Building a little more on this "idea", I accidentally discovered a circuit modification using the iCircuit simulator on the iPad which gave a phenomenal result. I don't know if it's a valid simulation (seeing as there is no component that can perfectly simulate a water cell) but I thought I would share this.
The attached image shows a 0.001 resistor directly inline with both sides of the "Excitor Array". I built the "Excitor Array" to have a positive side capacitor, a negative side capacitor and a capacitor to represent a water molecule with the smallest capacitance I could make work (0.01pF).
The result is, within a second the voltage charge on the "water molecule" goes up to 30kV!! Before 2.5 seconds it's in the GV range. Before 3.5 seconds it's in the TV range! This effect will NOT happen if the resistors are removed or if the resistance value is much higher or if the resistors leads are not in direct contact with the capacitor leads.
What I perceive is happening, is that the very small resistance is allowing the voltage to get into the "ER", but, once it's in, the "ER" doesn't want to let it out, again because of that small resistance.
I have not been able to replicate the result with the Falstad simulator (which iCircuit is based on). The components seem to work a little differently there. If anyone else has other circuit simulators, please try this configuration and let us know the result.
UPDATE: I think I have managed to sort of duplicate the effect of climbing voltage in Falstad using 0.0001 resistors instead of 0.001. Over 150kV in less than 4 seconds. I have attached the configs for both the Falstad and iCircuit.
TS