What the material actually changes
A surface electrode is an interface between a metal and an electrolyte. At that interface a contact potential forms, and its stability decides what you can record.
At high frequencies all materials are roughly equivalent. At low frequencies, below 0.5 Hz, the differences become visible: an unstable interface drifts, and the drift merges with the signal you are looking for. That is why slow potentials and newborn traces are done in Ag/AgCl.
The second variable is mechanical. A department electrode goes through hundreds of cleaning cycles, and the active layer eventually comes off. Sintering answers that: the layer is not deposited on the surface, it runs through the bulk.
| Material | Symbol | Low frequencies | Cleaning durability | When to choose it |
|---|---|---|---|---|
| Tin | Sn | Fair | Good | Routine EEG at usual frequencies, and pre-wired caps where cost per electrode matters. |
| Silver chloride | Ag/AgCl | Excellent | Fair | Slow potentials, newborn traces, polysomnography. The reference material as soon as the bandwidth goes below 0.5 Hz. |
| Sintered Ag/AgCl | Ag/AgCl Sint | Excellent | Very good | The same uses as Ag/AgCl, in a fleet that turns over a lot. The active layer runs through the bulk and does not wear off in cleaning. |
| Gold | Au | Good | Excellent | Long montages and heavily used fleets. Mechanically the toughest, and unaffected by the chloride pastes that corrode the others. |
Do not mix two materials on one montage
Two electrodes of different metals joined by an electrolyte form a battery. The resulting voltage is small, but it is of the same order as the slow potentials you record, and it drifts with temperature.
Concretely: a gold cup on Fp1 and a tin cup on Fp2 give a frontal asymmetry that does not exist in the patient. On a bipolar montage the artefact travels along the whole chain.
The rule is simple and has no useful exception: one montage, one material. Including the auricular references and the ground.