High temperature water electrolysis has a tremendous potential in enabling a cost-effective and sustainable production of synthetic fuels like hydrogen by using renewable heat and electricity from concentrating solar thermal (CSP) plants [1]. Most efforts in this research area are currently directed towards investigating solid oxide electrolysis cells (SOECs) based on fuel cell technology [2,3], although this approach is not optimal for integration with CSP plants because of electrolysis temperatures (i.e., above 800°C) significantly exceeding those of most solar heat storage systems. In order to improve thermal matching with current solar thermal fluids, development of electrolysis processes within the 500-600°C range would be a desirable option. In this context, molten salts could be seen as an ideal medium to lower process temperatures with respect to solid oxide electrolyzers since overall ionic conductivity and transport numbers of liquid salts are usually higher than solid-type electrolytes. Recently, alkali molten carbonate salts have gained a return of attention as a promising and attractive electrolyte for electrochemical conversion processes of mineral ores and CO2 gas at moderate temperatures [4,5]. At the same time, water electrolysis in alkali molten carbonates has been recently mentioned as a feasible process [2], although no systematic studies on this process are available in literature. The overall electrolysis reactions can be written as: cathode, H2O+CO2 + 2e = H2+CO3 2- (1)anode, CO3 2- = CO2 +0.5O2 +2e (2) The anode gas mixture made of 33 % O2 + 67% CO2 is ideal for oxy-combustion processes. Thus, a zero CO2 emission system could be realized by integrating a molten carbonate electrolysis with an oxycombustion process. The combined process could enable a CO2 closed-loop recycling scheme with CO2 capture. Part of the post-combustion CO2 could be, in fact, re-injected to the cathode, whereas the excess CO2 could be easily captured.From a mechanistic point of view, the exact cathode reduction reaction is not well understood. Although the cathode reaction can be interpreted in terms of a conventional water reduction process (eq. (1)), there are indications in literature that bicarbonate anions are likely the reactive species rather than dissolved water according to [6]: 2HCO3 - + 2e =H2+2CO3 2- (3) The presence of bicarbonate is possible because, in presence of a water-containing atmosphere, molten carbonates cannot be considered as a pure solvent, but rather as a mixture of three anionic constituents, namely carbonate, bicarbonate and hydroxide ions, according to the following chemical equilibrium: H2O+CO3 2- = HCO3 - + OH- (4) Thermodynamic calculations indicate that bicarbonate concentration is lower than hydroxide, but it is not negligible below 600°C. Initial lab-scale work has confirmed the effective presence of hydrogen as main cathode reaction product in a electrolysis conducted in ternary alkali carbonate eutectic mixture Li2CO3-Na2CO3-K2CO3 (43.5-31.5-25.0 mol %) despite the relatively low values of steady-state current density observed within the 525-600°C range (see Fig.1). Cyclic voltammetry experiments are also underway to study in more detail kinetics and mechanisms of the cathodic reaction, which is an essential aspect for interpreting and modeling the overall electrolysis process. Detailed results of these exploratory investigations will be shown at the time of Conference.