The increasing penetration of solar generated electricity in the United States will eventually lead to a glut of grid power during times of peak solar intensities. We discuss and present preliminary studies of an entirely new approach for converting and storing intermittent solar energy by directly storing the sunlight as electrochemical energy. This solar charged redox flow battery has the additional advantage of being able to capture and store the heat energy, that is 80% of the incoming solar flux, that is not used in the electrochemical reactions to then be used for space and water heating. This hybrid system could replace both a rooftop solar water heating systems and some photovoltaic panels resulting in substantial cost savings and wider implementation of solar energy and load leveling of the electrical grid. Compared to photoelectrochemical water splitting this approach has several other advantages including not having to deal with gases and bubbles that block light, many possible semiconductor redox couple combinatons rather than being restricted to semiconductors with band edges that match only water oxidation and water reduction potentials and use fast one-electron couples require no multielectron electrocatalysts reducing overpotential losses. We will present results using p-type InP and n-type GaP photoelectrodes with oxide protection layers to drive the well studied V(III) reduction and V(IV) oxidation reactions to directly store solar energy.