Ion-selective electrodes are widely used for the detection of ions in aqueous solutions such as natural waters. Their origin traces back to 1909 with the invention of the pH glass electrode. Nowadays, routine pH measurements are still performed by potentiometric measurements with glass electrodes. The phase-boundary potential difference at the glass membrane-sample interface, measured against a reference electrode, relates to solution pH following the Nernst equation. While being user-friendly, they suffer from multiple drawbacks. Firstly, their sensitivity is intrinsically dictated by the Nernst equation and is limited to 59.2 mV/pH at 25 °C. This might not be sufficient for applications where high precision pH sensing is required, such as ocean acidification monitoring. Secondly, a Nernstian response can only be obtained if all the other potential differences in the overall electrochemical cell are constant over the whole experimental procedure. This is not the case when, for example, the temperature or the ionic strength of the sample change during the measurement routine. The former influences the glass electrode itself, while the latter rather affects the reference electrode via liquid junction potential variations.

This thesis presents enhancements to the potentiometric experimental setup for pH sensing with glass or polymeric membrane pH electrodes, achieved through the integration of electronic components, chemical symmetry and open liquid junctions. A dynamic electrochemical readout called constant potential coulometry is explored for in situ pH sensing in coastal waters by implementation in a submersible probe deployed in the Krka River estuary in April 2025.