J. ROZMAN et al.: MEASUREMENT OF BIO-IMPEDANCE ON AN ISOLATED RAT SCIATIC NERVE ...
797–804
MEASUREMENT OF BIO-IMPEDANCE ON AN ISOLATED RAT
SCIATIC NERVE OBTAINED WITH SPECIFIC CURRENT
STIMULATING PULSES
MERITEV BIOIMPEDANCE NA IZOLIRANEM @IVCU
ISCHIADICUS PRI PODGANI, VZBUJENEM S POSEBNIMI
TOKOVNIMI STIMULACIJSKIMI IMPULZI
Janez Rozman1,3, Monika C. @u`ek2, Robert Frange`2, Samo Ribari~3
1Center for Implantable Technology and Sensors, ITIS d. o. o., Lepi pot 11, 1000 Ljubljana, Slovenia
2Institute of Physiology, Pharmacology and Toxicology, Veterinary Faculty, University of Ljubljana, Gerbi~eva 60, 1000 Ljubljana, Slovenia
3Institute of Pathophysiology, University of Ljubljana, ZaMedical Faculty, lo{ka 4, 1000 Ljubljana, Slovenia
samo.ribaric@mf.uni-lj.si
Prejem rokopisa – received: 2015-09-30; sprejem za objavo – accepted for publication: 2015-10-26
doi:10.17222/mit.2015.307
In this study we designed and tested a four-probe, bio-impedance measurement set-up for peripheral nerves, based on the Red
Pitaya open-source measurement and control tool. The set-up was tested on an isolated rat sciatic nerve (RSN) while it was
stimulated with specific current stimulating pulses. The measurements tested the hypothesis that the specific waveform of a
stimulating pulse elicits current differences at the double layer (DL) interface between the platinum (Pt) stimulating electrode
and the nerve tissue. Impedance spectroscopy was used to electrically characterize the interface between the Pt electrode and the
nerve tissue and measure the interface electrical impedance (Z). An analysis of the frequency response and the impedance, with
specific current stimulating pulses, characterised the structure and the composition-related electrical properties of the RSN. An
analysis of the voltage responses (VRs), measured at the same time, showed that the maximum negative polarization across the
electrode-electrolyte interface (Emc) and the maximum positive polarization across the electrode-electrolyte interface (Ema) did
not exceed the safety limits for water electrolysis. We conclude that the voltage and current changes, elicited at the DL of the in-
terface between the Pt stimulating electrode and the nerve tissue, do not lead to tissue damage. Based on the obtained
electrophysiological results we conclude that the developed stimulating electrodes and the stimulus pattern could act as a useful
tool for developing nerve-stimulating electrodes.
Keywords: electrical impedance, electrical impedance spectroscopy, bio-impedance, voltage response, current source, platinum,
rat sciatic nerve, electrode-nerve tissue interface
Izdelali in preizkusili smo napravo za {tiri-to~kovno merjenje elektri~ne impedance (Z) perifernih `ivcev, katere osnova je
odprtokodna merilna in kontrolna konzola (Red Pitaya). Naprava je bila preizku{ana na izoliranem `ivcu ischiadicus pri podgani
(RSN) tako, da smo `ivec lahko dra`ili z izbranimi tokovnimi stimulacijskimi impulzi. Preverili smo hipotezo, da izbrani
stimulacijski impulzi izzovejo razlike na dvojni plasti (DL) prehoda med platinovo (Pt) stimulacijsko elektrodo in `iv~nim
tkivom. Z metodo impedan~ne spektroskopije smo izmerili Z na prehodu med Pt elektrodo in `iv~nim tkivom. Analiza
frekven~nega odgovora RSN, medtem ko je bil stimuliran s posebnimi tokovnimi stimulacijskimi impulzi, je podala od sestave
in zgradbe odvisne elektri~ne lastnosti `ivca. Nadalje je analiza so~asno merjenega napetostnega odgovora (VR) pokazala, da
nobena od obeh, tako najve~ja negativna polarizacija Emc kot tudi najve~ja pozitivna polarizacija Ema, na prehodu med elektrodo
in elektrolitom, ni presegla varne meje pri kateri lahko pride do elektrolize vode. Zaklju~ujemo, da napetostne in tokovne
razlike, ki so nastale na DL prehodu med Pt stimulacijsko elektrodo in `iv~nim tkivom, nimajo {kodljivega u~inka na `iv~no
tkivo. Kon~ni sklep raziskave je, da je mogo~e pridobljene elektrofiziolo{ke rezultate koristno uporabiti pri nadaljnem
na~rtovanju in razvoju stimulacijskih elektrod.
Klju~ne besede: elektri~na impedanca, elektri~na impedan~na spektroskopija, bioimpedanca, napetostni odgovor, tokovni vir,
platina, kol~ni `ivec podgane, prehod med elektrodo in `iv~nim tkivom
1 INTRODUCTION
Tissue-characterizing techniques based on electrical
impedance (Z) are being used to study the Z variations of
biological tissues over a range of frequencies. Some of
these analysing techniques provide Z values of the tissue
sample as a lumped estimation at a suitable frequency
range and the information on several complex bioelec-
trical phenomena occurring in cells and tissues under an
alternating (AC) electric current signal.1,2 In this regard,
a measurement of the complex Z in biological systems is
one of the developing technologies for monitoring and
determining the pathological and physiological status of
biological tissues.3 Namely, with AC electrical
stimulation, biological tissues produce a complex Z that
depends on the tissue composition, structures, health
status, and applied stimulus frequency.4
In biological tissues, plasma membranes of the cells
are composed of electrically non-conducting lipid
bilayers and an ion-conducting proteins channels.5,6 This
structure provides a capacitance to the applied AC signal
and contributes to the overall response of the biological
tissues by producing a complex Z, which is a function of
the tissue composition as well as the frequency of the
applied AC signal. Z is a measurement of the overall
Materiali in tehnologije / Materials and technology 50 (2016) 5, 797–804 797
UDK 621.317.33:616.833.58:599.323.452:543.42 ISSN 1580-2949
Original scientific article/Izvirni znanstveni ~lanek MTAEC9, 50(5)797(2016)
ability of a medium to conduct electrical current, defined
as the ratio of the voltage in an object to the current in a
conductive medium. Z is a complex quantity, which
consists of a real part (resistance) and an imaginary part
(reactance).7 Its units are ohms (
). In this regard,
Wtorek et al.8 presented the construction of a probe for
immittance spectroscopy based on the four-electrode
technique. They tested the probe in "in-vivo"
measurements on swine gluteal tissue.
The simplest set-up for measuring Z is the so-called
two-probe set-up, where a current source is used in order
to feed a known current (I) into the unknown Z. A volt-
age measurement using a voltage-measurement device
determines the voltage drop, which is assumed to be pro-
portional to the unknown Z. However, the resistivity of
cables and the contact resistance also fully contribute to
the voltage drop and to the measured voltage. As a re-
sult, the measurement result Z is actually higher than the
true Z.
In the diagnosis of neuromuscular diseases, for in-
stance, the Z of muscles can be measured to monitor pa-
tients with amyotrophic lateral sclerosis, muscular dys-
trophy and inflammatory myopathies. In clinical
practice, a surface-electrode Z measurement is used for Z
measurements of the skin tissue, which can be used for
monitoring various conditions relating to the physical or
medical state of a patient. The accuracy of a Z measure-
ment with surface electrodes can be degraded by un-
known contact impedances.
In order to avoid the disadvantages of the two-probe,
Z measurement set-up, the four-probe, Z measurement
set-up employs an additional pair of electrodes. In the
four-probe, Z measurement set-up, the current is fed
through two feeding electrodes1,4 (considered as drive
terminals) within the line of four equidistantly placed
electrodes, while the voltage drop is measured between
two measurement electrodes2,3 (considered as
measurement terminals). The voltage is measured with
an instrument that has a very high input Z, so that almost
no current flows through the measurement electrodes,
i.e., their cables and contact resistance play almost no
role in the measurement. Therefore, the measured
voltage is almost identical to the voltage drop at the
unknown Z.9 For many applications, the four-probe, Z
measurement set-up thus provides sufficient accuracy.
The measurement of Z in biological systems, such as
nerve tissue, with the four-probe measurement set-up,
eliminates the influence of the Z of the electrode/nerve
tissue interface on the nerve tissue Z. Namely, electrode
polarization is avoided and the contact Z is eliminated
from the measurement.10,11 However, the Z measurement
error cannot be fully eliminated due to current flow
through the measurement electrodes. As the use of Pt
electrodes in selective nerve-stimulation applications
(SNS) became indispensable, significant research was
dedicated to understanding the electrode-electrolyte
interface in “in vivo” and the electrode-nerve tissue
interface in "in vivo". Of particular importance to SNS is
Schwan’s limit of linearity: the voltage at which the
electrode system’s Z becomes nonlinear and is often
exceeded during SNS.12,13
The main goal of this study was to use the electrical
impedance spectroscopy (EIS) and voltage response
(VR) experimental results to enhance our understanding
of the effect of physical processes at the interface Z, with
the expressed purpose of improving the interface design
of future multi-electrode stimulating systems for SNS.
The measurements were performed to test the hypothesis
that a specific waveform of current stimulating pulse
elicits controlled and tissue-safe voltage differences at
the double layer (DL) of the interface between the Pt
stimulating electrode and the nerve tissue. These repeti-
tive voltage differences are potentially harmful since
over time they can prevent effective excitation of the
nerve tissue at the stimulating electrode due to electroly-
sis-induced nerve-tissue damage.
2 MATERIAL AND METHODS
2.1 Experimental design and set-up
Among the several custom-designed pieces of equip-
ment, the most important was a temperature-controlled
measuring chamber and proprietary stimulation/record-
ing cables (Figure 1d). The chamber was developed for
the two-probe and the four-probe electrophysiological
measurements. The bulk body of the chamber was ma-
chined out of Plexiglas, using a computerized numerical
control milling machine. The main part of the chamber
was a ladder with seven Pt hook-shape electrodes
mounted horizontally over the chamber aperture and
filled with a Krebs-Ringer solution (154-mM NaCl,
5-mM KCl, 2-mM CaCl2, 1-mM MgCl2, 5-mM HEPES,
11-mM D-glucose, pH 7.4). The distance between the
electrodes was 4.8 mm.
The hook-shape electrodes were manufactured from a
cold-drawn Pt wire (99.99 % purity) with a thickness of
1 mm. Pure Pt is commonly used as a stimulating elec-
trode material because it can effectively supply high-
density electrical charge to activate the neural tissue. Pt
is normally used in the form of a pure metal because im-
purities and alloying elements may adversely affect its
mechanical characteristics, working characteristics and
its stability against corrosion in physiological media. Pt
also has many physical properties that are of great value
for their use in the technology of implantable stimulating
electrodes. The contact surface of the Pt electrodes was
increased, relative to the electrodes’ geometric area, by
sanding with a metallurgical grade sandpaper number
1000.
Consequently, in the developed measuring system,
the two Pt hooks (electrode 1 and 7) of the ladder,
representing the first two points of the four-probe
measuring set-up, are used as shown in Figure 1d. For
measurements of the voltage drop elicited by the current
J. ROZMAN et al.: MEASUREMENT OF BIO-IMPEDANCE ON AN ISOLATED RAT SCIATIC NERVE ...
798 Materiali in tehnologije / Materials and technology 50 (2016) 5, 797–804
pulse, the two hooks (electrode 3 and 5) representing the
second two probes of the four-probe measuring set-up,
were used. The hook electrode 1 is connected to the high
current source output (IH) of the bio-impedance front-end
device (Figure 1c), while the hook electrode 7 is
connected to the low current source output (IL) of the
bio-impedance front-end device. Electrode 3 was con-
nected to the high-voltage input (VH) of the bio impe-
dance front-end device and electrode 5 was connected to
the low-voltage input (VL) of the bio impedance
front-end device. Before the rat’s sciatic nerve (RSN)
was placed on the ladder of electrodes, the chamber
canal was filled with saline to prevent the drying out of
the RSN. To reduce the polarization and voltage shifts,
the electrodes were soaked for 15 min to 30 min.
2.2 Nerve dissection and preparation
Z measurements were performed on two isolated
RSNs harvested from adult male Wistar rats, weighing
300 g to 350 g, purchased from Charles River. Animal
handling complied with the Prevention of Cruelty to
Animals law, which is consistent with the European
Community Directive 2010/63/EU and was approved by
the Institutional Review Board. The rats are euthanized
by CO2 and exsanguination. Dissection began with an
incision through the skin on the lateral side of the hind
limb from the knee to the hip region. To expose the RSN,
the superficial layer of the thigh muscle is carefully
dissected between the m. gluteus superficialis and m.
biceps femoris. The length of RSN available for
measurement depends on the point at which the RSN
divides into the peroneal, tibial and sural nerves. To
prevent nerve injury during handling, a thread was tied
around the distal end of the RSN. After the attached
tissue was removed from the RSN, the RSN was cut as
close to the knee joint as possible. Similarly, at the
proximal end, the RSN was cut as close as possible to
the spinal column. By doing so, the RSN obtained was
long enough (between 30 mm and 35 mm) to be in
contact with all the electrodes in the ladder.
Afterwards, the RSN was placed in a glass sili-
con-lined Petri dish filled with Krebs-Ringer solution.
As the buffer was used for cell physiology experiments,
it was gassed with 100 % oxygen for 10–20 min, so that
the RSN could recover from the dissection. The RSN
was transferred from the Petri dish into the chamber and
placed on the ladder so that it touched all the electrodes.
The RSN was kept moist with saline to prevent it from
drying out and rendering it useless for the experiment.
Finally, to prevent evaporation and drying of the tissue
the chamber was covered with a lid. All of the recordings
were made at an ambient temperature of 22–23 °C.
2.3 Measuring system
The Red Pitaya (Red Pitaya, Solkan, Slovenia) devel-
opment platform was used for the stimulating/measuring
set-up shown schematically in Figure 1. This platform
enables signal generation, signal acquisition and signal
processing.
The set-up used two built in D/A and A/D converter
inputs enabling measurements with a 14-bit resolution at
a 125 MHz/s sampling rate. This sampling rate could be
adjusted with a decimal/dividing factor to 1, 8, 64, 1024,
8192 and 65536, respectively. Red Pitaya was attached to
the Z front-end device as shown in Figure 1c. This de-
vice was composed of a voltage-controlled current
source (VCCS) based on the Howland method. The
VCCS was designed to provide a constant current signal
(ic), up to 4 mA, independent of the value of the attached
load within a frequency range from 1 Hz to 1 MHz. The
Z front-end device enabled simultaneous, four- and
two-probe Z measurements.3,7
2.4 Bio-impedance spectroscopy measurements
The EIS-based frequency response studies of Z, of
any material, can provide the material’s structure- and
J. ROZMAN et al.: MEASUREMENT OF BIO-IMPEDANCE ON AN ISOLATED RAT SCIATIC NERVE ...
Materiali in tehnologije / Materials and technology 50 (2016) 5, 797–804 799
Figure 1: Schematic diagram of the stimulating/measuring set-up: a) programming part of the Z measurement set-up, b) Red Pitaya,
c) Z front-end device, d) 3D view of the measuring chamber with a ladder of electrodes
Slika 1: Shematski prikaz stimulacijsko/merilne naprave: a) programska enota naprave za merjenje Z, b) Red Pitaya, c) vhodno izhodna naprava
(Z), d) tridimenzionalna slika merilne celice z lestvico elektrod
composition-dependent electrical properties as well as its
frequency response.14 In this study, EIS was used to esti-
mate the complex (Z(f)) and phase angle ((f)) of the
RSN at different frequencies fi (fi: f1, f2, f3,..., fn) by mea-
suring the surface voltage (V(f)) developed for a constant
low-amplitude, low-frequency alternating sinusoidal cur-
rent (I(f)) injected into a RSN using both the two- and
four-probe methods.15–18 Afterwards, the frequency-de-
pendent electrical bio-impedance Z(fi) was found as the
transfer function of the RSN, and thus the Z(fi) was cal-
culated by dividing the voltage data (V(fi)) measurement
by the applied current (I(fi)), as shown in the following
Equation (1):
Z(fi) = V(fi)/I(fi) (1)
The programming part of the Z measurement set-up,
shown in Figure 1a, was performed with a Lenovo T420
portable computer (Lenovo, Singapore) and MATLAB
R2007a software (The Mathworks Inc., USA). In this
way, communication and concomitant functional control
of the Red Pitaya was made and monophasic and
biphasic voltage pulses to drive (driving pulse) a VCCS
within the front-end device, having a specific waveform
shown in Figures 2a and 2b, were generated. An afore-
mentioned waveform of the stimulating pulse was
adopted from a recent study where this waveform was
tested by selective nerve stimulation of the isolated por-
cine left cervical vagus nerve segment.19 Afterwards,
driving pulses were delivered to the VCCS, where they
were converted into constant current pulses by means of
the high-output-impedance current generator. To avoid a
measurement error in developing the presented Z mea-
surement set-up, it was essential that a high-output Z was
maintained over the operating frequency range so that
the injected current was constant and that the current
source did not load the sample.20 The stimulating current
pulses, having precisely the same waveform as the driv-
ing pulses that were delivered to the VCCS, were con-
stant current, biphasic, partially charge-balanced and
asymmetric, composed of a precisely determined
quasi-trapezoidal cathodic phase with a square leading
edge of intensity ic, a plateau of the cathodic phase with
the width of tc, followed by an exponentially decaying
phase with the width of texp and the time constant ôexp,
and ended by a wide, rectangular, anodic phase with the
width ta and intensity of ia.
In the four-probe measurement, the generated current
stimulating pulse waveform ic was applied to an isolated
RSN via the two outer hook electrodes within the ladder,
electrode number 1 and electrode number 7 that served
as current feeding probes IH and IL. They were separated
from the electrodes number 3 and 5 that served as volt-
age probes VH and VL, respectively. In the two-probe
measurement, the generated current stimulating pulse
was applied on an isolated RSN via the electrode number
1, serving as a current feeding probe IH, and connected to
the electrode number 3 that served as a voltage probe VH
and via the electrode number 7, that served as a current
feeding probe IL and was connected to electrode number
5 that served as a voltage probe VL.
These measurements were used to obtain the Bode
and Nyquist diagrams with which the equivalent circuit
model elements could be calculated for each of the sam-
ples. The Nyquist diagram consists of a Zimaginary vs.
Zreal plot, representing the imaginary and real values for
the Z. This paper however, only reports an “in-vitro”
study of RSN samples conducted using EIS to find the
frequency-dependent Z.
2.5 Voltage-response (VR) measurements
The presented set-up also enabled measurements of
VRs that were elicited in the RSN using generated
quasi-trapezoidal monophasic and biphasic pulses. The
generated stimulating pulse was measured indirectly as a
voltage drop at the reference resistor with a trains imped-
ance amplifier to assess the value of the phase angle .
Later on, this value was used in the processing of data
for the Z calculation. The aforementioned stimulating
pulse was used for excitation of the RSN that represented
a floating load; therefore, the elicited voltage drop was
measured differentially using an instrumental amplifier.
The RSN was stimulated with the generated current
pulses while information on the VR and stimulating cur-
rent was recorded simultaneously. In this regard, ic was
1.8 mA while tc was 60 μs. In each measurement, where
50 pulses were generated, the last pulse was saved and a
time to gap of 250 μs was introduced. The EIS was per-
J. ROZMAN et al.: MEASUREMENT OF BIO-IMPEDANCE ON AN ISOLATED RAT SCIATIC NERVE ...
800 Materiali in tehnologije / Materials and technology 50 (2016) 5, 797–804
Figure 2: Voltage driving pulses to control the VCCS current source:
a) a monophasic pulse and b) a biphasic pulse
Slika 2: Napetostna krmilna pulza za kontrolo VCCS tokovnega vira:
a) monopolarni pulz in b) bipolarni pulz
formed using current sinusoidal signals with an ampli-
tude of 400 μA that were generated at frequencies evenly
distributed within the frequency range between 1 Hz and
500 kHz and delivered to the RSN. While the RSN was
stimulated with the aforementioned current sinusoidal
signals, the stimulating current and VR were saved again.
All of the data were analysed off-line and Z was calcu-
lated with the lock-in method.
3 RESULTS
3.1 Bio-impedance spectroscopy
Figure 3a shows the Bode diagram of the absolute
value of Z, measured on the RSN, using the two- and
four-probe set-ups while the RSN is stimulated with two
specific monophasic and biphasic excitation pulses, re-
spectively. Figure 3b shows the  obtained with the two-
and four-probe set-ups expressed in the Bode diagram
during RSN stimulation with monophasic and biphasic
stimuli. Figure 3c shows the Z spectrum measured with
the two- and four-probe set-ups, presented with the
Nyquist diagram (Cole-Cole diagram), while the RSN is
stimulated with monophasic and biphasic stimuli.
The absolute Z value, expressed in the Bode diagram
while the RSN is stimulated with monophasic and
biphasic stimuli and measured using the two-probe
set-up, decreased almost exponentially from approxi-
mately 60 k
 measured at 1 Hz to slightly below 800 

measured at 20 kHz (Figure 3a). At frequencies between
20 kHz and 100 kHz, Z remained almost unchanged; at
frequencies above 100 kHz, however, Z began to
decrease rapidly. It was noted that traces of Z belonging
to stimulation with monophasic and biphasic stimuli
were almost identical.
Figure 3b, representing the  expressed in the Bode
diagram, shows that for the two-probe measurement, the
capacitive nature of an interface between RSN and plati-
num electrodes prevailed at frequencies below 500 Hz.
Namely, a corresponding  measured at 1 Hz was –60 °C
and  measured at 500 Hz was –10 °C, respectively. At
frequencies between 500 Hz and 30 kHz, the nature of
the interface remained still slightly capacitive, express-
ing the same value of the phase angle , i.e., –10 °C.
Within this frequency range, the ohmic nature of the in-
terface between the RSN and platinum electrodes pre-
vailed. At frequencies above 100 kHz, however, the na-
ture of the interface between the RSN and the platinum
electrodes started to be progressively capacitive again,
expressing the sharp decrease of  towards negative val-
ues. The traces of , belonging to the stimulation with
the monophasic and biphasic stimuli, were almost identi-
cal.
In Figure 3c, the Z spectra measured with two- and
four-probe set-ups expressed in a Nyquist diagram
(Cole-Cole diagram) while the RSN is stimulated with
monophasic and biphasic stimuli differ considerably.
This diagram represents the relationship between the
imaginary and the real component of Z, as measured at
the interface between the RSN and the particular plati-
num electrodes using the two- versus four-probe set-up
and monophasic versus biphasic stimuli. In other words,
this diagram represents the relationship between capaci-
tive and ohmic natures of the interface between the RSN
and the platinum electrodes. The difference was not sig-
nificant for monophasic versus biphasic stimuli but was
for the two- versus four-probe measurements. Traces of
Z spectra, belonging to stimulation with monophasic and
biphasic stimuli, were almost identical for the two- as
well as for the four-probe measurements. For the two-
J. ROZMAN et al.: MEASUREMENT OF BIO-IMPEDANCE ON AN ISOLATED RAT SCIATIC NERVE ...
Materiali in tehnologije / Materials and technology 50 (2016) 5, 797–804 801
Figure 3: Results of Z spectroscopy measurements while the RSN is
stimulated with monophasic and biphasic stimuli: a) Absolute Z value
measured with a two- and four-probe set-ups expressed in a Bode dia-
gram, b)  obtained with a two- and four-probe set-ups expressed in a
Bode diagram, c) Z spectrum measured with two- and four-probe
set-ups expressed in a Nyquist diagram (Cole-Cole diagram)
Slika 3: Rezultati spektroskopskih meritev Z, medtem, ko je bil RSN
stimuliran z monofaznimi in bifaznimi pulzi: a) absolutna vrednost Z
merjena z dvo in {tirito~kovno metodo, prikazana z Bodejevim
diagramom, b)  merjen z dvo in {tirito~kovno metodo, prikazan z
Bodejevim diagramom, c) spekter Z merjen z dvo in {tirito~kovno
metodo prikazana z Nyquistovim diagramom (Cole-Cole diagram)
and four-probe measurement, the Z spectrum below 10
k
 of the real/ohmic component, showed an almost
identical relationship between the imaginary/capacitive
and the real/ohmic nature of the interface between the
RSN and the Pt electrodes. Above 10 k
 of real/ohmic
component, however, the two-probe measurement
showed an almost linear decrease of an imaginary/capac-
itive nature and, at the same time, an almost linear in-
crease of the real/ohmic nature, while the four-probe
measurement showed no further increase in the
real/ohmic nature.
3.2 Voltage-response (VR) measurements
VRs, during the excitation of the RSN with the
quasi-trapezoidal monophasic and biphasic pulses using
the two- and four-point measurement set-ups, are shown
in the Figures 4a to 4c. Figure 4a shows the VRs elic-
ited with a monophasic pulse and measured with the
two- and four-point set-ups. VRs measured with the
four-point set-up follow the quasi-trapezoidal pulse more
accurately than the VRs measured using the two-point
set-up. This was in accordance with Figure 3 where the
real/ohmic nature of the interface between the RSN and
the Pt electrode in the four-point measurement prevails
over the imaginary/capacitive nature of the interface be-
tween the RSN and the Pt electrode.
Figure 4b shows VRs measured with the two- and
four-point set-ups elicited with the biphasic pulse. VRs
measured with the two-point set-up follow the
quasi-trapezoidal pulse more accurately than the VRs
measured with the four-point set-up. This is in accor-
dance with Figure 3 where the real/ohmic nature of the
interface between the RSN and the Pt electrode in the
four-point measurement prevails over the imaginary/ca-
pacitive nature of the interface between the RSN and the
Pt electrode.
As seen in all three panels of Figure 3, the leading
edge of the cathodic phase ic and the plateau of the cath-
odic phase with the width tc of the generated stimulating
current pulses were not of exactly the same shape as the
waveform of pulses that were delivered to the VCCS.
Other parts of the generated current pulses, such as expo-
nentially decaying phase with the width of texp and the
time constant exp in both monophasic and biphasic
pulses as well as the rectangular, anodic phase with the
width ta and intensity of ia in the biphasic pulse, are al-
most identical. It was presumed that in calculations of Z,
after a monophasic and biphasic pulse, the aforemen-
tioned dissimilarity did not influence Z significantly.
Figure 4c shows VRs measured using the four-point
set-up, elicited with a train of generated biphasic pulses
with a cathodic intensity (ic) of 1.8 mA and the elicited
VRs with the entire voltage drop of V = 1.7 V. As
shown in Figure 4c, an onset of ic elicited the near-in-
stantaneous voltage increase with the same time course.
However, when ic was terminated, the course of the volt-
age in the exponential decay region where ic exponen-
tially approached the lowest value, the voltage also expo-
nentially approached the value of approximately -0.15 V.
Almost at the same time as the cathodic pulse ic was ter-
minated, the onset of an anodic intensity (ia) of 0.6 mA
elicited the near-instantaneous positive voltage of ap-
proximately 0.65 V, having a slightly different course.
As in the VR of the tested stimulation pulse, the max-
imum negative polarization across the electrode-nerve
tissue interface (Emc) and the maximum positive polariza-
tion across the electrode-nerve tissue interface (Ema)
reached -0.15 V and 0.65 V, respectively. None of them
exceed the safety limit range for water electrolysis, from
-0.60 V to +0.85 V.21
J. ROZMAN et al.: MEASUREMENT OF BIO-IMPEDANCE ON AN ISOLATED RAT SCIATIC NERVE ...
802 Materiali in tehnologije / Materials and technology 50 (2016) 5, 797–804
Figure 4: Stimulation pulses and corresponding VRs: a) VRs measured
using a two- and four-point measurements elicited with a monophasic
pulse, b) VRs measured using a two- and four-point measurements
elicited with a biphasic pulse, c) VRs measured using a four-point
measurement elicited with a train of biphasic pulses
Slika 4: Stimulacijski pulzi in pripadajo~i VR-ji: a) VR-ji merjeni z
dvo in {tirito~kovno metodo, izzvani z monofaznim pulzom, b) VR-ji
merjeni z dvo in {tirito~kovno metodo, izzvani z bifaznim pulzom, c)
VR-ji merjeni z dvo- in {tirito~kovno metodo, izzvani z vlakom
bifaznih pulzov
4 DISCUSSION
The presented work demonstrates the feasibility of a
Z measurement on an isolated RSN with quasi-trape-
zoidal current biphasic stimulating pulses. The feasibility
of the stimulating paradigm, where the specific stimulus
waveform was used to selectively stimulate different
types of nerve fibres within the cervical segment of por-
cine vagus nerve, has already been published.19
In the last two decades, particular attention has been
paid to vagus nerve stimulation techniques that are used
to treat, among others, a number of nervous-system dis-
orders, neuropsychiatric disorders, eating disorders,
sleep disorders, cardiac disorders, endocrine disorders,
and pain.22–25 Considerable scientific and technological
efforts have been devoted to developing systems of elec-
trodes that interface the human vagus nerve with elec-
tronic implantable devices.26,27
The Z characterization of the electrode-electrolyte in-
terface is of paramount importance in the field of
neuroprotheses, where small electrodes are required for
SNS and recording of superficial regions of peripheral
nerves via multi-electrode systems. The electrode-elec-
trolyte interface is determined from the known electrode
Z, which should be as low as possible to avoid nerve tis-
sue damage. A high Z would result in a large applied
electrode voltage leading to undesirable electrochemical
reactions that damage the nerve tissue.28 As shown in the
presented study, neither the maximum negative polariza-
tion across the electrode-electrolyte interface (Emc), nor
the maximum positive polarization across the elec-
trode-electrolyte interface (Ema), exceeded the safety lim-
its for water electrolysis.21
Nevertheless, large surface area stimulating elec-
trodes (due to a roughened surface), with a relatively
much smaller geometric surface area and thus a much
lower Z than current designs, could be beneficial, espe-
cially where a larger number of stimulating electrodes
within the multi-electrode system29 is needed. The au-
thors demonstrated “in vitro” that by increasing the real
surface area of stimulating electrodes, a significantly
lower polarization impedance and electrode impedance,
as well as a low residual direct current, could be ob-
tained. Also, all three electrical parameters, electrode im-
pedance, access resistance and polarization, could be cal-
culated from the current and voltage measurements.
Thus, a theoretical equivalent circuit model of the inter-
face between the RSN and Pt electrodes could be mod-
elled by fitting the equivalent circuit model elements to
the EIS-based frequency-response spectroscopy experi-
mental results.28,30,31
5 CONCLUSIONS
The first conclusion is that at frequency range
between 0 kHz and 2.5 kHz, which confines the power
spectrum density of the proposed quasi-trapezoidal
stimulating pulse (power spectrum density not shown in
this paper), the ohmic nature of the interface between
RSN and Pt electrodes prevailed over the imaginary/
capacitive nature, yielding the relative low Z of approxi-
mately 1 k
.
The second conclusion is that the differences elicited
at the DL of the interface between the Pt stimulating
electrode and the nerve tissue, while the RSN was stimu-
lated using biphasic quasi-trapezoidal stimuli, have not
exceeded the voltage range considered safe for nerve tis-
sue.
The third conclusion is that the designed and tested
four-probe bio-impedance measurement set-up, based on
the Red Pitaya open-source measurement and control
tool, provided data on the structure- and composition-re-
lated electrical properties of the RSN tissue that contrib-
ute to the development of multi-electrode nerve stimulat-
ing systems.
Definitions of field-specific terms
SNS – selective nerve stimulation
RSN – segment of a rat sciatic nerve
AC – alternating current
Z – electrical impedance
VCCS – voltage controlled current source
driving pulse – voltage pulses to drive a VCCS
EIS – Electrical impedance spectroscopy
Pt – platinum
chamber – temperature-controlled measuring chamber
CNC – computerized numerical control
ic – intensity of the cathodic phase
tc – width of the cathodic phase
texp – width of the cathodic exponential decay
exp – time constant of the exponential decay
ia – intensity of the anodic phase
ta – width of the anodic phase
IH and IL – current feeding probes
VH and VL – voltage probes
VR – voltage response
R – reference resistor
 – phase angle
DL – double layer
Ema – maximum positive polarization across the elec-
trode-electrolyte interface
Emc – maximum negative polarization across the elec-
trode-electrolyte interface
Acknowledgement
This work was financed by the research grant
P3-0171 and P4-0053 from the Slovenian Research
Agency (ARRS), Ministry of Education, Science and
Sport, Ljubljana, Republic of Slovenia. The authors ac-
knowledge Professor Dejan Kri`aj from the Faculty of
Electrical Engineering, University of Ljubljana, for the
J. ROZMAN et al.: MEASUREMENT OF BIO-IMPEDANCE ON AN ISOLATED RAT SCIATIC NERVE ...
Materiali in tehnologije / Materials and technology 50 (2016) 5, 797–804 803
useful discussion and enabling us to use a measuring set
up developed in his laboratory.
Ethical approval
The neural tissue was treated in accordance with the
approval guidelines of the ethics committee of the Veteri-
nary Administration, Ministry of Agriculture, Forestry
and Food, Republic of Slovenia.
Conflict of interest statement
The authors declare that there are no conflicts of in-
terest regarding the publication of this paper. None of the
authors received any financial or other compensation for
undertaking this study or during its execution. No per-
sonal relationships with other people or organizations in-
appropriately influenced the work.
6 REFERENCES
1 T. K. Bera, J. Nagaraju, Electrical impedance spectroscopic study of
broiler chicken tissues suitable for the development of practical
phantoms in multifrequency EIT, J Electr Bioimp, 2 (2011), 48–63,
doi:10.5617/jeb.174
2 H. P. Schwan, Electrical properties of tissue and cell suspensions,
Adv Biol Med Phy, 5 (1957), 147–209
3 H. Solmaz, Y. Ülgen, M. Tümer, Design of a Cole-Cole Meter for
Complex Impedance Measurement of Living Tissues, Biomedical
Engineering, 7 (2010), doi:10.2316/Journal.216.2010.7.680-137
4 J. J. Ackmann, Complex bioelectric impedance measurement system
for the frequency range from 5 Hz to 1 MHz, Ann Biomed Eng, 21
(1993) 2, 135–146
5 R. W. Griffiths, M. E. Philpot, B. J. Chapman, K. A. Munday, Imped-
ance cardiography: non-invasive cardiac output measurement after
burn injury, Int J Tissue React, 3 (1981) 1, 47–55
6 C. J. Schuster, H. P. Schuster, Application of impedance cardiogra-
phy in critical care medicine, Resuscitation, 11 (1984) 3–4, 255–274
7 H. Solmaz, Y. Ülgen, M. Tümer, Design of a microcontroller based
Cole-Cole impedance meter for testing biological tissues, World
Congress on Medical Physics and Biomedical Engineering, 2009,
doi:10.1007/978-3-642-03885-3_135
8 J. Wtorek, A. Bujnowski, A. Poliñski, L. Józefiak, B. Truyen, A
probe for immittance spectroscopy based on the parallel electrode
technique, Physiol Meas, 25 (2004) 5, 1249–1260, doi:10.1088/
0967-3334/25/5/014
9 R. Pinter, R. Schmidt, Impedance Measurement Circuit and Method.
U.S. Patent No. 8831898, 2010
URL:http://patft.uspto.gov/netacgi/nphParser?Sect1=PTO1&Sect2=
HITOFF&d=PALL&p=1&u=%2Fnetahtml%2FPTO%2Fsrchnum.ht
m&r=1&f=G&l=50&s1=8831898.PN.&OS=PN/8831898RS=PN/88
31898
10 T. Palko, F. Bialokoz, J. Weglarz, Multifrequency device for mea-
surement of the complex electrical bio-impedance – design and ap-
plication, Proc. of the Engineering in Medicine and Biology Society,
1995 and 14th Conference of the Biomedical Engineering Society of
India, An International Meeting, the first regional conference, New
Delhi, 1995, doi:10.1109/RCEMBS.1995.508682
11 Y. Yang, J. Wang, G. Yu, F. Niu, P. He, Design and preliminary eval-
uation of a portable device for the measurement of bioimpedance
spectroscopy, Physiol Meas, 27 (2006) 12, 1293–1310
12 H. P. Schwan, Electrode polarization impedance and measurements
in biological materials, Ann N Y Acad Sci, 148 (1968), 191–209
13 H. P. Schwan, Linear and nonlinear electrode polarization and bio-
logical materials, Ann Biomed Eng, 20 (1992), 269-288
14 E. Barsoukov, J. R. Macdonald, Impedance spectroscopy: theory, ex-
periment and applications, 2nd ed., John Wiley & sons, Hoboken,
New Jersey 2005
15 A. J. Bard, L. R. Faulkner, Electrochemical methods: fundamentals
and applications, 2nd ed., John Wiley & sons, New York 2001
16 J. O’M. Bockris, A. K. N. Reddy, Modern Electrochemistry, vol. 2,
Plenum publishing corporation, New York 1970
17 J. R. Macdonald, Impedance spectroscopy, John Wiley & sons, New
York 1987
18 T. K. Bera, Bioelectrical impedance methods for noninvasive health
monitoring: a review, Journal of Medical Engineering, 2014 (2014),
Article ID 381251, 28 pages, doi:10.1155/2014/381251
19 P. Pe~lin, J. Rozman, Alternative paradigm of selective vagus nerve
stimulation tested on an isolated porcine vagus nerve, Scientific
World Journal, 2014 (2014), 310283, doi:10.1155/2014/310283
20 A. A. Silverio, A. A. Silverio, A high output impedance current
source for wideband bioimpedance spectroscopy using 0.35μM
TSMC CMOS technology, International Journal of Engineering and
Applied Sciences, 1 (2012) 2, 68–75
21 E. M. Hudak, J. T. Mortimer, H. B. Martin, Platinum for neural stim-
ulation: voltammetry considerations, J Neural Eng, 7 (2010) 2,
26005, doi:10.1088/1741-2560/7/2/026005
22 T. A. Anholt, S. Ayal, J. A. Goldberg, Recruitment and blocking pro-
perties of the CardioFit stimulation lead, J Neural Eng, 8 (2011) 3,
034004, doi:10.1088/1741-2560/8/3/034004
23 A. M. Bilgutay, I. M. Bilgutay, F. K. Merkel, C. W. Lillehei, Vagal
tuning: a new concept in the treatment of supraventricular
arrhythmias, angina pectoris, and heart failure, J Thorac Cardiovasc
Surg, 56 (1968), 71–82
24 C. Heck, S. L. Helmers, C. M. DeGiorgio, Vagus nerve stimulation
therapy, epilepsy, and device parameters: scientific basis and recom-
mendations for use, Neurology, 59 (2002), S31–S37
25 D. M. Labiner, G. L. Ahern, Vagus nerve stimulation therapy in de-
pression and epilepsy: therapeutic parameter settings, Acta Neurol
Scand, 115 (2007), 23–33
26 J. D. Sweeney, D. A. Ksienski, J. T. Mortimer, A nerve cuff tech-
nique for selective excitation of peripheral nerve trunk regions, IEEE
Trans Biomed Eng, 37 (1990) 7, 706–715
27 A. Q. Choi, J. K. Cavanaugh, D. M. Durand, Selectivity of multi-
ple-contact nerve cuff electrodes: asimulation analysis, IEEE Trans
Biomed Eng, 48 (2001), 165–172
28 W. Franks, I. Schenker, P. Schmutz, A. Hierlemann, Impedance char-
acterization and modeling of electrodes for biomedical applications,
IEEE Trans Biom Eng, 52 (2005) 7, 1295–1302
29 M. Tykocinski, Y. Duan, B. Tabor, R. S. Cowan, Chronic electrical
stimulation of the auditory nerve using high surface area (HiQ) plati-
num electrodes, Hear Res, 159 (2001) 1-2, 53–68
30 D. R. Merrill, The electrochemistry of charge injection at the elec-
trode/tissue interface, In: D. Zhou, E. Greenbaum, Implantable Neu-
ral Prostheses 2, Springer, New York 2010, 85–138, doi:10.1007/
978-0-387-98120-8_4
31 L. A. Geddes, Historical evolution of circuit models for the elec-
trode–electrolyte interface, Ann Biomed Eng, 25 (1997), 1–14
J. ROZMAN et al.: MEASUREMENT OF BIO-IMPEDANCE ON AN ISOLATED RAT SCIATIC NERVE ...
804 Materiali in tehnologije / Materials and technology 50 (2016) 5, 797–804