Anal. Chem. 2000, 72, 2050-2054
Direct Potentiometric Information on Total Ionic Concentrations Alan Ceresa,† Erno 1 Pretsch,*,† and Eric Bakker*,‡
Department of Organic Chemistry, Swiss Federal Institute of Technology (ETH), Universita¨tstrasse 16, CH-8092 Zu¨rich, Switzerland, and Department of Chemistry, Auburn University, Auburn, Alabama 36849
Polymeric membrane ion-selective electrodes exhibit an apparently super-Nernstian response at low sample activities if inner solutions are used that induce strong zerocurrent fluxes of primary ions toward the inner compartment. This is due to the limited ion fluxes in the aqueous boundary layer near the membrane. In the presence of labile complexes, the effective flux rate is increased and the emf depends on the total concentration of the ions. The concept is illustrated experimentally with calciumselective electrodes based on the ionophore N,N-dicyclohexyl-N′,N′-dioctadecyl-3-oxapentanediamide (ETH 5234) that either respond to total or free ion concentrations. Samples can be distinguished that contain varying levels of total calcium but are all buffered with EDTA to the same free calcium concentration of 5 × 10-8 M. It is common knowledge that, in variance to most analytical methods, ion-selective electrodes (ISEs) respond to activities and not concentrations of ions in the sample.1 However, it has been documented in the literature that the activities at the membrane surface and in the bulk of the solution may considerably differ both for solid-state2-6 and solvent polymeric membranes.7,8 Such differences are often responsible for the lower detection limits since the dissolution of the solid membrane4 or leaching from the polymeric membrane9,10 induces local activities that are independent of further dilution of the sample. Leaching from polymeric membrane electrodes has been avoided by using inner reference solutions that, through ion exchange at the inner membrane surface, induce gradients and concomitant fluxes of primary ions toward the inner compartment.11 Too large fluxes, †
Swiss Federal Institute of Technology (ETH). Auburn University. (1) Morf, W. E. The Principles of Ion-Selective Electrodes and of Membrane Transport; Elsevier: New York, 1981. (2) Morf, W. E.; Kahr, G.; Simon, W. Anal. Chem. 1974, 46, 1538-1543. (3) Buffle, J.; Parthasarathy, N. Anal. Chim. Acta 1977, 93, 111-120. (4) Parthasarathy, N.; Buffle, J.; Haerdi, W. Anal. Chim. Acta 1977, 93, 121128. (5) Lindner, E.; To´th, K.; Pungor, E. Anal. Chem. 1982, 54, 72-76. (6) Morf, W. E. Anal. Chem. 1983, 55, 1165-1168. (7) Bakker, E. Sens. Actuators B 1996, 35, 20-25. (8) Morf, W. E.; Badertscher, M.; Zwickl, T.; de Rooij, N. F.; Pretsch, E. J. Phys. Chem. B 1999, 103, 11346-11356. (9) Maj-Zurawska, M.; Erne, D.; Ammann, D.; Simon, W. Helv. Chim. Acta 1982, 65, 55-62. (10) Mathison, S.; Bakker, E. Anal. Chem. 1998, 70, 303-309. (11) Sokalski, T.; Ceresa, A.; Zwickl, T.; Pretsch, E. J. Am. Chem. Soc. 1997, 119, 11347-11348. ‡
2050 Analytical Chemistry, Vol. 72, No. 9, May 1, 2000
however, lead to a reduced concentration at the membrane surface of the sample side.7,12 Such effects are especially strong at low sample activities where super-Nernstian response curves with drifting signals and an apparent loss of selectivity are observed. Therefore, design of ISEs for low detection limits usually aims at avoiding such super-Nernstian responses.12,13 Here we show that ISEs with super-Nernstian response can also provide useful analytical information because their response depends on total concentrations rather than on the activities of the uncomplexed ions if the sample ions are partly complexed with a chelator. The essence of this new method is that in the presence of a complexing agent the net flux of the ions is larger than expected from the diffusion of the uncomplexed ions alone, and as a consequence, the activities at the membrane surface layer depend on the total concentration rather than the activity of the uncomplexed ions. This principle is akin to the function of amperometric metal electrodes, where the observed current is mass transport limited but uses zero-current potentiometry as a transduction principle. EXPERIMENTAL SECTION Reagents. Poly(vinyl chloride) (PVC), bis(2-ethylhexyl) sebacate (DOS), sodium tetrakis[3,5-bis(trifluoromethyl)phenyl]borate (NaTFPB), the ionophore N,N-dicyclohexyl-N′,N′-dioctadecyl3-oxapentanediamide (ETH 5234), and tetrahydrofuran (THF) were from Fluka AG (CH-8071 Buchs, Switzerland). Aqueous solutions were prepared with freshly deionized water (18.0 MΩ cm specific resistance) obtained with a Nanopure reagent-grade water system (Barnstead, CH-4009 Basel, Switzerland). Nitric acid and NaCl were of Suprapur quality from Merck (Darmstadt, Germany), and CaCl2, Ca(NO3)2, disodium ethylenediaminetetraacetic acid (Na2EDTA), HCl, and NaOH were of puriss. p.a., or Microselect quality from Fluka. Membranes. ISE membranes contained 1.04 wt % (13.0 mmol kg-1) ETH 5234, 0.41 wt % (4.6 mmol kg-1) NaTFPB, 66.3 wt % DOS, and 32.2 wt % PVC. Membranes of ∼150 µm thickness were obtained by casting a solution of ∼400 mg of the membrane components, dissolved in ∼5 mL of THF, into a glass ring (45 mm i.d.) fixed on a glass plate. ISEs were prepared by punching a disk of 4 mm diameter from the above ion-selective membrane and affixing it to plasticized PVC tubing with a THF/PVC slurry. The respective internal filling solutions were in contact with the (12) Sokalski, T.; Ceresa, A.; Fibbioli, M.; Zwickl, T.; Bakker, E.; Pretsch, E. Anal. Chem. 1999, 71, 1210-1214. (13) Sokalski, T.; Zwickl, T.; Bakker, E.; Pretsch, E. Anal. Chem. 1999, 71, 1204-1209. 10.1021/ac991092h CCC: $19.00
© 2000 American Chemical Society Published on Web 03/31/2000
reference half-cell (Ag/AgCl in 0.1 M NaCl) through a 0.1 M NaCl bridge electrolyte. These were 10-3 M CaCl2 and 0.1 M NaCl (conventional ISE), 10-3 M CaCl2, 5 × 10-2 M Na2EDTA, adjusted to pH 9.0 with 6 × 10-2 M NaOH, calculated aCa2+ ) 3 × 10-12 M (ISE with Ca2+ uptake), and 1 M Ca(NO3)2 (ISE with Ca2+ release). The following constants were used for calculating the activities: EDTA-Ca2+, log K 10.9, pK1 2.0, pK2 2.76, pK3 6.16, and pK4 10.26.14 The ISEs were initially conditioned in 10-3 M CaCl2, and 10-5 M HNO3 for 3 days and then before measurements in 10-5 M CaCl2, 2 × 10-4 M NaCl, and 10-5 M HNO3 for 2 days. EMF Measurements. The ISE and a free-flowing calomel reference electrode with 1 M KCl as bridge electrolyte15 were mounted into a 30 × 30 × 60 mm Plexiglas flow-through cell with a horizontal 3-mm-diameter flow channel. The ISE and reference electrode were accommodated in vertical holes (6 mm diameter) perpendicular to the flow channel, facing each other at a distance of 2 mm. The solutions were pumped with a peristaltic pump at a flow rate of 2.5 mL/min. Calibration solutions of various CaCl2 concentrations contained a background of 2 × 10-4 M NaCl. Ion buffered samples contained 10-4 M Na2EDTA and 10-5, 10-6, 5 × 10-7, 3 × 10-7, 2 × 10-7, and 10-7 M CaCl2 with pH values adjusted with NaOH and HCl to 6.05, 5.45, 5.25, 5.10, 5.00, and 4.80, respectively. The free Ca2+ concentration was 5 × 10-8 M in all samples. Potentials were measured with a custom-made 16channel electrode monitor at room temperature (20-21 °C) in an alternating sequence of samples (10 min) and a reference solution (10-5 M CaCl2, 2 × 10-4 M NaCl, and 10-5 M HNO3; 20 min). Typical drifts in the transient response range were -1.7 mV/min, otherwise
