Extracellular electrophysiological recordings of rodent brains using tetrodes requires plating them with gold solution in order to achieve appropriate impedances.

How does the ideal plating impedance depends on the size, depth of the desired recording structure, and the packing density of the objective cells?

  • $\begingroup$ Sorry, why is this statement, made by Nachosan, true? "...having a very low impedance electrode does not allow you to record single units. And if it gets too low, you are mostly recording LFP." Why does having very low impedance (assuming there are no shorts) not allow you to record single units? $\endgroup$
    – nn123
    Jun 10, 2021 at 11:57

1 Answer 1


The ideal impedance of any electrode aimed at recording low-amplitude neural signals (barred suction glass-pipette electrodes for patch clamping) is simply: as low as possible.

In case of tetrodes, i.e., electrodes consisting of a bundle of four microwires that can record from multiple neurons simultaneously in the brain of freely moving animal, they are usually electroplated to reduce their impedance from 2-3 MΩ to 200-500 kΩ (measured at 1 kHz). Low impedances increase the signal-to-noise ratio and allow for the recording of small-amplitude signals. Tetrodes with lower impedances could theoretically further improve SNR, but this is technically not possible with the standard electroplating methods, per an article from 2009 (Ferguson et al., 2009).

A straightforward way of reducing impedance is to increase the surface of the electrode. Increasing the absolute size may very well be detrimental to the purpose of collecting single unit recordings, as larger electrodes inherently pick up the activity more neurons. A more feasible way is to increase the surface by means of altering the microstructure of the electrode, i.e., increase its roughness (Fig. 1).

electrode surface
Fig. 1. smooth (A) and rough electrode surface (B). source: Jeff-Du et al (2015)

A method where tetrodes were plated with a gold solution with added polyethylene glycol (PEG) or multi-walled carbon nanotube (MWCNT) solutions was reported to to lower the impedance further to 30-70 kΩ. MWCNTs and PEG reportedly act as inhibitors in the electroplating process and create a larger surface area, thereby lowering the impedance on the tetrode tips (Ferguson et al., 2009).

- Ferguson et al., Sens Actuators A Phys (2009); 156(2): 388–93
- Jeff Du et al., J Mater Chem C (2015); 3: 6515-24

  • 1
    $\begingroup$ I will edit my unclear question. Your answer has classic references about tetrode plating. I should have put them myself. However I'm looking for real experience based insight into which are the best configurations to target different areas cortex vs hippocampus vs amygdala for instance. As you mention picking up many cells is a problem when you are trying to record single units, but this not only depends on the tetrode itself but on the way the cells are packed in the brain and the chance of recording them (how big an extracellular signal they make). This is where the question is aiming at. $\endgroup$
    – nachosan
    Sep 8, 2016 at 7:51
  • $\begingroup$ I think you should probably remove the patch clamp part. It's confusing, it's a completely different type of electrode and technique with completely different objectives. $\endgroup$
    – nachosan
    Sep 8, 2016 at 7:53
  • $\begingroup$ @nachosan I make a general statement that impedances should always be low, which is not true for patch clamp, so I'd rather leave it there. I hope you get more useful answers soon. Good luck! $\endgroup$
    – AliceD
    Sep 8, 2016 at 8:00
  • $\begingroup$ @Christian Well, yes, while true, the thing is that they are not being used for recording the same signal. Also, its not simply as low as possible. As you state later having a very low impedance electrode does not allow you to record single units. And if it gets too low, you are mostly recording LFP. So although I get the point, I think its a little bit more complicated than that. Both for the rule and for the counterexample. $\endgroup$
    – nachosan
    Sep 8, 2016 at 8:23
  • $\begingroup$ @nachosan - I can't help you any further. I just noticed your excellent question didn't receive answers and decided to write up what I know. If it's important to you, you could consider putting a bounty on the question of +50, after edits. Also don't hold back downvoting my answer if you consider it irrelevant. $\endgroup$
    – AliceD
    Sep 8, 2016 at 8:47

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.