Biosensors and Bioelectronics 64 (2015) 554–559

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

journal homepage: www.elsevier.com/locate/bios

A novel type of electrochemical sensor based on ferromagnetic carbon-

encapsulated iron nanoparticles for direct determination of
hemoglobin in blood samples
Edyta Matysiak a, Mikolaj Donten a, Agata Kowalczyk a, Michal Bystrzejewski a,
Ireneusz P. Grudzinski b, Anna M. Nowicka a,n
a
Faculty of Chemistry, University of Warsaw, ul. Pasteura 1, PL-02-093 Warsaw, Poland
b
Faculty of Pharmacy, Medical University of Warsaw, ul. Banacha 1, PL-02-097 Warsaw, Poland

art ic l e i nf o a b s t r a c t

Article history: An effective, fast, facile and direct electrochemical method of determination of hemoglobin (Hb) in blood
Received 11 August 2014 sample without any sample preparation is described. The method is accomplished by using the
Received in revised form ferromagnetic electrode modifier (carbon-encapsulated iron nanoparticles) and an external magnetic
12 September 2014
field. The successful voltammetric determination of hemoglobin is achieved in PBS buffer as well as in
Accepted 19 September 2014
Available online 2 October 2014
the whole blood sample. The obtained results show the excellent electroactivity of Hb. The measure-
ments are of high sensitivity and good reproducibility. The detection limit is estimated to be 0.7 pM. The
Keywords: electrochemical determination data were compared with the gravimetric data obtained with a quartz
Hemoglobin crystal microbalance. The agreement between these results is very good. The changes of the electrode
Electrochemistry
surface morphology before and after Hb detection are monitored by electron microscopy. The
Direct determination
functionality of the electrochemical sensor is tested with human and rat blood samples. The concentra-
Core–shell type ferromagnetic
nanoparticles tion of hemoglobin in the blood samples determined by using voltammetric/gravimetric detection is in
Human blood perfect agreement with the data obtained from typical clinical analysis.
Rat blood & 2015 Elsevier B.V. All rights reserved.

1. Introduction
The major role of hemoglobin is to carry oxygen from the lungs
to the tissues and return carbon dioxide from the tissue to the
lungs. So, the changes of Hb concentration in the blood can cause
several diseases such as anemia and even death (Anand, 2008;
Chonchol and Nielson, 2008; Weissbluth, 1973). One has to be
noted that hemoglobin is a key hematological metric in medical
diagnostics, therefore, the accurate determination of this hemo-
protein is very essential in a number of human pathologies. In
clinical analysis the determination of Hb is carried out using
spectrophotometric method such as KCN-based and KCN-free
assays (Zwart et al., 1984; Zwart, 1993) or immunoassay proce-
dures (Darain et al., 2009; Suprun et al., 2010). However, most of
these methods suffer from utilization of some toxic substances and
low sensitivity to describe discrete changes in Hb levels in blood if
any due to some fatalities.
The electrochemical detection of hemoglobin seems to be very
competitive in relation to the present clinical methods used to
n
Corresponding author.
E-mail address: anowicka@chem.uw.edu.pl (A.M. Nowicka).
date due to the low cost, no need to use the toxic substances and
short analysis time. However, the facilitation of electron transfer
between the heme centers in the large three-dimensional struc-
ture of Hb and the electrodes is still a challenging task in
electronanochemistry. The Hb protein does not undergo facile
heterogeneous redox reaction at the electrode surface because its
electroactive centers (heme groups) are deeply embedded in the
protein shell and in consequence the electron transfer rate is too
slow (Xu et al., 2007). Moreover, the direct adsorption of large
protein onto the metallic electrode surface would lead to its
denaturation, which in turn efficiently blocks the electron com-
munication between the heme centers and the electrode (George
and Lee, 2009; Heller, 1990). Many efforts have been made to
improve the electron transfer of Hb by using mediators and
promoters, especially by modifying the electrode with the desir-
able matrix, which allowed to the effective immobilization of Hb
on the electrode surface (Chen et al., 2011; George and Lee, 2009;
Lojou and Bianco, 2004; Sun et al., 2010; Xie et al., 2013).
Due to the fact that hemoglobin and some of its modified forms
have paramagnetic properties the application of magnetic nano-
particles as the electrode modifier seems to be promising. Addi-
tionally, the introduction of an external magnetic field should lead
to: (i) the higher specific interaction between Hb and magnetic

http://dx.doi.org/10.1016/j.bios.2014.09.079
0956-5663/& 2015 Elsevier B.V. All rights reserved.
 E. Matysiak et al. / Biosensors and Bioelectronics 64 (2015) 554–559
nanoparticles, (ii) accelerate the mass transfer rate, (iii) lower
charge transfer resistance, (iiii) the selective separation of the
protein from the sample (Nowicka et al., 2014). The new class of
hybrid magnetic nanomaterials including carbon-encapsulated
iron nanoparticles (Fe@C Nps) is the promising choice for various
biomedical applications e.g.: drug delivery, imaging, detection of
paramagnetic substances (Pankhurst et al., 2003; Herrmann et al.,
2009), also for mobile platforms for catalysis (Ma et al., 2008) and
sorption (Koehler et al., 2009; Pyrzyńska and Bystrzejewski, 2010),
even magnetic data storage (Reiss and Hütten, 2005). Recent
studies demonstrate that core-shell type carbon-encapsulated
magnetic nanoparticles are interesting materials useful for the
modification of electrodes (Yu et al., 2013).
The direct voltammetry of Hb in solution at very low level
without using any mediator has not been observed to date. The
detection of hemoglobin in the solution at nM level is possible
only by application of an mediator (e.g. methylene blue)
(Pakapongpan et al., 2011). In this work a simple, fast and direct
electrochemical method of determination of Hb in natural samples
as whole human and rat blood at pM level without the mediator
and any special complicated treatment of the physiological sample
is presented. The effectiveness of the measurement is enhanced by
the magnetic properties of the selected Hb derivative and the
ferromagnetic electrode modifier, i.e. carbon-encapsulated iron
nanoparticles. The application of an external magnetic field
significantly enhances the Hb flux to the electrode surface and in
consequence the current of the reduction of Fe(III)–Fe(II) in the
heme group increases. The magnetic field influence on the
bioelectrochemical system is studied using the cyclic voltammetry
(CV), linear scan voltammetry (LSV), electrochemical quartz mi-
crobalance (EQCM) and scanning electron microscopy (SEM).
2. Experimental section
2.1. Materials
All chemicals were of the highest purity available. Human
hemoglobin was purchased from Sigma-Aldrich and was used as
received. A phosphate buffer solution (PBS; 20 mM, pH 6.0) was
prepared by mixing solutions of NaH2PO4, Na2HPO4 with subse-
quent addition of 2 mM KCl and 150 mM NaCl. All solutions were
prepared using Milli-Q water of conductivity 0.056 μS cm
1. The
ferromagnetic electrode modifier, carbon-encapsulated iron nano-
particles (Fe@C Nps), was synthesized according to the procedure
described elsewhere (Bystrzejewski et al., 2011).
2.2. Electrochemical quartz microbalance (EQCM) measurements
An electrochemical quartz crystal microbalance, model Auto-
lab-EQCM (Eco Chemie B.V., Utrecht, Netherlands) with 6 MHz
Au/TiO2 quartz crystal resonators (A ¼0.435 cm2), was used in this
study. Initially the Au-EQCM electrode was electrochemically
pretreated by cycling first between 0 V and 1.8 V (hold 10 s at
1.8 V) in 0.1 M NaOH with the scan rate 50 mV/s, and then by
cycling between
0.3 and 1.5 V (vs Ag/AgCl) in 0.1 M H2SO4
solution until the stable voltammogram typical for a clean gold
electrode was observed (Finklea et al., 1987). Then the thin smooth
film of carbon (  25 nm) was deposited on the gold quartz surface
(Au-EQCM/C). For the vapor deposition of carbon a Polaron
evaporator, (model CA7625), was used. The presence of an external
magnetic field did not influence on the work of the quartz crystal
microbalance.
555
2.3. Voltammetric measurements
Cyclic voltammetry and linear scan voltammetry (CV, LSV)
were performed using an Autolab, model PGSTAT 12 potentiostat
equipped with an ECD amplifier module (RC time settings: 0 s for
the scan rates 4 10 mV/s and 0.1 s for the scan rates o 10 mV/s)
and the electrochemical analysis system based on the GPES soft-
ware package (Eco Chemie B.V., Utrecht, Netherlands). For each
voltammetric measurement the three-electrode system consisting
of a quartz crystal resonator covered by carbon layer (Au-EQCM/C)
used as the working electrode, a reference electrode (Ag/AgCl/3 M
KCl) and a gold wire used as the auxiliary electrode were
employed. In all experiments, the electrochemical cell was kept
in a Faraday cage to minimize the electrical noise. As a source of
magnetic field (40 mT) a large plate Fe14Nd2B magnet, placed
parallel under the electrochemical cell, was used. The magnetic
field induction vector was normal to the electrode surface.
Additionally the surface of the magnet was at least 20-times
larger than the electrode surface, what should provide the uniform
external magnetic field in the region of the electrode surface.
2.4. Electron microscopy
Scanning Electron Microscope (SEM) images were obtained
with a Merlin (Zeiss, Germany) field emission scanning electron
microscope. In order to enhance the material contrast the images
obtained using InLens secondary electron detector and Energy
Filtered Back Scattered electron (also in-lens) were mixed. The
combination of signals from these two detectors allowed to
distinguish the hemoglobin and carbon present on the substrate
or carbon shell of magnetic Fe@C Nps. All images were taken at
low EHT (below 3 kV) and low beam current ca. 30 pA preventing
any degradation of the sample.
2.5. Preparation of modified electrode
Carbon-encapsulated iron nanoparticles, which were applied as
the electrode modifier have typical weak ferromagnetic character-
istics. The magnetic hysteresis loop of Fe@C Nps (at 25 °C) is
shown in Fig. S1. The material attains the maximum achievable
magnetic moment (121 emu/g) at relatively low magnetic field (ca.
8 kOe). The weak ferromagnetic performance is appeared by the
low coercive field, which is found to be 165 Oe. The remnant
magnetization is only 6 emu/g and this value corresponds to less
than 5% of the saturation magnetization. The magnetic field
induction measured by the Hall magnetometer within the electro-
chemical cell was 40 mT. In air, as well is water (which has nearly
the same magnetic permeability in comparison to air) the magni-
tude of the magnetic induction and the magnetic field are
numerically equal (1 mT ¼10 Gs ¼10 Oe; hence 40 mT ¼400 Oe).
According to the hysteresis loop the net magnetic moment of Fe@C
NPs that cover the glassy carbon electrode is ca. 21 emu/g. The
diameter of Fe@C Nps ranges between 5 and 65 nm. The major
fraction of nanoparticles has relatively narrow size distribution
(5C30 nm), so the mean value of Fe@C Nps diameter is estimated
to be ca. 20 nm.
Carbon-encapsulated iron nanoparticles were dispersed in
water (0.125 mg/ml) with ultrasonic irradiation and then, depos-
ited on the surface of the electrode. A droplet (100 μl) of the Fe@C
Nps suspension was placed on the electrode surface and the
electrode was left to dry at room temperature. The droplet volume
was sufficient to cover the electrode surface completely. Since the
nanoparticles are covered by a thin carbon shell the electrical
conductivity of the electrode surface should not be harmed. To
prepare the suspension with uniformly distributed Nps a highly-
efficient ultrasound homogenizer instead of commonly used
556 E. Matysiak et al. / Biosensors and Bioelectronics 64 (2015) 554–559

simple ultrasonic bath was applied. In the first step the remnant
magnetization of Fe@C Nps was eliminated by continuous and
intensive (50 W) ultrasound irradiation of the suspension under
slowly decaying magnetic field. The second step – the exact
dispersing of the Nps was done by application of alternating
symmetric (1 s) ON–OFF ultrasound 80 W in peak pulses for 90 s.
A uniform aqueous suspension of the nanoparticles prepared in
this simple two-step procedure appeared a significantly reduced
tendency to agglomeration of the solid phase and in consequence
it allowed to obtain a thin and relatively smooth layer of the
ferromagnetic modifier on the surface the electrode. The typical
SEM images of the Fe@C agglomerates attached to the electrode
surface after evaporation of the liquid (water) phase from the
suspensions are presented in Fig. S2. Fig. S2b and c present the
differences in morphology of ferromagnetic layers formed after
evaporation of water from the suspensions dispersed in the
regular method (using ultrasonic bath) and improved the method
with the ultrasonic homogenizer, respectively. The appearance of a
surface of EQCM working electrode device after vapor-deposited Fig. 1. UV–vis spectra of pure Hb (0.10 μM) and after addition of H2O2 (1 μM) in
carbon film ca. 10 nm (Fig. S2a) indicates an ordered coverage of 0.02 M PBS buffer (pH 6.0). Inset: enlarged selected part of UV–vis spectra
the electroactive surface with an amorphous carbon layer. (460C650 nm).

Additionally the intensity of this band decreases but the band does
3. Results and discussion not disappear, so the heme prostetic group destruction does not
take place (Li et al., 2006).
The aim of the experiments is to develop a relay-able fast and
simple method of the direct electrochemical determination of 3.1. Analytical performance
hemoglobin in human and rat blood samples. The main idea of
the proposed method utilizes the magnetic properties of Hb and The linear scan voltammograms of 6.45  10
5 g dl
1 (10 nM) of
ferromagnetism of the nanostructured electrode modifier (Fe@C paramagnetic species of Hb in 0.02 M PBS buffer (pH 6.0) on the
Nps). It is known that hemoglobin may exist in solution in two gold quartz crystal covered by carbon layer and modified with
forms: as diamagnetic oxyhemoglobin (oxyHb, HbFe(II)O2) and carbon-encapsulated iron nanoparticles (Au-EQCM/C/Fe@C Nps) in
paramagnetic deoxyhemoglobin (deoxyHb, HbFe(II)). OxyHb is a the absence and presence of magnetic field (40 mT) are shown in
fairly stable molecule and does slowly auto-oxidize to methemo- Fig. 2. The use of ferromagnetic electrode modifier and an external
globin (metHb, HbFe(III)) with the yield of  0.5–3% per day magnetic field facilitates the direct electron transfer between Hb
(Umbreit, 2007). The autoxidation of oxyHb involves the dissocia- dissolved in the solution and the electrode surface. The observed
tion of the oxygen without electron transfer to superoxide (O2
)
reduction peak of Hb is well visible. The conformation of oxyhe-
and metHb (Misra and Fridovich, 1972). Under the physiological
moglobin and its oxidation products (metHB/oxoferrylHb) is
conditions oxyhemoglobin can be oxidized by hydrogen peroxide
defined as a state of R – relaxed, in which the heme groups
(at very low concentration of H2O2; r1 μM) to paramagnetic
(electroactive prosthetic groups) are more exposed on the outside
metHb containing Fe(III) ions (Kanias and Acker, 2010). The
concentration of H2O2 in normal human plasma is 4–5 μM
(Yamamoto et al., 1987). In turn, the reaction of highly concen-
trated hydrogen peroxide (order of mM) with oxyHb results in
formation of paramagnetic very reactive species oxoferrylhemo-
globin (oxoferrylHb, HbFe(IV)¼O) (Nagababu and Rifkind, 2000).
OxoferrylHb can react with hydrogen peroxide via reaction
(∙Hb[Fe(IV)¼O] þH2O2-Hb[Fe(III)] þO2 þH2O) leading to the for-
mation of methemoglobin (Nagababu and Rifkind, 2000).
Due to the fact, that the examined solutions were not deox-
ygenated the Hb protein existed in the oxygenated state (oxyHb).
Before further examinations the oxyHb solutions were treated by
excess of H2O2 in order to obtain the paramagnetic derivatives of
Hb, methemoglobin. The ratio of Hb and peroxide concentration
was 1:10 (in the total volume of 3 ml; e.g. for CHb equaled 0.1 μM
the CH2O2 was equaled 1 μM) (Paulus et al., 2012; Svistunenko
et al., 2002). In order to get the information about the hemoglobin
derivative present in the solution after the addition of H2O2 the
UV–vis measurements were performed. The obtained UV–vis
spectra are presented in Fig. 1. Addition of H2O2 leads to a decrease Fig. 2. Background subtracted linear scan voltammograms recorded in 0.02 M PBS
in the absorption at 408, 540 and 630 nm. The increase of the buffer (pH 6.0) containing: only 10 μM H2O2 (black lines); and Hb
absorbance of the band located at 580 nm might indicate a (6.45  10
5 g dl
1≡10 nM) with addition of 100 nM H2O2 (red lines) in the
composite of HbFe(IV) ¼O/∙HbFe(IV) ¼O (Giulivi and Davies, presence (solid lines) and absence (dashed lines) of magnetic field (40 mT). Inset:
Background subtracted cyclic voltammogram f Hb (6.45  10
5 g dl
1) recorded in
1990). The shift in the absorbance band at 408 nm towards the 0.02 M PBS buffer in the presence of magnetic field; scan rate 20 mV s
1. (For
lower wavelengths confirms the appearance of metHb in the interpretation of the references to color in this figure legend, the reader is referred
solution after reaction with H2O2 (Gebicka and Banasiak, 2009). to the web version of this article.)
 E. Matysiak et al. / Biosensors and Bioelectronics 64 (2015) 554–559 557

of the protein shell (Harmening, 2002). In effect, the transport of

paramagnetic molecules to the electrode is enhanced in the
presence of magnetic field, and the molecules of hemoglobin that
approach the electrode surface are in optimal orientation for the
electrochemical reaction. In this case the distance between the
electrode and the heme group is likely the smallest. This is
manifested by an increase in the reduction current of Hb and the
shift of the voltammetric peak potential by circa 120 mV towards
more positive potentials. Moreover, the oxidiation signal of Hb is
not visible. The obtained CV voltammogram presented in the inset
in Fig. 2 does not contain the reversal peak, and this observation
confirms that the degradation of the heme moiety by addition of
hydrogen peroxide does not take place. Since the presence of H2O2
in the solution may give an electrochemical response in the similar
potential range as Hb, the control experiments in the 0.02 M PBS
buffer containing H2O2 (10 μM) were carried out. The recorded
LSV curves presented in Fig. 2 (black lines) clearly indicate that the
addition of H2O2 to the solution does not have visible impact on
current at the potential range where the Hb reduction signal is
observed.
The developed direct electrochemical method was tested in the
concentration range of Hb 3.2  10
9–6.5  10
4 g dl
1 (from
0.5 pM to 0.1 μM). The effectiveness of the electrochemical mea-
surements is also evaluated by applying the EQCM technique,
which allows for the simultaneous measurements of changes in
the peak current and the corresponding frequency shifts caused by
an increase of weight of hemoglobin which reaches the surface of
the electrode. With increasing the concentration of Hb in the
solution an increase of the respective signals are observed (see
Fig. 3A and B). Additionally, the reduction peak height increases
Fig. 3. (A) Background subtracted linear scan voltammograms of Hb at different
with increasing the scan rate and the peak position shifts towards concentrations recorded in 0.02 M PBS buffer (pH 6.0) containing H2O2 in the
more negative potentials. The increased cathodic peak current is appropriate concentration, in the presence of magnetic field (40 mT); scan rate
linearly proportional to the square root of the scan rate with 20 mV s
1. (B) Frequency shifts observed during reduction of Hb at different
ranging 5–500 mV s
1. The correlation coefficient calculated from concentrations in 0.02 M PBS buffer (pH 6.0) in the presence of magnetic field
(40 mT). Blue line – hemolyzed human blood; red line – hemolyzed rat blood. (For
this plot is 0.997. These results confirm that the redox process of
interpretation of the references to color in this figure legend, the reader is referred
Hb at the modified electrode (Au-EQCM/C/Fe@C Nps) in the to the web version of this article.)
presence of an external magnetic field is strictly transport-con-
trolled process when Hb concentration in the solution is low. By
comparing the mass of adsorbed hemoglobin after the electro-
reduction process and the charge corresponding to the voltammo-
gram it was concluded that the electrode reaction was one-
electron process per one iron center. Therefore, the reactant of
the electrode reaction is metHb.
According to the data obtained from LSV and EQCM experi-
ments the calibration curves were plotted vs. logarithm of Hb
concentration in the solution. The plots shown in Fig. 4 are
linear in the broad concentration range from 6.5  10
9 to
6.5  10
4 g dl
1 (from 1 pM to 0.1 μM). The first part of the I vs.
CHb relation is strictly linear. When the concentration of deter-
mined Hb increases the relation of the signal (reduction current)
vs. Hb concentration fails to maintain a linear relationship. This
deviation can be explained by an effect of blocking of the electrode
surface by the reduction product of metHB, which irreversibly
adsorbed at Fe@C Nps. Such blocking will limit the possibility (in a Fig. 4. Calibration plots of change in reduction current of Hb (IHb) and frequency
shifts ( ). Empty points correspond to LSV data and a filled point corresponds to
smaller degree) of reaching the electrode surface by Hb molecules EQCM data. Blue squares – hemolyzed human blood; red cycles – hemolyzed rat
at low concentration. The significant deviation from the linear plot blood. Other conditions as in Fig. 3. (For interpretation of the references to color in
is observed for Hb concentrations over 0.1 nM. However, the this figure legend, the reader is referred to the web version of this article.)
linearity of the calibration plot is maintained in the whole
analyzed concentration range when the analytical signal is plotted
vs. logarithm of Hb concentration. The limit of detection (DL) is (n ¼5). The electrode-to-electrode reproducibility is also examined
circa 5  10-9 g dl
1 (0.7 pM) Hb in the sample. The DL was for six different independently modified electrodes, and the
estimated taking into account the variability of the background relative standard deviation found as low as 5.2%.
current measured in the pure buffer solution. The value of the After electrochemical measurements the electrode surface was
detection limit was determined as 3 standard deviation of the six examined by SEM microscopy. The obtained images show that the
controls. For all applied concentrations the repeatability was protein particles containing Hb stick to the electrode after their
satisfactory; the relative standard deviation was circa 9.5% reduction, see Fig. 5. Additionally it is clearly seen that the
558 E. Matysiak et al. / Biosensors and Bioelectronics 64 (2015) 554–559
Fig. 5. SEM images of the electrode surface in an external magnetic field,  40 mT, after voltammetric measurements of (A) hemolyzed (10 min) and diluted blood;
(B) hemolyzed (20 min) and diluted blood. RBC-red blood cells.
hemoglobin molecules are adsorbed more readily on the magnetic
nanocapsules than on the electrode surface covered by pure
carbon film. Due to the adsorption process each measurement
were done on the freshly prepared electrode only.
3.2. Detection of HB in whole Blood
The usefulness of the developed method was tested using the
natural samples (human and rat blood). To prepare the hemoglo-
bin solutions, the blood was hemolyzed by diluting (106 times)
and incubation for 20 min with 0.02 mM PBS buffer. The progress
of this process depends on its duration. Fig. 5A shows that the time
of 10 min is not enough to completely release the Hb from red
blood cells (RBC); not all RBC were decomposed. After these steps
H2O2 was added (0.1 μM.) and the measurements were carried
out. The obtained results are presented in Figs 3 and 4 (see blue
and red lines). For both real samples the reduction current signals
are shifted towards more negative potentials by circa 200 mV
compared to the isolated Hb. The matrix of the real sample is
much more complex in terms of composition, which may affects
the shape and position of the recorded electrochemical signal. The
presence of other surface active agents, which are presented in the
plasma, is particularly undesirable from the electrochemical point
of view. The complexity of the matrix of blood samples slow down
the electrode-reaction rate, increases reduction overpotential
what affects the position of the current signals, but in our opinion
does not change their intensity. So, at slightly higher overpotential
the Hb released from whole blood sample undergoes the some
electrode reaction as the isolated Hb. Therefore the calibration plot
obtained for the isolated Hb can be used for quantitative analysis
of Hb in whole blood sample. It was proved by independent
analysis of Hb (gravimetric and standard clinical analysis; see the
Table 1).
The quantitative determination of hemoglobin in human and
rat blood is presented in Table 1. To be sure that the constructed
sensor works correctly the level of Hb in the tested bloods samples
was evaluated via typical clinical analysis. To data, human (IPG as a
volunteer blood donor) and rat Wistar (n¼ 3) blood samples were
collected into sodium citrate-coated plastic tubes. The preclinical
experiments were performed in compliance with the Medical
University of Warsaw Institutional Animal Care and Use Commit-
tee (Warsaw, Poland). The hemoglobin level was determined
colorimetrically at 546 nm using a compact hematology analyzer
type BC-2800 VET (Mindray) and ADVIA 120 Hematology System
(Siemens) for rat and human whole blood samples, respectively.
The determined concentrations of Hb in tested blood samples by
applying our detector and typical clinical analysis are in very good
agreement.
Table 1
Quantitative results of total Hb from human and rat blood.
Human blood [g dl
1] Rat blood [g dl
1]
Clinical analysisa 14.5 70.2 13.2 7 0.4
Direct voltammetric detection
LSV 13.9 70.8 13.17 0.6
EQCM 13.17 1.4 13.2 7 1.2
a
Officially approved colorimetric method.
4. Conclusions
Herein, we demonstrate a simple, rapid and direct electroche-
mical method of determination of hemoglobin in human and rat
blood samples. The application of magnetic nanoparticles compos-
ing of the iron core as an electrode modifier and the use of an
external magnetic field provides an appropriate stereochemic
position of the analyzed molecule (Hb) vs. the electrode surface
and facilitates the direct electron transfer between the Hb protein
and the electrode surface. In a consequence the direct electro-
chemical determination of Hb present in the solution is readily
possible. The constructed sensor allows for the straight and fast
voltammetric detection of hemoglobin in blood samples without
any special treatment of the samples and using special mediators.
Moreover, the presented studies show very good direct electro-
chemical characteristics and performances for determination of Hb
with high sensitivity, good reproducibility and very low detection
limit (pM level). Additionally the fabrication procedure as well as
the determination step does not require any specific toxic re-
agents. It should be highlighted that, the combination of two
techniques (LSV and EQCM) gives broader and more complete
information and can be applied successfully for different types of
paramagnetic proteins analysis. The SEM microscopy analysis
showed that due to the adsorption process the electrode is in fact
disposable. It also showed the efficiency and very useful selectivity
vs. hemoglobin in blood samples of the applied ferromagnetic
electrode modifier.
The developed method can be also successfully applied for
analyzing different types of paramagnetic proteins analysis in the
buffer as well as in the body fluids.
Acknowledgments
Support for this work by a Polish NCN Grant no. 2011/01/B/ST4/
03032 is gratefully acknowledged. The synthesis of carbon-en-
capsulated iron nanoparticles was supported by the National
 E. Matysiak et al. / Biosensors and Bioelectronics 64 (2015) 554–559
Centre for Research and Development (Poland) through the
Project LIDER 527/L-4/2012. The SEM images were obtained using
the equipment purchased within CePt Project no. POIG.02.02.00-
14-024/08-00.
Appendix A. Supplementary information
Supplementary data associated with this article can be found in
the online version at http://dx.doi.org/10.1016/j.bios.2014.09.079.
References
Anand, I.S., 2008. J. Am. Coll. Cardiol. 52, 501–511.
Bystrzejewski, M., Karoly, Z., Szepvolgyi, J., Huczko, A., Lange, H., 2011. Mater. Res.
Bull. 46, 2408–2417.
Chen, S.-P., Yu, X.-D., Xu, J.-J., Chen, H.-Y., 2011. Biosens. Bioelectron. 26, 4779–4784.
Chonchol, M., Nielson, C., 2008. Am. Heart J. 155, 494–498.
Darain, F., Yager, P., Gan, K.L., Tjin, S.C., 2009. Biosens. Bioelectron. 24, 1744–1750.
Finklea, H.O., Avery, S., Lynch, M., Furtsch, T., 1987. Langmuir 3, 409–413.
Gebicka, L., Banasiak, E., 2009. Acta Biochim. Pol. 56, 509–513.
George, S., Lee, H.K., 2009. J. Phys. Chem. B 113, 15445–15454.
Giulivi, C., Davies, K.J.A., 1990. J. Biol. Chem. 265, 19453–19460.
Harmening, D.M., 2002. Clinical Hematology and Fundamentals of Hemostasis. F.A.
Davis Company, Philadelphia.
Heller, A., 1990. Acc. Chem. Res. 23, 128–134.
Herrmann, K.I., Grass, N.R., Stark, W.J., 2009. Nanomedicine 4, 787–798.
Kanias, T., Acker, J.P., 2010. FEBS J. 277, 343–356.
Koehler, F.M., Rossier, M., Waelle, M., Athanassiou, E.K., Limbach, L.K., Grass, R.N.,
Gunther, D., Stark, W.J., 2009. Chem. Commun. 32, 4862–4864.
559
Li, D., Zhang, X., Long, Y., Sun, X., 2006. Life Sci. J. 3, 52–58.
Lojou, E.A., Bianco, P., 2004. Electroanalysis 16, 1113–1121.
Ma, Y., Yue, B., Yu, L., Wang, X., Hu, Z., Fan, Y., Chen, Y., Lin, W., Lu, Y., Hu, J., 2008. J.
Phys. Chem. C 112, 472–475.
Misra, H.P., Fridovich, I., 1972. J. Biol. Chem. 247, 6960–6962.
Nagababu, E., Rifkind, J.M., 2000. Biochemistry 39, 12503–12511.
Nowicka, A.M., Kowalczyk, A., Donten, M.L., Donten, M., Bystrzejewski, M., Stojek,
Z., 2014. Electrochim. Acta 126, 115–121.
Pakapongpan, S., Palangsuntikul, R., Surareungchai, W., 2011. Electrochim. Acta 56,
6831–6836.
Pankhurst, Q.A., Connolly, J., Jones, S.K., Dobson, J., 2003. J. Phys. D 36, R167–R181.
Paulus, A., Gijsbertus, S., Rossius, H., Dijk, M., de Vries, S., 2012. J. Biol. Chem. 287,
8830–8838.
Pyrzyńska, K., Bystrzejewski, M., 2010. Colloids Surf. A 362, 102–109.
Reiss, G., Hütten, A., 2005. Nat. Mater. 4, 725–726.
Sun, W., Qin, P., Zhao, R., Jiao, K., 2010. Talanta 80, 2177–2181.
Suprun, E., Bulko, T., Lisitsa, A., Gendenko, O., Ivanov, A., Shumyantseva, V.,
Archakov, A., 2010. Biosens. Bioelectron. 25, 1694–1698.
Svistunenko, D.A., Dunne, J., Fryer, M., Nicholls, P., Reeder, B.J., Wilson, M.T., Bigotti,
M.G., Cutruzzolà, F., Cooper, C.E., 2002. Biophys. J. 83, 2845–2855.
Umbreit, J., 2007. Am. J. Hematol. 82, 134–144.
Weissbluth, M., 1973. Molecular Biology: Bichemistry and Biophysics. Springer-
Verlag, New York.
Xie, L., Xu, Y., Cao, X., 2013. Colloids Surf. B: Biointerfaces 107, 245–250.
Xu, J., Li, W., Yin, Q., Zhong, H., Zhu, Y., Jin, L., 2007. J. Colloid. Interf. Sci. 315,
170–176.
Yamamoto, Y., Brodsky, M.H., Baker, J.C., Ames, B.N., 1987. Anal. Biochem. 160, 7–13.
Yu, Ch., Wang, Y., Wang, L., Zhu, Z., Bao, N., Gu, H., 2013. Colloids Surf. B 103,
231–237.
Zwart, A., Buursma, A., Van Kampen, E.J., Zijlstra, W.G., 1984. Clin. Chem. 30,
373–379.
Zwart, A., 1993. Clin. Chem. 39, 1570–1572.