Photoswitchable Azobis(benzo-15-crown-5) Ionophores as a Molecular Probe for Phase Boundary Potentials at Ion-Selective Poly(vinyl chloride) Liquid Membranes

To quantitatively investigate the relationship between the surface charge densities and the phase boundary potentials at poly(vinyl chloride)-based liquid membranes of ion-selective electrodes, lipophilic derivatives of photoresponsive azobis(benzo-15-crown-5) were used as the probe molecule. The photoinduced change in the number of primary cations permselectively extracted into the membrane side of the interface was estimated from the corresponding concentrations of the cis and trans isomers of the photoswitchable ionophores and the ratio of their complexation stability constants. To accurately estimate the changes in the phase boundary potential that arises at the interface of the membrane and the aqueous sample solution, an electrode system without an inner filling solution was employed. The direction as well as the magnitude of the photoinduced change in phase boundary potential depended on the change in the number of uptake of primary cations into the membrane side of the interface: Membranes containing sufficient amount of ionophores to give Nernstian slopes simply exhibited a parallel shift in phase boundary potentials, keeping Nernstian slopes upon UV irradiation. The membranes containing relatively low concentrations of ionophores exhibited only sub-Nernstian responses, in which case a photoinduced increase in the response slope was observed. The photoinduced changes in the phase boundary potentials thus observed were analyzed by a surface model based on electrical diffuse layers at both membrane and aqueous sides of the interface. In this model, the phase boundary potential is correlated to the number of surface charges due to the formation of cationic complexes at the membrane side of the interface. The observed photoinduced changes in the phase boundary potential and the response slope were in good agreement with the values calculated from the proposed surface model. It is therefore concluded that (1) the change in the phase boundary potential can be quantitatively correlated to that in the number of the primary cations permselectively extracted into the membrane side of the interface, which determines the surface charge density, and (2) the sub-Nernstian response slope is attributed to a low surface charge density due to insufficient number of permselectively extracted primary cations. It is estimated on the basis of the present surface model that, to obtain a Nernstian potentiometric response, a sufficient amount of primary cation to produce a net positive surface charge density of > 2 × 10 - 7 C cm - 2 must be uptaken into the membrane side of the interface by complexation with the ionophore.