First of all, Thank you everyone for sharing such valuable ideas and opinions with me. As I said in the beginning, I am not an expert nor majoring in electronics, so I am slowly getting the hang of what to do now. It is a bit frustrating since it slows down my research but I just need to get this to work as a polymer engineer.
If the system were to require a very low current in the range of tens of mA, I would have just borrowed an Arbitrary Waveform Generator from the department of EE. Straightforward. However, I speculate that there is no way that such equipment can handle high currents (If you know cheap equipment that can handle high currents over 10A, please let me know!)
This is clearly seen when measuring the conductivity of a salt solution in water: over the first few seconds you apply a DC voltage the current drops drastically as the ions migrate and find a new equilibrium around the probes. You have to apply at least 1 kHz to have no effect from this migration.
As a mechanistician.. Happy to see chemistry parts.. It has been awhile.. Anyway, in our system, when the current is high, over the first few seconds under an applied DC current, the voltage increases dramatically. The same as your description.
In a metal (or alloy) there is indeed a cloud of electrons that moves around more or less freely through the material, but the metal atoms as such are neutrally charged and not be moving around in the field.
You may be right. I think it is possible that metallic ions in the lattice will eventually diffuse when the momentum transfer between electrons and metallic ions exceeds the metallic ionic activation energy.
Do you actually see such an effect in your liquid metal when applying a current?
When the liquid metal is in acidic/base conditions, even under very low currents, the LM moves around very rapidly. There are hundreds of research articles about the behavior. However, there is little work about what happens when LM interfaces with air or other solids. Such an effect, electromigration in LM, has been just studied last year (published this year). From the paper, electromigration occurs especially at the spot where high current density is observed. In my work, since I am not doing rigorous studies on the phenomena, I did not try to see the electromigration under a microscope. Electromigration is speculation. I believe it might be true though...
Indeed the heating of the metal increases resistance as well, and will affect your measurements.
Also very interesting... The resistance... decreases! when the LM is heated up. Not only in our lab, there are several articles reporting the behavior. Very fascinating... So I need to keep increasing the current to increase the temperature. If I keep the current constant, the temperature won't go beyond a certain point due to the reduction of resistance.
For the polarity reversals: I agree that you will need a controller (Arduino sounds like a good one) for this. That way you can control your period accurately, and you don't get tired doing it a million times by hand. An Arduino can do this much faster than you can do by hand, you possibly you have to go to frequencies of 10-100 kHz to see effects.
Also you will have to use at electronic switches as no relay can get near such speeds.
Your final challenge is the measuring of the actual current, not trivial for AC, especially high frequency high current AC. Building an H-bridge switching 10A at high frequency will also not be trivial. Look at motor controllers, they can do just that.
Sounds great. I will start with an Arduino to go to frequencies of 10-100 kHz. Yeah, according to simple google search, relays typically have 20Hz (50ms) switching speeds.. I will look for high current rated transistors and motor controllers.
