD found at 6.76 ± 0.03 × 10−6 M, whereas 56.3 ± 0.5 mV fluxes.
decade−1 slope with LOD of 6.31 ± 0.01 × 10−7 M was obtained for To assess the degree of water layer formation at the
NO3−-ISM electrodes. electrode/membrane interface, we performed contact angle
measurements and carried out a potentiometric water layer test.
obtained for the three MOF analogues included in this study, Contact angle measurements were performed by dropping 1.0
demonstrating the general applicability of conductive MOFs in μL water aliquot onto the surface of MOF-coated GCE (MOF
the development of ISEs (Figure S10). Performance character- layer thickness60 μm; membrane thickness110 ± 10 μm).
istics of GCE/MOF/K+-ISM-II devices were similar to other The contact angle obtained for this electrode configuration was
potentiometric devices employing analogous K+-ISM-II ISMs.78 74 ± 0.4°, indicating hydrophobicity of the studied surface
Small differences between the reported detection limits of (Figure S18) in contrast to unmodified GCE with contact angle
devices with similar ISM and those reported here may arise estimated at approximately 50°.80 Even though the observed
from the presence of transmembrane ion fluxes (from the inner contact angle is lower than 104° and 132° reported for CIM
filling solution to the sample) that are readily observed in ISEs carbon 18 and PEDOT-C 14 (tetrakis(penta¯uorophenyl)-
based on the liquid contact. borate),19 respectively, we anticipated that the extent of water
To confirm that the response characteristics (slope, detection layer formation at the electrode/membrane interface could be
limits) of potassium sensors were primarily a function of the sufficiently diminished.
ISM, rather than the choice of the ion-to-electron transducer, A potentiometric water layer test was conducted by placing
we performed additional sensing experiments by using the GCE/MOF/NO3−-ISM electrode sequentially into 0.1 M
valinomycin as an ionophore within the ISM to generate K+- NH4NO3, 0.1 M CaCl2, and 0.1 M NH4NO3 solutions.
ISEs-I (Figure 5).8 We observed a near-Nernstian slope of 58.2 Immediately after the primary solution was replaced by 0.1 M
± 1.0 mV decade−1 and an LOD of 5.01 ± 0.01 × 10−7 M, CaCl2, a large change in the electromotive force was observed
which showed an improvement in the detection limit compared directly reflecting the selectivity of the electrodes (Figure 6A).
to potentiometric devices employing this ionophore in The same electrode, also in 0.1 M CaCl2, showed minimal
conjunction with using single-walled CNTs (5.00 × 10−6 potential drift, indicating that the NO3− ions at the membrane/
19254 DOI: 10.1021/acsami.8b03956
ACS Appl. Mater. Interfaces 2018, 10, 19248−19257
ACS Applied Materials & Interfaces Research Article

We anticipate that this study will be of value to the scientific

community interested in innovation at the interface of materials
and analytical chemistry. Because applications of conductive
polymers and porous carbons as ion-to-electron transducers in
ISEs have undergone tremendous progress to achieve current
performance levels, we believe that further advancements in the
design, synthesis, and device integration of conductive MOFs
can enable future improvements beyond the proof-of-concept
demonstration described in this study. We also expect that the
discoveries and the fundamental electrochemical character-
ization of conductive MOFs reported herein will serve as a
gateway to further fundamental and applied advances in the
Figure 6. (A) Potentiometric water layer test of GCE/MOF/NO3−- development of solid-state chemical sensors.
ISM electrodes. At t = 1 h, the primary analyte solution (0.1 M
NH4NO3) was exchanged to 0.1 M CaCl2, and after 3 h, the sample
was replaced by the initial solution for 8 h. (B) Water film tests of
■
*
ASSOCIATED CONTENT
S Supporting Information
solid-contact K+-ISM-II electrodes based on the GCE/MOF/K+-ISM- The Supporting Information is available free of charge on the
II configuration. At t = 1 h, the primary ion solution (0.1 M KCl) was ACS Publications website at DOI: 10.1021/acsami.8b03956.
exchanged to 0.1 M NaCl, and after 3 h, the sample was replaced by
the initial solution for 8 h. Experimental details; potentiometric data; and SEM,
EDX, X-ray photoelectron spectra, and CV results (PDF)
electrode interface (water layer) were not replaced to a
significant extent by Cl− ions. After placing the electrodes back
into the primary ion solution, the MOF-based devices exhibited
■ AUTHOR INFORMATION
Corresponding Author
very small potential drift, demonstrating that the extent of *E-mail: [email protected].
water layer formation was significantly reduced for the ISEs ORCID
utilizing MOF as a conductive layer in between the polymeric Katherine A. Mirica: 0000-0002-1779-7568
membrane and solid contact. Similarly, potentiometric water
Author Contributions
test of GCE/MOF/K+-ISM-II electrodes (Figure 6B) revealed
The manuscript was written through contributions of all
only very small potential drift upon changing from primary ion
authors. All authors have given approval to the final version of
solution (0.1 M KCl) to interfering ion solution (0.1 M NaCl)
the manuscript.
and back to targeted analyte, indicating good ability of
Ni3HHTP2 MOFs to minimize the extent of water layer Notes
The authors declare no competing financial interest.

■
formation at the electrode/ISM interface.

■ CONCLUSIONS
This paper describes the first experimental demonstration of
ACKNOWLEDGMENTS
The authors acknowledge support from startup funds provided
efficient ion-to-electron transduction enabled by conductive by Dartmouth College, from Walter and Constance Burke
MOFs. This material property enables the implementation of Research Initiation Award, and from the Army Research Office
conductive MOFs in potentiometric detection. The generality Young Investigator Program Grant no. W911NF-17-1-0398.
of this approach is established by integrating several analogues
of 2D conductive MOFs into solid-state potentiometric devices
(coated with ISM) to achieve potentiometric determination of
■ REFERENCES
(1) Bakker, E.; Bühlmann, P.; Pretsch, E. Carrier-Based Ion-Selective
K+ and NO3− ions in aqueous solutions. The sensors exhibited Electrodes and Bulk Optodes. 1. General Characteristics. Chem. Rev.
excellent performance characteristics including near-Nernstian 1997, 97, 3083−3132.
behavior (54.1−58.2 mV/s), wide dynamic range (mM−nM), (2) Bühlmann, P.; Pretsch, E.; Bakker, E. Carrier-Based Ion-Selective
and good signal stability (11.1 ± 0.5 μA/h). Electrodes and Bulk Optodes. 2. Ionophores for Potentiometric and
Optical Sensors. Chem. Rev. 1998, 98, 1593−1688.
Conductive MOFs possess at least five advantageous (3) Bobacka, J.; Ivaska, A.; Lewenstam, A. Potentiometric Ion
characteristics as ion-to-electron transducers with promising Sensors. Chem. Rev. 2008, 108, 329−351.
utility in potentiometric detection: (i) large bulk capacitance; (4) Ding, J.; Qin, W. Current-Driven Ion Fluxes of Polymeric
(ii) low contact resistance at the MOF/ISM interface; (iii) Membrane Ion-Selective Electrode for Potentiometric Biosensing. J.
compatibility with the detection of both cations and anions; Am. Chem. Soc. 2009, 131, 14640−14641.
(iv) high potential stability enabled by the suppression of the (5) Johnson, R. D.; Bachas, L. G. Ionophore-Based Ion-Selective
formation of water layer at MOF interface; and (v) high Potentiometric and Optical Sensors. Anal. Bioanal. Chem. 2003, 376,
structural modularity and ease of synthesis of these materials 328−341.
accessible through aqueous self-assembly. However, the (6) Bakker, E.; Pretsch, E. The New Wave of Ion-Selective
magnitude of capacitance that dictates the stability of Electrodes. Anal. Chem. 2002, 74, 420A−426A.
(7) Malon, A.; Vigassy, T.; Bakker, E.; Pretsch, E. Potentiometry at
potentiometric sensors is currently limited by the intrinsic Trace Levels in Confined Samples: Ion-Selective Electrodes with
porosity and conductivity of the MOFs examined in this study. Subfemtomole Detection Limits. J. Am. Chem. Soc. 2006, 128, 8154−
Maximizing potential utility of MOFs as functional components 8155.
in potentiometric sensors requires further research and (8) Hu, J.; Stein, A.; Bühlmann, P. Rational Design of All-Solid-State
development centered on maximizing conductivity, porosity, Ion-Selective Electrodes and Reference Electrodes. TrAC, Trends Anal.
and hydrophobicity of these materials. Chem. 2016, 76, 102−114.

19255 DOI: 10.1021/acsami.8b03956

ACS Appl. Mater. Interfaces 2018, 10, 19248−19257
ACS Applied Materials & Interfaces Research Article

(9) Hu, J.; Stein, A.; Bühlmann, P. A Disposable Planar Paper-Based (29) He, N.; Papp, S.; Lindfors, T.; Höfler, L.; Latonen, R.-M.;
Potentiometric Ion-Sensing Platform. Angew. Chem., Int. Ed. 2016, 55, Gyurcsányi, R. E. Pre-Polarized Hydrophobic Conducting Polymer
7544−7547. Solid-Contact Ion-Selective Electrodes with Improved Potential
(10) Lan, W.-J.; Zou, X. U.; Hamedi, M. M.; Hu, J.; Parolo, C.; Reproducibility. Anal. Chem. 2017, 89, 2598−2605.
Maxwell, E. J.; Bühlmann, P.; Whitesides, G. M. Paper-Based (30) Bobacka, J. Conducting Polymer-Based Solid-State Ion-Selective
Potentiometric Ion Sensing. Anal. Chem. 2014, 86, 9548−9553. Electrodes. Electroanalysis 2006, 18, 7−18.
(11) Chen, L. D.; Mandal, D.; Pozzi, G.; Gladysz, J. A.; Bühlmann, P. (31) Abad, J.; Espinosa, N.; Ferrer, P.; García-Valverde, R.; Miguel,
Potentiometric Sensors Based on Fluorous Membranes Doped with C.; Padilla, J.; Alcolea, A.; Castro, G. R.; Colchero, J.; Urbina, A.
Highly Selective Ionophores for Carbonate. J. Am. Chem. Soc. 2011, Molecular Structure of poly(3-Alkyl-Thiophenes) Investigated by
133, 20869−20877. Calorimetry and Grazing Incidence X-Ray Scattering. Sol. Energy
(12) Hu, J.; Ho, K. T.; Zou, X. U.; Smyrl, W. H.; Stein, A.; Mater. Sol. Cells 2012, 97, 109−118.
Bühlmann, P. All-Solid-State Reference Electrodes Based on Colloid- (32) Chen, S.-A.; Ni, J.-M.; Hua, M.-Y. Thermal Undoping Behavior
Imprinted Mesoporous Carbon and Their Application in Disposable of FeCl3-Doped poly(3-Octylthiophene). J. Polym. Res. 1997, 4, 261−
Paper-Based Potentiometric Sensing Devices. Anal. Chem. 2015, 87, 265.
2981−2987. (33) Lindfors, T. Light Sensitivity and Potential Stability of
(13) Cattrall, R. W.; Freiser, H. Coated Wire Ion-Selective Electrically Conducting Polymers Commonly Used in Solid Contact
Electrodes. Anal. Chem. 1971, 43, 1905−1906. Ion-Selective Electrodes. J. Solid State Electrochem. 2009, 13, 77−89.
(14) Nikolskii, B. P.; Materova, E. A. Solid Contact in Membrane (34) Csahók, E.; Vieil, E.; Inzelt, G. In Situ Dc Conductivity Study of
Ion-Selective Electrodes. In Ion-Selective Electrode Reviews; Elsevier, the Redox Transformations and Relaxation of Polyaniline Films. J.
1985; Vol. 7, pp 3−39. Electroanal. Chem. 2000, 482, 168−177.
(15) Veder, J.-P.; De Marco, R.; Clarke, G.; Chester, R.; Nelson, A.; (35) Crespo, G. A.; Macho, S.; Bobacka, J.; Rius, F. X. Transduction
Prince, K.; Pretsch, E.; Bakker, E. Elimination of Undesirable Water Mechanism of Carbon Nanotubes in Solid-Contact Ion-Selective
Layers in Solid-Contact Polymeric Ion-Selective Electrodes. Anal. Electrodes. Anal. Chem. 2009, 81, 676−681.
Chem. 2008, 80, 6731−6740. (36) Lai, C.-Z.; Fierke, M. A.; Stein, A.; Bühlmann, P. Ion-Selective
(16) Fibbioli, M.; Morf, W. E.; Badertscher, M.; de Rooij, N. F.; Electrodes with Three-Dimensionally Ordered Macroporous Carbon
Pretsch, E. Potential Drifts of Solid-Contacted Ion-Selective Electrodes as the Solid Contact. Anal. Chem. 2007, 79, 4621−4626.
Due to Zero-Current Ion Fluxes Through the Sensor Membrane. (37) Bandaru, P. R. Electrical Properties and Applications of Carbon
Electroanalysis 2000, 12, 1286−1292. Nanotube Structures. J. Nanosci. Nanotechnol. 2007, 7, 1239−1267.
(17) De Marco, R.; Veder, J.-P.; Clarke, G.; Nelson, A.; Prince, K.; (38) Novoselov, K. S.; Fal’ko, V. I.; Colombo, L.; Gellert, P. R.;
Pretsch, E.; Bakker, E. Evidence of a Water Layer in Solid-Contact Schwab, M. G.; Kim, K. A. Roadmap for Graphene. Nature 2012, 490,
Polymeric Ion Sensors. Phys. Chem. Chem. Phys. 2008, 10, 73−76. 192−200.
(18) Hu, J.; Zou, X. U.; Stein, A.; Bühlmann, P. Ion-Selective (39) Prasek, J.; Drbohlavova, J.; Chomoucka, J.; Hubalek, J.; Jasek,
Electrodes with Colloid-Imprinted Mesoporous Carbon as Solid
O.; Adam, V.; Kizek, R. Methods for Carbon Nanotubes Synthesis
Contact. Anal. Chem. 2014, 86, 7111−7118.
review. J. Mater. Chem. 2011, 21, 15872.
(19) Guzinski, M.; Jarvis, J. M.; D’Orazio, P.; Izadyar, A.; Pendley, B.
(40) Charlier, J.-C. Defects in Carbon Nanotubes. Acc. Chem. Res.
D.; Lindner, E. Solid-Contact pH Sensor without CO 2 Interference
2002, 35, 1063−1069.
with a Superhydrophobic PEDOT-C14 as Solid Contact: The Ultimate
(41) Athavale, R.; Dinkel, C.; Wehrli, B.; Bakker, E.; Crespo, G. A.;
“Water Layer” Test. Anal. Chem. 2017, 89, 8468−8475.
Brand, A. Robust Solid-Contact Ion Selective Electrodes for High-
(20) Crespo, G. A.; Cuartero, M.; Bakker, E. Thin Layer Ionophore-
Resolution In Situ Measurements in Fresh Water Systems. Environ. Sci.
Based Membrane for Multianalyte Ion Activity Detection. Anal. Chem.
Technol. Lett. 2017, 4, 286−291.
2015, 87, 7729−7737.
(42) Hmadeh, M.; Lu, Z.; Liu, Z.; Gándara, F.; Furukawa, H.; Wan,
(21) Pławińska, Ż .; Michalska, A.; Maksymiuk, K. Optimization of
Capacitance of Conducting Polymer Solid Contact in Ion-Selective S.; Augustyn, V.; Chang, R.; Liao, L.; Zhou, F.; Perre, E.; Ozolins, V.;
Electrodes. Electrochim. Acta 2016, 187, 397−405. Suenaga, K.; Duan, X.; Dunn, B.; Yamamto, Y.; Terasaki, O.; Yaghi, O.
(22) Fibbioli, M.; Enger, O.; Diederich, F.; Pretsch, E.; M. New Porous Crystals of Extended Metal-Catecholates. Chem.
Bandyopadhyay, K.; Liu, S.-G.; Echegoyen, L.; Bühlmann, P. Redox- Mater. 2012, 24, 3511−3513.
Active Self-Assembled Monolayers as Novel Solid Contacts for Ion- (43) Sun, L.; Campbell, M. G.; Dincă, M. Electrically Conductive
Selective Electrodes. Chem. Commun. 2000, 5, 339−340. Porous Metal−Organic Frameworks. Angew. Chem., Int. Ed. 2016, 55,
(23) Bobacka, J. Potential Stability of All-Solid-State Ion-Selective 3566−3579.
Electrodes Using Conducting Polymers as Ion-to-Electron Trans- (44) Clough, A. J.; Skelton, J. M.; Downes, C. A.; de la Rosa, A. A.;
ducers. Anal. Chem. 1999, 71, 4932−4937. Yoo, J. W.; Walsh, A.; Melot, B. C.; Marinescu, S. C. Metallic
(24) Migdalski, J.; Blaz, T.; Lewenstam, A. Conducting Polymer- Conductivity in a Two-Dimensional Cobalt Dithiolene Metal−
Based Ion-Selective Electrodes. Anal. Chim. Acta 1996, 322, 141−149. Organic Framework. J. Am. Chem. Soc. 2017, 139, 10863−10867.
(25) Crespo, G. A.; Macho, S.; Rius, F. X. Ion-Selective Electrodes (45) Dou, J.-H.; Sun, L.; Ge, Y.; Li, W.; Hendon, C. H.; Li, J.; Gul, S.;
Using Carbon Nanotubes as Ion-to-Electron Transducers. Anal. Chem. Yano, J.; Stach, E. A.; Dincă, M. Signature of Metallic Behavior in the
2008, 80, 1316−1322. Metal−Organic Frameworks M3(Hexaiminobenzene)2 (M = Ni, Cu).
(26) Bobacka, J.; McCarrick, M.; Lewenstam, A.; Ivaska, A. All Solid- J. Am. Chem. Soc. 2017, 139, 13608−13611.
State Poly(vinyl Chloride) Membrane Ion-Selective Electrodes with (46) Huang, X.; Sheng, P.; Tu, Z.; Zhang, F.; Wang, J.; Geng, H.;
poly(3-Octylthiophene) Solid Internal Contact. Analyst 1994, 119, Zou, Y.; Di, C.-A.; Yi, Y.; Sun, Y.; Zhu, D. Two-Dimensional π−d
1985. Conjugated Coordination Polymer with Extremely High Electrical
(27) Michalska, A. Optimizing the Analytical Performance and Conductivity and Ambipolar Transport Behavior. Nat. Commun. 2015,
Construction of Ion-Selective Electrodes with Conducting Polymer- 6, 7408.
Based Ion-to-Electron Transducers. Anal. Bioanal. Chem. 2005, 384, (47) Campbell, M. G.; Sheberla, D.; Liu, S. F.; Swager, T. M.; Dincă,
391−406. M. Cu3(Hexaiminotriphenylene)2: An Electrically Conductive 2D
(28) Guzinski, M.; Jarvis, J. M.; Perez, F.; Pendley, B. D.; Lindner, E.; Metal−Organic Frameworks for Chemiresistive Sensing. Angew.
De Marco, R.; Crespo, G. A.; Acres, R. G.; Walker, R.; Bishop, J. Chem., Int. Ed. 2015, 54, 4349−4352.
PEDOT(PSS) as Solid Contact for Ion-Selective Electrodes: The (48) Campbell, M. G.; Liu, S. F.; Swager, T. M.; Dincă, M.
Influence of the PEDOT(PSS) Film Thickness on the Equilibration Chemiresistive Sensor Arrays from Conductive 2D Metal−Organic
Times. Anal. Chem. 2017, 89, 3508−3516. Frameworks. J. Am. Chem. Soc. 2015, 137, 13780−13783.

19256 DOI: 10.1021/acsami.8b03956

ACS Appl. Mater. Interfaces 2018, 10, 19248−19257
ACS Applied Materials & Interfaces Research Article

(49) Smith, M. K.; Mirica, K. A. Self-Organized Frameworks on (66) Jiang, H.-L.; Liu, B.; Lan, Y.-Q.; Kuratani, K.; Akita, T.;
Textiles (SOFT): Conductive Fabrics for Simultaneous Sensing, Shioyama, H.; Zong, F.; Xu, Q. From Metal−Organic Framework to
Capture, and Filtration of Gases. J. Am. Chem. Soc. 2017, 139, 16759− Nanoporous Carbon: Toward a Very High Surface Area and
16767. Hydrogen Uptake. J. Am. Chem. Soc. 2011, 133, 11854−11857.
(50) Smith, M. K.; Jensen, K. E.; Pivak, P. A.; Mirica, K. A. Direct (67) Campbell, M.; Dincă, M. Metal−Organic Frameworks as Active
Self-Assembly of Conductive Nanorods of Metal−Organic Frame- Materials in Electronic Sensor Devices. Sensors 2017, 17, 1108.
works into Chemiresistive Devices on Shrinkable Polymer Films. (68) Hendon, C. H.; Rieth, A. J.; Korzyński, M. D.; Dincă, M. Grand
Chem. Mater. 2016, 28, 5264−5268. Challenges and Future Opportunities for Metal−Organic Frameworks.
(51) Ko, M.; Aykanat, A.; Smith, M. K.; Mirica, K. A. Drawing ACS Cent. Sci. 2017, 3, 554−563.
Sensors with Ball-Milled Blends of Metal−Organic Frameworks and (69) Miner, E. M.; Dincă, M. Metal−Organic Frameworks: Evolved
Graphite. Sensors 2017, 17, 2192. Oxygen Evolution Catalysts. Nat. Energy 2016, 1, 16186.
(52) Yao, M.-S.; Lv, X.-J.; Fu, Z.-H.; Li, W.-H.; Deng, W.-H.; Wu, G.- (70) Fierke, M. A.; Lai, C.-Z.; Bühlmann, P.; Stein, A. Effects of
D.; Xu, G. Layer-by-Layer Assembled Conductive Metal−Organic Architecture and Surface Chemistry of Three-Dimensionally Ordered
Framework Nanofilms for Room-Temperature Chemiresistive Sens- Macroporous Carbon Solid Contacts on Performance of Ion-Selective
ing. Angew. Chem. 2017, 56, 16510−16514. Electrodes. Anal. Chem. 2010, 82, 680−688.
(53) Li, W.-H.; Ding, K.; Tian, H.-R.; Yao, M.-S.; Nath, B.; Deng, W.- (71) Ji, H.; Zhao, X.; Qiao, Z.; Jung, J.; Zhu, Y.; Lu, Y.; Zhang, L. L.;
H.; Wang, Y.; Xu, G. Conductive Metal−Organic Framework MacDonald, A. H.; Ruoff, R. S. Capacitance of Carbon-Based Electrical
Nanowire Array Electrodes for High-Performance Solid-State Super- Double-Layer Capacitors. Nat. Commun. 2014, 5, 3317.
capacitors. Adv. Funct. Mater. 2017, 27, 1702067. (72) Naidek, K. P.; Zuconelli, C. R.; Cruz, O. M.; Ribeiro, R.;
(54) Sheberla, D.; Bachman, J. C.; Elias, J. S.; Sun, C.-J.; Shao-Horn, Winnischofer, S. M. B.; Winnischofer, H. Characterization of
Y.; Dincă, M. Conductive MOF Electrodes for Stable Supercapacitors 2,3,6,7,10,11-Hexahydroxytriphenylene and Its Effects on Cell Viability
with High Areal Capacitance. Nat. Mater. 2016, 16, 220−224. in Human Cancer Cell Lines. Biochem. Cell Biol. 2016, 94, 205−211.
(55) Feng, D.; Lei, T.; Lukatskaya, M. R.; Park, J.; Huang, Z.; Lee, (73) Punckt, C.; Pope, M. A.; Aksay, I. A. On the Electrochemical
M.; Shaw, L.; Chen, S.; Yakovenko, A. A.; Kulkarni, A.; Xiao, J.; Response of Porous Functionalized Graphene Electrodes. J. Phys.
Fredrickson, K.; Tok, J. B.; Zou, X.; Cui, Y.; Bao, Z. Robust and Chem. C 2013, 117, 19253.
Conductive Two-Dimensional Metal−organic Frameworks with (74) Bryan, A. M.; Santino, L. M.; Lu, Y.; Acharya, S.; D’Arcy, J. M.
Exceptionally High Volumetric and Areal Capacitance. Nat. Energy Conducting Polymers for Pseudocapacitive Energy Storage. Chem.
2018, 3, 30−36. Mater. 2016, 28, 5989−5998.
(56) Downes, C. A.; Clough, A. J.; Chen, K.; Yoo, J. W.; Marinescu, (75) Ping, J.; Wang, Y.; Wu, J.; Ying, Y. Development of an All-Solid-
S. C. Evaluation of the H2 Evolving Activity of Benzenehexathiolate State Potassium Ion-Selective Electrode Using Graphene as the Solid-
Coordination Frameworks and the Effect of Film Thickness on H2 Contact Transducer. Electrochem. Commun. 2011, 13, 1529−1532.
Production. ACS Appl. Mater. Interfaces 2018, 10, 1719−1727. (76) Lee, J.; Kim, H.-J.; Kim, J. Polydiacetylene Liposome Arrays for
(57) Clough, A. J.; Yoo, J. W.; Mecklenburg, M. H.; Marinescu, S. C. Selective Potassium Detection. J. Am. Chem. Soc. 2008, 130, 5010−
Two-Dimensional Metal−Organic Surfaces for Efficient Hydrogen 5011.
Evolution from Water. J. Am. Chem. Soc. 2015, 137, 118−121. (77) Lundberg, J. O.; Weitzberg, E.; Cole, J. A.; Benjamin, N.
(58) Dong, R.; Pfeffermann, M.; Liang, H.; Zheng, Z.; Zhu, X.; Opinion: Nitrate, Bacteria and Human Health. Nat. Rev. Microbiol.
Zhang, J.; Feng, X. Large-Area, Free-Standing, Two-Dimensional 2004, 2, 593−602.
Supramolecular Polymer Single-Layer Sheets for Highly Efficient (78) Tamura, H.; Kumami, K.; Kimura, K.; Shono, T. Simultaneous
Electrocatalytic Hydrogen Evolution. Angew. Chem., Int. Ed. 2015, 54, Determination of Sodium and Potassium in Human Urine or Serum
12058−12063. Using Coated-Wire Ion-Selective Electrodes Based on Bis(crown
(59) Miner, E. M.; Fukushima, T.; Sheberla, D.; Sun, L.; Ether)s. Mikrochim. Acta 1983, 80, 287−296.
Surendranath, Y.; Dincă, M. Electrochemical Oxygen Reduction (79) Bakker, E. Determination of Unbiased Selectivity Coefficients of
Catalysed by Ni3(hexaiminotriphenylene)2. Nat. Commun. 2016, 7, Neutral Carrier-Based Cation-Selective Electrodes. Anal. Chem. 1997,
10942. 69, 1061−1069.
(60) Huang, X.; Yao, H.; Cui, Y.; Hao, W.; Zhu, J.; Xu, W.; Zhu, D. (80) Taner, B.; Deveci, P.; Ü stündaǧ, Z.; Keskin, S.; Ö zcan, E.; Solak,
Conductive Copper Benzenehexathiol Coordination Polymer as a A. O. Modification of Glassy Carbon Electrode by the Electrochemical
Hydrogen Evolution Catalyst. ACS Appl. Mater. Interfaces 2017, 9, Oxidation of 3-aminophenylcalix[4]pyrrole in Nonaqueous Media.
40752−40759. Surf. Interface Anal. 2012, 44, 185−191.
(61) Sun, X.; Wu, K.-H.; Sakamoto, R.; Kusamoto, T.; Maeda, H.; Ni,
X.; Jiang, W.; Liu, F.; Sasaki, S.; Masunaga, H.; Nishihara, H.
Bis(aminothiolato)nickel Nanosheet as a Redox Switch for Con-
ductivity and an Electrocatalyst for the Hydrogen Evolution Reaction.
Chem. Sci. 2017, 8, 8078−8085.
(62) Jia, H.; Yao, Y.; Zhao, J.; Gao, Y.; Luo, Z.; Du, P. A Novel Two-
Dimensional Nickel Phthalocyanine-Based Metal−Organic Framework
for Highly Efficient Water Oxidation Catalysis. J. Mater. Chem. A 2018,
6, 1188−1195.
(63) Mendecki, L.; Ko, M.; Zhang, X.; Meng, Z.; Mirica, K. A. Porous
Scaffolds for Electrochemically Controlled Reversible Capture and
Release of Ethylene. J. Am. Chem. Soc. 2017, 139, 17229−17232.
(64) Sheberla, D.; Sun, L.; Blood-Forsythe, M. A.; Er, S.; Wade, C.
R.; Brozek, C. K.; Aspuru-Guzik, A.; Dincă, M. High Electrical
Conductivity in Ni3(2,3,6,7,10,11-Hexaiminotriphenylene)2, a Semi-
conducting Metal−Organic Graphene Analogue. J. Am. Chem. Soc.
2014, 136, 8859−8862.
(65) To, J. W. F.; Chen, Z.; Yao, H.; He, J.; Kim, K.; Chou, H.-H.;
Pan, L.; Wilcox, J.; Cui, Y.; Bao, Z. Ultrahigh Surface Area Three-
Dimensional Porous Graphitic Carbon from Conjugated Polymeric
Molecular Framework. ACS Cent. Sci. 2015, 1, 68−76.

19257 DOI: 10.1021/acsami.8b03956

ACS Appl. Mater. Interfaces 2018, 10, 19248−19257