[Eeglablist] Movement artifact = Liquid junction potential

[Makoto Miyakoshi]

Dear colleagues,

Let me share this info.

E. Huigen (2000) Noise in biopotential recording using surface electrodes
https://www.semanticscholar.org/paper/Noise-in-biopotential-recording-using-surface-Huigen/8fc0837f7af0a36b19799139d9b763969057b985

*%%%%%%%%%%%%%%%%%%%%%*

*2.4 Motion artifact*
Movement can cause changes in the potentials that are created when an
electrode is applied to the skin. Normally, when the patient is relaxed,
and high quality electrodes are used, the recording is not distorted by
motion artifact. Brinkman (1993) has found no significant correlation
between the intentional movement of the arm and the noise signal. The
mechanisms that can cause motion artifact are described next.

*Liquid junction potential variations by electrode movement*
The various phase junctions in the electrode-electrolyte-skin interface all
cause junction potentials, sensitive to motion artifact. The
skin-electrolyte interface can cause artifacts of 400-600 μV when the
electrode is moved parallel to the skin surface (Smith and Wace, 1995).
When the electrode is moved perpendicular to the skin the potential changes
can be up to 900 μV. Firm attachment to the skin can reduce the potential
changes. The electrode-electrolyte interface also produces artifacts when
mechanically disturbed. Gatzke (1974, as described in Webster, 1984)
measured a 15 mV potential change when a pure silver electrode is moved in
electrolyte. Coating with silver chloride, thus creating a stable double
layer, produces a 10-fold reduction of the artifact. Further reduction (up
to negligible value) can be achieved by recessing the electrode-electrolyte
interface in a protective cup, in which a sponge soaked in gel is placed
(figure 2-2).

*Skin potential changes*
Earlier, the stratum corneum and the barrier layer have been identified as
the major sources of the impedance of the skin. Webster (1984) has also
observed a potential difference between the inside and the outside of the
barrier layer. Van Wijk van Brievingh (1988) however states that this skin
potential is a liquid junction potential between deeper skin layers and the
electrolyte. The skin potential has a typical value of 30 mV at the thorax.
Stretching of the skin decreases the skin potential to about 25 mV. This
artifact can be reduced to negligible value by abrading the skin. Webster
gives 20 strokes with fine sandpaper as an indication. Shackel (1959, as
described in Geddes and Baker, 1989) developed a method for shorting the
skin potential, called the skin-drilling technique. The epidermis is eroded
using a dental burr. The capillaries remain undisturbed, so no blood is
drawn. Unfortunately, the epidermis is also the layer that protects the
skin from irritation. A mild electrode gel has to be used to prevent
unwanted effects.

%%%%%%%%%%%%%%%%%%%%%

Wikipedia 'liquid junction potential'
https://en.wikipedia.org/wiki/Liquid_junction_potential

In Huigen, Peper, Grimbergen (2002) Med. biol. Eng, they described:

*RECORDINGS  OF  biomedical  signals  from  the  body  surface  often
 contain  a  substantial  noise  component. This  noise  signal  can
 severely  impair  the  resolution  of  biomedical  recordings  as  its
 magnitude  can  be  as high  as  10-60  uVp_p  (GONDRAN  et  aL,  1996).
This  is  in  the  range  of  EEG  potentials  and  is  at  least ten
 times  as  high  as  several  types  of  evoked  potentials  (e.g.  visual
 evoked  potentials  or  somatosensory evoked  potentials).*

By the way I could not find Smith and Wave (1995)* European Journal of
Anaesthesiology. *If you have a copy, please let me know.

Makoto
-- 
Makoto Miyakoshi
Swartz Center for Computational Neuroscience
Institute for Neural Computation, University of California San Diego
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20180503/754bc936/attachment.html>