Jou nal o Neu al Enginee ing
PAPER • OPEN ACCESS
Di ec measu emen o oxygen educ ion eac ions
a neu os imula ion elec odes
To ci e his a icle: Jií Ehlich e al 2022 J. Neu al Eng. 19 036045
View he a icle online o upda es and enhancemen s.
You may also like
Ex ension o he ECRH ope a ional space
wi h O2 and X3 hea ing schemes o
con ol ungs en accumula ion in ASDEX
Upg ade
H. Höhnle, J. S obe , A. He mann e al.
-
Measu emen o he O2O3O4 and
O3O4O5 supe Cos e –K onig a es in
ungs en ia asymme ic di ac ion
spec ome y
John Seely, Jack L Glo e , Law ence
Hudson e al.
-
LIGO de ec o cha ac e iza ion in he
second and hi d obse ing uns
D Da is, J S A eeda, B K Be ge e al.
-
This con en was downloaded om IP add ess 130.236.88.56 on 27/06/2022 a 15:15
J. Neu al Eng. 19 (2022) 036045 h ps://doi.o g/10.1088/1741-2552/ac77c0
Jou nal o Neu al Enginee ing
OPEN ACCESS
RECEIVED
23 Decembe 2021
REVISED
1 June 2022
ACCEPTED FOR PUBLICATION
10 June 2022
PUBLISHED
27 June 2022
O iginal con en om
his wo k may be used
unde he e ms o he
C ea i e Commons
A ibu ion 4.0 licence.
Any u he dis ibu ion
o his wo k mus
main ain a ibu ion o
he au ho (s) and he i le
o he wo k, jou nal
ci a ion and DOI.
PAPER
Di ec measu emen o oxygen educ ion eac ions a
neu os imula ion elec odes
Jiˇ
í Ehlich1,4, Ludo ico Migliaccio1,4, Iho Sahaliano 1, Ma a Niki´
c1,2, Jan B odský1, Im ich Gablech1,
Xuan Thang Vu3, S en Ingeb and 3and E ic Daniel Głowacki1,∗
1Bioelec onics Ma e ials and De ices Labo a o y, Cen al Eu opean Ins i u e o Technology CEITEC, B no Uni e si y o Technology,
Pu kyˇ
no a 123, 61200 B no, Czech Republic
2Ins i u e o Neu oelec onics, Technical Uni e si y o Munich, Munich, Ge many
3Ins i u e o Ma e ials in Elec ical Enginee ing 1, RWTH Aachen Uni e si y, 52074 Aachen, Ge many
4These au ho s a e con ibu ed equally.
∗Au ho o whom any co espondence should be add essed.
E-mail: e ic.daniel.glowac[email p o ec ed].cz
Keywo ds: bioelec onics, neu os imula ion, a adaic eac ions, hypoxia, eac i e oxygen species
Supplemen a y ma e ial o his a icle is a ailable online
Abs ac
Objec i e. Elec ic s imula ion deli e ed by implan able elec odes is a key componen o neu al
enginee ing. While ac o s a ec ing long- e m s abili y, sa e y, and biocompa ibili y a e a opic o
con inuous in es iga ion, a widely-accep ed p inciple is ha cha ge injec ion should be e e sible,
wi h no ne elec ochemical p oduc s o ming. We wan o e alua e oxygen educ ion eac ions
(ORR) occu ing a di e en elec ode ma e ials when using es ablished ma e ials and s imula ion
p o ocols. App oach. As s imula ion elec odes, we ha e es ed pla inum, gold, ungs en, nich ome,
i idium oxide, i anium, i anium ni ide, and poly(3,4-e hylenedioxy hiophene):poly(s y ene
sul ona e). We use cyclic ol amme y and ol age-s ep ampe ome y in oxygena ed e sus ine
condi ions o es ablish a which po en ials ORR occu s, and he magni udes o di usion-limi ed
ORR cu en s. We also benchma k he a eal capaci ance o each elec ode ma e ial. We use
ampe ome ic p obes (Cla k- ype elec odes) o quan i y he O2and H2O2concen a ions
in he icini y o he elec ode su ace. O2and H2O2concen a ions a e measu ed while
applying DC cu en , o a ious biphasic cha ge-balanced pulses o ampli ude in he ange
10–30 µC cm−2/phase. To co obo a e expe imen al measu emen s, we employ ini e elemen
modelling o ec ea e 3D g adien s o O2and H2O2.Main esul s. All elec ode ma e ials suppo
ORR and can c ea e hypoxic condi ions nea he elec ode su ace. We ind ha elec ode ma e ials
di e signi ican ly in hei onse po en ials o ORR, and in he ex en o which hey p oduce H2O2
as a by-p oduc . A key esul is ha ypical cha ge-balanced biphasic pulse p o ocols do lead o
i e e sible ORR. Some elec odes induce se e ely hypoxic condi ions, o he s addi ionally p oduce
an accumula ion o hyd ogen pe oxide in o he mM ange. Signi icance. Ou indings highligh
a adaic ORR as a c i ical conside a ion o neu al in e ace de ices and show ha he es ablished
biphasic/cha ge-balanced app oach does no p e en i e e sible changes in O2concen a ions.
Hypoxia and H2O2can esul in di e en (elec o)physiological consequences.
1. In oduc ion
Elec ical neu os imula ion is an es ablished pa o
nume ous implan able bioelec onics de ices like
deep b ain s imula o s [1], pe iphe al ne e s imula-
ion de ices [2], spinal co d s imula o s [3], and e -
inal p os he ics [4,5]. Elec ical s imula ion in ol es
cha ge injec ion om an elec ode su ace in o
physiological elec oly e. The sa e y and eliabili y
o his p ocess is a i al conside a ion. Mechan-
isms o cha ge exchange a he elec ode/elec oly e
in e ace a e he subjec o ex ensi e esea ch, and
can be di ided in o capaci i e, pseudo-capaci i e,
and a adaic [6–8]. Capaci i e cha ge injec ion
© 2022 The Au ho (s). Published by IOP Publishing L d
J. Neu al Eng. 19 (2022) 036045 J Ehlich e al
in ol es he cha ging and discha ging o elec oly ic
double-laye s, and no cha ge is ans e ed o spe-
cies in solu ion. The pseudo-capaci i e mechanism,
also known as pseudo- a adaic, ea u es edox eac-
ions o he elec ode ma e ial i sel . These eac ions
can esul in high densi y o cha ge ans e , and his
ans e can be highly e e sible. The hi d ca ego y
is a adaic, whe e cha ge is ans e ed o a species
in solu ion ia a edox p ocess occu ing a he elec-
ode/elec oly e in e ace. This p ocess may o may
no be e e sible, depending on kine ic ac o s like
ac i a ion ba ie s (aka o e po en ials) o a gi en
eac ion as well as di usion o eac an s/p oduc s. A
s imula ion elec ode wo ks by injec ing cu en in o
a physiological medium and he esul an elec ic
ields modula e he memb ane po en ial o nea by
exci able cells. Fo ins ance, a ca hodic pulse will
e icien ly depola ize cell memb anes and ac i a e
ol age-ga ed sodium channels, igge ing ac ion
po en ials. Ca hodic, i.e. nega i e cu en s, ha e been
long ecognized as being mos e icien a elici ing
ac ion po en ials. Howe e , i ollows ha cha ge
a i icially injec ed in o a physiological en i onmen
should be subsequen ly emo ed, ha is no ne cha -
ging o he sys em should occu . This is because ne
cha ge emaining would co espond o pola iza ion
o he elec ode, o o elec ochemical changes in he
biological su ounding i sel . These changes could
po en ially be oxic o o he wise spu ious. Fo his
eason, i ually all neu os imula ion p o ocols o
bo h basic esea ch and implan ed biomedical de ices
ely on biphasic ope a ion [6]. Cha ge-balanced,
ca hodic-leading pulses a e he s anda d [9]. This
way, he o al injec ed ca hodic cha ge in he i s
phase is equalized by an equal-cha ge anodic phase.
In p inciple, any elec ochemical eac ion p oduc s
o med on he elec ode du ing he leading phase
should be eoxidized du ing he second, anodic, phase
[7,10]. De e mining sa e limi s o cha ge injec ion
is he opic o deba e and se e al empi ical no ms
ha e been sugges ed o a ious in i o applica ions.
A numbe o s udies ha e conside ed he e e s-
ibili y o a adaic elec ochemis y du ing biphasic
pulsing [11]. Nea ly all o hese s udies ocus on es -
ing co osion o he me al elec ode i sel , o ol age
excu sions beyond he wa e -spli ing window, and
hus hyd ogen e olu ion o oxygen e olu ion eac-
ions [7,11,12]. The possibili y o oxygen educ ion,
occu ing du ing he ca hodic s imula ion phase, has
ecei ed ela i ely li le a en ion [13,14], despi e he
ac ha he modynamically i is much mo e a ou ed
han hyd ogen e olu ion (by a leas 1.23 V). Oxygen
is p esen in all physiological luids. While i is ans-
po ed in heme-bound o m in he ascula u e, in
exci able neu al issues, oxygen a els h ough he
ex acellula and in acellula space by passi e di u-
sion. Neu al issue oxygena ion le els a y depending
on species, anaes hesia, loca ion. The highes possible
equilib ium concen a ion o dissol ed O2is abou
250 µM. This is he same le el o oxygena ion as wha
is expec ed o a con aine o wa e open o ambi-
en 21% a mosphe ic oxygen. This condi ion also
applies o mos in i o expe imen s. We he e o e
pe o m his s udy wi h his 21% e e ence poin ,
since i applies o in i o condi ions and wi h espec
o in i o his assump ion mi o s he ‘bes case’
scena io.
To ou knowledge, he e a e ou published s ud-
ies which conside oxygen educ ion eac ions (ORR)
on neu os imula ion elec odes, p ima ily on P and
Au [13–16]. These all ely on some o m o ansien
elec ochemical measu emen echniques o es ima e
i e e sible cha ge ans e o O2, and all hese s ud-
ies ag ee ha a subs an ial ac ion o ca hodic cu -
en can be i e e sibly ans e ed o oxygen ( epo -
ed anges be ween 5% and 80%). None o hese
s udies quan i ied he gene a ion o pe oxide ia
wo-elec on educ ion, o ac ual oxygen concen a-
ion changes. In con as o hese wo ks, we ha e
designed ou s udy o p obe bo h O2and H2O2con-
cen a ions in he icini y o he elec ode su ace o
quan i y he e ec s o ORR di ec ly, and ha e used
he same echnique o compa e eigh di e en neu -
os imula ion elec ode ma e ials we ha e p epa ed in
hin- ilm om: Ti, TiN, Au, I Ox, P , W, NiC , and
poly(3,4-e hylenedioxy hiophene):poly(s y ene sul -
ona e), sho ened as PEDOT:PSS. This lis is chosen
i ep esen s common elec ode ma e ials used in
bo h neu oscience/elec ophysiology esea ch as well
as biomedical de ices. In addi ion, a sample o com-
me cial high-su ace a ea TiN was es ed [17]. This
way, we can es ablish o wha ex en i e e sible ORR
can occu a neu os imula ion elec odes, and com-
pa e and con as di e en elec ode ma e ials. To
co obo a e expe imen ally-measu ed O2and H2O2
concen a ions, we ha e also es ablished ini e ele-
men simula ions o unde s and he geome y o con-
cen a ion g adien s o e ime.
2. Ma e ials and me hods
2.1. Model s imula ion elec ode p epa a ion
Mic oscope slides (3 ×1) inch2we e cleaned acco d-
ing o es ablished me hods, ea ed wi h oxygen
plasma, and hen spu e coa ed wi h a 100 nm laye
o Ti using a Kau man ion-beam sou ce (IBS). The
Ti ac s as he common conduc ing laye below all
s udied samples, as i has excellen adhesion on glass
and is a sui able unde laye o all he s udied ma e -
ials. Pla inum (60 nm) is deposi ed using DC mag-
ne on spu e ing. W (60 nm) was deposi ed using
he same spu e ing sys em. TiN (60 nm) is eac -
i ely spu e ed om a Ti a ge using wo Kau man
IBSs, and will be e e ed o in his a icle as IBS_TiN.
The p ima y IBS is used o spu e ing om Ti a ge
employing A and N2plasma, while he seconda y
2
J. Neu al Eng. 19 (2022) 036045 J Ehlich e al
IBS is used o subs a e bomba dmen wi h ions
om pu e N2plasma. Au is also deposi ed using
p ima y IBS, o a hickness o 60 nm. NiC (60 nm)
was p epa ed by he same p ima y IBS, using pu e
A . I Oxwas ob ained ia DC eac i e magne on
spu e ing in an A /O2plasma (100 nm) acco d-
ing o p e ious published me hods [18]. PEDOT:PSS
(PH1000 o mula ion om Cle ios, plus 5 w % e hyl-
ene glycol, 0.1 w % 4-dodecylbenzenesul onic acid
and 1 w % (3-glycidyloxyp opyl) ime hoxysilane)
was spin-coa ed a 3000 pm and annealed o 1 h a
130 ◦C (gi ing oughly 100 nm hickness as measu ed
by s ylus p o ilome y). As a e e ence, we cha ac-
e ize comme cial TiN used in mul ielec ode a ays
(Mul ichannel Sys ems GmbH). These ilms a e abou
600 nm hick. To dis inguish his TiN om ou
in-house p epa ed samples, we will e e o i as
MCS_TiN.
2.2. Elec ochemical cell
A homemade cus om elec ochemical cell allowed
co ec posi ioning and cha ac e iza ion o he elec-
odes unde in es iga ion. The cell was made om
clea 5 mm hick ac ylic shee s and a mic oscope
glass slide as a on window. The in e nal olume
o he cell was 9.3 ml. Mic oscope slides wi h depos-
i ed model s imula ion elec odes we e cu o (1 ×1)
inch2pieces and ho izon ally inse ed h ough a igh
opening on he side o he cell. Polydime hylsilox-
ane was used o ix and seal he sample in posi ion
and o p e en any elec oly e leakage. The exposed
a ea o each elec ode was masked using a 70 µm
hick poly inylchlo ide oil (Mini onic elek onik
GmbH). The ci cula opening in he oil had a dia-
me e o 3 mm, de ining he elec ode unde es
a ea (ac i e elec ode a ea =0.0706 cm2). The cell
was equipped wi h a P wi e coil as coun e (ac i e
a ea ∼7 cm2) and an Ag/AgCl as e e ence elec odes,
wo openings p o ided access o he O2o H2O2
senso and a e lon ube o O2/N2pu ging. The
senso was placed in icini y o he exposed ac i e
elec ode ma e ial (dis ance =200 µm). The coun e
elec ode is oughly 5 mm away om his a ea, a he
op o he cell. Ca e should be aken ha his dis ance
is su icien o ensu e ha any e en ual p oduc s on
he coun e elec ode do no a ec he measu emen .
The cell was moun ed on a submic ome e -p ecision
XYZ s age (Tho Labs). Using a digi al mic oscope
(Q-SCOPE 20200-P), he senso ip was posi ioned
using XYZ s age o he poin o ligh con ac in he
middle o he sample and hen mo ed o he 200 µm
dis ance in he Zdi ec ion ( igu es 1(b) and S1). The
cyclic ol amme y (CV) elec ochemical cha ac e iz-
a ion was ca ied ou in a la ge cell ( om Redox.me)
using an elec ochemical ac i e a ea o 1 cm2, in a
h ee-elec ode con igu a ion ha ing Ag/AgCl as e -
e ence elec ode and P wi e as coun e elec ode in
a ange o po en ials using 0.1 M phospha e bu e
saline (PBS) solu ion as elec oly e.
2.3. Elec ochemical measu emen s—DC and AC
condi ions
CV and ch onoampe ome y (DC condi ions) we e
applied using an I ium Pocke STAT2 po en ios a .
Fo biphasic pulsing, a Digi ime DS4 biphasic con-
s an cu en isola ed s imula o was used, wa e-
o m was d i en by a PicoScope 3404D oscilloscope
wi h a buil -in unc ion gene a o . T ansien ol ages
we e eco ded du ing biphasic pulsing by measu -
ing ol age using he oscilloscope inpu (1 MΩinpu
impedance) be ween he elec ode unde es and
an Ag/AgCl e e ence elec ode. Da a we e collec-
ed in oxygena ed and deoxygena ed condi ions. The
ela i ely low inpu impedance is selec ed in o de
o p o ide a shun esis ance o p e en po en ial
a che ing and hus p e en he in oduc ion o a DC
bias o se ol age du ing AC pulsing expe imen s.
2.4. Cla k elec ode O2and H2O2quan i ica ion
Local oxygen and pe oxide concen a ions we e
measu ed in si u du ing elec ochemical meas-
u emen s desc ibed abo e in sec ion 2.2 using a
ou -channel mic oampe ome ic ampli ie sys em
(TBR4100, Wo ld Scien i ic Ins umen s), wi h ou -
channel analog-digi al con e e boa d (LabT ax,
Wo ld Scien i ic Ins umen s). The espec i e senso
p obes used we e ISO-HPO-2 and ISO-OXY-2. The
O2senso was kep cons an ly pola ized a a bias
o 700 mV, meanwhile he H2O2speci ic senso a
450 mV. The senso s we e always calib a ed be o e
he measu emen o an indi idual ma e ial ollowing
he p ocedu e epo ed in he ins uc ion manual.
The d op in O2concen a ion o inc ease in H2O2
was acked by LabSc ibe so wa e (Wo ld Scien i ic
Ins umen s). I should be no ed ha he O2senso
unc ions along he classic Cla k-elec ode mech-
anism whe e oxygen is educed a he sensing elec-
ode, and oxygen eaches he senso ia an oxygen-
pe meable memb ane. The pe oxide senso , on he
o he hand, ope a es ia he oxida ion o H2O2, and is
he e o e also c oss-sensi i e o dissol ed H2( ia he
H2oxida ion eac ion). Fo expe imen s whe e ca h-
odic pola iza ion o he elec ode unde es esul s
in H2e olu ion, he senso signal can egis e a alse
posi i e H2O2signal. Cau ion should he e o e be
aken o es elec odes in deoxygena ed elec oly es
o es ablish he ca hodic wa e -spli ing onse . Pe ox-
ide senso eadings should only be accep ed wi hin
he ange be o e ca hodic wa e spli ing begins.
2.5. Fini e elemen analysis o ORR a s imula ion
elec odes
Simula ion o oxygen and pe oxide di usion was
conduc ed wi h ini e elemen me hod, implemen ed
in COMSOL 5.5 so wa e package and anspo o
dilu ed species module (www.comsol.com/p oduc -
download).
The expe imen al se up was ep oduced wi hin
an axisymme ic 2D model. A e he e olu ion, he
3
J. Neu al Eng. 19 (2022) 036045 J Ehlich e al
Figu e 1. (a) Schema ic o capaci i e and a adaic cha ge ans e ha can occu a an elec ode in e ace du ing he ca hodic
phase. A a ca hodically-pola ized elec ode, cha ge can accumula e in a capaci i e double laye (Cdl), o can be ans e ed in a
a adaic eac ion o a species in solu ion (Z x). Oxygen educ ion eac ions (ORR), Z 1−Z 3, a e he modynamically a ou ed
o e he hyd ogen e olu ion eac ion Z 4. Oxygen can unde go a ou -elec on, Z 1, o 2 +2 elec on educ ion pa hway
(Z 2+Z 3) o yield wa e as a p oduc . The wo-elec on pa hway Z 2p oduces hyd ogen pe oxide, while Z 3consumes hyd ogen
pe oxide. All ORR pa hways Z 1−Z 3, lead o deple ion o oxygen concen a ion nea he elec ode, and Z 2can esul in ne
accumula ion o hyd ogen pe oxide. (b) Expe imen al se up o es ing ORR occu ing a an elec ode in e ace. In his
con igu a ion, he ol age/cu en o he elec ode unde es is con olled in a h ee-elec ode con igu a ion wi h a po en ios a ,
o wi h a biphasic cu en s imula o . A Cla k elec ode senso is used o p obing he oxygen o pe oxide concen a ion a a ixed
posi ion nea he elec ode unde es (200 µm om he su ace). The whole chambe is enclosed o allow expe imen s unde ai
(21% O2), 100% O2, o 100% N2. Pho og aphs o he se up wi h he adjus able posi ion be ween he elec ode unde es and
senso a e shown below he schema ic. (c) DC mode p o ocol used o es ing elec odes in ol es s eps o ca hodic po en ials om
0 o −0.7 V s. Ag/AgCl. These example esul s a e o an Au elec ode. Po en ial is held o 600 s and cu en is egis e ed. ORR
cu en s s a a highly ca hodic alues and dec ease o di usion-limi ed equilib ium alues wi hin ens o seconds. O e he
cou se o 600 s, he senso is used o egis e he local O2o H2O2concen a ion. (d) AC mode in ol es cha ge-balanced
ca hodic-leading cu en pulses 250 µs pe phase, 10 Hz modula ion. In e pulse ime is a ied 0, 20, 50 µs; and h ee ampli udes
a e es ed: 10, 20, 30 µC cm−2/phase. These example aces a e eco ded o Au elec odes.
compu a ional model ob ained a cylind ical shape
illed wi h wa e -based elec oly es ( igu e 5(a1)).
Fa adaic eac ions occu on he a ea o a ca hodic
pixel loca ed on he bo om. The diame e o pixel
d=3 mm was adop ed om he expe imen . The
wa e elec oly e domain sp eads on 5 mm om he
s imula ion elec ode, hus making he diame e o
he elec oly e cylinde 13 mm and i s heigh 5 mm
( igu e 5(a1)). The model con ained wo a iables: cO2
and cH2O2, which goes o a concen a ion o dissol ed
4

J. Neu al Eng. 19 (2022) 036045 J Ehlich e al
oxygen and hyd ogen pe oxide. Ini ial alues we e
assigned as csa u a ed
O2=300µM and cH2O2=0µM.
We simula ed bo h wo- and ou -elec on eac-
ion pa hways sepa a ely. In case o wo-elec on case,
we conside ed he possibili y o u he educ ion o
pe oxide in o wa e , o so called 2 +2 eac ion pa h-
way. ORR was modelled ia he in oduc ion [19] o
he empi ical a adaic e iciency unc ion (1), which
modi ies luxes o oxygen and pe oxide:
e = 1+ 2
Cnea pixel
O2
Csa u a ed
O2
,(1)
whe e Cnea pixel
O2is oxygen concen a ion, measu ed
a 10 nm om he s imula ion elec ode. The al-
ues o cons an s 1and 2 ange be ween 0 and 1
( 1+ 2=1) and depend on he ype o ca hode.
I he cons an 2is bigge han ze o, he esul ing
lux o hyd ogen pe oxide will dec ease wi h oxygen
deple ion.
The di usion equa ions go e ned he change o
oxygen and pe oxide concen a ion
dci
d +∇ · Ji=0,(2)
Ji=−Di∇ci,(3)
whe e igoes o O2and H2O2and Jia e luxes o oxy-
gen and pe oxide. Di usion coe icien s a 23 ◦C a e:
DH2O2=1.8 ×10−9m2s−1;DO2=2.5 ×10−9m2s−1
[20,21].
P oduc ion and consump ion o dilu ed O2and
H2O2molecules we e ca ied ou h ough bounda y
luxes and can be ound wi h all o he bounda y con-
di ions in igu e 5(a2). The model conside ed oxygen
educ ion in o hyd ogen pe oxide in a io 1:1 wi h
a possible co ec ion by a adaic e iciency unc ion.
Bounda y luxes RO2and RH2O2we e de ined as in he
ecen wo k o Abdullae a e al [19]:
RH2O2=FI
2A× e ,(4)
RO2=−FI
2A×( e +1− e
2),(5)
whe e Iis applied cu en (di e en in case o
DC o AC simula ions), Fis Fa aday cons an ,
A=3.14 ×1.52mm2is he s imula ion elec ode
a ea, e is dimensionless a adaic e iciency unc ion,
de ined by cons an s 1and 2and measu ed O2con-
cen a ion Cnea pixel
O2on he 10 nm dis ance om a p o-
duc ion pixel.
In he case o he ou -elec on ORR eac ion
pa hway, H2O2does no pa icipa e in he p ocess,
and di usion equa ions (2) and (3) is sol ed only o a
concen a ion o oxygen wi h luxes o O2molecules.
Assuming he absence o he e e sed eac ion, oxy-
gen consump ion is de ined by
RO2=−FI
4A.(6)
Mo e de ails on he model cons uc ion can be ound
in he supplemen a y in o ma ion appendix 1.
3. Resul s
3.1. CV cha ac e iza ion o elec odes
CV was used o cha ac e ize bo h capaci i e cha -
ging and a adaic p ocess occu ing a each elec-
ode ma e ial in PBS solu ion ( igu e 2). The sys-
em has been oxygena ed and de-oxygena ed (N2gas
low) o moni o he elec ochemical beha iou o he
ma e ials unde h ee di e en condi ions (21% O2
(a mosphe ic p essu e), 100% N2and 100% O2).
Compa ing CV cu es om de-oxygena ed condi-
ions e sus oxygena ed ones can e eal which peaks
a e o igina ing om ORR. The a adaic/capaci i e
cha ging beha iou di e s ma kedly depending on
he elec ode ma e ial, ne e heless in all cases i is
possible o obse e an inc ease in ca hodic cu en
co esponding o oxygena ion. ORR is appa en in
CVs o ol ages lowe han +100 mV. Fo each
ma e ial, we chose a smalle ol age window o meas-
u e in a egion whe e capaci i e cha ging domina es,
allowing es ima ion o he double-laye capaci ance
o each elec ode ma e ial ( able 1; supplemen a y
igu e S2). CV scans o capaci ance de e mina ion
we e ob ained using a 1 mV s ep and cu en a e -
aging acco ding o me hods desc ibed by Wel in and
Kieninge [22].
3.2. DC ch onoampe ome y and di ec
measu emen s o ORR p ocesses
While CV is use ul o sc een o possible ORR and
o he a adaic eac ions, i does no e eal he mag-
ni ude o di usion-limi ed oxygen educ ion cu -
en s, o quan i y he concen a ion o ORR p oduc s.
To accomplish his, we combine DC ch onoampe o-
me y expe imen s wi h simul aneous eco ding o
oxygen and hyd ogen pe oxide concen a ions a a
ixed poin nea he s imula ion elec ode su ace (a
a heigh o 200 µm). In hese expe imen s, we meas-
u e cu en o e ime du ing he applica ion o a con-
s an po en ial ( om +0.1 V o −0.9 V, by 0.1 V
s eps, depending on he ma e ial in es iga ed) o e
a de ined ime (600 s) on each di e en s imula ion
elec ode. The p o ocol and example esul s a e plo -
ed in igu e 1(C). A po en ials which a e oo anodic
o educe oxygen a a gi en elec ode ma e ial, no
sus ained ch onoampe ome ic cu en s a e measu -
able. Once he onse po en ial is eached, sus ained
ca hodic ORR cu en is clea . Onse po en ials o
ORR p ocesses can be ound in able 1. Fo e e ence,
5
J. Neu al Eng. 19 (2022) 036045 J Ehlich e al
Figu e 2. Cyclic ol amme y o neu os imula ion elec odes in PBS solu ion. CVs o Ti, IBS_TiN, MCS_TiN, P , Au, NiC , W,
I Oxand PEDOT:PSS in 100% oxygena ed (blue), 21% oxygena ed (ambien ai , black), and 0% oxygena ed (100% N2pu ged,
ed) condi ions. Scan a e =100 mV s−1; Po en ial, E, ange +0.7 V o −0.9 V e sus Ag/AgCl. Due o la ge di e ences be ween
ma e ials in e ms o hei capaci i e and a adaic cu en s, he espec i e plo s ha e di e en cu en densi y y-axes.
Table 1. ORR p ocess onse po en ials o each elec ode ma e ials in PBS solu ion, and elec ochemical double-laye capaci ance
es ima ed o CVs measu ed in a non a adaic po en ial window (CVs shown in igu e S2). Onse po en ials o ORR and ca hodic
wa e -spli ing a e es ima ed om he ch onoampe ome y ol age-s ep p o ile and co esponding ampe ome ic con i ma ion o
oxygen educ ion/pe oxide gene a ion. Po en ials a e gi en e sus Ag/AgCl, [Cl−]=0.12 M.
Ma e ial
Oxygen
educ ion onse
(mV s. Ag/AgCl)
Hyd ogen pe oxide
e olu ion onse
(mV s. Ag/AgCl)
Hyd ogen e olu ion
eac ion onse
(mV s. Ag/AgCl)
Double-laye
capaci ance
(µF cm−2)
Ti −700 −700 −900 21
IBS_TiN −500 −600 −1400 22
MCS_TiN −500 −500 −1400 669
P +100 0a−700 97
Au −300 −300 −700 56
NiC −400 −400 −1000 24
W−500 −600 −1000 69
I Ox0 0 −700 530
PEDOT:PSS −600 −600 −1600 197
aOnly ace amoun s o pe oxide de ec ed.
each sample is measu ed also in deoxygena ed elec-
oly e o es ablish he ca hodic onse po en ial o
wa e spli ing/H2e olu ion. These H2e olu ion
onse s, also gi en in able 1, a e always mo e nega i e
han measu ed ORR po en ials. Ch onoampe ome y
e eals a peak ca hodic ORR cu en which hen
decays o a s eady-s a e ca hodic cu en . We de ine
he s eady-s a e cu en alue as he cu en measu ed
a he =600 s imepoin . The s eady-s a e cu en
as a unc ion o applied po en ial o each ma e ial
6
J. Neu al Eng. 19 (2022) 036045 J Ehlich e al
Figu e 3. DC ch onoampe ome y wi h ol age-s ep p o ocol and esul an O2d op and H2O2inc ease o each elec ode
ma e ial. All po en ials a e e sus Ag/AgCl. Each poin along he J(V) ace (black) is he equilib ium cu en a =600 s. The ed
ace is he lowes measu ed %O2a a posi ion 200 µm abo e he elec ode su ace o e he 600 s pe iod. A d op o −21% is
comple e deoxygena ion, while 0% co esponds o a no mal 21% oxygen sa u a ion. The blue aces ma k he [H2O2] in mM
measu ed a each po en ial a e 600 s, a a posi ion 200 µm abo e he elec ode su ace. Pe oxide eco dings a ol ages whe e
ca hodic H2e olu ion occu s on a gi en ma e ial a e no plo ed, since H2e olu ion in e e es wi h he pe oxide senso .
is plo ed in he black aces shown in igu e 3.
Simul aneously o eco ding ch onoampe ome y,
he O2/H2O2ampe ome ic senso is ac i e and
eco ding he espec i e O2o H2O2concen a ion
alue a a poin in he elec oly e di ec ly abo e he
elec ode-unde - es . The measu ed alues o O2o
H2O2concen a ion a he =600 s imepoin a e
shown in igu e 3as he ed and blue aces, espec -
i ely. A change o −21% O2co esponds o a si ua ion
whe e he senso eco ds 0% oxygen, i.e. he calib-
a ed lowes limi o a ully-deoxygena ed solu ion. In
be ween each ol age-s ep, he elec oly e in he cell is
eplaced wi h esh, oxygena ed elec oly e. I is no e-
wo hy ha all elec ode ma e ials can p oduce qui e
hypoxic condi ions, wi h some such as P , Au, and
PEDOT:PSS eaching nea ly comple e deoxygena ion
a highe ca hodic po en ials. The ma e ials di e in
hei abili y o p oduce pe oxide om ORR. Au and
PEDOT:PSS, o ins ance, p oduce concen a ions
in o he millimola ange, while P gene a es ba ely-
de ec able ace amoun s o pe oxide. The eco ded
ch onoampe ome ic cu en s and accompanying
oxygen and pe oxide measu emen s used o cons uc
he plo s in igu e 3can be ound in supplemen a y
igu es S3–S11.
3.3. Di ec measu emen s o ORR p ocesses du ing
AC s imula ion p o ocols
Using he same con igu a ion o ampe ome ic sens-
ing o oxygen and pe oxide, we nex applied cha ge
balanced ca hodic-leading pulses o he elec odes
unde es . The pulse du a ion pe phase o cha ge
balanced ca hodic leading pulses used du ing expe -
imen s is kep cons an (250 µs) and he in e pulse
spacing is changed o ha e h ee di e en du a-
ions (0, 20, o 50 µs). We use a pe iod o 100 ms
(ƒ=10 Hz). We es ed h ee cha ge densi y alues:
10, 20, o 30 µC cm−2/phase. The o al ime o each
AC expe imen is 1200 s, compa ed wi h he DC ch o-
noampe ome y which was 600 s. This AC s imula-
ion p o ocol and accompanying examples o meas-
u ed ampe ome ic ansien s a e gi en in igu e 1(d).
The aw da a o he ampe ome ic aces o oxy-
gen and pe oxide o e he cou se o AC pulsing a e
shown in igu es S3–S11. In all cases, cha ge-balanced
biphasic pulses led o d ops in oxygen concen a ion
compa able in magni ude o hose ound unde he
mos ca hodic DC condi ions. The gene a ion o pe -
oxide as a byp oduc a ied based on ma e ial, wi h
I Oxand P p oducing only ace quan i ies, and Au
and PEDOT:PSS p oducing he mos . Su p isingly,
7
J. Neu al Eng. 19 (2022) 036045 J Ehlich e al
Figu e 4. Cha ge-balanced ca hodic-leading biphasic pulses 250/x/250, a 10, 20, o 30 µC cm−2 esul in ne oxygen educ ion,
and in he case o some ma e ials, app eciable H2O2gene a ion. The in e pulse spacing, x, o 0, 20, o 50 µs has li le e ec on he
ou come. The cha ge densi y also has minimal impac . The e o ba s ep esen one s anda d de ia ion a e aking he mean o
all condi ions ( h ee di e en cha ge densi ies × h ee di e en in e pulse spacings). (a) Peak O2change du ing he 1200 s pe iod;
(b) To al in eg a ed a ea-unde -cu e o consumed O2o e he cou se o he 1200 s pe iod; (c) Peak pe oxide concen a ion
eco ding o e he 1200 s pe iod; (d) To al p oduced pe oxide a ea-unde -cu e o e 1200 s. While O2concen a ion alues (a),
(b) should be conside ed an accu a e, quan i a i e esul , he pe oxide alues in (c), (d) should be conside ed semi-quan i a i e,
due o pe oxide senso c oss-sensi i i y o H2. Ma e ials like Au do p oduce some amoun o H2du ing he ca hodic phase.
he magni ude o he cha ge densi y applied in he
pulse had li le e ec on he peak and o al amoun s o
ORR obse ed. The in e pulse spacing has a minimal
con ibu ion on he o e all change in %O2o H2O2
concen a ion. Since bo h pulse ampli ude and in e -
pulse ha e no de ini i e impac on ORR le els, he
da a p esen ed in igu e 4 ea u e he nine di e en
AC condi ions all pooled o gi e a mean o a gi en
ma e ial, wi h he s anda d de ia ion be ween con-
di ions exp essed by he e o ba . Oxygen deple ion
and pe oxide gene a ion a e exp essed by bo h peak
alues, as well as he a ea-unde -cu e (AUC), he
in eg al o he measu ed ampe ome ic signal o e
ime (1200 s) signi ying he o al amoun o oxygen
educed/pe oxide gene a ed. Du ing biphasic cu en
pulses, ol age ansien s we e collec ed ( igu e S12).
4. Discussion
4.1. ORR on elec ode ma e ials du ing CV and DC
measu emen s
In his s udy, we ha e decided o in es iga e ORRs on
commonly-used neu os imula ion elec ode ma e -
ials, aiming o ind ou o wha ex en hese eac-
ions a ec concen a ions o dissol ed oxygen and
hyd ogen pe oxide nea he elec ode. We selec ed
eigh ep esen a i e elec ode ma e ials which we ab-
ica ed in hin- ilm o m. Addi ionally, as a nin h
ma e ial we s udied a comme cial TiN sample wi h
high cha ge-injec ion capaci y. The logic behind his
s udy was o i s measu e CV in oxygena ed e sus
deoxygena ed elec oly es. This kind o cha ac e iz-
a ion unambiguously e eals he p esence o ORR.
Nex , we pe o med s ep- ol age ch onoampe ome -
ic measu emen s o es ablish he magni ude o equi-
lib ium ORR cu en densi y ha is possible a a
gi en elec ode ma e ial. Du ing he ch onoampe o-
me ic measu emen s, O2o H2O2concen a ion is
egis e ed in eal ime in he solu ion abo e he elec-
ode su ace. The magni ude o ORR cu en s, as
well as he onse po en ials, a ied g ea ly be ween he
samples, which is why he cu en densi y y-axes plo -
ed in igu es 2and 3a e all di e en . These di e -
ences o igina e because o he elec oca aly ic p ope -
ies o each ma e ial wi h espec o ORR. Measu ed
pe oxide concen a ions in his s udy a ied o e ou
o de s o magni ude, om 1 µM o 10 mM. The use
o a Cla k- ype ch onoampe ome ic de ec ion sys-
em is, o he bes o ou knowledge, he only me hod
o measu e pe oxide concen a ions o e such a la ge
8
J. Neu al Eng. 19 (2022) 036045 J Ehlich e al
[12] Shephe d R K, Ca e P, Dal ymple A, Enke Y L, Wise A K,
Nguyen T, Fi h J, Thompson A and Fallon J B 2021
Pla inum dissolu ion and issue esponse ollowing
long- e m elec ical s imula ion a high cha ge densi ies J.
Neu al Eng. 18 036021
[13] Mo on S L, Da oux M and Mo ime J T 1991 The ole o
oxygen educ ion in elec ical s imula ion o ne ous issue
P oc. Annual Con . Enginee ing in Medicine and Biology
Socie y ol 13 pp 552–3
[14] Cogan S F, Eh lich J, Plan e T D, Ginge ich M D and
Shi e D B 2010 Con ibu ion o oxygen educ ion o cha ge
injec ion on pla inum and spu e ed i idium oxide neu al
s imula ion elec odes IEEE T ans. Biomed. Eng. 57 2313–21
[15] Musa S, Rand D R, Ba ic C, Ebe le W, Nu in B and
Bo ghs G 2011 Coulome ic de ec ion o i e e sible
elec ochemical eac ions occu ing a P mic oelec odes
used o neu al s imula ion Anal. Chem. 83 4012–22
[16] Dijk G, Ruig ok H J and Conno R P O 2021
PEDOT:PSS-coa ed s imula ion elec odes a enua e
i e e sible elec ochemical e en s and educe cell
elec ope meabiliza ion Ad . Ma e . In e aces 82100214
[17] Ege U, Schlosshaue B, Fenn ich S, Nisch W, Fej l M,
Kno T, Mülle T and Hämme le H 1998 A no el
o gano ypic long- e m cul u e o he a hippocampus on
subs a e-in eg a ed mul ielec ode a ays B ain Res.
P o ocols 2229–42
[18] an Ooyen A, Topalo G, Ganske G, Mokwa W and
Schnakenbe g U 2009 I idium oxide deposi ed by pulsed
dc-spu e ing o s imula ion elec odes J. Mic omech.
Mic oeng. 19 074009
[19] Abdullae a O S, Sahaliano I, Ejneby M S, Jakešo á M,
Zozoulenko I, Liin S I and Głowacki E D 2022 Fa adaic pixels
o p ecise hyd ogen pe oxide deli e y o con ol M- ype
ol age-ga ed po assium channels Ad . Sci. 92103132
[20] Wise D L and Hough on G 1966 The di usion coe icien s o
en sligh ly soluble gases in wa e a 10–60 ◦CChem. Eng.
Sci. 21 999–1010
[21] an S oe-biezen S A M M, E e ae s F M, Janssen L J J J and
Tacken R A 1993 Di usion coe icien s o oxygen, hyd ogen
pe oxide and glucose in a hyd ogel Anal. Chim. Ac a
273 553–60
[22] Wel in A and Kieninge J 2021 Elec ochemical me hods o
neu al in e ace elec odes J. Neu al Eng. 18 052001
[23] Song C and Zhang J 2008 Elec oca aly ic Oxygen Reduc ion
Reac ion PEM Fuel Cell Elec oca alys s and Ca alys Laye s:
Fundamen als and Applica ions ed J Zhang (London:
Sp inge ) pp 89–134
[24] Donahue M J, Sanchez-Sanchez A, Inal S, Qu J, Owens R M,
Mece eyes D, Mallia as G G and Ma in D C 2020 Tailo ing
PEDOT p ope ies o applica ions in bioelec onics Ma e .
Sci. Eng. R140 100546
[25] Mi aka E e al 2019 Elec oca aly ic p oduc ion o hyd ogen
pe oxide wi h Poly(3,4-e hylenedioxy hiophene) elec odes
Ad . Sus ain. Sys . 31800110
[26] Schaldach M, Hubmann M, Weikl A and Ha d R 1990
Spu e -deposi ed TiN elec ode coa ings o supe io
sensing and pacing pe o mance Pacing Clin. Elec ophysiol.
13 1891–5
[27] Ka souna os I, Schneide W B, Meie J C, Benedik U,
Biede mann P U, Aue A A and May ho e K J J 2012
Hyd ogen pe oxide elec ochemis y on pla inum: owa ds
unde s anding he oxygen educ ion eac ion mechanism
Phys. Chem. Chem. Phys. 14 7384
[28] Ki ino T 2000 Delayed neu onal dea h Neu opa hology
20 95–7
[29] Lip on P 1999 Ischemic cell dea h in b ain neu ons Physiol.
Re . 79 1431–568
[30] Sies H and Jones D P 2020 Reac i e oxygen species (ROS) as
pleio opic physiological signalling agen s Na . Re . Mol. Cell
Biol. 21 363–83
[31] Sies H 2017 Hyd ogen pe oxide as a cen al edox signaling
molecule in physiological oxida i e s ess: oxida i e eus ess
Redox Biol. 11 613–9
[32] Gampe N, Zaika O, Li Y, Ma in P, He nandez C C,
Pe ez M R, Wang A Y C, Ja e D B and Shapi o M S 2006
Oxida i e modi ica ion o M- ype K +channels as a
mechanism o cy op o ec i e neu onal silencing EMBO J.
25 4996–5004
[33] Lodola F, Ros i V, Tullii G, Desii A, Tapella L, Ca a si P,
Lim D, Moccia F and An ognazza M R 2019 Conjuga ed
polyme s op ically egula e he a e o endo helial
colony- o ming cells Sci. Ad . 5eaa 4620
15