My son tried it on small parts and loves the method. However, his process involved sodium carbonate (washing soda), not sodium bicarbonate (baking soda).
Baking soda, sodium bicarbonate = NaHCO3
Washing soda, sodium carbonate = Na2CO3
The references I've seen seem to prefer washing soda, but I can't see that it would make much difference. The idea is to get monatomic hydrogen onto the cathode (negative, the part to be de-rusted) so that it will reduce the iron oxide to iron particles and water.
With washing soda, (Na2CO3), two sodium ions will plate out on the part, but will immediately react with the water in the bath to form monatomic hydrogen and sodium hydroxide. (Na + H2O > NaOH + H.)
With baking soda, (NaHCO3) both a sodium ion and a hydrogen will plate out, the hydrogen being monatomic, and the sodium immediately reacting with the water in the bath (as noted above), to form monatomic hydrogen anyhow.
Monatomic: A single hydrogen atom, like many gases, will immediately react with another single hydrogen atom nearby to form H2... in the same way that oxygen forms O2 and chlorine forms Cl2.
But if you can get the hydrogen in its initial state, as a single hydrogen atom all by itself, H, it will be far more reactive than in its H2 state, where it has already reacted with itself. That's the "monatomic" state.
This higher reactivity is common to most monatomic gases (which don't last long, since there's usually another monatomic atom nearby to react with immediately).
As an example, when hydrogen peroxide, H2O2, decomposes into water (H2O) and a single O, oxygen, that monatomic oxygen is very reactive and will oxidize almost anything (like a germ) immediatly. That's why hydrogen peroxide is so valuably as a germ-killer. You will see it fizz when you put it on a wound --most things, like blood and tissue, decompose it pretty quickly. (However by the time you see it forming bubbles, it has already formed up as O2.)
"Plating out": In solutions like of baking soda or washing soda, the sodium is ionized as Na+, having broken away from the (-HCO3) or (-CO3) parts. Water breaks stuff apart like that a lot, and it's called "ionization." So the positive sodium ion drifts toward the negative electrode (cathode) and when it contacts the cathode, it picks up a negative electron and voila! becomes neutralized and turns into an actual metallic atom of sodium. (As noted, it then reacts with water in the solution to form hydrogen, which escapes as bubbles.)
This is the plating process and is the way chrome or copper is plated onto other things, if you have the right baths and voltages and whatnot.
Hydrogen is actually a sort of metal (except with a really, really low boiling point) and "plates out" in the same way. Except that at ordinary temperatures, it just rapidly forms H2 bubbles instead of sticking to the cathode.
Terry, 230RN