Article pubs.acs.org/JPCB
Improved Ion-Selective Detection Method Using Nanopipette with Poly(vinyl chloride)-Based Membrane Eun Ji Kang,† Tomohide Takami,† Xiao Long Deng, Jong Wan Son, Tomoji Kawai, and Bae Ho Park* Division of Quantum Phases and Devices, Department of Physics, Konkuk University, Seoul 143-701, Republic of Korea
ABSTRACT: Ion-selective electrodes (ISEs) are widely used to detect targeted ions in solution selectively. Application of an ISE to a small area detection system with a nanopipette requires a special measurement method in order to avoid the enhanced background signal problem caused by a cation-rich layer near the charged inner surface of the nanopipette and the selectivity change problem due to relatively fast saturation of the ISE inside the nanopipette. We developed a novel ion-selective detection system using a nanopipette that measures an alternating current (AC) signal mediated by saturated ionophores in a poly(vinyl chloride) (PVC) membrane located at the conical shank of the nanopipette to solve the above problems. Small but reliable K+ and Na+ ionic current passing through a PVC membrane containing saturated bis(benzo-15-crown-5) and bis(12-crown-4) ionophore, respectively, could be selectively detected using the AC signal measurement system equipped with a lock-in amplifier.
1. INTRODUCTION Ion-selective electrodes (ISEs) have been widely used to detect selectively cations such as sodium, potassium, and calcium since it was developed by Štefanac and Simon almost a half century ago.1−3 It has been applied to many fields, including pollution monitoring, agriculture, food processing, and cosmetics.2,3 On the other hand, local concentration of ions can be measured using scanning ion conductance microscopy (SICM)4 with a nanopipette whose position is controlled on the nanoscale.5 However, SICM is unable to map the distribution of a specific ion because of its observation of total ionic current. Our purpose is to realize selective detection of a specific ionic current on the nanoscale, which can be used for the mapping with a specific ion above living cells to elucidate the ion channel dynamics.6 Such a selective mapping technique can provide different information from that obtained by fluorescent microscopy.7 Nano- or micropipettes have been widely used for sensors in the biological field. ZnO nanorods on the tip of a borosilicate glass capillary could be used as an intracellular sensor for pH measurement8 and a Ca2+ selective sensor,9 although the insertion of the nanorods might damage cells. Reid and Zhao observed specific ion flux using Na+ ionophore at the top of the micropipet.10 The diameter of the micropipette was 3−4 μm, and smaller tips had higher resistance, which made them more susceptible to electronic noise. Recently, the nanopipette © 2014 American Chemical Society
functionalized with probe materials inside it has been used to selectively detect proteins, calcium ions, and copper ions.11 Conventional ISEs cannot be applied to small area detection because the typical opening diameter of ISEs is on the order of several millimeters. Amemiya and Bard succeeded in fabricating a valinomycin-based potassium ion-selective membrane in a micropipette by obtaining steady current only when the concentration of the ionophore was much higher than the cation concentration.12 The authors tried to apply this method to nanopipette with size around 100 nm and found the problem of signal detection due to the negative surface charges on the inner wall of the nanopipette. In contrast, we developed a method to detect ionic current selectively using a nanopipette with a membrane filter containing ionophores by capturing a specific ion at an ionophore, as shown in Figure 1,13−15 and we succeeded in observing local potassium ion concentration in living HeLa cells.16 However, the nanopipette with filter does not work when more than three kinds of cations exist in the sample solution because all cations other than the specific cation selected by the filter can penetrate the filter. Moreover, the nanopipette shows change in selectivity because of relatively Received: March 19, 2014 Revised: April 25, 2014 Published: April 25, 2014 5130
dx.doi.org/10.1021/jp502715q | J. Phys. Chem. B 2014, 118, 5130−5134
The Journal of Physical Chemistry B
Article
from which the inner diameter of the fabricated nanopipette was confirmed. The plasticized poly(vinyl chloride) (PVC) membrane was prepared using the process described in previous studies.13,14 Briefly, the membrane was prepared from the solution that contains ionophore [bis(12-crown-4) (Dojindo CAS 80403-594) or bis(benzo-15-crown-5) (Dojindo CAS 69271-98-3)], PVC (Aldrich, CAS 9002-86-2), plasticizer [o-nitrophenyloctylether (Aldrich, CAS 3244-41-5)], and anion repeller (tetraphenylborate) whose weight ratio was 5:32:62:1. These compounds were dissolved in 1 mL of tetrahydrofuran (Aldrich, CAS 3244-42-1) to prepare the membrane solution. We prepared nanopipettes with PVC membranes, which are located (i) in the middle or (ii) at the conical shank of the nanopipettes, by the following two methods. (i) The deionized (DI) water (about 0.2 μL) was injected into the nanopipette, and then the prepared membrane solution (about 0.4 μL) was filled onto the DI water. The nanopipette was dried in a box with 98% humidity for 2 to 3 h. Finally, 4 × 10−3 M electrolyte (KCl or NaCl aqueous solution) was filled into the nanopipette which was kept in the high-humidity box for 2 days. (ii) The prepared membrane solution (about 0.6 μL) was filled into the nanopipette which was dried in the box with 98% humidity. During the drying process, the top of the nanopipette was immersed in DI water to prevent the membrane from being dried completely. Finally, 4 × 10−3 M electrolyte (KCl or NaCl aqueous solution) was filled into the nanopipette which was kept in the highhumidity box for 2 days. Figure 2c shows the schematic diagram of our ionic current detection system used in this work. The signals were recorded by a digital multimeter (Fluke 287) using a detection system that is the same as that used in previous work.15 The 0.1 mm diameter Ag electrode inserted into the ion-selective nanopipette was used to obtain the ion current signal. The obtained current was converted to voltage by current−voltage (I−V) converter. Both the 66 mV AC voltage from the function generator in the lock-in amplifier (Stanford SR 830) and the 100 mV direct current (DC) voltage from UM-1 battery in a
Figure 1. Schematic cross section of the nanopipette in which a poly(vinyl chloride) membrane is formed. The membrane shown in red contains a bis(benzo-15-crown-5) ionophore which preferably captures a K+ ion.
fast saturation of the ionophores in the membrane filter located near the middle of the nanopipette. In this study, we have developed a novel method to selectively detect a small alternating current (AC) signal of a specific ion passing through a membrane located at the conical shank of a nanopipette. The developed ion-selective nanopipette shows very reproducible selectivity for the specific ion because it exploits saturated ionophores instead of nonsaturated ionphores used in the nanopipette with a filter membrane.13−15 The stability of the measured ionic current is much enhanced because of the membrane position at the conical shank, which may reduce the ionic current oscillation caused by the electric double layer.
2. EXPERIMENTAL DETAILS Among all the pipettes with inner diameters of 50 to 1000 nm, the pipette with an inner diameter of 120 nm was that with the smallest inner diameter that showed detectable and stable ionic current values. Therefore, pipettes with an inner diameter of 120 nm were used in this study because we were interested in detection of a specific ionic current on the nanoscale. They were fabricated from borosilicate glass capillaries with filaments (GD-1, Narishige) using a laser-based puller (Model P-2000, Sutter Instrument).17 Scanning electron microscope (SEM) images of the fabricated nanopipette are shown in Figure 2a,b
Figure 2. Scanning electron microscope images of the nanopipette used in this study: (a) top view and (b) side view. (c) Schematic diagram of the signal detection system. (d) Schematic illustration of our experimental procedure; adding one droplet (40 μL of 0.1 M KCl or NaCl) near the nanopipette containing the ion-selective membrane. The tip of the nanopipette was immersed in the glass dish with 10 mL of deionized water. 5131
dx.doi.org/10.1021/jp502715q | J. Phys. Chem. B 2014, 118, 5130−5134
The Journal of Physical Chemistry B
Article
Figure 3. Ion-selective detection of potassium and sodium ions with a membrane located at the middle in the nanopipette. (a) Schematic illustration of the ion-selective nanopipette in which the membrane is located at the middle. The ion currents measured with the ion-selective nanopipettes in which the membrane contains the ionophores of (b) bis(benzo-15-crown-5) and (c) bis(12-crown-4).
Figure 4. Ion-selective detection of potassium and sodium ions with a membrane located at the conical shank in the nanopipette. (a) Schematic illustration of the ion selective nanopipette in which the membrane is located at the conical shank. The ion currents measured with the ion-selective nanopipettes in which the membrane contains the ionophores of (b) bis(benzo-15-crown-5) and (c) bis(12-crown-4).
ionophores of bis(12-crown-4). The selective detection of a specific ionic current using a membrane containing saturated ionophores can be explained by an ion-exchange process between solutions and a membrane.19,20 Saturated ionophores contribute to the ion-exchange process, i.e., cation (aq) ⇆ cation (org), at both sample−membrane and membrane−inner solution interfaces. The background signals of Figure 3b,c are not so stable because of the motion of ions from the sample dish to the membrane through the narrow conical shank of the nanopipette, in which the negatively charged inner wall produces an electric double layer, causing the fluctuation in ionic current.18 The background noise level depends on the condition of the inner wall of the nanopipette, especially at the narrow conical shank, which is a reason why the obtained data show fluctuating and high noise levels in the case of the geometry shown in Figure 3a. Next we demonstrate how the ion-selective membrane works when the membrane is located at the conical shank of the nanopipette (Figure 4a). Panels b and c of Figure 4 demonstrate the selective detection of potassium and sodium ionic currents through membranes containing saturated ionophores of bis(benzo-15-crown-5) and bis(12-crown-4), respectively. The background noise levels decrease significantly compared with those in Figure 3b,c. The improvement is attributed to the position of the membrane, being closer to the sample solution, which excludes the fluctuation effect from the
homemade voltage controller were applied to the 1 mm diameter Ag/AgCl counter electrode in the glass dish (50 mm diameter) filled with 10 mL of DI water. We conducted the measurement for ion-selective detection by adding one droplet (40 μL) of each sample solution (0.1 M KCl or NaCl) from a different micropipette, as shown in Figure 2d.
3. RESULTS AND DISCUSSION Figures 3 and 4 show the selective detection of ions using nanopipettes with saturated ionophores in the membranes prepared by two different methods described in Experimental Details. First we demonstrate how the ion-selective membrane works when the membrane is located in the middle of the nanopipette (Figure 3a). This configuration is the same as that in our previous studies,13−15 except that a DI water inner liquid and nonsaturated ionophores in the membranes were used in the previous studies. In this configuration, with electrolyte inner liquid and saturated ionophores, the membrane works in way that is opposite to that of the previous studies where a membrane works as a filter for a specific ion.13−15 Figure 3b demonstrates the selective detection of potassium ionic current passing through a membrane containing the saturated ionophores of bis(benzo-15-crown-5), whereas potassium ions were selectively captured by the nonsaturated ionophores of bis(benzo-15-crown-5) inside a filter membrane in the previous studies.13−15 Figure 3c shows the selective detection of sodium ionic current using a membrane containing the saturated 5132
dx.doi.org/10.1021/jp502715q | J. Phys. Chem. B 2014, 118, 5130−5134
The Journal of Physical Chemistry B
Article
Figure 5. Durability test of ion-selective detection with the membrane at the conical shank. (a) Case of bis(benzo-15-crown-5) ionophores detecting K+ ions. The top of the ion-selective nanopipette was kept immersed in 4 × 10−3 M electrolyte (KCl) for 2, 7, and 10 days. (b) Case of bis(benzo15-crown-5) ionophores detecting Na+ ions. The top of the ion-selective nanopipette was kept immersed in 4 × 10−3 M electrolyte (NaCl) for 2, 7, and 10 days, respectively.
electric double layer in the solution at the narrow shank.18 The spike noise shown in Figures 3 and 4 depends on how the droplet is added to the dish; thus, keeping the spike noise always at the same level is difficult. However, we confirmed that the spike noise did not affect the signal after the spike diminishes. To confirm the improved performance of a nanopipette with a membrane at the conical shank (Figure 3a) compared to that of a nanopipette with a membrane in the middle (Figure 4a), we calculate the signal-to-noise (S/N) ratio. The S/N ratios of the nanopipette with a membrane in the middle for detecting potassium and sodium ions are 1.8 and 1.1, respectively. In the case of the nanopipette with a membrane at the conical shank, the S/N ratios for detecting potassium and sodium ions are 22 and 17, respectively. These results illustrate the more than 15-fold improvement in the S/N ratio by changing the position of the membrane to the conical shank in a nanopipette. The ionic current noise is mainly ascribed to the fluctuation of ionic current caused by the electric double layer formed inside the nanopipette. In our previous study, the nonlinear oscillating behavior of the ionic current was observed through a nanopipette with specific inner diameter in the 0.1 M KCl solution.18 In the case of the nanopipette with the membrane at the middle (Figure 3a), the bottom electrolyte part of the nanopipette could produce ionic current oscillation because of the oscillation of the Debye length of the electric double layer when described by the Gouy−Chapman−Stern model. On the other hand, the nanopipette with membrane at the conical shank is free from the ionic oscillation because the electric double layer does not significantly affect ionic current in the middle of the nanopipette with diameter much larger than 100 nm. Therefore, the absence of ionic current oscillation results in
the reduction of the background noise in the nanopipette with membrane at the conical shank (Figure 4). Finally, the durability of the ion-selectivity is demonstrated. Panels a and b of Figure 5 show the ionic current signals measured using nanopipettes with membranes containing saturated ionophores of bis(benzo-15-crown-5) and bis(12crown-4), respectively, 2, 7, and 10 days after the nanopipettes with the membranes are immersed into 4 × 10−3 M KCl electrolyte. The nanopipettes preserve the selectivity for potassium and sodium ionic currents, respectively, even after 10 days, in contrast to the case of the nanopipettes with filter membranes containing unsaturated ionophores.13−15 The filter membranes revealed the change in selectivity because of the saturation of the ionophores. In addition, our nanopipettes showed similar selective detection behaviors even after several tens of measurements over 10 days. It seems that the membranes and ionophores are not damaged by repetitive detection of ionic current. Therefore, our experimental data confirm that stable specific ionic current can be reproducibly detected with the aid of saturated ionophores in a membrane placed at the conical shank of a nanopipette.
4. CONCLUSION In summary, we have improved the stability and reproducibility of an ion-selective nanopipette by detecting ionic current mediated through ion-selective saturated ionophores in a membrane located at the conical shank of a nanopipette. We demonstrate the effect of the membrane position on ionselective detection using a nanopipette. The ion-selective nanopipette with a membrane at the conical shank exhibits more stable ionic current signal and higher S/N ratio than that with a membrane in the middle. The saturated ionophores 5133
dx.doi.org/10.1021/jp502715q | J. Phys. Chem. B 2014, 118, 5130−5134
The Journal of Physical Chemistry B
Article
(14) Takami, T.; Son, J. W.; Lee, J. K.; Park, B. H.; Kawai, T. Separate Detection of Sodium and Potassium Ions with SubMicropipette Probe. Jpn. J. Appl. Phys. 2011, 50, 08LB13_1− 08LB13_4. (15) Deng, X. L.; Takami, T.; Son, J. W.; Kawai, T.; Park, B. H. Selective Measurement of Calcium and Sodium Ion Conductance Using Sub-Micropipette Probes with Ion Filters. Appl. Phys. Express 2012, 5, 027001_1−027001_3. (16) Takami, T.; Iwata, F.; Yamazaki, K.; Son, J. W.; Lee, J. K.; Park, B. H.; Kawai, T. Direct Observation of Potassium Ions in HeLa Cell with Ion-Selective Nano-Pipette Probe. J. Appl. Phys. 2012, 111, 044702_1−044702_5. (17) The pulling parameters were as follows: Heat = 350, Fil = 3, Vel = 30, Del = 190, Pull = 0; Heat = 330, Fil = 2, Vel = 27, Del = 180, Pull = 250. (18) Deng, X. L.; Takami, T.; Son, J. W.; Kawai, T.; Park, B. H. Ion Current Oscillation in Glass Nanopipettes. J. Phys. Chem. C 2012, 116, 14857−14862. (19) Samec, Z.; Trojánek, A.; Langmaier, J.; Samcová, E. Cyclic and Convolution Potential Sweep Voltammetry of Reversible Ion Transfer across a Liquid Membrane. J. Electroanal. Chem. 2000, 481, 1−6. Samec, Z.; Trojánek, A.; Langmaier, J.; Samcová, E. Cyclic Voltammetry of Biopolymer Heparin at PVC Plasticized Liquid Membrane. Electrochem. Commun. 2003, 5, 867−870. (20) Deng, X. L.; Takami, T.; Son, J. W.; Kang, E. J.; Kawai, T.; Park, B. H. Ion-Selective Detection by Plasticized Poly(vinyl chloride) Membrane in Glass Nanopipette with Alternating Voltage Modulation. J. Nanosci. Nanotechnol. 2013, 13, 5413−5419.
contribute to the enhanced reproducibility; however, they cause the reduction of the measured current level, which can be overcome by forming the ion-selective membrane at the conical shank in the nanopipette.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Author Contributions †
These authors contributed equally to this work.
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the National Research Foundation of Korea (NRF) grants funded by the Korea government (MSIP) (2013R1A3A2042120, 2010-0024525, 2011-0030228, and 2008-0061893 (QMMRC)).
■
REFERENCES
(1) Štefanac, Z.; Simon, W. In-Vitro-Verhalten von Makrotetroliden in Membranen als Grundlage für hochselektive kationenspezifische Elektrodensysteme. Chimia 1966, 20, 436−451. Štefanac, Z.; Simon, W. Ion Specific Electrochemical Behavior of Macrotetrolides in Membranes. Microchem. J. 1967, 12, 125−132. (2) Bakker, E.; Bühlmann, P.; Pretsch, E. Carrier-Based Ion-Selective Electrodes and Bulk Optodes. 1. General Characteristics. Chem. Rev. (Washington, DC, U.S.) 1997, 97, 3083−3132. (3) Horváth, V.; Horvai, G.; Pungor, E. Ion Exchange at Neutral Carrier Ion-Selective Electrode Membranes. Fresenius J. Anal. Chem. 1993, 346, 569−571. (4) Hansma, P. K.; Drake, B.; Marti, O.; Gould, S. A. C.; Prater, C. B. The Scanning Ion-Conductance Microscope. Science 1989, 243, 641− 643. (5) Novak, P.; Li, C.; Shevchuk, A. I.; Stepanyan, R.; Caldwell, M.; Hughes, S.; Smart, T. G.; Gorelik, J.; Ostanin, V. P.; Lab, M. J.; Moss, G. W. J.; Frolenkov, G. I.; Klenerman, D.; Korchev, Y. E. Nanoscale Live Cell Imaging Using Hopping Probe Ion Conductance Microscopy. Nature Meth. 2009, 6, 279−281. (6) Smyth, J. W.; Shaw, R. M. Chapter ten − Visualizing Cardiac Ion Channel Trafficking Pathways. Methods Enzymol. 2012, 505, 187−202. (7) Schneckenburger, H. Total Internal Reflection Fluorescence Microscopy: Technical Innovations and Novel Applications. Curr. Opin. Biotechnol. 2005, 16, 13−18. (8) Al-Hilli, S. M.; Willander, M.; Ost, A.; Stralfors, P. ZnO Nanorods as an Intracellular Sensor for pH Measurements. J. Appl. Phys. 2007, 102, 084304_1−084304_5. (9) Asif, M. H.; Fulati, A.; Nur, O.; Willander, M.; Brannmark, C.; Stralfors, P.; Borjesson, S. I.; Elinder, F. Functionalized Zinc Oxide Nanorod with Ionophore-Membrane Coating as an Intracellular Calssium Selective Sensor. Appl. Phys. Lett. 2009, 95, 023703_1− 023703_3. (10) Reid, B.; Zhao, M. Ion-Selective Self-Referencing Probes for Measuring Specific Ion Flux. Commun. Integr. Biol. 2011, 4, 524−527. (11) Vilozny, B.; Actis, P.; Seger, R. A.; Vallmajo-Martin, Q.; Pourmand, N. Reversible Cation Response with a Protein-Modified Nanopipette. Anal. Chem. 2011, 83, 6121−6126. (12) Amemiya, S.; Bard, A. Scanning Electrochemical Microscopy. 40. Voltammetric Ion-Selective Micropipet Electrodes for Probing Ion Transfer at Bilayer Lipid Membranes. J. Anal. Chem. 2000, 72, 4940− 4948. (13) Son, J. W.; Takami, T.; Lee, J. K.; Park, B. H.; Kawai, T. Concentration Dependence of Ionic Conductance Measured with IonSelective Sub-Micro Pipette Probes in Aqueous Sodium and Potassium Chloride Solutions. Appl. Phys. Lett. 2011, 99, 033701_1−033701_3. 5134
dx.doi.org/10.1021/jp502715q | J. Phys. Chem. B 2014, 118, 5130−5134