Post on 29-May-2020
transcript
Publications of the University of Eastern FinlandDissertations in Forestry and Natural Sciences
Publications of the University of Eastern Finland
Dissertations in Forestry and Natural Sciences
isbn 978-�952-�61-�0216-�0
Sami Myllymaa
Novel Micro- and Nano-technological Approaches for Improving the Performance of Implantable Biomedical Devices
Recent advances in micro- and
nanotechnology offer a great oppor-
tunity to develop intelligent bioma-
terials and the next generation of
implantable devices for diagnostics,
therapeutics, and tissue engineering.
This dissertation is focusing on the
development of novel polymer-based
microelectrode arrays suitable for
use in intracranial electroencepha-
lographic recordings. Moreover,
the performances of novel thin film
materials and their surface modifica-
tions at micro- and nanoscales were
studied with physicochemical and
cellular experiments in order to de-
vise new solutions for further devel-
opment of biomedical microdevices.
dissertatio
ns | 014 | S
am
i Myllym
aa | N
ovel M
icro- a
nd
Na
no
-techn
ologica
l Ap
pro
ach
es for Im
pro
ving th
e Perfo
rma
nce of...
Sami MyllymaaNovel Micro- and
Nano-technological Approaches for Improving
the Performance of Implantable
Biomedical Devices
SAMI MYLLYMAA
Novel Micro- and Nano-
technological Approaches for
Improving the Performance
of Implantable Biomedical
Devices
Publications of the University of Eastern Finland
Dissertations in Forestry and Natural Sciences
Number 14
Academic Dissertation
To be presented by permission of the Faculty on Sciences and Forestry for public
examination in the Auditorium, Mediteknia Building at the University of Eastern
Finland, Kuopio, on Friday 19th November 2010, at 2 p.m.
Department of Physics and Mathematics
Kopijyvä Oy
Kuopio, 2010
Editors: Prof. Pertti Pasanen,
Prof. Tarja Lehto, Prof. Kai Peiponen
Distribution:
Eastern Finland University Library/Sales of Publications
P.O. Box 107, FI-80101 Joensuu, Finland
Tel: +358-50-3058396
http://www.uef.fi/kirjasto
ISBN 978-952-61-0216-0
ISSN 1798-5668
ISSNL 1798-5668
ISBN 978-952-61-0217-7 (PDF)
ISSN 1798-5676 (PDF)
Author’s address: Department of Physics and Mathematics
University of Eastern Finland
P.O. Box 1627
FI-70211 KUOPIO
FINLAND
E-mail: sami.myllymaa@uef.fi
Supervisors: Prof. Reijo Lappalainen, Ph.D.
Department of Physics and Mathematics
University of Eastern Finland
Prof. Juha Töyräs, Ph.D.
Department of Physics and Mathematics
University of Eastern Finland
Reviewers: Prof. Timo Jämsä, Ph.D.
Department of Medical Technology
University of Oulu
Prof. Sami Franssila, Ph.D.
Department of Material Science and Engineering
Aalto University School of Science and Technology
Opponent: Prof. Jukka Lekkala, Ph.D.
Tampere University of Technology
Department of Biomedical Engineering
ABSTRACT
Biomaterials are used in a wide variety of in vivo applications,
ranging from joint and dental implants to neural prostheses. The
ultimate success or failure of implants mainly depends on the
biological interactions (molecular, cellular, tissue) at the
implant/tissue interface. Recent advances in micro- and
nanotechnology offer a great opportunity to develop intelligent
biomaterials and the next generation of implantable devices
may well be of achieving the desired tissue-implant interaction
and resolving various biomedical problems.
The main aim of this thesis work was to design, fabricate and
evaluate a novel flexible microelectrode array suitable for use in
sub- or epidural electroencephalographic recordings. Other aims
were to investigate the opportunities to improve the
electrochemical and biological properties of neural interfaces
using modern micro- and nanotechnology tools as well as to test
whether the micropatterning of thin films can be used to guide
the cellular response on biomaterial surface.
The developed microelectrode array was implemented on
polyimide with platinum which achieved both mechanical
flexibility and high quality electrochemical characteristics as
demonstrated via impedance spectroscopy. Somatosensory and
auditory evoked potentials were successfully recorded with
epidurally implanted array in rats with excellent signal stability
over two weeks. Subsequently, the signal levels declined, most
probably due to the thickening of dura and the growth of scar
tissue around the electrodes. It was hypothesized that one
obvious reason for this limited life-span was the poor
biocompatibility of photosensitive polyimide used as an
insulation material in these arrays. However, this possibility
was excluded by in vitro cytotoxicity studies according to ISO
10993-5 standard. Furthermore, ultra-short pulsed laser
deposition was demonstrated to be an effective method to
produce nanotextured platinum surfaces as well as ultrasmooth
insulators for further development of neural interfaces.
Experiments with osteoblast-like cells and mesenchymal
stem cells on micropatterned biomaterial surfaces indicated that
even partial coating of silicon with a biocompatible material is
an effective way to enhance the cytocompatibility of silicon-
based biomedical micro-electromechanical systems. Moreover, it
was demonstrated that not only the chemical composition of the
materials, but also the shape, edges (height) and size of the
features used for surface patterning have a remarkable effect on
cell guidance.
Overall, the results of the present thesis provide a solid basis
for the further development of neural interfaces as well as other
types of implantable devices.
National Library of Medicine Classification: QT 36, QT 37, WL 102,
WV 270
INSPEC Thesaurus: biomedical materials; biomedical electrodes; thin
films; microelectrodes; electroencephalography; bioelectric potentials;
polymer films; silicon; platinum; metals; adhesion; surface texture;
surface topography; surface energy; nanopatterning; nanotechnology;
microfabrication; vapour deposited coatings; pulsed laser deposition
Yleinen suomalainen asiasanasto: biomateriaalit; implantit; polymeerit;
pii; platina; pinnoitus; pinnoitteet; pintarakenteet; pintailmiöt; kuviot;
mikrotekniikka; nanotekniikka; mikroelektrodit; EEG
Preface
This study was carried out during the years 2005-2010 in the
Department of Physics and Mathematics, the Department of
Neurobiology and BioMater Centre, University of Eastern
Finland, Microsensor Laboratory, Savonia University of Applied
Sciences, Kuopio, and the Department of Medicine, Helsinki
University Central Hospital.
I wish to express my deepest gratitude to everyone who has
contributed to the studies included in this thesis. Especially, I
wish to mention the following persons. First of all I would like
to thank my principal supervisor, Professor Reijo Lappalainen
for giving me the opportunity to work in his research group and
providing great facilities and scientific guidance for research
work. I am also grateful to my second supervisor, Professor Juha
Töyräs for his professional guidance and never-ending
enthusiasm and optimism during this thesis work.
I want to thank my official reviewers Professor Timo Jämsä
and Professor Sami Franssila for their constructive criticism and
valuable suggestions. I am also grateful to Ewen MacDonald for
linguistic review of this thesis.
The studies included in this interdisciplinary thesis would
not have been possible without the excellent collaboration with
colleagues in the several research units. Professor Heikki Tanila
is gratefully acknowledged for his valuable ideas and guidance
related to neuroscience. Kaj Djupsund is gratefully thanked for
his patient testing of microelectrode array prototypes in the
animal model. Professor Yrjö T. Konttinen, Emilia Kaivosoja and
Vesa-Petteri Kouri are sincerely thanked for great co-operation
in studies aimed to clarify the interaction phenomena between
cells and engineered biomaterial surfaces. Professor Mikko
Lammi, Virpi Tiitu, Sanna Miettinen and Aila Seppänen are
gratefully acknowledged for their contributions in cytotoxicity
testing. Moreover, I would like to thank all other co-authors in
the original publications of this thesis including Professor Juha E.
Jääskeläinen, Irina Gureviciene and Tarvo Sillat.
I sincerely thank all of the members in the Biomaterial
Technology Research Group and BioMater Centre for providing
supportive working atmosphere. In particular, the contributions
of Hannu Korhonen, Markku Tiitta, and Juhani Hakala have
been valuable. Aimo Tiihonen is sincerely thanked for technical
assistance in electronics. The former and current staff of
Microsensor Laboratory, including Matti Sipilä, Pasi Kivinen,
Mikko Laasanen and Ari Halvari is greatly acknowledged for
providing excellent microfabrication facilities and giving their
contributions to my work. Picodeon Ltd. Oy is acknowledged
for providing nanotextured depositions.
I am greatly indebted to my parents Seija and Heimo for their
love and support during my life. I want to thank my parents-in-
law, Anneli and Jukka, for their help and encouragements. I am
also grateful to my friends and relatives for their support.
Finally, I owe my deepest thanks to my beloved wife Katja, for
her irreplaceable love, support, and understanding. As a
research colleague, your contribution to my research has been
invaluable. Words are not enough for me to describe how
important you are as a wife and as the mother of our two little
daughters. Dearest Lumia and Neelia, you are the sunshine of
my life and have helped me to remember that there are more
important and valuable things in life than work.
This study was financially supported by Finnish Funding
Agency for Technology and Innovation (TEKES), the National
Graduate School of Musculoskeletal Diseases and Biomaterials,
Otto A. Malm Foundation, Ulla Tuominen Foundation,
Foundation for Advanced Technology of Eastern Finland and
COST, European Cooperation in Science and Technology.
Kuopio, October 2010
Sami Myllymaa
ABBREVIATIONS AND SYMBOLS
µCP microcontact printing
AEP auditory evoked potential
AFM atomic force microscopy
Ag silver
Ag-AgCl silver-silver chloride
Al aluminium
Al2O3 alumina (aluminium oxide)
ALP alkaline phosphatase
ANOVA analysis of variance
AP action potential
Au gold
BHK-21 baby hamster kidney fibroblast
Bio-MEMS biomedical micro-electromechanical systems
Ca calcium
Cl chlorine
CN carbon nitride
CNS central nervous system
Cr chromium
CVD chemical vapour deposition
CZ Czochralski process for silicon crystallization
DBS deep brain stimulation
DC direct current
DLC diamond-like carbon
DMEM Dulbecco’s modified Eagle’s medium
DNA deoxyribonucleic acid
ECG electrocardiography
EEG electroencephalography
EIS electrical impedance spectroscopy
EMG electromyography
ESEM environmental scanning electron microscope
EP evoked potential
FCS fetal calf serum
FDA Food and Drug Administration
FZ float-zone process for silicon crystallization
HMDS hexamethyldisilazane
hMSC human mesenchymal stem cell
Ir iridium
ISO International Organization for Standardization
ITO indium tin oxide
K potassium
KOH potassium hydroxide
MEA microelectrode array
MSC mesenchymal stem cell
MEMS micro-electromechanical systems
MS mineral staining
MSCGM mesenchymal stem cell growth medium
MTS cell proliferation assay based on (3-(4,5-
dimethylthiazol-2-yl)-5-(3 carboxymethoxy-
phenyl)-2-(4-sulfophenyl)-2H-tetrazolium)-salt
Na sodium
OC osteocalcin
PGMEA propylene glycol methyl ether acetate
PBS phosphate buffered saline
PCB printed circuit board
PDMS polydimethylsiloxane
PE polyethylene
PI polyimide
PNS peripheral nervous system
PSPI photosensitive polyimide
Pt platinum
PVD physical vapour deposition
RIE reactive ion etching
SAM self assembled monolayer
SD standard deviation of the mean
SEM scanning electron microscopy
SEM standard error of the mean
SEP somatosensory evoked potential
SFE surface free energy
Si silicon
SS stainless steel
Ta tantalum
Ti titanium
TiN titanium nitride
USPLD ultra-short pulsed laser deposition
UV ultraviolet
VEP visual evoked potential
ZIF zero insertion force
WC Warburg (polarization) capacitance
f frequency
aR average surface roughness as arithmetic mean
deviation of surface
pvR peak-to-valley roughness
SR resistance of electrode solution
pvR Warburg (polarization) resistance
mV transmembrane potential
ppV peak-to-peak signal amplitude
Z impedance
S total surface free energy D
S dispersive component of total surface free energy P
S polar component of total surface free energy
zeta potential
LIST OF ORIGINAL PUBLICATIONS
This thesis is based on, but not limited to, results presented in
the following original publications, which are referred in the
text by their Roman numerals (I-VI):
I Myllymaa, S., Myllymaa, K., Korhonen, H., Gureviciene, I.,
Djupsund, K., Tanila, H. & Lappalainen, R. (2008) Development
of flexible microelectrode arrays for recording cortical surface
field potentials. In: Proceedings of the 30th Annual International
Conference of the IEEE engineering in Medicine and Biology Society,
Vancouver, British Columbia, Canada, 20-24 August, 2008, pp.
3200-3203.
II Myllymaa, S., Myllymaa, K., Korhonen, H., Töyräs, J.,
Jääskeläinen, J.E., Djupsund, K., Tanila, H. & Lappalainen, R.
(2009) Fabrication and testing of polyimide-based micro-
electrode arrays for cortical mapping of evoked potentials.
Biosensors and Bioelectronics 24, 3067-3072.
III Myllymaa, S., Myllymaa, K., Korhonen, H., Lammi, M.J., Tiitu,
V. & Lappalainen, R. (2010) Surface characterization and in vitro
biocompatibility assessment of photosensitive polyimide films.
Colloids and Surfaces B: Biointerfaces 76, 505-511.
IV Myllymaa, S., Myllymaa, K. & Lappalainen, R. (2009) Flexible
implantable thin film neural electrodes. In: Recent Advances in
Biomedical Engineering, In-Tech, Vukovar, Croatia, Editor:
Ganesh R. Naik, ISBN 978-953-307-004-9, Chapter 9, pp. 165-189.
V Myllymaa, S.*, Kaivosoja, E.*, Myllymaa, K., Sillat, T., Korhonen,
H., Lappalainen, R. & Konttinen, Y.T. (2010) Adhesion,
spreading and osteogenic differentiation of mesenchymal stem
cells cultured on micropatterned amorphous diamond, titanium,
tantalum and chromium coatings on silicon. Journal of Materials
Science: Materials in Medicine 21 (1), 329-341.
(* equal contribution)
VI Kaivosoja, E.*, Myllymaa, S.*, Kouri, V.-P., Myllymaa, K.,
Lappalainen, R. & Konttinen, Y.T. (2010) Enhancement of silicon
using micropatterned surfaces of thin films. European Cells and
Materials, 19, 147-157. (* equal contribution).
The publications are reprinted with the kind permission of the
copyright holders. This thesis also contains unpublished results
related to publications III and VI as well as unpublished data
focused on the electrochemical characterization of nanotextured
bioelectrode surfaces.
AUTHOR’S CONTRIBUTION
Publications I and II concern the development of flexible
microelectrode array suitable for recording cortical surface field
potentials. The original idea of this electrode system was
proposed by H. Tanila. The author of this thesis carried out most
of the development work related to the electrode design,
material selection and fabrication processes. Apart from
depositions of metal thin-films, which were performed by H.
Korhonen and R. Lappalainen, the author carried out all array
fabrication steps and characterization experiments with a
contribution from K. Myllymaa. In vivo testing of electrodes was
performed by K. Djupsund and I. Gureviciene. The author wrote
the articles, after receiving constructive comments from the co-
authors. R. Lappalainen supervised the work.
Publication III concerns the cytocompatibility of the novel
photosensitive polyimide used as an insulator material in our
sensor prototypes (papers I and II). The design and fabrication
of test samples were performed by the author. The author also
conducted the surface characterization together with H.
Korhonen and K. Myllymaa. The author planned cell
experiments together with V. Tiitu, who also performed the
experimental studies. The author was responsible for analyzing
the data, presenting the results and writing the article, while
receiving constructive comments from the co-authors. R.
Lappalainen, V. Tiitu and M.J. Lammi supervised the study.
The author was mainly responsible for writing publication IV
(book chapter), which is a literature review discussing the main
requirements and features of flexible thin film neural electrodes,
this being supplemented with personal results in the area of
sensor development.
Publications V and VI concern the interactions between cells
and biomaterial surfaces. R. Lappalainen and Y.T. Konttinen
originally presented the idea of studying the effect of surface
micropatterning on the cellular response. Apart from thin film
depositions, performed by H. Korhonen and R. Lappalainen, the
author carried out all sample design and sample fabrication
steps in collaboration with K. Myllymaa. The surface
characterization analysis was mainly performed by the author
with some contributions from H. Korhonen and K. Myllymaa.
Experimental work and data analysis on cellular studies was
planned and carried out by E. Kaivosoja, with contribution of
V.-P. Kouri in paper VI, and Y.T. Konttinen. The author and E.
Kaivosoja contributed equally to preparing results and writing
paper V, assisted by R. Lappalainen and Y. T. Konttinen. E.
Kaivosoja and Y.T. Konttinen chiefly wrote paper VI, assisted by
S. Myllymaa and R. Lappalainen. These studies were supervised
by R. Lappalainen and Y.T. Konttinen.
Contents
1 INTRODUCTION ........................................................................... 21
2 BIOMEDICAL MICRO-ELECTROMECHANICAL
SYSTEMS .......................................................................................... 25
2.1 Microfabrication techniques ........................................................... 26
2.1.1 Photolithography ...................................................................... 27
2.1.2 Soft lithography ........................................................................ 29
2.1.3 Thin film deposition ................................................................. 32
2.1.4 Lift-off processing ..................................................................... 34
2.1.5 Etching ..................................................................................... 35
2.2 Substrate materials ........................................................................... 37
2.2.1 Silicon ...................................................................................... 37
2.2.2 Glass ......................................................................................... 38
2.2.3 Polymers................................................................................... 38
2.3 Advantages of microfabrication .................................................... 40
2.4 Bio-MEMS applications ................................................................... 42
3 ORIGIN OF BIOELECTRIC SIGNALS ...................................... 45
3.1 Excitable nerve cell .......................................................................... 45
3.1.1 Cell membrane and ion channels ............................................. 47
3.1.2 Transmembrane potential and equilibrium potentials ............. 47
3.2 Synaptic potentials and action potentials ..................................... 49
3.3 Recording electrical activity of the brain ...................................... 51
4 BIOELECTRODES .......................................................................... 55
4.1 Implantable electrodes .................................................................... 56
4.2 Material requirements for implantable electrodes ...................... 59
4.2.1 Electrode materials ................................................................... 60
4.2.2 Substrate materials .................................................................. 63
5 BIOCOMPATIBILITY .................................................................... 67
5.1 Biocompatibility testing .................................................................. 70
5.2 Cell-biomaterial interactions .......................................................... 71
5.2.1 Effect of surface topography on cellular responses ................... 71
5.2.2 Effect of wettability properties on cell-biomaterial
interactions .............................................................................. 75
5.2.3 Effect of charge distribution on cell-biomaterial interactions .. 76
5.2.4 Effect of surface micropatterning on cellular responses ........... 78
6 AIMS OF THE PRESENT STUDY ............................................... 79
7 MATERIALS AND METHODS ................................................... 81
7.1 Micro- and nanofabrication of electrodes and other samples ... 82
7.1.1 Preparation of photomasks ....................................................... 82
7.1.2 Flexible polyimide-based microelectrode arrays ....................... 82
7.1.3 Micropatterned biomaterial surfaces on silicon ....................... 86
7.1.4 Tailored surfaces produced by ultra-short pulsed laser
deposition ................................................................................. 88
7.1.5 Spin-coated polyimide films for cytotoxicity testing ................ 89
7.2 Microscopic characterization of surfaces ...................................... 90
7.2.1 Scanning electron microscopy ................................................. 90
7.2.2 Atomic force microscopy .......................................................... 90
7.2.3 Contact angle measurements and determination of surface
free energies.............................................................................. 91
7.3 Electrochemical characterization of surfaces ............................... 92
7.3.1 Electrochemical impedance spectroscopy ................................. 92
7.3.2 Zeta potential measurements ................................................... 94
7.4 Recording of evoked potentials in rats ......................................... 95
7.5 Cell culture studies .......................................................................... 96
7.5.1 Cell lines ................................................................................... 96
7.5.2 MTS assay ................................................................................ 98
7.5.3 Scanning electron microscopy of cultured cells ....................... 98
7.5.4 Immunofluorescence and confocal laser scanning
microscopy ............................................................................... 99
7.6 Statistical analyses ......................................................................... 101
8 RESULTS ........................................................................................ 103
8.1 Evaluation of microelectrode arrays ........................................... 103
8.2 Evaluation of nanorough electrode surfaces .............................. 109
8.3 Cytocompatibility testing of insulator materials ....................... 111
8.4 Micropatterned biomaterial coatings .......................................... 115
8.4.1 Behaviour of human mesenchymal stem cells ........................ 116
8.4.2 Behaviour of osteoblast-like (SaOS-2) cells ............................ 120
9 DISCUSSION ................................................................................ 125
9.1 Microelectrode arrays for intracranial recordings .................... 125
9.2 Material science strategies to improve the performance of
microelectrode arrays .......................................................................... 128
9.3 Micropatterned biomaterial surfaces .......................................... 130
10 CONCLUSIONS .......................................................................... 133
11 REFERENCES .............................................................................. 135
APPENDIX: ORIGINAL PUBLICATIONS I-VI
21
1 Introduction
Developments in micro-electromechanical systems (MEMS)
technology have introduced a variety of breakthrough products
in the fields of microelectronics, telecommunications and the
automotive industry (Judy 2001, Madou 2002). In recent years,
the biomedical applications of micro-electromechanical systems
(bio-MEMS) have attracted growing interest in many
application areas such as diagnostics, therapeutics and tissue
engineering (Saliterman 2006, Grayson et al. 2004, Nuxoll &
Siegel 2009, Betancourt & Brannon-Peppas 2006, Bashir 2004,
Urban 2006). Bio-MEMS is a powerful technology capable of
ever-greater functionality and cost reduction in smaller
biomedical devices for improved diagnostics and treatment. The
merging of biology with micro/nanotechnology has been
postulated to trigger a scientific and technological revolution in
the future (Kotov et al. 2009). Although some applications such
as blood analysis cartridges, catheter pressure sensors and
cochlear implants have been already commercialized, the vast
majority of biomedical applications are still being investigated
or undergoing clinical trials (Saliterman 2006, Urban 2006).
Brain research can be considered as a one of the most
challenging scientific areas. The very first biological MEMS
devices, i.e. multi-sensor neural probes, were developed for
neuroscientists in order to facilitate studying of neuronal
activities at the tissue and cellular level (Urban et al. 2006). In
addition to pure scientific research, there has been a growing
interest in the clinical applications of stimulating and recording
neural electrodes and prostheses (DiLorenzo & Bronzino 2008).
These neural interfaces can benefit many patients with neural
disorders such as impaired hearing (cochlear implant) or
neuropathic pain (deep brain stimulators). Some extremely
challenging applications such as retina implants which may be
able to restore the vision or sieve and cuff electrodes potentially
1 - Introduction
22
evoking nerve regeneration in paralyzed patients are still under
intensive development (Stieglitz & Mayer 2006b, DiLorenzo &
Bronzino 2008).
To date, a wide variety of neural electrodes have been
utilized in neural interfaces starting with the early electrolyte-
filled micropipettes and subsequent metal electrodes to the
current emerging MEMS-based electrodes. The vast majority of
these MEMS devices have been fabricated on silicon (Si) even
though there is a huge mismatch between mechanical properties
of the neural tissue and Si causing many adverse tissue effects
(Polikov et al. 2005). Moreover, Si is not per se a biocompatible
material (Voskerician et al. 2003), and it is a poor substrate for
cell adhesion, even being slightly cytotoxic (Liu et al. 2007).
These weaknesses of Si could limit the integration of Si-based
devices into the human body. Recently, there has been a
growing interest toward developing polymer-based interfaces
that could be flexible enough to mimic biological tissue,
reducing mechanical damage and evoking less adverse tissue
reactions (Stieglitz & Mayer 2006b, Cheung 2007, HajjHassan et
al. 2008, DiLorenzo & Bronzino 2008).
Biomaterials are used in many in vivo applications, ranging
from joint and dental implants to neuroprosthetic devices.
Depending on particular specifications for each such application,
a set of different material characteristics, such as mechanical,
chemical and electrical properties, influence the performance of
biomaterial/prosthetic devices in a specific manner (Williams
2008). In all cases, however, the ultimate success or failure
depends on the biological interactions (molecular, cellular,
tissue) at the implant-tissue interface (Puleo & Nanci 1999,
Williams 2008, Navarro et al. 2008, Polikov et al. 2005).
Therefore, several approaches have been introduced to modify
surface properties, such as surface topography on the micro-
and nanoscales (Flemming et al. 1999, Martinez et al. 2009),
surface energy (van Kooten et al. 1992, Hallab et al. 2001, Lim et
al. 2008) and surface charge (Krajewski et al. 1996, Krajewski et
al. 1998, Cai et al. 2006), in order to achieve significant effects on
the functions of proteins and thus on cells. These enhanced
1 - Introduction
23
effects on cellular functions are crucial in subsequent
improvements in new tissue formation and integration of
implants in the surrounding tissues. Recently, much attention
has been paid to mesenchymal stem cells (MSCs) due to their
role in tissue repair and their potential in regenerative medicine
and tissue engineering in which these kinds of cells are grown
on biomaterial scaffold, which provides structural support and
substrate for cellular adhesion (Zippel et al. 2010, Pittenger et al.
1999, Stoddart et al. 2009). The ability to influence the adhesion,
distribution and behaviour of cells on biomaterial surfaces and
the knowledge of the nature of cell-substrate interaction has
therefore become increasingly important.
In this thesis, novel flexible polyimide (PI)-based
microelectrode arrays (MEA) suitable for sub- or epidural
electroencephalographic (EEG) recordings were developed. The
suitabilities of different MEMS materials as well as their surface
modifications for implantable applications were investigated
using electrochemical testings and cell experiments. Moreover,
the effect of surface micropatterning achieved using photo-
lithographic and thin film techniques on the behaviour of MCSs
and osteoblast-like cells was studied.
1 - Introduction
24
25
2 Biomedical micro-
electromechanical systems
Micro-electromechanical systems refer to miniature devices that
are fabricated using techniques originally developed and widely
utilized in the microelectronics (integrated circuits) industry,
and then modified, e.g. by adding mechanical components, such
as beams, gears, and springs, for the creation of microstructures
and microdevices such as sensors and actuators. MEMS, as well
as its interchangeable acronyms, Micromachines (popular in
Asia) and Microsystems (popular in Europe), refer to the
miniature devices which have at least some of their dimensions
in the micrometer range. According to the widest definition,
MEMS comprises all devices and systems produced by
micromachining other than integrated circuits or other
conventional semiconductor devices (Judy 2001). At the
beginning of MEMS era, in the 1970s and 1980s, the field was
dominated by mechanical applications, but today most new
applications are either communication/information-related or
chemical and biological in nature (Madou 2002). MEMS
technology has enabled low-cost, high-functionality micro-
devices in some widespread application areas, such as printer
cartridges for ink jet printing and the accelerometers responsible
for deployment of airbags in modern automobiles (Judy 2001).
In recent years, the biological and biomedical applications of
MEMS, which are commonly referred to as bio-MEMS, have
gained increasing world-wide interest (Saliterman 2006, Nuxoll
& Siegel 2009, Urban 2006). Bio-MEMS are generally defined as
‘‘devices or systems, constructed using techniques inspired from
micro/nanoscale fabrication, that are used for processing,
delivery, manipulation, analysis, or construction of biological
and chemical entities’’ (Bashir 2004). At the present moment,
bio-MEMS technology is a topic of intense research and
2 – Biomedical micro-electromechanical systems
26
development activity with a wide variety of applications in the
fields of cell and molecular biology, pharmaceutics and
medicine so that it includes the areas of therapeutics,
diagnostics and tissue engineering (Saliterman 2006, Grayson et
al. 2004, Nuxoll & Siegel 2009, Betancourt & Brannon-Peppas
2006, Voldman et al. 1999, Bashir 2004).
2.1 MICROFABRICATION TECHNIQUES
A number of techniques are used to form bio-MEMS objects
with dimensions ranging from the micrometer to millimeter
scale. Some of these techniques have been adopted directly from
the industry of integrated circuits whereas some others have
been specifically developed for this novel purpose (Judy 2001,
Voldman et al. 1999, Saliterman 2006, Li et al. 2003). The
microfabrication process is typically a process flow that utilized
these techniques in a sequential manner to produce the desired
structure. MEMS devices can be built within the bulk of a
substrate material in what is referred to as bulk micromachining,
or if it is on the surface of the substrate it is known as surface
micromachining (Madou 2002). However, a combination of bulk
and surface micromachining is commonly used (Voldman et al.
1999). The most important microfabrication techniques in bio-
MEMS are photolithography, soft lithography, thin film
deposition and etching (Saliterman 2006, Li et al. 2003).
Photolithography is used to transfer the desired shape onto a
material through the selective exposure of a photosensitive
polymer (Madou 2002). Soft lithography is a set of different
patterning techniques based on polymer block used as a stamp,
mold or stencil for carrying out surface patterns and micro-
structures (Xia & Whitesides 1998, Li et al. 2003, Saliterman
2006). Thin film deposition which consists of numerous different
techniques is used to form thin layers with thickness ranging
from atomic layer to a few microns on the surface of a substrate.
Etching is used to remove material selectively from the surface
2 – Biomedical micro-electromechanical systems
27
of the microsystem by either chemical or physical processes
(Madou 2002).
2.1.1 Photolithography
Photolithography is a basic microfabrication technique widely
employed to create the desired patterns onto a material. The
photolithographic patterning is composed of a number of
process steps (Fig. 1).
photoresist
substrate
photomask
UV exposure
resist deposition
and soft baking
latent image
mask alignment
(x, y, φ)
positive photoresist negative photoresist
development
Figure 1: Photolithographic patterning process. A photomask with opaque regions in
the desired pattern is used to selectively expose a photosensitive polymer (photoresist).
Depending on the polarity of the resist used, it will become more soluble (positive-tone
photoresist) or crosslinked (negative photoresist) after ultraviolet light illumination,
thus generating the desired pattern after developing in liquid chemicals.
Firstly, a pattern is drawn using computer assisted design
software and this is transferred onto a mask. The photomasks
are typically transparent glass plate blanks covered with an
opaque material (usually chromium, Cr) in the defined pattern
(Madou 2002). The masks are usually prepared by a commercial
mask manufacturer using electron beam or laser writing (Wu et
2 – Biomedical micro-electromechanical systems
28
al. 2002). If the feature sizes and tolerances in the mask are
relatively large, an option may be to utilize low-cost
transparency films as a photomask (Deng et al. 1999, 2000, Wu
et al. 2002, Gale et al. 2008). After producing the mask, a
photosensitive polymer, i.e. photoresist, is spin-coated onto a
substrate material, such as a Si wafer. The resulting photoresist
thickness, t, is a function of spinner rotational speed, solution
concentration, and molecular weight (determined by intrinsic
viscosity) and can be estimated by (Madou 2002):
.
KCt (1)
In this equation, K is the overall calibration constant, C is the
polymer concentration (g/100 ml), η is the intrinsic viscosity and
ω is the spinner rotational speed (rotations per minute, rpm). α,
β and γ are exponential factors dependent on process.
Typically, spinning speeds of 1500-8000 rpm are used to
achieve resist thicknesses of about 0.5-2 µm (Madou 2002). The
photoresist film is then prebaked at 75 to 100°C on a hotplate or
in an oven to remove solvents and to promote adhesion of the
resist layer to the substrate (Saliterman 2006, Madou 2002). In
the following step of exposure, the photomask is placed on top
of the photoresist-coated wafer in close proximity or even in
contact with it and then ultraviolet (UV) light is used to
illuminate the photoresist film through the photomask. With
this procedure, the solubility difference between the exposed
and unexposed regions is achieved, and depending on the
polarity of the photoresist used, exposed or unexposed areas can
then be removed by dissolving them in a developing solution. In
the case of a positive-tone photoresist, the exposed regions
break down and become more soluble in the developing
solution, whereas in a negative-tone photoresist, the exposed
areas become crosslinked and insoluble in the developing
solution. The resulting photoresist pattern can now be used as a
protective mask in following microfabrication processes such as
in the etching step to prevent the covered substrate to be
2 – Biomedical micro-electromechanical systems
29
removed and in thin film deposition to pattern metallization
layer on the surface of substrate. After the followed process is
finished, the photoresist can be removed, leaving the pattern
design on the substrate. The removal of photoresist is usually
performed by sonication in an organic solvent, e.g. acetone.
(Madou 2002)
The resolution of photolithography depends on the quality of
mask and the wavelength of applied electromagnetic radiation
(light). The typical wavelengths used are 436 nm, 365 nm, 248
nm and 193 nm (Madou 2002). The shorter the wavelength of
light used, the higher resolution, i.e. smaller feature sizes are
possible. Although photolithography is the main technology in
the production of microscale features in MEMS, it has also a few
short-comings. It requires clean-room facilities, and therefore it
is not an inexpensive technology; it is poorly suited for
patterning nonplanar surfaces and it is directly applicable only
to a limited set of photosensitive polymers (Xia & Whitesides
1998). Furthermore, photolithography requires the use of strong
solvents, meaning that it is poorly compatible with the
patterning biological molecules used in many applications e.g. in
biosensing, medical implants and the control of cell adhesion
and growth (James et al. 1998, Li et al. 2003, St. John et al. 1997).
2.1.2 Soft lithography
Soft lithography consists of a set of non-photolithographic
patterning methods based on self-assembly and replica molding
for carrying out micro- and nanofabrication (Saliterman 2006). It
provides a low-cost, effective, and biocompatible strategy for
the formation and manufacturing of surface patterns and
structures with feature dimensions ranging from 30 nm to 100
µm (Xia & Whitesides 1998). The key element of soft
lithography is an elastomeric block with patterned relief
structures on its surface that subsequently utilized in different
soft lithographic techniques such as in microcontact printing,
stencil patterning and microfluidic patterning (Li et al. 2003).
The commonest material used for the production of elastomeric
block is polydimethylsiloxane (PDMS). This material has several
2 – Biomedical micro-electromechanical systems
30
desirable properties such as durability, optical transparency,
biocompatibility, flexibility, gas permeability and amenability to
different surface modifications (Whitesides et al. 2001, Li et al.
2003). The major advantages of soft lithography compared to
conventional photolithography are that numerous solid and
liquid materials other than photoresists can be patterned, and it
is suitable for curved or complicated surface geometrics and
even 3D structures (Xia & Whitesides 1998, Gale et al. 2008).
The soft lithography process starts with conventional
photolithographic steps to create a PDMS block which will be
used later as a mold or a stamp (Fig. 2a).
Si
PDMS
Si
photoresist
Pour PDMS over master
PDMS
Cure and peel off stamp
Alkanethiol ”ink”
PDMS
Substrate
Au
PDMS
Microcontact print
Substrate
Substrate
Remove stamp
SAM (2-3 nm)
A B
Figure 2: (a) Fabrication of a polydimethylsiloxane (PDMS) stamp. PDMS is poured
onto a silicon wafer coated with patterned photoresist. After the curing step, the stamp
can be removed. (b) Microcontact printing with the PDMS stamp. The stamp is
coated with alkenethiol ink, and then it is pressed on gold coated substrate. After
application of gently pressure for a few seconds, a self-assembly monolayer (SAM) is
formed on gold surface. Figure is inspired by Xia & Whitesides (1998).
Photoresist is spin-coated onto a substrate (e.g. Si wafer). The
height of the resist layer (i.e. pattern relief) can be controlled by
the viscosity of the solution and the spin speed (Madou 2002).
EPON™ SU-8, originally developed by IBM, and recently
marketed by MicroChem Corp. (Newton, MA, USA), is one the
most popular photoresists being used in stamp fabrication (Li et
al. 2003). This negative-tone epoxy-resist permits the fabrication
2 – Biomedical micro-electromechanical systems
31
of tall structures, even more than 1 mm in height with a superior
aspect ratio (Lorenz et al. 1997, 1998). The photoresist layer is
illuminated with UV light through a photomask with the
desired pattern, and then developed. In the case of SU-8, the
resist areas exposed to UV light will remain in the development.
Liquid PDMS precursor is then cast over the patterned substrate
and cured at an elevated temperature. After cooling, the PDMS
stamp can be peeled off from the master. A replica of the
original template in a functional material can be generated by
molding against the patterned PDMS block. (Li et al. 2003, Xia &
Whitesides 1998, Whitesides et al. 2001)
The microcontact printing (µCP), also known as
microstamping, is based on the patterned transfer of the
material (‚ink‛) from the surface of the PDMS stamp onto the
receiving surface on a sub-micrometer scale (Fig. 2b) (Qin et al.
2010, Xia & Whitesides 1998, Kumar & Whitesides 1993). The
material of interest to be transferred, e.g. alkanethiol, is applied
onto the surface of the stamp with patterned surface relief. The
dried stamp is then placed face down on the substrate, e.g. gold
(Au)-coated glass, and gentle pressure is applied for a few
seconds (Qin et al. 2010). The elastic PDMS material enables an
excellent contact between the stamp and the substrate surface,
and molecules that touch the surface are transferred from the
stamp to the substrate, forming a self-assembly monolayer
(SAM). The formed SAM layer can be used as a protective layer
in subsequent microfabrication steps such as etching or
deposition (Xia & Whitesides 1998). The chemicals used in the
µCP to form SAMs typically have a chemical formula of
Y(CH2)nX, where Y is the anchor and X is the headgroup
(Saliterman 2006). A typical anchor group in alkanethiols is
sulfide since this promotes the binding of thiol very tightly to
Au. Typically, CH3 and COOH are used as the headgroups. The
choice of headgroup has a great influence on the wettability
properties, i.e. hydrophobicity/hydrophilicity, of the surface and
subsequently to protein and cell binding. Furthermore, the
headgroups can be chemically modified, e.g. an arginine-
glycine-aspartate peptide that enhances cell attachment (Hersel
2 – Biomedical micro-electromechanical systems
32
et al. 2003, LeBaron & Athanasiou 2000, Shi et al. 2008, Chua et
al. 2008). The µCP offers the advantage over traditional photo-
lithographic patterning methods in that it requires no strong
chemicals, making it suitable for patterning biologically active
layers (Xia & Whitesides 1998, Whitesides et al. 2001, Saliterman
2006). In addition to alkanethiols, other peptides, proteins, poly-
saccharides and other molecules can also be stamped (Bernard
et al. 1998, Branch et al. 1998, James et al. 1998, Li et al. 2003).
Stencil patterning is the second soft lithography technique in
which a membrane stencil is fabricated by casting PDMS to the
top of a photoresist master, which creates holes with the shape
of the master features. The PDMS stencil can be used as a mask
for selective adsorption of cell-adhesive proteins to promote
cellular patterning. (Folch et al. 2000, Li et al. 2003, Folch &
Toner 2000)
Micromoulding in capillaries, the third soft lithographic
technique, employs a PDMS mold to build up microchannels
against a substrate. These microchannels can be used to pattern
fluid materials onto a substrate (Kim et al. 1995, 1996). A low
viscosity prepolymer is applied at the open ends of the channels.
The channels become filled with polymer due to capillary forces.
After curing of the prepolymer, the PDMS mold is removed and
the three-dimensional polymer microstructures are revealed.
Subsequently, these microstructures can be used for selective
delivery of different cell suspensions to specific locations of a
substrate resulting in micropatterns of attached cells (Folch &
Toner 1998, Tan & Desai 2003).
2.1.3 Thin film deposition
The application of thin layers of materials is a common
procedure in microfabrication. Thin films refer to thin material
layers ranging from the thickness of atomic layers to several
micrometers in thickness (Madou 2002). Thin films can play a
structural or functional role in the device process (Smith 1995,
Hsu 2008). All material classes, i.e. metals, ceramics, polymers,
composites and biological compounds, can be deposited. They
can be used either as sacrificial layers or mask layers that
2 – Biomedical micro-electromechanical systems
33
selectively protect the substrate material during etching or they
can be used as electrical insulators or conductors such as
electrodes and transmission lines in sensors. Two common thin
film materials used as insulators in Si-based MEMS are silicon
dioxide (SiO2) and silicon nitride (Si3N4) (Voldman et al. 1999).
Deposition techniques can be divided into physical vapour
deposition (PVD) and chemical vapour deposition (CVD)
methods (Madou 2002). In general, the PVD methods are based
on the production of a condensable vapour by physical means
and its deposition on a substrate as a thin film. Vacuum or low
pressure gaseous environment is used in the PVD processing, in
which deposition species are atoms or molecules or both. The
most common PVD techniques are (1) evaporation in a vacuum,
(2) sputter deposition, (3) arc-vapour deposition, (4) laser
ablation, and (5) ion plating (Saliterman 2006). On the other
hand, the CVD methods are based on chemical reactions which
take place at a heated substrate surface to deposit a solid film.
Gas phase reactions between chamber gases are not desirable
because they often lead to poor adhesion, low density and high
detect films. The main advantages of the CVD methods are the
possibility to fill holes, cavities and other 3D structures, good
adhesion between coating and substrate, and excellent thickness
uniformity of coating. The major shortcomings are the toxicity of
the by-products as well as a high processing temperature, being
above 600°C which eliminates the use of CVD with heat
sensitive materials like polymers (Hsu 2008, Saliterman 2006,
Madou 2002).
In addition, thin films can be produced by other techniques,
such as spin coating, electrolytic deposition and electroless
deposition. Spin-coating is typically used to deposit thin
polymer films such as photoresists in photolithography. In the
electroplating process, the substrate is immersed in an
electrolyte solution. When a voltage is applied between a
substrate (working electrode) and an inert counter electrode,
such as platinum (Pt) in the liquid, chemical reduction-oxidation
processes take place resulting in the formation of a layer of
material on the surface of the substrate (Paunovic & Schlesinger
2 – Biomedical micro-electromechanical systems
34
2006, Schlesinger & Paunovic 2000). An electroless plating
process does not require any external electrical potential but the
deposition spontaneously happens on any surface which forms
a sufficiently high electrochemical potential with the electrolyte
solution. Although electroless deposition (compared to the
electroplating) is much easier to set-up and use, without the
need for an electrical supply and its connections to electrodes, it
is also more difficult to control with regards to film thickness
and uniformity. Electroplating and electroless plating are
typically used in MEMS to form conductive metal (e.g. platinum,
gold, copper, nickel) or conductive polymer (e.g. polypyrrole,
polyaniline) thin films (Madou 2002).
2.1.4 Lift-off processing
Lift-off is a simple and easy method for patterning thin film
metal layers. It is especially useful for patterning catalytic metals,
such as Pt, which do not lend themselves well to direct wet
etching (Madou 2002, Hsu 2008). The lift-off process starts with
a photolithographic step, in which the photoresist layer is
deposited and patterned as described in chapter 2.1.1. Thin film
of desired material is then deposited all over the substrate,
covering the photoresist and areas in which the photoresist has
been cleared. Thereafter, the substrate is immersed in a solvent
that dissolves the remaining soluble photoresist underneath the
metal, and lifts off the metal. Only the metal which has been
deposited directly on the substrate leaves and forms the final
pattern on the substrate. In order to achieve clean results in lift-
off processing, photolithographic and thin film deposition
processes need to be optimized (Franssila 2004). Undercut
photoresist patterns are beneficial and can be realized by
reducing exposure dose and/or increasing the developing time
of negative photoresist. The substrates should be kept at
temperatures below 200 to 300°C during metal deposition to
avoid hardening and flow of photoresist (Madou 2002).
2 – Biomedical micro-electromechanical systems
35
2.1.5 Etching
Etching techniques are used to create topographical patterns on
a surface via the selective removal of material. These techniques
can be roughly divided into wet and dry etching which are
based on the utilization of liquid chemicals (etchants) or reactive
ions/vapour-phase etchant, respectively (Madou 2002). The
etching process is isotropic in its nature if it removes material in
all directions equally, leading to mask undercutting and a
rounded etch profile (Fig. 3a), or it is anisotropic if removal
occurs with different etch rates in different directions in the
material (Fig. 3b,c) (Voldman et al. 1999). In the simple wet
etching technique, the sample is immersed in a container filled
with a liquid solution (etchant). The sample is covered by a
mask which leads to the selective removal of material (Madou
2002). The most critical task is to find a mask material that will
not dissolve or at least is etched much more slowly than the
sample material in question. A wide variety of etchants such as
buffered hydrofluoric acid, Aqua regia, Piranha solution (i.e.
mixture of sulfuric acid and hydrogen peroxide), potassium
hydroxide (KOH), tetramethylammonium hydroxide and
hydrochloric acid can be used to etch different MEMS materials
(Williams & Muller 1996, Williams et al. 2003). Unfortunately,
etchants are usually isotropic leading to etch profiles with large
undercuts particularly when one is machining thick films.
Additionally many wet etch chemicals are toxic and, thus they
are poorly suited for the state-of-the-art bio-MEMS processes.
Anisotropic wet etch is possible on some single crystal materials,
such as Si in which anisotropic etchants (e.g. KOH) dissolve Si
rapidly in the direction <100>, but almost not at all in the
direction of <111>. Thus cavities with trapezoidal cross-sections
with a characteristic 54.7° sidewall will be created on a [100]-
oriented wafer (Fig. 3c) (Madou 2002, Voldman et al. 1999).
2 – Biomedical micro-electromechanical systems
36
etching mask
substrate
54.7
A B C
Figure 3: Different etching profiles resulted in (a) isotropic wet etching, (b)
anisotropic dry etching, and (c) anisotropic wet etching of silicon using potassium
hydroxide. Figure is modified from Voldman et al. (1999).
In order to achieve anisotropic etching with high resolution,
different dry etching methods are used. Reactive ion etching
(RIE) is one of the most popular dry etching methods in which
the substrate is placed inside a reactor and several gases are
introduced (Kovacs et al. 1998). The gas molecules can be made
to disintegrate into ions using a radiofrequency (13.56 MHz)
power source. Accelerated ions collide onto the surface of the
material to be etched, react with the surface molecules forming
another gaseous material. In addition to this chemical aspect,
there is also a physical part present in the process. If the plasma
particles, i.e. neutral radicals and ions, have enough energy, they
can remove atoms out of the materials without any chemical
reaction. This process can be quite complicated since there are
many parameters to be adjusted. However, with optimal
adjustment, it is possible to etch almost straight down without
undercutting, which provides much higher resolution compared
to wet etching. (Madou 2002)
A special improvement of RIE technique is the deep RIE,
based on the so called ‚Bosch process‛ (Laermer 1996) in which
an etch depth of hundreds of micrometers can be achieved with
vertical sidewalls. The deep RIE process is based on two
different alternating gas compositions in the reactor. The first
gas composition forms a polymer film on the surface of the
substrate, and the second gas composition etches the substrate
(Saliterman 2006). The polymer is immediately sputtered away
by the second etching gas, but this happens only on the
2 – Biomedical micro-electromechanical systems
37
horizontal surfaces and not the sidewalls protecting them from
etching (Laermer 1996). Very high etching aspect ratios, even
50:1, can be achieved (Kovacs et al. 1998). Dry etching is a
widely used method in the integrated circuit industry to achieve
high resolution features, but in many cases it is not essential in
the production of bio-MEMS.
2.2 SUBSTRATE MATERIALS
Traditionally, MEMS devices have been fabricated on
microelectronics related materials, such as Si and glass. Recently,
however, there has been a growing interest towards rubber and
plastic substrate materials due to their suitable mechanical
properties, enhanced biocompatibility, rapid prototyping and
inexpensive manufacturing techniques available (Wilson et al.
2007).
2.2.1 Silicon
Silicon is the most widely used material in microchips and
MEMS devices since it is straightforward to grow oxide layers to
form dielectrics on its surface and it has excellent semiconductor
properties over a wide temperature range. Silicon wafers used
in semiconductor/MEMS industry is produced using two
alternative crystallization methods: Czochralski process (CZ)
and float-zone (FZ) process (Pearce 1988). In the CZ method, a
seed crystal is dipped into molten Si and pulled upwards with
simultaneous rotation. By optimizing the rate of pulling and the
speed of rotation as well as other relevant parameters such as
temperature, it is possible to extract a large-diameter cylindrical
single crystal ingot from the melt. Czochralski process is a more
common and cheaper method compared to FZ, but the wafers
produced using the CZ method contain slightly more impurities.
Impurities in the CZ wafers originate from the quartz crucible,
containing the Si melt that dissolves during the process. In the
FZ method, a local melted zone produced by radiofrequency
field is slowly passed along a polycrystalline rod. The seed
2 – Biomedical micro-electromechanical systems
38
crystal is used at one end in order to launch the single
crystalline growth. Impurities in the molten zone tend stay in
the molten zone rather than being incorporated into the
solidified region, hence allowing an extremely pure single
crystal region being left after the molten zone has passed. The
largest Si ingots, produced using CZ method, are 300-450 mm in
diameter and 1 to 2 meters in length today. The diameters of the
FZ ingots are typically less than 200 mm due to the surface
tension limitations encountered during the crystallization
process (Franssila 2004). Thin Si wafers with a typical thickness
of 0.25-1.0 mm, are cut from these ingots and then polished to a
very high smoothness. Silicon can be doped with impurity
atoms such as boron or phosphorous, i.e. changing it into n-type
or p-type Si, in order to enhance the electrical conductivity.
(Madou 2002, Franssila 2004)
In addition to the excellent semiconductor properties of Si, it
has also superb mechanical properties, enabling the design of
MEMS structures (Petersen 1982). Silicon micromachining
techniques are established and widely available. The major
drawbacks of Si for bio-MEMS applications include its limited
biocompatibility, unfavourable mechanical properties (rigidity,
fragility) as well as the relatively high material and processing
cost (Cheung 2007), which makes it less attractive for use in
disposable biomedical devices.
2.2.2 Glass
In spite of the limited number of micromachining techniques
available for glass substrates (compared to Si), this material is
used in bio-MEMS applications due to some unique properties,
most notably optical transparency. Glasses with a wide variety
of compositions can be used. Fused silica (pure amorphous SiO2)
and borosilicate (e.g. Pyrex®) are examples from commonly
used glass materials (Voldman et al. 1999).
2.2.3 Polymers
Polymers are very attractive for bio-MEMS applications due to
their suitable mechanical properties, enhanced biocompatibility,
2 – Biomedical micro-electromechanical systems
39
optical transparency, ability to modify their bulk and surface
properties in order to improve their functionality, and the
availability of rapid prototyping and mass production processes
(e.g. injection molding, hot embossing) (Wilson et al. 2007).
These issues enable the fabrication of low-cost disposable
microdevices for clinical applications. Polydimethylsiloxane,
polyimide, epoxy resins and parylene are some examples of
widely utilized polymers in bio-MEMS (Seymour et al. 2009,
Cheung 2007, Saliterman 2006).
Polyimide has a long history of use in microelectronics e.g. as
an encapsulation and stress buffer material on semiconductor
chips as well as a substrate material in flexible printed
boards/cables (Ghosh & Mittal 1996). It possesses numerous
desirable properties such as excellent resistance to solvents,
strong adhesion to metals and metal oxides and good dielectric
properties (Wilson et al. 2007). Recently, PI has been introduced
as a substrate/insulating material in numerous bio-MEMS
applications, such as neural interfaces (Boppart et al. 1992,
Rousche et al. 2001, Hollenberg et al. 2006, Molina-Luna et al.
2007, Takahashi et al. 2003, Cheung et al. 2007, Owens et al.
1995, Stieglitz 2001, Mercanzini et al. 2008, Patrick et al. 2008,
Spence et al. 2007) due to its desirable mechanical and dielectric
properties and good biocompatibility (Wilson et al. 2007,
Stieglitz et al. 2000, Richardson et al. 1993).
Polydimethylsiloxane, also familiar as a soft lithographic
stamp material, is known to be biocompatible and it is approved
for use as an implanted material in medical devices, i.e.
approved by the Food and Drug Administration (FDA). It has
been used as a biomaterial in catheters, pacemakers, and ear and
nose implants (Visser et al. 1996). PDMS is a highly flexible and
optically transparent material and it is permeable to gases. It is
also amenable to different surface modifications, which can be
used to tailor the biochemical functionality of the PDMS
(McDonald & Whitesides 2002, Mata et al. 2005). PDMS can be
conformed to submicron features to develop different
microstructures (Xia & Whitesides 1998) such as microfluidic
components (Ng et al. 2002, Folch et al. 1999) for bio-MEMS.
2 – Biomedical micro-electromechanical systems
40
An epoxy-based negative photoresist SU-8 is one of the most
widely used thick-film photoresists in MEMS (Lorenz et al. 1997,
1998, Wilson et al. 2007). It has been utilized as a structural
material because it can be produced with a wide range of
thicknesses from below 1 µm to even more than 1 mm (Lorenz
et al. 1998). Recently, it has been employed in numerous bio-
MEMS applications, including analytical microfluidic systems
(Tuomikoski 2007) and neural sensors (Hollenberg et al. 2006,
Tijero et al. 2009, Altuna et al. 2010). Parylene is a thermoplastic
polymer that can be vapour-deposited at room temperature to
create pin-hole free, optically transparent barrier coatings that
are stress-free, chemically and biologically inert, and minimally
permeable to moisture (Saliterman 2006, Seymour et al. 2009).
2.3 ADVANTAGES OF MICROFABRICATION
Recent developments of MEMS and associated nanotechnology
represent great opportunities in the design of miniature, smart,
and low-cost biomedical devices that could revolutionize
biomedical research and clinical practice (Urban 2006,
DiLorenzo & Bronzino 2008, Nuxoll & Siegel 2009). Micro- and
nanotechnology can be used either to improve the performance
of an existing device or to enable development of an entirely
new device. The bio-MEMS devices have many advantages over
their macroscopic counterparts. The MEMS techniques enable
development of small size devices that may be easier to use (e.g.
portable and hand-held devices), or they can be truly innovative
(e.g. implantable or even injectable devices) (Saliterman 2006,
Grayson et al. 2004). The small size often saves on the costs of
reagents, time and money (Voldman et al. 1999). For example,
portable ‚point-of-care‛ hematological devices and test kits
permit physicians to diagnose the patient’s condition more
rapidly (Betancourt & Brannon-Peppas 2006). Diagnostic bio-
MEMS devices, also known as lab-on-a-chips or micro-total
analysis systems, can be used to detect cells, microorganisms,
viruses, proteins, deoxyribonucleic acid (DNA) and related
2 – Biomedical micro-electromechanical systems
41
nucleic acid (Bashir 2004, Guber et al. 2004). In general,
miniaturization offers several advantages over conventional
analytical methods. Smaller sample volumes enable reducing
assay cost and less waste disposal (Voldman et al. 1999). Due to
the low sample volume, short diffusion distances and fast
heating, less time is required for diagnostics (Saliterman 2006).
Furthermore, miniaturized parallel operation allows high-
throughput analysis, which plays an important role in genomic
research and drug discovery (Bashir 2004). On the other hand,
miniaturization of devices increases their surface area to volume
ratio, leading often to situations where surface effects dominate
volume effects. The larger surface area is essential in some
applications e.g. to ensure heat removal avoiding adverse heat
effects when high electric fields are used (Voldman et al. 1999).
In addition, MEMS fabricated electrodes and sensor systems
allow measurements and stimulations with higher spatial and
temporal resolution than with conventional macroscale counter-
parts, and hence enable more precise biomedical research or
clinical diagnostic and therapy (DiLorenzo & Bronzino 2008,
Stieglitz & Mayer 2006a).
Most MEMS fabrication processes can be performed
simultaneously on many, even thousands of devices, similarly
to the case of manufacturing of microchips on Si wafers. This
kind of batch processing enables high volume manufacturing at
low unit cost (Judy 2001). Processes are very reproducible,
minimizing variations between objects that easily arise among
individually constructed devices. Lastly, polymers and other
cheap materials can be utilized to provide bio-MEMS devices,
making feasible the manufacturing of disposable devices.
Disposable devices are particularly crucial when handling blood
and other biological fluids that may contain hazardous
substances such as human immunodeficiency virus or hepatitis
virus. Disposable devices also eliminate the risk of cross-
contamination and subsequent analysis errors which are
common in re-used devices (Wang & Soper 2007).
2 – Biomedical micro-electromechanical systems
42
2.4 BIO-MEMS APPLICATIONS
During recent years, the bio-MEMS technology has been under
intensive research and development activity. Numerous bio-
MEMS applications have been implemented in the fields of cell
and molecular biology, pharmaceutics and medicine, including
the areas of therapeutics, diagnostics and tissue engineering
(Voldman et al. 1999, Bashir 2004, Cheung & Renaud 2006,
Nuxoll & Siegel 2009, Saliterman 2006, Betancourt & Brannon-
Peppas 2006, Grayson et al. 2004). Although some applications,
such as blood analysis cartridges, cochlear implants and deep
brain stimulators (DBS) have been already commercialized, the
vast majority of bio-MEMS objects are under research or
undergoing clinical trials (Stieglitz & Mayer 2006a, DiLorenzo &
Bronzino 2008). On the other hand, most of the commercially
available systems are designed for in vitro diagnostics. The
development cycle from the lab prototype to commercial
manufacturing of implantable devices is long, perhaps even 10-
20 years, due to the extremely challenging environment inside
the human body as well as strict requirements to pass the
approval process (classified as FDA class III devices) (Ratner
1996). Examples of implantable bio-MEMS devices are listed in
Table 1. A more detailed description of the electrodes used in
neural prosthesis is given in Section 4 of the thesis.
2 – Biomedical micro-electromechanical systems
43
Table 1: Examples of implantable Bio-MEMS devices.
Device Brief description of its function Manufacturer(s) General
reference(s)
Cardiac
pacemakers
Regulation of the heart beating by delivering electrical impulses to
the electrodes in contact with
heart muscles
Boston Scientific, USA; Medtronic,
USA; St. Jude
Medical, USA
Das et al. 2009
Cochlear
implants
Electrical stimulation of auditory
nerve through microelectrodes implanted in the inner ear to
restore hearing
AllHear, USA;
Advanced Bionics, USA; Cochlear,
Australia; MED-
EL, Austria;
Neurelec, France
Loizou 1999,
Eddington 2008, Wilson & Dorman
2008a,b
Deep-brain
stimulators
Through intracortically implanted electrodes electrical stimulation is
delivered to targeted regions of
the brain to treat severe neurological disorders, such as
Parkinson disease
Medtronic, USA Kumar et al. 1997, Kern &
Kumar 2007,
Richardson 2008
Spinal cord
stimulators
Electrical stimulation of spinal cord through electrodes implanted in
epidural space in order to treat
intractable pain and motor
disorders
Advances Neuromodulation
Systems, USA;
Medtronic, USA
Waltz 1997, Mailis-Gagnon et
al. 2004, Burton
& Phan 2008
Vagus nerve
stimulators
Electrical stimulation of vagus
nerve in the neck via an implanted
lead wire to treat certain types of intractable epilepsy and major
depression
Cyberonics, USA Ardesch et al.
2007, Boon et al.
2001
Drug delivery
silicon chips
Controlled drug release from microfabricated Si reservoirs by
applying electrical potential
Microchips, USA Santini et al.
1999, 2000
Visual
prosthesis
Electrical stimulation of neural tissue of blind patients to restore
vision. Different approaches in
which microelectrode arrays are implanted on or under the retina,
around the optic nerve, and on or
in the visual cortex are under
development
Second Sight, USA; Retina
Implant AG,
Germany
Humayun et al. 2003, Stieglitz
2009, Greenberg
2008
2 – Biomedical micro-electromechanical systems
44
45
3 Origin of bioelectric
signals
Bioelectric signals are generated as a result of electrochemical
activity in certain type of cells, referred to as excitable cells
which are the basic components of nervous, muscular and
glandular tissue (Malmivuo & Plonsey 1995). An electrical
potential difference, i.e. the transmembrane potential, exists
between the internal and external environment of the cell (Hille
1992). When appropriately stimulated to cross a limiting
threshold, the excitable cell will generate an action potential due
to a flow of charged ions across the permeable cell membrane
with the ions passing through ion channels. These action
potentials are transferred from one cell to adjacent cells via their
cellular extensions, i.e. axons and dendrites due to synaptic
transmission (Tortora & Grabowski 2000). The activity of a
group of excitable cells (i.e. population activity) generates an
electric field that propagates through the biological tissue and
this can be measured by field-potential measurements with
surface electrodes attached to the surface of tissue, e.g.
cerebellum cortex, or skin (Nunez & Srinivasan 2006, McAdams
2006). The field-potentials commonly monitored in a patient are
those producing electrocardiograms (ECG) from the heart, EEG
from the brain and electromyograms (EMG) from contraction of
muscles.
3.1 EXCITABLE NERVE CELL
A nerve cell, also known as a neuron, is an excitable cell in the
nervous system whose main function is to receive, handle and
transmit cellular information via electrochemical phenomena.
Neurons are the basic components of the brain, the spinal cord
3 – Origin of bioelectric signals
46
and the peripheral nerves. They can be divided into two
categories: (1) principal cells that receive input from nearby or
distant neurons and project to neurons in other nuclei; (2)
interneurons that modulate the activity of principal cells in the
same structure. Principal cells are usually excitatory and use
glutamate as the neurotransmitter, whereas interneurons are
usually inhibitory and use gamma-amino butyric acid as the
neurotransmitter (Kandel et al. 2000). There are a huge variety
of different nerve cells with sizes in the range of 600 – 70 000
µm3 (Schadé & Ford 1973). Most nerve cells consist of three
main parts: a cell body and two kinds of cell extensions –
dendrites and axons. The cell body (soma) is similar to that of all
other cells containing the nucleus, mitochondria, endoplasmic
reticulum, ribosomes, and other organelles (Tortora &
Grabowski 2000). Most of the protein synthesis happens in this
part of cell. Dendrites are typically short and highly-branched
cellular extensions, through which the vast majority of signal
input to the neuron occurs (afferent signals). Due to numerous
branches (‚dendrite tree‛), a neuron may receive nerve impulses
from thousands of other neurons (Nunez & Srinivasan 2006).
The axon of a neuron is a single, cylindrical-shaped projection
which carries nerve impulses away from the cell body toward
the axon terminals passing the message on to other neurons.
Although axons are very thin, typically about 1-20 µm in
diameter, they can be very long (Malmivuo & Plonsey 1995).
The length of axons, e.g. the human sciatic nerves, may be more
than one meter long. The axons of many neurons are covered
with an insulating layer called the myelin sheath, which is
formed by either of two types of glial cells: Schwann cells or
oligodendrocytes which insulate peripheral neurons or those of
the central nervous system, respectively (Tortora & Grabowski
2000). The myelin sheath is not continuous but the gaps, known
as nodes of Ranvier occur at regular intervals, dividing the
sheath into short sections (Malmivuo & Plonsey 1995). This
myelination structure enables much faster conduction (i.e.
saltatory conduction) of nerve impulses so the impulse is
transported faster than in nonmyelinated axons.
3 – Origin of bioelectric signals
47
3.1.1 Cell membrane and ion channels
The living cell is surrounded by the cell membrane which is an
approximately 5 nm thick phospholipid bilayer where the
hydrophobic tails of phospholipid molecules are pointing inside
the membrane whereas the negatively polarized, hydrophilic
heads of these molecules are directed outward of the membrane
(Alberts et al. 1994). The main functions of the cell membrane
are to form the boundaries for the intracellular components of
the cell, maintaining cell homeostatis and regulating the
incoming and outgoing substances (Tortora & Grabowski 2000).
The cell membrane is permeable to only specific types of ions
and molecules and thus the concept of selective permeability is
used to describe this property of the membrane. The excitability
of the membrane is based on embedded protein molecules,
which form tiny ion channels and allow the flow of sodium
(Na+), potassium (K+), calcium (Ca2+) and chloride (Cl-) ions
through the membrane (Hille 1992). Ion channels can be divided
into leakage channels and gated channels (Tortora & Grabowski
2000). The difference between them is that leakage channels are
always open, whereas gated channels open and close in
response to some kind of stimulus. The gated channels, on
which the excitability of neural cells is based, can be divided
into three groups according to the stimulus involved: (1)
voltage-gated ion channels, (2) ligand-gated ion channels and (3)
mechanically gated ion channels. The state (i.e. open or closed)
of voltage-gated channels is adjusted by a change in the
transmembrane potential. Ligand-gated ion channels open and
close in response to specific chemical stimuli including
neurotransmitters, hormones and ions, whereas mechanically
gated ion channels are regulated by mechanical stimuli such as
vibration, pressure or tissue stretching (Tortora & Grabowski
2000, Kandel et al. 2000).
3.1.2 Transmembrane potential and equilibrium potentials
The selective permeability of the cell membrane maintains ion
concentration difference between the inside and outside of the
cell membrane (i.e. concentration gradient). For mammalian
3 – Origin of bioelectric signals
48
muscle, the K+ concentration of the intracellular space is 155
mmol/l, while that within the extracellular space is only 4
mmol/l (Hille 1992). In addition, Na+, Ca2+ and Cl- ionic
concentrations inside the cell differ greatly from the outside
(Table 2). To balance the positive and negative charges inside
the cell, there are also other negatively charged molecules,
which are too large to permeate through the membrane.
Table 2: Ion concentrations for mammalian muscle cells (Hille 1992).
Ion Intracellular concentration (mM/l)
Extracellular concentration (mM/l)
Na+ 12 145
K+ 155 4
Ca2+ 10-4 1.5
Cl- 4.2 123
This existing concentration gradient promotes the outflow of
K+ ions and inflow of Na+, Ca2+ and Cl- ions in order to balance
the ion concentrations. Potassium ions diffuse from inside to
outside, making the interior of the cell more negative than the
outside. Additionally, there is an active transportation of ions
against a concentration gradient. These Na-K pumps transport
two K+ ions into the cell for every three Na+ ions that the enzyme
pumps out of the cell, and thus there is a net loss of positive
charges within the cell (Hille 1992). The transmembrane
potential, Vm, is defined as a potential difference between the
internal and external environment of a cell. The resting
membrane potential corresponds to the value of Vm when the
cell is in the resting state in its natural, physiological
environment. In neurons, the typical resting potential is around
-70 mV (Tortora & Grabowski 2000). The membrane potential
that just balances the ion concentration difference, i.e.
equilibrium potential when no current flows across the
membrane, can be determined for each specific ion by using the
Nernst equation (derived by Water Hermann Nernst in 1888):
3 – Origin of bioelectric signals
49
.ln
o
i
C
C
zF
RTV (2)
Here R is the gas constant (8.314510 J/mol x K), T is absolute
temperature, z is the valence of an ion, F is the Faraday’s
constant and Ci and Co are ion concentrations inside and outside
of the cell, respectively.
It can be easily calculated by using the ion concentrations
(Table 2) that the equilibrium potential for K+ ions at 37 °C is
approximately -98 mV. The corresponding values for Na+ and
Ca2+ ions are +67 mV and +128 mV, respectively. Since the
equilibrium potential of K+ is quite close but not same as the
value of the resting potential (typically -70 mV), at rest the cell
membrane is much more permeable to K+ ions than to other ions.
The Nernst equation can be extended to Golzman-Hodgkin-
Katz equation, which produces the equilibrium potential when
there are several types of ions in the intracellular and
extracellular space, and when the membrane is permeable to all
of them (Hille 1992). Then, the resting membrane potential can
be estimated by:
,ln
,,,
,,,
CliClNaoNaKoK
CloClNaiNaKiK
RESTcPcPcP
cPcPcP
F
RTV
(3)
where PK, PNa and PCl are membrane permeabilities of K+ , Na+,
Cl- ions, respectively.
3.2 SYNAPTIC POTENTIALS AND ACTION POTENTIALS
The transfer of information from one neuron to the next is based
on junctions between an axon and the neighboring cell, called
synapses. They can be divided into electrical and chemical
synapses (Tortora & Grabowski 2000). An electrical synapse is a
mechanical, electrically conductive link between two neurons
which can transmit nerve impulses very quickly (Hormuzdi et
al. 2004). Electrical synapses are common in the neural systems
3 – Origin of bioelectric signals
50
that require the fastest possible response, such as defensive
reflexes. However, it is more common to operate with chemical
synapses. The end of the axon has branching terminals through
which neurotransmitter chemicals (e.g. dopamine, glutamate,
serotonin) packaged into presynaptic vesicles, are released into
a gap called the synaptic cleft in order to communicate with an
adjacent neuron (Kandel et al. 2000). Most synapses are between
axon and dendrite whereas also other combinations like axon-
to-axon and dendrite-to-dendrite exist (Tortora & Grabowski
2000). The width of synaptic cleft is only about 20-40 nm
(Hormuzdi et al. 2004). Due to the small volume of the cleft, the
neurotransmitter concentration can be raised and lowered
rapidly. There are certain types of receptor molecules located on
the cell membrane of the postsynaptic cell which after being
occupied by neurotransmitter molecules, respond by opening
nearby ion channels in the membrane, allowing charged ions to
rush in or out. As a result, the local transmembrane potential of
the cell is changed and thus is referred to as the post-synaptic
potential. The result can be either excitatory (depolarizing) or
inhibitory (hyperpolarizing) depending on the amount and type
of neurotransmitter as well as the type of receptor molecules
(Kandel et al. 2000). In an excitatory reaction, a net movement of
positive ions into the cell changes the transmembrane potential
making it less negative. Hyperpolarization is the exact opposite
to depolarization, in which the change of Vm occurs in the
negative direction. If the depolarizing response is great enough,
the cell membrane is able to produce a characteristic bioelectric
impulse called an action potential (Fig. 4) (Tortora & Grabowski
2000). At the beginning of depolarizing stimulus, the voltage-
gated Na+-selective ion channels become opened, allowing the
flow of Na+ ions from outside to inside. The inside of the cell
becomes more positive reaching a potential value of
approximately +30 mV. Subsequently, more K+-selective ion
channels will open allowing the flow of K+ ions from inside to
outside. The flow of K+ ions returns the transmembrane
potential to its resting value. The duration of the action potential
is around 1 millisecond (Tortora & Grabowski 2000).
3 – Origin of bioelectric signals
51
1 2 3 4 5 Time (ms)
Tra
nsm
em
bra
ne
po
ten
tia
l(m
V)
+30
0
-70
Resting stateRefractory
period
Figure 4: The shape of an action potential. When a stimulus depolarizes the membrane
beyond a threshold, an action potential is generated. At the end of depolarization
phase, a membrane potential of 30 mV is reached. The flow of potassium ions returns
the membrane potential to its resting value (-70 mV) during a short refractory period.
3.3 RECORDING ELECTRICAL ACTIVITY OF THE BRAIN
The human nervous system is a complex, highly organized
network of billions of neurons and even more neuroglia (Tortora
& Grabowski 2000). It is composed of two sub-systems: the
central nervous system (CNS) and peripheral nervous system
(PNS). Anatomically, the CNS contains the cerebrum,
cerebellum, brainstem and spinal cord (Heimer 1995). The brain
stem is a structure through which nerve fibers relay action
potentials in both directions between spinal cord and higher
brain centers. The cerebrum is divided almost equally into two
halves, which both have subdivisions of the frontal, temporal,
parietal and occipital lobes. The cerebral cortex, about 2-4 mm
thick folded structure, forms the outer portion of the cerebrum
(Tortora & Grabowski 2000). The surface area of the cerebral
cortex varies from 1600 to 4000 cm2 and it contains perhaps as
3 – Origin of bioelectric signals
52
many as 10 billion neurons (Nunez & Srinivasan 2006). Those
neurons are strongly interconnected. The surface of a large
cortical neuron may be covered by as many as 105 synapses that
transmit inputs from other neurons (Tortora & Grabowski 2000).
Post-synaptic potentials in thousands of synchronously active
pyramidal neurons produce an electric field propagating
through the biological tissue and these can be measured on the
scalp or on the surface of brain (subdural electrodes) or even
inside the brain (probe electrodes) with an amplifier through
appropriate electrodes (Nunez & Srinivasan 2006, Niedermayer
& Lopes da Silva 2005). Scalp-EEG is a standard method in
clinical neurophysiology to study the function of the brain. Both
spontaneous activity and evoked potentials (EPs) can be
measured. Spontaneous scalp-EEG is routinely utilized in many
clinical cases, ranging from diagnosis of epilepsy to the analysis
of sleep and the depth of anaesthesia. EPs are produced as a
response to given sensory stimuli. The EPs can be classified
according to the applied stimulus into three categories:
somatosensory (SEP), visual (VEP) and auditory (AEP) evoked
potentials (Niedermayer & Lopes da Silva 2005). The stimulus
evokes electrical activity with a very short latency (delay) period
over a limited area of the cerebral cortex. For example, VEPs are
detected from the occipital region when a visual stimulus is
presented, whereas AEPs are detected from the temporal region
when an auditory stimulus is presented. Evoked potential
recordings can be utilized to localize lesions, or monitor a
sensory pathway during surgical procedures (Chiappa 1997).
Stimulus is usually applied many times (even thousands), and
then signal averaging is undertaken to achieve clean EP curves
with a reduced effect of noise (Nunez & Srinivasan 2006).
In clinical practice, scalp-EEG is typically measured using the
standard international 10-20 electrode placement system in
which 21 electrodes are attached on the scalp with exact
intervals and potential differences between electrodes is
recorded. The amplitudes of the scalp EEG varies from 10 to 100
µV (Niedermayer & Lopes da Silva 2005). The frequency band
from 0.16 to 100 Hz is commonly adequate, but sometimes a
3 – Origin of bioelectric signals
53
much lower frequency response, down to direct current (DC)
(Vanhatalo et al. 2003) or much higher (e.g. up to 3 kHz) is
useful (Curio 2005). The scalp-EEG possesses good temporal
resolution on a sub-millisecond scale, but the spatial resolution
is poor. Spatial resolution can be improved by increasing the
electrode numbers. Nowadays, multi-channel EEG caps with up
to 256 electrodes are applied in experiments aimed at obtaining
detailed information about brain sources (Väisänen 2008).
Another choice to improve the source localization is to place the
electrodes closer to the brain tissue. Both intracortical depth
electrodes and subdural strip electrodes placed on the surface of
cortex are used in intracranial recordings of local field potentials
and occasionally spikes (Grill 2008). Compared to scalp-EEG,
intracranial recordings are worthwhile in terms of their
increased spatial resolution, signal amplitude level (0.5 – 4 mVpp
vs. 10 – 100 µVpp), and reduced noise level, meaning that one
can acquire more detailed maps directly of the brain (Lachaux et
al. 2003, Niedermayer & Lopes da Silva 2005, Leuthardt et al.
2008). However, the use of penetrating electrode probes in
humans is limited to the areas destined for surgical resection
due to the risk of damaging brain tissue. Flexible subdural strip
and grid electrodes are utilized as a less invasive alternative in
some clinical cases, e.g. in mapping cerebellum cortex in patients
undergoing epilepsy or brain tumor surgery. The signals
recorded with subdural electrodes are considerably higher in
amplitude because they are much closer to the source of the
activity, and are separated from it by only a relatively high
conductivity media (cerebrospinal fluid, brain parenchyma)
(Nair et al. 2008, Leuthardt et al. 2008). The resistivity of the
skull is estimated to be even 80 times higher than that of brain
tissue (Rush & Driscoll 1968) that causes a decline in the signal
amplitude and signal blurring in scalp-EEG recordings.
Moreover, recordings made by subdural electrodes are almost
free of artifacts (e.g. not affected by electrical signals originating
from muscles) that are seen in scalp-EEG, and thus they yield a
much higher signal-to-noise ratio (Nair et al. 2008).
3 – Origin of bioelectric signals
54
55
4 Bioelectrodes
Biomedical electrodes are extensively utilized in numerous
biomedical applications (McAdams 2006). Firstly, they are used
to record bioelectrical events such as the ECG, EEG and EMG
that intrinsically arise due to electrical activity of excitable cells.
Secondly, bioelectrodes are utilized in applying electrical
impulses to the body for treating neurological pain in many
therapeutic methods such as in transcutaneous electrical nerve
stimulation or deep brain stimulation. Deep brain stimulators
consist of intracortically implanted electrodes through which a
high frequency electrical stimulation is delivered to a targeted
region of the brain. DBS is used in treating several neurological
disorders, e.g. Parkinson’s disease (Kumar et al. 1997, Kern &
Kumar 2007, Perlmutter & Mink 2006). Thirdly, they are used in
alternating current impedance characterization of body tissues.
These methods such as electrical impedance plethysmography
and electrical impedance tomography (Cheney et al. 1999) do
not monitor intrinsic biosignals emanating from the body, but
inject small alternating currents/voltages and record the
resultant voltages/currents. The electrical properties of the body
can then be determined. Finally, bioelectrodes are utilized in
electrotherapy by applying electrical potentials in order to
facilitate the transdermal delivery of ionized molecules for
therapeutic effect (iontophoresis). Moreover, recent activity on
neuroengineering research has focused on implantable neural
prosthetics to help restore the functions of the damaged nervous
system (DiLorenzo & Bronzino 2006, Stieglitz & Mayer 2006b).
The most successful examples of commercialized neural
prosthesis are cardiac pacemakers and cochlear implants.
Cochlear implant is a surgically implanted microelectronic
device that utilizes an electrode array inserted into the scala
tympani of the cochlea to stimulate the auditory nerve through
4 – Bioelectrodes
56
the bone, enabling the deaf to hear sounds (Loizou 1999,
Federspil & Plinkert 2004, Wilson & Dorman 2008a,b).
4.1 IMPLANTABLE ELECTRODES
The commercially available clinic subdural strip and grid
electrodes (Fig. 5) typically consist of a number of conductive
metal such as Pt, stainless steel (SS), and silver (Ag) disks, which
are partially embedded in a thin sheet of biomedical grade
silicone rubber (e.g., AD Tech Medical Instrument Corp., Racine,
WI, USA; PMT Corp., Chanhassen, MN, USA). Typical diameter
of disk contacts is about 2-3 mm, and the distance between
adjacent electrodes around 10 mm.
Figure 5: Examples of clinical subdural strip and grid electrodes.
Despite major advances in microsystem technology and its
great potential in the development of novel high-density MEAs,
most human intracranial electrodes used to record brain activity
are still handcrafted from Pt foils and silicone rubber sheet and
therefore are quite bulky (see Fig. 5). Miniaturized neural
microelectrodes are extensively developed approaches in
neurophysiological (animal) studies aimed at understanding
4 - Bioelectrodes
57
behaviour and function of the brain at the tissue and cellular
level, but their human counterparts are still mainly under
investigation or at best are undergoing clinical trials.
Already over forty years ago it was discovered that the
photolithographic, Si etching and thin-film deposition
techniques originally developed for the production of integrated
circuits would be also applicable for providing novel improved
recording and stimulating electrodes for experiments with
animal models (Wise 2005). Compared to the traditional
approach of single microwires or microwire bundles (Nicolelis
et al. 1997, Williams et al. 1999), microfabricated electrodes
offered many advantages including precise control of the
electrode sizes and separations between electrodes as well as a
high degree of reproducibility in physical, chemical and
electrical characteristics (Rutten 2002). Microfabrication is also a
high yield, low-cost process once the design and processing
steps have been optimized. Much of the pioneer work of Si-
based microelectrodes has been done in three research centers in
the USA, i.e. at the University of Stanford, University of
Michigan, and University of Utah. Today Utah arrays (Campbell
et al. 1991, Rousche & Normann 1998) and Michigan probes
(Wise & Angell 1975, Najafi & Wise 1986, Vetter et al. 2004) are
perhaps the best-known commercially available Si-based
microelectrodes. A three dimensional Utah electrode array (Fig.
6a) consists of one hundred 1.5 mm long Si needles that project
out from a 4 mm x 4 mm substrate (Campbell et al. 1991,
Normann et al. 1999). The sharpened tip of each Si needle is
coated with Pt. An advanced version of Utah array, i.e. Utah
slanted electrode array, has also been developed (Fig. 6b). In this
model, the length of Si needles can be increased from 0.5 to 1.5
mm permitting focal excitation of nerve fibers at different
depths (Badi et al. 2003). Utah arrays have been used in studies
of parallel information processing by the CNS at the visual and
auditory cortex (Rousche & Normann 1998), and the control of
muscle force and limb position by the PNS (Normann 2007). The
typical Michigan probes are single-shank or multi-shank Si
penetrating systems (Fig. 6c) where the electrodes are placed
4 – Bioelectrodes
58
along the length of the shank allowing the measurement of
neuronal activity at various depths of the brain tissue. Michigan
probes have been successfully used in a number of neuroscience
experiments throughout the CNS, PNS and even cardiac muscle
(Vetter et al. 2004, Kipke et al. 2003, Wise 2005).
Figure 6: Silicon–based neural interfaces: (a) Utah array; (b) Slanted Utah array.
These electrode arrays which both contain 100 microneedle-shaped electrodes are
designed to be implanted in cerebral cortex or peripherally. (c) Michigan multi-shank
probe contains active electronics integrated into the array. Reprinted by permission
from Macmillan Publishers: Nature Clinical Practice Neurology (Normann 2007),
copyright (2007).
Although rigid microwires and Si-based microelectrodes
have long been used for recording of biosignals, the mechanical
mismatch between the rigid electrode and soft biological tissue
may evoke many adverse effects such as tissue damage,
inflammatory reactions and scar formation (Polikov et al. 2005).
Thus, there has been a growing interest in developing polymer-
4 - Bioelectrodes
59
based (e.g. polyimide, benzocyclobutene, silicone rubbers and
parylene-C) implants that could be sufficiently flexible to mimic
biological tissue and to reduce adverse tissue reactions (Cheung
2007). The flexibility of polymer may decrease the mechanical
mismatch between nervous tissue and the rigid implant
reducing the risk of tissue damage and inflammation reactions.
PI is one of the most widely studied polymers due to its suitable
mechanical properties, biocompatibility and stability in wet
microfabrication processes (Rubehn & Stieglitz 2010, Stieglitz et
al. 2000). Several PI-based microelectrode constructions have
been developed in the fields of neural engineering, including
devices for in vitro recordings from brain slices (Boppart et al.
1992), cortical surface field potential recordings (Owens et al.
1995, Hollenberg et al. 2006, Takahashi et al. 2003, Hosp et al.
2008), intracortical multiunit neural activity recordings (Rousche
et al. 2001, Cheung et al. 2007, Mercanzini et al. 2008) and for
action potential recording in nerve and muscle tissues in vivo
(González & Rodríguez 1997, Rodríguez et al. 2000, Spence et al.
2007). PI has also been applied for different types of sieve and
cuff electrodes for interfacing with regenerating peripheral
nerves (Stieglitz et al. 1997, 2000, Stieglitz 2001).
4.2 MATERIAL REQUIREMENTS FOR IMPLANTABLE
ELECTRODES
The materials used in implantable neural electrodes and
prostheses must fulfill strict requirements in order to ensure
optimal performance and longevity. According to Geddes and
Roeder (2003), there are four criteria that must be considered
when selecting a material for implantable use: (1) tissue
response; (2) allergic response; (3) electrode-tissue impedance;
and (4) radiographic visibility. An ideal implanted electrode
and/or its insulation should be biocompatible without triggering
any vigorous local or generalized host response or allergic
reactions. The electrode-tissue impedance must be stable and
low enough to enable high-resolution measurements. Moreover,
4 – Bioelectrodes
60
the materials should be radiographically visible and be
compatible with magnetic resonance imaging.
It has been concluded in many studies that in spite of the
extremely challenging processes used in fabrication of
miniaturized electrodes, electrode failures due technical reasons,
such as short-circuits and broken signal transmission lines, are
rare, and the stable, long-term functioning of neural implant is
mainly determined by the tissue response to chronically
implanted electrodes (Polikov et al. 2005, Nicolelis et al. 2003,
Rousche & Normann 1998, Williams et al. 1999). A general
immune activation of the brain in response to the presence of a
foreign body implant has been commonly believed to be the
main reason for signal deterioration of chronically implanted
electrodes (Nicolelis et al. 2003, Edell et al. 1992, Schmidt et al.
1993). The biocompatibility aspects of electrode materials will
therefore be very crucial in the development of new generation
chronic implantable devices.
4.2.1 Electrode materials
Recording bioelectrodes are functional elements that convert the
ionic current around the electrode site in the body into electron
current in electronic circuitry. Electrodes can be roughly divided
into polarizable and non-polarizable electrodes. Perfectly
polarizable electrodes are those in which no actual charge
crosses through the electrode-electrolytic interface, but changes
the charge distribution within the solution near the electrode
when a current is applied. This kind of electrode acts like a
capacitor, and only the displacement current crosses the
interface. In contrast, perfectly non-polarizable electrodes allow
the current to pass freely through the electrode-electrolytic
interface without changing the charge distribution in the
electrolytic solution. However, neither of these ‚perfect‛
electrode types can be fabricated, but some practical electrodes
can come close to possessing their features. (McAdams 2006,
Neuman 1978)
4 - Bioelectrodes
61
A silver-silver chloride (Ag-AgCl) electrode is an almost
perfectly non-polarizable electrode. Due to many desirable
properties, such as small and stable offset potential, low contact
impedance and low intrinsic noise, this type of electrode has
been found to be an excellent electrode material for surface (skin)
biopotential recordings (McAdams 2006). Unfortunately, it Ag-
AgCl cannot be used as an implantable electrode because of the
rapid dissolution of Ag. Moreover, AgCl as well as pure Ag has
a toxic effect on neural tissue (Yuen et al. 1987).
Noble metals such as platinum, gold, iridium, rhodium and
palladium are commonly used highly polarizable electrode
materials in neural applications due to their excellent corrosion
resistance and biocompatibility (Geddes & Roeder 2003,
McAdams 2006). In the extensive study carried out by Stensaas
and Stensaas (1978), twenty-seven different potential neural
electrode materials were tested by implanting them into the
cortices of rabbits. After 30 days, the histological examination
revealed no adverse reaction in response to gold, platinum and
tungsten. Unfortunately, noble metals usually tend to give rise
to high electrode impedances and unstable potentials
(McAdams 2006). Recently, different kinds of micro- and
nanostructuring techniques have been investigated to increase
the effective surface of noble metals and subsequently reduce
the electrode-electrolyte interface impedance. Pt has been the
most widely used material in implantable electrodes. Clinically,
it has been utilized as an electrode material in various
applications such as in cardiac pacemakers, cochlear implants
and subdural strip and grid electrodes. Pt suffers from corrosion
only slightly in both recording and stimulating use. Due to the
mechanical softness of Pt, it is sometimes alloyed with iridium
(Ir) to achieve electrodes for applications requiring high
mechanical strength (McAdams 2006). Gold is also often used in
recording electrodes, but it suffers from corrosion if used for
stimulating purposes. On the contrary, Ir is mainly utilized as a
stimulating electrode. The electrode properties of Ir can be
further developed by electrochemical activation that leads to the
adhesion of a porous and very stable oxide layer on the
4 – Bioelectrodes
62
electrode surface. Activation provides much higher charge
injection levels and also reduces the electrode contact
impedance due to the porous microstructure (Cogan 2008, Blau
et al. 1997).
Titanium nitride (TiN) is another novel candidate suitable for
use in stimulating electrodes, and it is already being utilized in
cardiac pacemakers. The high charge delivery capacity of TiN is
based on its high effective surface area due to the columnar
structure of this material (Cogan 2008). Indium tin oxide (ITO) is
a widely used electrode material in electrochemistry (Huang et
al. 2009), but so far its use in neurophysiology is limited to in
vitro MEAs for recording electrical signals from neurons. The
biocompatibility of ITO is not completely understood although
recent in vitro investigations on cellular responses (Bogner et al.
2006) and protein adsorption on ITO (Selvakumaran et al. 2008)
have displayed promising results.
The impedance of an electrode-electrolyte interface depends
on the species of metal, the type of electrolyte used, the surface
area of electrode, and the temperature (Geddes & Roeder 2001,
2003). Magnitude of impedance decreases with expanding
surface area and increasing roughness. It also decreases with
increasing frequency and increasing current density. Several
different equivalent circuit models for the electrode-electrolyte
have been suggested (reviewed by Geddes 1997). One of the
most sophisticated models is presented in Fig. 7a. There are two
identical electrodes in contact with electrolyte. Electrode
polarization impedance consists of Warburg (polarization)
resistance Rw and capacitance Cw. The Faradic resistance, Rf as a
parallel with Warburg components, provides a route for direct
current to pass through the interface. The resistance of the
electrolyte solution is marked with Rs. Both Rw and the
polarization reactance (Xw = 1/2πfCw) are frequency-dependent:
they decrease as ( f/1 ) with increasing frequency ( f ). At a
sufficiently high frequency, the impedance between two
electrodes in saline approaches asymptotically the value of Rs
(Fig. 7b).
4 - Bioelectrodes
63
A B
Frequency (Hz)
Rs
Rw1
Rw2
Cw1
Cw2
Rf2
Rf1
Rf1+ Rf2 + Rs
Rs
1
2
Z
Imped
ance
|Z|(Ω
)
Figure 7: (a) The equivalent circuit model for two identical bioelectrodes immersed in
saline. (b) Typical electrode-electrolyte interface impedance curve measured between
terminals 1 and 2 as a function of frequency. The curve approaches asymptotically the
resistance value of the solution.
Due to the inverse relationship between electrode impedance
and its surface area, it is advantageous to develop
microelectrodes with high nanoscale surface topography since
in this way it will be possible to achieve a high effective surface
area without changing their geometrical dimensions (Kotov et al.
2009). Different techniques are used to increase the effective
surface area including surface roughening, micro- and nano-
patterning and chemical modification. In micro/nanopatterning
techniques, different etching and machining techniques are
utilized to form different micro/nanostructures, e.g. grooves and
wells onto the surface of the substrate upon which the electrode
material can then be deposited using traditional thin film
techniques (Cheung 2002). Some examples of widely used
surface modification techniques used to enhance nanoscale
roughness includes Pt electroplating (Pt black) (de Haro et al.
2002) wet etching of Au and activation of Ir (Cogan 2008, Blau et
al. 1997) to form a nanoporous iridium oxide layer as already
discussed.
4.2.2 Substrate materials
The surface area of substrate/encapsulation material is usually
much larger compared to areas of the active recording sites.
Therefore, biocompatibility aspects of substrate and insulator
materials are very crucial in the development of neural
4 – Bioelectrodes
64
electrodes and prosthesis which require long-term stability and
functionality. These materials need to be biocompatible,
biostable and possess good dielectric properties. An artificial
material or implant can be considered to be biocompatible if (1)
it does not evoke a toxic, allergenic or immunologic reaction, (2)
it is not harmful to enzymes, cells and tissues, (3) it does not
evoke thrombosis or tumors, and (4) it remains for a long term
within the body without encapsulation or rejection (Heiduschka
& Thanos 1998). On the other hand, biostability means that the
implant material must not be susceptible to attack by biological
fluids or any metabolic substances. Unfortunately, all implanted
artificial materials in contact with neural tissue have an
inconvenient tendency to induce significant glial scar tissue
formation (Polikov et al. 2005). In response to implantation, glial
cells such as astrocytes become activated and they transform
into reactive phenotype which produces much larger size glial
fibrillary acid protein filaments. Cell proliferation and migration
capacity of astrocytes is also enhanced. The resultant scar tissue
formation causes serious impairment of implant performance
due to decreased local density of neurons and the formation of
an encapsulation layer that increases electrode impedance and
decreases the signal amplitudes. It has been noted that the
astrocyte response to electrode implantation can be divided into
two phases, the early response that occurs immediately after
electrode implantation and the long-term chronic response.
Within a few hours after implantation, the number of glial cells
is significantly enhanced in the area surrounding the Si probes
(Szarowski et al. 2003, Turner et al. 1999). The chronic response,
starting approximately one week after implantation (Norton et
al. 1992), consists of a compact glial scar tissue formation
surrounding the electrode, which ultimately isolates the
microelectrodes from neurons increasing the electrode contact
impedance and causing signal amplitude deterioration (Edell et
al. 1992, Nicolelis et al. 2003, Schmidt et al. 1993).
Silicon with its common insulators, SiO2 and silicon nitride
(Si3N4), have been the most widely used materials in neural
interfaces as well as in the bio-MEMS technology overall. In
4 - Bioelectrodes
65
spite of numerous desirable features of Si, there are at least two
major weaknesses related to its use in neural applications.
Firstly, there is a huge mismatch between the mechanical
properties of the brain tissue and Si. The values of Young’s
modulus for Si and brain tissue are 170 GPa and 3 kPa,
respectively. The rigidity of Si may cause many adverse effects
such as tissue damage, inflammation reactions and scar
formation (Polikov et al. 2005). Secondly, the results from
studies related to cytocompatibility of Si are somewhat
contradictory (Kotzar et al. 2002). Some studies have shown that
Si is a rather poor substrate for cell adhesion and even slightly
cytotoxic (Liu et al. 2007, Madou & Tierney 1993), whereas some
others have claimed that Si is almost as cytocompatible and
immunogenic as tissue culture polystyrene (Ainslie et al. 2008).
This uncertainty could limit the much-anticipated integration of
Si-based neural implants into the human body.
There is a growing interest in developing polymer-based
implant materials that could be flexible enough to mimic
biological tissue and to reduce mechanical damage, but not
evoking any adverse tissue reactions. A wide variety of polymer
materials, such as polyimide, benzocyclobutene, silicone rubbers,
parylene-C and epoxy resins have been investigated as substrate
and encapsulating materials in implantable neural interfaces
(Cheung 2007, HajjHassan et al. 2008, Polikov et al. 2005).
Polyimide possesses numerous desirable properties when
used as a neural implant material. It has appropriate mechanical
properties (Rubehn & Stieglitz 2010) and its biocompatibility has
been demonstrated in many studies in vitro and in vivo (Stieglitz
et al. 2000, Richardson et al. 1993, Seo et al. 2004, Lee et al. 2004,
Yuen et al. 1987, Yuen & Agnew 1995). Stieglitz et al. (2000)
evaluated the cytotoxicity of three commercial PI grades
(Pyralin PI 2611, PI 2556, PI 2566; HD Microsystems GmbH, Bad
Homburg, Germany) and reported excellent biocompatibility for
PI 2611 and PI 2556 and good results for PI 2566. This last PI,
since it is fluorinated, differs from the others with respect to its
chemical structure (Kanno et al. 2002). In another study (Lee et
al. 2004), it has been shown that fibroblasts attach, spread out
4 – Bioelectrodes
66
and grow on the PI surfaces in a manner corresponding to the
behaviour of control cells growing on the surface of polystyrene,
i.e. the plastic that is used in cell culture well plates and flasks.
Furthermore, Yuen et al. (1987) demonstrated that subdural
implantations of two common insulators, parylene-C and PI
(DuPont PI-2555), into the cerebral cortex of the cat evoked only
minimal tissue reactions after 16 weeks. Tissue responses to PI
and parylene-C were comparable to pure Pt (used as a negative
control) whereas Ag-AgCl (positive control) triggered chronic
inflammatory reactions (Yuen et al. 1987). In contrast, PI-Pt
electrodes implanted into a rat sciatic nerve for 3 months have
been shown to induce only mild scar tissue formation and local
inflammation reactions, with these reactions being limited to a
small area around the electrode (Lago et al. 2007). One
shortcoming of PI is related to its permeability to environmental
moisture and ions that could be crucial in terms of short
circuiting and reducing the life-time of an implantable device.
There are two common types of PIs on the market: non-
photosensitive PI and photosensitive polyimide (PSPI) (Ghosh &
Mittal 1996). A special feature of PSPI, i.e. photopatternability,
simplifies the fabrication process of the microdevice compared
to conventional PI by eliminating the need for complex
multilevel processes, i.e. oxygen RIE through a lithographically
patterned metal aluminum (Al) mask (Wilson et al. 2007). PSPI
is advantageous in electrode fabrication in which electrode sites
and contact pads can be easily opened though an encapsulation
layer while still insulating the transmission lines as well as in
precision feature defining of the device’s outer shape. Although
mechanical and electrical properties of PSPI are comparable to
conventional PI, there will be a degree of uncertainty related to
its biocompatibility due to the different chemical composition
(e.g. existence of photoinitiators), imidization kinetics and
solvent processes. So far very little data have been published
addressing the biocompatibility of PSPI (Sun et al. 2009).
67
5 Biocompatibility
The main requirement for the successful clinical and
experimental application of implantable devices constructed
from artificial materials is that the surrounding tissues accept
the implant, i.e. that the implant is biocompatible.
Biocompatibility can be defined as ‚the ability of a material to
perform with an appropriate host response in a specific
situation‛ (Williams 1999). It is worth noting that the host
response to specific individual materials could vary from one
application site to another. An increasing number of
applications require that the material should actively react in a
desired manner with the tissue, or even degrade over time in the
body, rather than remain constantly as an inert material.
Biomaterials can be classified according to the chemical
composition to metals, polymers, ceramics, composites and
materials of biological origin (Navarro et al. 2008). Another
classification is based on historical stages of biomaterial
development. The first generation of implantable devices,
developed during the years between 1940 and 1980, were
constructed using materials that were chemically least reactive.
It was thought that the best performance of implants could be
achieved with a material that would be as passive as possible
regarding adsorption of host proteins and cellular adhesion.
Examples are inert metal materials that are very resistant to
corrosion such as stainless steel, cobalt-chromium alloys,
titanium alloys and the platinum group metals. With respect to
the polymers, polytetrafluoroethylene, polymethylmethacrylate,
polyethylene (PE) and silicones were widely used due to their
high resistance to degradation (Williams 2008). The second
generation biomaterials were specifically designed to be
‚bioactive‛, i.e. they could elicit a controlled host reaction and
reaction in their physiological environment to achieve the
desired result of the implantation. For example, certain
5 – Biocompatibility
68
compositions of bioactive glasses promote the desired cellular
responses, like cellular adhesion, proliferation and
differentiation into specific cell types, e.g. bone cells that will
form new bone tissue and, thus, integrate the implant strongly
into the natural peri-implant tissues. Recently, the third
generation biomaterials have been and are being actively
developed especially for in situ regeneration of damaged tissue
and different tissue engineering applications. In these materials,
living cells are used in combination with artificial materials
(Navarro et al. 2008, Hench & Polak 2002, Ratner & Bryant 2004).
There is a wide variety of chemical, biochemical,
physiological, physical and other mechanisms which are
involved in the contact between biomaterials and various tissues
and thus will affect the biocompatibility (Williams 2008,
Vadgama 2005). The concept of biocompatibility is not limited to
bulk characteristics of the material used, but also includes
physical and chemical surface properties as well as various
decomposition products. In some cases, chemical composition of
implant surface is of major importance, but in other cases some
physical features such as stiffness and shape of the implant are
the most important determinants of biocompatibility. The major
material variables that may influence on the host response are
summarized in Table 3.
Table 3: Major material characteristics that affect the host response, modified from
Williams (2008).
Material variable
The composition and structure of bulk material
Crystallinity and crystallography
Elastic constants
Porosity
Chemical composition of surface
Surface roughness, texture
Surface charge and electrical properties
Surface wettability (hydrophilicity, hydrophobicity)
Corrosion parameters, ion release profile and ion toxicity
Degradation profile and toxicity of degradation products
5 – Biocompatibility
69
The interaction between an artificial material and a living
system can be considered to consist of the local and systemic
response of host tissue to the material as well as the response of
the material to the host tissue (Williams 2008). After
implantation of a biomaterial into the body, the host response
involves a series of complex protein and cellular events.
According to the ‚race for the surface‛ concept (Gristina 1987),
there is in many occasions a critical competition between the
microbial and host tissue cells for the implant surface. If this
race is won by tissue cells, the implant surface is covered by
tissue which provides resistance against bacterial colonization,
and promotes tissue integration of the implant. On the other
hand, if the race is won by microbes, the implant surface will be
covered by bacteria and/or bacterial biofilm. In this case, the
host cellular functions are hindered, resulting in adverse tissue
reactions, and in the worst scenario, the implant has to be
removed to protect the host.
When a biomaterial is implanted into the living tissue or
placed in cell culture, water molecules bind within nanoseconds
to its surface. The extent and pattern of the interaction between
the water molecules with the surface of implant depends on the
surface properties of the material, such as surface free energy
(SFE) (Roach et al. 2007, Kasemo 1998). Subsequently, within
seconds to hours after implantation, the implant surface
becomes covered by a monolayer of adsorbed serum or
interstitial proteins. Since there are many hundreds of different
proteins in blood and interstitial tissue fluid, there is a race for
the surface between these molecules. Typically, the more
abundant small proteins lead this race due to their rapid
transport and attachment to the surface. Later, larger proteins
with higher affinity towards the surface replace these small
proteins (Roach et al. 2007). The protein layer formation process
involves a very rapid initial phase, followed by a slower
maturation phase as the protein coating of the implant
approaches a steady-state phase. In the third stage, that occurs
from as early as minutes after implantation, lasting up to several
days, cells eventually adhere, spread out and differentiate on
5 – Biocompatibility
70
the protein molecules coating the biomaterial surface. These
cellular processes are influenced by biological molecules such as
extracellular matrix proteins, cell membrane proteins and
cytoskeleton proteins as well as by the physicochemical
properties of the surface such as its chemistry, topography,
surface energy and electric charge (Puleo & Nanci 1999, Roach et
al. 2007). Finally, the desired integration of implant will occur
with the surrounding tissue. In biomedical implants, the final
stage which can last from a few days (rapidly biodegradable
devices) up to several decades (total hip replacement), can be
seen as the continuum of the early implant stages. During this
last stage, also adverse responses, such as clots, fibrous capsule
formation, wear, corrosion, late hematogenic infections and
device failure may occur.
5.1 BIOCOMPATIBILITY TESTING
Rigorous biocompatibility testing, both in vitro and in vivo, is
needed when novel materials that are aimed for use in
implantable medical devices, are developed. The International
Organization for Standardization (ISO) has published the
standard series ISO 10993 for biological evaluation of medical
devices. This series of standards is currently composed of 18
parts.
In vitro cytotoxicity tests (ISO 10993-5) are usually performed
as the first stage acute toxicity and cytocompatibility testing for
materials planned for use in medical devices. These tests are
rapid to perform, sensitive, well standardized and cheap.
Unnecessary use of animal models can be also avoided. In vitro
tests are convenient in studies in which the adhesion,
proliferation, differentiation and contact guidance of cells as
well as the mineralization of matrix on biomaterial surface are
investigated. These standardized and easily quantifiable tests
(Pearce et al. 2007, Pizzoferrato et al. 1994) are also useful in
studies aimed to clarify the individual effects of different surface
5 – Biocompatibility
71
parameters such as SFE, surface charge and topography on the
cellular responses.
Various cell lines can be used for cytotoxicity testing; the cell
line should be chosen to represent the cells that will be in
contact with the biomaterial in its predicted in vivo use. Tests
can be performed either via direct contact with a material and/or
with extracts of the material. The direct contact test simulates
the near-surface effects of the implanted biomaterial on tissues,
whereas the extracts mimic the effect of the particles that are
released from the surface. An incubation time ranging from 24
hours to 96 hours is recommended.
5.2 CELL-BIOMATERIAL INTERACTIONS
The clinical success of implants is largely based on adaptive and
integrative interactions between the implant material and tissue
cells. Although appropriate mechanical properties, durability
and functionality of modern biomaterials and related devices
are primarily derived from the bulk properties of materials, the
biological responses to materials are affected by their physical
and chemical surface properties. Thus, the interactions between
cells and implants are regulated to a large extent by the surface
properties of the implant material, although properties like
elasticity can also affect host responses. The tendency of cells to
adhere or not to adhere to the surface of implant is crucial for
their performance, for example in total joint replacement
prostheses or urinary stents (Tomlins et al. 2005). In order to
improve the biological performance of implants, the effects of
modified surface properties on cellular functions have been
widely studied (Ratner & Bryant 2004, Williams 2008). Surface
chemistry, roughness, texture, wettability and surface charge are
among the characteristics that have an effect on the cellular
response to a biomaterial.
5.2.1 Effect of surface topography on cellular responses
Surface roughness is a measure of the random topography of a
surface and it is quantified by the vertical variations of a real
5 – Biocompatibility
72
surface from its ideal form. Amplitude parameters that are used
in quantification of vertical deviations of the roughness profile
from the midline can be grouped into two subclasses: averaging
parameters and peak-to-valley parameters.
The arithmetic mean deviation of the absolute value of the
profile from a midline, Ra, is the most widely used parameter to
quantify surface roughness. Ra is determined as:
lr
a dxxzlr
R0
.1
(4)
In this equation, z(x) is the absolute profile value from a midline
and lr is the sampling length over which the surface profile has
been measured (Tomlins et al. 2005).
However, the Ra values are not always useful parameters for
assessing biomaterial surfaces, because of the lack of detailed
information about the geometry of the surface or the variations
in peak heights or valley depths. Different peak-valley
parameters have been developed to solve this problem
including statistical distribution parameters called skewness
and kurtosis. The simplest approach is to determine the highest
and lowest points within the overall measuring length, and then
the Rpv (peak-to-valley roughness) value can be estimated as the
vertical distance between the farthest points.
Both contact and non-contact techniques are available for
measuring surface roughness. Contact methods are usually
based on the use of stylus-based instruments in which a
measurement needle is traversed over a test surface and the
movement of the stylus arm can be detected e.g. using a
piezoelectric crystal. However, these devices are not suitable for
assessing the surface texture of soft materials, as they tend to
damage the surface to be analyzed. Non-contact optical
techniques such as laser profilometry and white light
interferometry are especially useful for assessing the surface
roughness of sensitive soft materials. Furthermore, different
scanning probe microscopy techniques such as atomic force
microscopy (AFM) and scanning tunneling microscopy can be
5 – Biocompatibility
73
used for imaging surfaces with resolution at the atomic level.
(Tomlins et al. 2005, Giessibl 2003, von Recum et al. 1996)
The effect of surface roughness on cell functions has been
studied in biomaterial surfaces which can be modified by
several techniques such as sandblasting (Lampin et al. 1997,
Iwaya et al. 2008, Orsini et al. 2000, Anselme & Bigerelle 2005),
plasma treatment (Chu et al. 2002, Nebe et al. 2007, Saldana et al.
2006), mechanical polishing (Ponsonnet et al. 2003), acid etching
(Iwaya et al. 2008, Orsini et al. 2000, Anselme & Bigerelle 2005)
or laser treatment (Teixeira et al. 2007). It has been reported that
surface roughness characteristics have an effect on the
morphology, adhesion, proliferation and differentiation of
osteoblast-like cells (Puckett et al. 2008, Kim et al. 2005, Khang et
al. 2008, Das et al. 2009) as well as on fibroblasts (Könönen et al.
1992, Kunzler et al. 2007, Ponsonnet et al. 2003, Wirth et al. 2005).
It is generally recognized that fibroblasts (connective tissue cells)
prefer smooth surfaces whereas osteoblastic cells favor rougher
surfaces (Kunzler et al. 2007, Ponsonnet et al. 2003, Wirth et al.
2005). Furthermore, enhanced differentiation and bone-specific
gene expression of adhered human mesenchymal stem cells
(hMSCs) have been demonstrated on nanorough surfaces (Ra <
50 nm) compared to rougher, machined surfaces (Ra ≈ 100 - 400
nm) (Mendonca et al. 2009, 2010, Dalby et al. 2006). However,
the effects of surface roughness are also dependent on the
substrate material being used (Hallab et al. 2001, Jinno et al.
1998). Moreover, it is important to note that a change in the
surface roughness is also related to changes in the wettability
properties of the surface. This is partly due to the fact that
machining techniques commonly used to roughen or polish the
surfaces often also introduce changes in the surface charges and
tensions. Therefore, it may be difficult to conclude which
mechanism is involved in modulating the cell behaviour at these
instances (Khang et al. 2008, Das et al. 2009, Ponsonnet et al.
2003).
Surface texture refers to a surface topography in which
patterns of features are placed deliberately on the surface.
Several microfabrication techniques, mainly originating from
5 – Biocompatibility
74
the production of integrated circuits, have been recently utilized
in the biomaterial research to study cellular reactions to a
substrate having a three-dimensional structured surface
topography in the micro- or nanoscale. The major part of
published literature has focused on micro- and nanostructures
produced through photolithography or electron beam
lithography in conjunction with reactive ion etching processes
on the surfaces of Si, quartz or Ti (Curtis & Wilkinsson 1997,
Flemming et al. 1999, Martinez et al. 2009). More recently,
polymer microfabrication techniques, such as hot embossing
(Charest et al. 2004), injection molding (Myllymaa et al. 2009b)
and soft lithography (Mata et al. 2002a,b) have been increasingly
used. Different pattern feature types such as grooves/ridges,
wells, pits and pillars with a wide variety of pattern dimensions
have been examined (Martinez et al. 2009). Substrate micro- and
nanotopography, independently of surface biochemistry, seems
to have significant effects on cell orientation and adhesion,
morphology, proliferation and differentiation. It has been
demonstrated that cells seeded onto microgrooved surface
aligned their shape and elongated in the direction of the grooves
(Brunette et al. 1983, Chou et al. 1995, 1998, Britland et al. 1996,
Curtis & Wilkinson 1997, Teixeira et al. 2003, Dalby et al. 2004,
Ber et al. 2005, Mwenifumbo et al. 2007, Loesberg et al. 2007, den
Braber et al. 1996, Clark et al. 1987, Charest et al. 2004, Hamilton
& Brunette 2005; Hamilton et al. 2006). The cell membrane
conformed to the widest grooves, but tended to bridge the
narrowest grooves (Matsuzaka et al. 2003). Other geometrical
features such as wells, pits and pillars with dimensions of less
than 5 µm seem to lead to smaller, rounded cells with less
organized cytoskeletons (Martinez et al. 2009, Andersson et al.
2003, Gallagher et al. 2002). The results from investigations
related to the effects of nanoscale topographical structures on
cell proliferation and differentiation are somewhat contradictory
and highly dependent on the cell line and structures being
investigated (Puckett et al. 2008, Andersson et al. 2003, Martinez
et al. 2009).
5 – Biocompatibility
75
5.2.2 Effect of wettability properties on cell-biomaterial
interactions
Surface energy of materials is of special interest in the
biomedical applications due to its obvious effect on cell-
biomaterial interactions. Surface wetting properties have a
significant effect on various biological events at the sub-cellular
and cellular level including protein adsorption, cell attachment
and spreading. Contact angle measurements are widely used in
the assessment of wettability properties. In this technique, a
small droplet (e.g. 15 µl) is placed on the surface and the angle
between the drop (liquid-vapour interface) and the substrate
(solid) is determined. Relatively more water wettable surfaces
are called hydrophilic (high SFE), while less wettable surfaces
are hydrophobic (low SFE) surfaces (Ratner 1996, Hasirci &
Hasirci 2005). Although these are relative terms, it has been
suggested that water contact angles smaller than 65° indicate
hydrophilic surface, while those larger than 65° indicate
hydrophobic surface (Vogler 1998).
Different approaches, such as the Owens-Wendt (Owens &
Wendt 1969) and van Oss (van Oss et al. 1988) models have been
developed to estimate SFE of materials and their components on
the basis of measured contact angles with polar and non-polar
liquids whose surface tension values are known. Using the
Owens-Wendt model, the polar PS and dispersive D
S
components can be estimated:
).)()((2)cos1( 2/12/1 P
L
P
S
D
L
D
SL
(5)
Here θ is the measured contact angle value and L , DL and P
L stand for the liquid’s total SFE and its dispersive and
polar components, respectively. The total SFE ( S ) is the sum of
its dispersive and polar components:
.P
S
D
SS
(6)
5 – Biocompatibility
76
The polar component of the SFE is related to hydrogen
bonding, dipole-dipole and other site-specific interactions,
whereas the dispersive component refers to the van der Waals
and other non-site specific interactions between the solid surface
and liquid applied on it.
Increased wettability with a higher SFE has been shown
generally to enhance the adhesion, and spreading of various cell
types such as fibroblasts (Wei et al. 2007, Altankov et al. 1996,
van Kooten et al. 1992), osteoblasts (Khang et al. 2008, Feng et al.
2003, Lim et al. 2008, Myllymaa et al. 2009b) and MSCs (Sawase
et al. 2008). Moreover, through the SFE modifications, it is
possible to influence the differentiation of MSCs into the desired
phenotypes (Curran et al. 2005, 2006, Marletta et al. 2007,
Calzado-Martin et al. 2010). It has been presented that the SFE
is an even more important surface characteristic than surface
roughness for cellular adhesion strength and proliferation and
might be a useful parameter for modifying cell adhesion and cell
colonization onto engineered tissue scaffolds (Hallab et al. 2001).
The wettability of a solid material can be modified either by
changing the surface roughness (Wang et al. 2006, Hao &
Lawrence 2007) or the surface chemistry (Lim et al. 2008,
Michiardi et al. 2007, Kennedy et al. 2006, Myllymaa et al. 2009b).
In hydrophobic materials, surface roughening increases the
hydrophobic nature of material, whereas in hydrophilic
materials surface roughening enhances hydrophilicity. Surface
chemistry modifications are usually performed by using
different plasma treatments (Lim et al. 2008, Michiardi et al.
2007), proper surface coatings (Khang et al. 2008, Myllymaa et al.
2009b), SAMs (Kennedy et al. 2006, Folch & Toner 2000) or
incorporating dopants into the bulk material (Myllymaa 2009a).
5.2.3 Effect of charge distribution on cell-biomaterial
interactions
The electric charge distribution in the vicinity of a biomaterial
surface can be considered to be one of the main factors affecting
the biological response of a biomaterial (Khorasani et al. 2006,
Cai et al. 2006, Krajewski et al. 1998, Altankov et al. 2003). This
5 – Biocompatibility
77
charge depends on several factors, such as the chemical
composition of the material surface, the composition of the
surrounding fluid and the environmental pH value (Krajewski
et al. 1998). The charge and its distribution are closely related to
protein adsorption, the strength of cellular adhesion and the
shape of the cells growing on the surface of the biomaterial
(Cheng et al. 2005, Bodhak et al. 2009, Thian et al. 2010).
Therefore, the investigations of the relationship between
electrical charge on a biomaterial surface and protein adsorption
as well as subsequent cellular interactions are particularly useful
ways to improve understanding of the mechanisms involved in
the biological integration of materials with tissues.
The interface between a solid surface and the surrounding
electrolyte establishes a charge distribution which is different
from the bulk phases of solid and liquid. In this electrochemical
double layer model, at the vicinity of a solid-liquid interface, the
charge carriers are fixed (a stationary layer), whereas they are
mobile in the liquid phase (a mobile layer) at a greater distance
The zeta potential (ζ) is termed as the potential decay between
the solid surface and the bulk liquid phase at the shear plane
(Lyklema 1995) (Fig. 8). Streaming current/streaming potential
measurement is a convenient method for examining the existing
interface charges of a solid biomaterial in an electrolyte solution
(Roessler et al. 2002). In this method, an electrolyte is circulated
through the measuring cell and forced under pressure to flow
through a small gap formed by two sample surfaces. This fluid
flow evokes a relative motion between the stationary and mobile
layer and this leads to a charge separation which provides
experimental access to the zeta potential.
Higher positive values of the zeta potential at a fixed pH
reflect a positive charge of the surface which would attract
negatively charged entities such as anions or charged proteins
while lower values of zeta potential at a fixed pH are evidence
of a negative charge of the surface which tends to attract
positively charged cations and particles (Cai et al. 2006). The
relationship between the magnitude of zeta potential and
cellular responses is not well understood (Altankov et al. 2003,
5 – Biocompatibility
78
Hamdan et al. 2006), although in recent studies it has been
demonstrated that negatively charged surfaces can enhance
osteoblastic cell – biomaterial interactions (Bodhak et al. 2010,
Thian et al. 2010). Instead, the effect of zeta potential on protein
adsorption has been more extensively studied. The lower the
absolute value of the zeta potential of the surface, the lower is
the adsorption of serum proteins (Krajewski et al. 1996, 1998, El-
Ghannam et al. 2001).
stationary layer mobile layer
-
-
-
-
-
-
-
-
+
+
+
+
+
+
+
+
-
-
-
+
Distance
Ele
ctro
kine
tic p
oten
tial
ζ
+ -
+
solid
Figure 8: The interface between a solid surface and a surrounding solution introduces
an electrochemical double layer. The charge distribution is divided into a stationary
and a mobile layer. These layers are separated by a plane of shear. The zeta potential (ζ )
is assigned to the potential decay at this shear plane.
5.2.4 Effect of surface micropatterning on cellular responses
Surface micropatterning has been extensively employed for
controlling the adhesion, proliferation, differentiation and
contact guidance of cells (Folch & Toner 2000, Falconnet et al.
2006, Ito 1999). Photolithography and soft lithography are
powerful techniques to produce surface patterns which differ
from the background e.g. in terms of surface chemistry (Kane et
al. 1999, Winkelmann et al. 2003, Scotchford et al. 2003, Levon et
al. 2009, Myllymaa et al. 2009a), surface charge (Soekarno et al.
1993), SFE (Matsuda & Sugawara 1995), or the attachment of
biological substances that can influence cellular function
(McBeath et al. 2004, Chen et al. 1997, Blawas & Reichert 1998).
79
6 Aims of the present study
In this study, micro- and nanofabrication methods have been
utilized in the development of novel neural interfaces as well as
intelligent implant materials capable of achieving the desired
cellular responses on a biomaterial surface. It was hypothesized
that these approaches would enhance the tissue response and
functionality of the implants used in contact with soft (neural)
or hard (bone) tissue. The main aims of this thesis were:
1. to design and fabricate novel thin film microelectrode arrays
for stable and high-resolution intracranial recordings,
2. to study the suitability of flexible polymers (e.g. polyimide)
as construction materials in bio-MEMS applications to
replace materials based on silicon technology,
3. to study the opportunities to improve the electrochemical
properties (stability, contact impedance) of bioelectrodes by
using modern micro- and nanostructuring methods,
4. to investigate the cytocompatibility of different dielectric
thin films produced by USPLD method and to clarify their
suitability for neural interfaces,
5. to test whether the surface micropatterning could be used to
enhance the cytocompatibility of cell unfriendly silicon-
based implants and improve their biocompatibility during
the initial phase of integration with tissue, and
6. to investigate the effect of surface micropatterning,
produced using photolithographic and PVD techniques, on
the behaviour (e.g. adhesion, spreading, and differentiation)
of osteoblast-like cells and human mesenchymal stem cells.
6 – Aims of the present study
80
81
7 Materials and methods
The present thesis consists of six studies, which are referred to
by the Roman numerals (I-VI). In addition, some unpublished
data are included. A summary of the materials and methods
used in this thesis is presented in Table 4.
Table 4: Overview of the materials and methods used in the present thesis.
Study Device/ sample Characterization Cell line/
Animal
Testing methods
I Microelectrode array
Optical microscopy, EIS
Wistar rat Acute cortical evoked potential recordings
II Microelectrode array
Optical microscopy, SEM,
AFM, EIS
Wistar rat Acute and chronic cortical evoked potential
recordings
III PI films on Si,
Al203 on Si (*)
C3N4 on Si (*)
AFM, Contact angle, SFE, Zeta
potential
BHK-21
Qualitative SEM analysis of cultured cells, MTS
assay, Fluorescence
microscopy (live/dead)
IV Review article
V Micropatterned Ti, Ta, Cr and DLC
coatings on Si
Optical microscopy, AFM,
Contact angle,
SFE, Zeta
potential (*)
hMSC Quantitative and qualitative SEM analysis
of cultured cells,
Fluorescence
microscopy of stained cells (focal adhesion,
ALP, OC, MS)
VI Micropatterned Ti and DLC coatings
on Si
Micropatterned Cr and Ta coatings
on Si (*)
(#) SaOS-2 Quantitative and qualitative SEM analysis
of cultured cells,
Confocal laser scanning
microscopy of stained
cells (focal adhesion)
Unpubl. data
nanorough USPLD Pt surfaces
EIS, AFM
(*) Additional materials/methods: not included in original publications III, V or VI;
(#) Sample surfaces were characterized in study V
Abbreviations: EIS: electrical impedance spectroscopy; SEM: scanning electron
microscopy; AFM: atomic force microscopy; PI: polyimide; Si: silicon; Al203: alumina;
C3N4: carbon nitride; SFE: surface free energy; BHK-21: baby hamster kidney fibroblast;
Ti: titanium; Ta: tantalum; Cr: chromium; DLC: diamond-like carbon; hMSC: human
mesenchymal stem cell; ALP: alkaline phosphatase; OC: osteocalcin; MS: mineral
staining; SaOS-2: sarcoma osteogenic cell; USPLD: ultra-short pulsed laser deposition;
Pt: platinum
7 – Materials and methods
82
7.1 MICRO- AND NANOFABRICATION OF ELECTRODES AND
OTHER SAMPLES
A wide variety of micro- and nanofabrication techniques as well
as materials were used in this thesis work. A description of
these fabrication processes is given in chapters of 7.1.1 – 7.1.5.
7.1.1 Preparation of photomasks
In the studies I, II, V and VI, photolithographic methods were
used in patterning thin deposition layers on PI or Si substrates.
The photomasks used in these studies were designed by CleWin
layout editor, version 4.0.2 (WieWeb software, Hengelo, The
Netherlands) and fabricated using the laser scanning technique
by Mikcell Oy (Ii, Finland) on 4-inches or 5-inches soda-lime
glass plates with a structured Cr layer. The photomask used in
studies I and II consists of two mask patterns both repeated five
times enabling the ‚batch fabrication‛ at a prototyping scale.
The first set of patterns was used in structuring the metallization
layer and the second set in patterning the PI insulating layer to
expose electrode sites. The photomasks used in studies V and VI
consist of micropatterned areas ( 4 mm x 4 mm or 8 mm x 8 mm)
containing regularly spaced microsquares (sides 5, 25, 75 or 125
µm) or microcircles (diameters 5, 25 or 125 µm).
7.1.2 Flexible polyimide-based microelectrode arrays
A fabrication process for flexible MEAs that follows photo-
lithographic and thin film procedures was developed and
iterated to a fully functional solution during a two year trial
period such that appropriate sensor layouts and materials with
their thicknesses for base, metallization and insulating layers
were optimized. The most relevant variations are presented in
Table 5. The developed arrays consisted of either 8 or 16 round-
shaped microelectrodes with diameters of 200 µm or 100 µm,
respectively, in an area of about 2 mm x 2 mm at the end of a PI
ribbon. The layouts of these electrodes are presented in Fig. 9.
7 – Materials and methods
83
Table 5: Developed microelectrode array prototypes.
Type Channels Substrate material
and thickness
Electrode material,
its diam.
Insulation layer, its
thickness
Reference
A 8 Polyimide PI-2525a: 3 layers, total 30 µm
Ti/Pt, 200 µm
PI-2771a, 3µm
Study I
B 8 Kapton HN filmb 25 µm + PI-2525 5 µm
Ti/Pt, 200 µm
PI-2771, 3µm
Study II
C 8 Kapton HN film 25 µm + PI-2525 5 µm
Ti/Au, 200 µm
PI-2771, 3µm
Myllymaa et al. 2008
D 16 Kapton HN film 25 µm + PI-2525 5 µm
Ti/Pt, 100 µm
PI-2771, 3µm
Myllymaa et al. 2008
E 16 Kapton HN film 25 µm
+ PI-2525 5 µm
Ti/Pt,
100 µm
SU-8c,
3 µm
Myllymaa
et al. 2008
Abbreviations: Ti: titanium; Pt: platinum; Au: gold a HD Microsystems GmbH, Bad Homburg, Germany; b Goodfellow Ltd., Cambridge, UK; c Microchem Corp., Newton, MA, USA
A
B C
32 m
m
1 mm
Electrode sites
Connector
pads
Figure 9: (a) The layout of the 16-channel microelectrode array. The length of arrays is
designed to be about 32 mm. There are connector pads at one end of the device allowing
connection of the array to miniature connector and further to recording
instrumentation. The magnified schema of the 16-channel (b) and 8-channel (c) array
viewed at the recording ends. The sizes of circular-shaped microelectrodes are 100 µm
(b) and 200 µm (c).
7 – Materials and methods
84
Similar processes were used in studies I and II except for the
construction of a base layer and the curing parameters for
insulating PI layer. In study I, a base layer was constructed from
three spin-coated PI (PI-2525, HD Microsystems) layers, i.e. the
thickness of the base was about 30 µm. In study II, a 25 µm thick
Kapton HN film (Goodfellow Ltd., Cambridge, UK) was utilized
together with a spin-coated single PI-2525 layer, resulting in a
base with total thickness of 30 µm. The process flow is
demonstrated in Fig. 10.
polyimide, d ~ 30 µm photoresist
Ti/Pt thin film, d ~ 200 nm
UVUV
glass slide
A
B
C
D
connector padelectrode opening
E
F
G
photosensitive polyimide, d ~ 3 µm
Figure 10: The fabrication process flow of polyimide-based microelectrode arrays. The
process is based on photolithographic and magnetron sputter deposition techniques.
Electrode sites, transmission lines and connector pads are sandwiched between two
layers of polyimide. See details in the text.
First, one of the above mentioned practices was used to form
a 30 µm thick PI layer onto 2‛ x 3‛ microscope glass slides
(Logitech Ltd., Glasgow, Scotland, UK) which were used to
ensure a rigid support throughout the process (Fig. 10a). The PI
layer was cured in an oven according to the manufacturer’s
recommendations. After cooling, 20 % hexamethyldisilazane
7 – Materials and methods
85
(HMDS) (Riedel-de-Haen AG, Seelze, Germany) in xylene was
spun upon the PI base layer to enhance adhesion between the PI
and the photoresist. A negative photoresist (ma-N 1420, Micro
Resist Technology GmbH, Berlin, Germany) was spun (5000
rpm, 30 s), pre-baked on a hot plate (100°C, 120 s), exposed with
i-line (365 nm) UV light (Karl Suss MA45, Suss Microtec Inc.,
Waterbury Center, VT, USA) for 30 s, developed in a ma-D 553s
developer (Micro Resist Technology), rinsed with deionized
water and post-baked in an oven (100°C, 30 min) (Fig. 10b). Thin
films of Ti and Pt were deposited at thicknesses of 20 nm and
200 nm, respectively, by using magnetron sputtering (Stiletto
Serie ST20, AJA International Inc., North Scituate, MA, USA)
and high purity (99.9 % or better) Ti and Pt targets (Goodfellow
Metals Ltd., Huntingdon, UK) (Fig. 10c). Ti serves as an
adhesion promoter between the PI and the Pt. The metallization
layer was patterned via a lift-off procedure in resist remover,
mr-Rem 660 (Micro Resist Technology) forming electrodes,
transmission lines and connector pads on the PI base (Fig. 10d).
Photosensitive polyimide PI-2771 (HD Microsystems) was used
as an insulation layer. PI-2771 was spun (5000 rpm, 60 s) and
pre-baked on the hot plate (Fig. 10e). After cooling, it was
exposed to UV light (12 s) and developed in cyclopentanone
(Sigma-Aldrich Corp., St. Louis, MO, USA) and rinsed with
propylene glycol methyl ether acetate (PGMEA, Sigma-Aldrich).
The PI-2771 layer was cured in an oven in which the
temperature was gradually raised either to 200 °C after which it
was maintained for 30 minutes (study I) or to 300 °C after which
it was maintained for 60 minutes (study II). PI-2771 formed
about a 3 µm thick insulation layer that was present on all
positions other than the areas of the electrode sites and
connector pads (Fig. 10f). Finally, the arrays were detached
from the glass slide (Fig. 10g) and cut into their final shapes
using scissors and a knife. Since only the edge regions were
tightly attached to glass slides (detailed descriptions of these
protocols are given in original papers I and II), it was possible to
achieve detachment easily with by immersing in deionized
water.
7 – Materials and methods
86
Thin film connector pads were designed to fit into a 16-
channel 0.5 mm pitch zero-insertion-force (ZIF) connector (JST
Ltd., Halesworth, UK). This ZIF connector was soldered with
lead-free solder to a thin printed circuit board (PCB) adapter
(Kytkentälevy Oy, Helsinki, Finland) containing also a 2 x 8
channel surface mount microsocket (CLM-serie, Samtec Inc.,
New Albany, IN, USA) through which the array was connected
to the recording instrumentation. Casco strong epoxy-resin
(Akzo Nobel Coating Oy, Vantaa, Finland) was used as an
encapsulant at the interface between the array and the ZIF
connector and the signal tracks in the PCB in order to avoid
short-circuiting when implanted onto the surface of rat cortex.
7.1.3 Micropatterned biomaterial surfaces on silicon
In studies V and VI, micropatterned titanium (Ti), tantalum (Ta),
chromium (Cr) and diamond-like carbon (DLC) coatings were
fabricated using photolithography and PVD methods. Si wafers
were used as substrates. Prior to application of the photoresist,
the Si wafers were dry-baked and coated with adhesion
promoter (20 % HDMS in xylene). Then, the chosen photoresist,
i.e. ma-N 1420 (Micro Resist Technology) in study V or SU-8
(MicroChem) in study VI was applied, patterned and cured
using the optimized process parameters described in the
original papers (V, VI). Two PVD methods were used to carry
out thin film depositions. Magnetron sputtering was used to
deposit Ta, Ti and Cr coatings and a filtered pulsed plasma arc
discharge method was used to produce DLC coatings onto the
surface of the patterned wafers. The process parameters were
optimized (details are described in the original articles V and VI)
to achieve well-adhering, smooth coatings with a thickness of
about 200 nm. The micropatterns were revealed via a lift-off
procedure by immersing the wafers in a resist remover (acetone
or mr-Rem 660) ultrasonic bath. The biomaterial coating
deposited on the top of the photoresist was removed together
with the dissolved resist, and the final microfeatures were
formed. The Si wafers were cut into individual samples (10 mm
x 10 mm) with a custom-made device using a diamond knife.
7 – Materials and methods
87
The size of each final micropatterned sample used in the
cellular studies was 10 mm x 10 mm. In study VI, three different
sample sets were utilized. In the first sample set, each chip
contained four 4 mm x 4 mm sample areas. Three of these areas
consisted of circles with diameters of 5, 25 or 125 µm or squares
with the length of the sides being 5, 25 or 125 µm composed of
either DLC or Ti on a Si background with the fourth sample area
containing Ti or DLC as a homogeneous surface (Fig. 11a,b).
Regardless of the size of the patterns, the circles and squares
covered 30.6 % and 25 % of the total sample surface area,
respectively. In the second set of samples, the pattern and
background were reserved (‚inverse samples‛) so that the
patterns were composed of uncoated Si squares or circles on a
DLC or Ti background. In the third set, each sample contained a
micropatterned area of 8 mm x 8 mm at the center of the sample.
Micropatterning in these areas consisted of 75 µm squares with
100 µm spacing between adjacent squares in two orthogonal
directions (Fig. 11c). Thus, the (Ti, Ta, Cr or DLC) patterns
occupied 18.4 % of the total sample surface area. This last
sample type was also used in study V.
Figure 11: Sample types used in studies V-VI to clarify cellular behaviour on
micropatterned surfaces: (a) circular patterning, diameters of circles correspond to 125
µm, 25 µm and 5 µm plus plain reference area (bottom right), (b) rectangular
patterning, sides of squares correspond to 125 µm, 25 µm and 5 µm plus plain
reference area (bottom right) and (c) rectangular patterning, sides of squares are 75
µm.
7 – Materials and methods
88
7.1.4 Tailored surfaces produced by ultra-short pulsed laser
deposition
Ultra-short pulsed laser deposition (USPLD) technique was
utilized in the creation of novel coatings for applications of bio-
MEMS. Firstly, ultrasmooth alumina (Al2O3) and carbon nitride
(C3N4) dielectric films were deposited on Si wafers to clarify
their cytocompatibility properties using similar fibroblast
experiments as in study III (unpublished findings). Secondly, Pt
depositions with varied nanotextured surface topography were
carried out on SS, Si and Kapton substrates aiming to find out
new solutions to improve the electrochemical properties of
bioelectrodes.
Neural electrodes need to be encapsulated in a material that
provides excellent barrier properties and shows acceptable
biocompatibility. In this context, Al2O3 and C3N4 were
considered as interesting coating candidates. Biomaterials
coated with Al2O3 have been used for a long time in medical
applications, particularly in fields of dentistry and orthopedics
(Szeitzer et al. 2006). The biocompatibility of Al2O3 has been
previously demonstrated as a bulk material (Heimke et al. 1998,
Takami et al. 1997) as well as an atomic layer deposited film
(Finch et al. 2008). The dielectric properties of pulsed laser
deposited Al2O3 are also superior (Katiyar et al. 2005) and
USPLD deposited Al2O3 provides good insulating properties
even in liquid environment (R. Lappalainen, personal
communication 2009). Carbon nitride is a promising novel
coating material for biomedical applications. It has superb
chemical inertness and its tribological and dielectric properties
are also favourable (Cui & Li 2000). The abbreviation ‚CN‛ is
commonly used to describe all carbon nitride coatings with
different C/N ratios. The bio- and hemocompatible nature of CN
has been also demonstrated (Du et al. 1998, Cui & Li 2000, Cui et
al. 2005). However, in neural applications, the testing of Al2O3
and CN films has been rare.
The USPLD method utilizes intense laser pulse (typically in
ps time scale) to erode a target and deposit the eroded material
in the form of high speed ionized plasma onto a substrate. Just
7 – Materials and methods
89
prior to the depositions, the sample surfaces were gently
cleaned using Ar+ ion sputtering (SAM-7KV, Minsk, Belarus). A
new type of mode-locked fiber laser (Corelase Oy, Tampere,
Finland) and the ColdabTM deposition technology developed by
Picodeon Ltd. Oy (Helsinki, Finland) were used for USPLD. A
high purity sintered target of C3N4 and Al2O3 was used for
deposition. In both cases, optimized laser parameters (Amberla
et al. 2006), such as pulse energy, pulse repetition rate and
scanning length were used. Maximum average power was 20 W
at 4 MHz, which results in 5 µJ pulse energy. The pulse length
was 15-20 ps. The laser beam was focused into a 15-25 micron
spot giving the maximum fluences about 4 J/cm2 at 5 µJ. Both
Al2O3 and C3N4 films were deposited at thicknesses of about 200
nm on p-type Si wafers.
The USPLD technique was also used to produce nanorough
Pt depositions on SS, Si and Kapton substrates. Laser deposition
parameters were modified to obtain thin (100-300 nm) Pt films
with a different nanotextured surface topography (unpublished
data). Due to the high energy of Pt plasma, good adhesion was
achieved without using any special intermediate layers or
special techniques. The main deposition parameter to control
the surface topography was the Ar gas atmosphere with a
pressure varied in the range 1-10-6 mbar.
7.1.5 Spin-coated polyimide films for cytotoxicity testing
In order to evaluate the cytocompatibility of different PI grades,
non-photosensitive PI-2525 (HD Microsystems) and photo-
sensitive PI-2771 (HD Microsystems) were deposited on Si
wafers by using a spin-coating technique. Prior to application of
the PI precursor, 2-inches, p-type <100> Si wafers (Si-Mat,
Landsberg am Lech, Germany) were baked for 20 min at 200 °C
and coated with VM-651 (HD Microsystems) to remove
moisture and to enhance the adhesion, respectively. Then, the
PI-2525 and PI-2771 precursors were spin-coated (5000 rpm, 60 s)
under low acceleration onto Si wafers. Subsequently, the PI-2771
deposits were pre-baked (100 °C, 90 s), exposed to 365nm UV
light (Karl Suss MA-45), developed in cyclopentanone (Sigma–
7 – Materials and methods
90
Aldrich) and rinsed with PGMEA (Sigma–Aldrich). Half of the
PI-2771 coated wafers and all of the PI-2525 coated wafers were
polymerized by curing them in an oven in which the
temperature was gradually raised to 350 °C over a period of 60
min and then maintained for 60 min at that temperature.
Another half of the PI-2771 coated wafers was cured at a lower
temperature (200 °C). Therefore, three different PI groups were
investigated: (1) non-photosensitive PI-2525, cured at 350°C; (2)
photosensitive PI-2771, cured at 200 °C and (3) photosensitive
PI-2771, cured at 350 °C.
7.2 MICROSCOPIC CHARACTERIZATION OF SURFACES
Different microscopic methods, including optical microscopy,
scanning electron microscopy (SEM), atomic force microscopy
(AFM) and contact angle measurements were used in this thesis
in quality control monitoring and in the characterization of the
produced samples and devices. Brief descriptions of these
characterization techniques are given in the following sections
of 7.2.1– 7.2.3.
7.2.1 Scanning electron microscopy
Scanning electron microscopy was used for examining the
quality of photolithographically patterned surfaces (studies V
and VI), spin-coated PI films (study III) and the MEAs (study II).
Prior to imaging with a Philips XL30 ESEM-TMP (Fei Company,
Eindhoven, The Netherlands), the samples were usually coated
with a thin layer of Au using a Sputter Coater E 5100 (Polaron
Equipment Ltd., Hertfordshire, UK).
7.2.2 Atomic force microscopy
Atomic force microscopy was used in this thesis (studies II, III,
and V) to analyze the surface topography (roughness) of
biomaterial samples and to determine the thickness of thin film
depositions. The characterization was performed using a PSIA
XE-100 (Park Systems Corp., Suwon, Korea) AFM system at
7 – Materials and methods
91
ambient temperature and humidity. Al-coated Si cantilevers
(Acta-10, ST Instruments B.V., LE Groot-Ammers, The
Netherlands) were used in a non-contact mode to probe the
surface across an area of 2 x 2 µm. The scanning rate of 0.25 Hz
was used. The thicknesses of thin film depositions as well as the
average surface roughness (Ra) and peak-to-valley roughness
(Rpv) values of sample surfaces were determined using the
analysis software (XIA) of the AFM instrument.
7.2.3 Contact angle measurements and determination of
surface free energies
The contact angle measurements were performed to clarify
wettability properties of material surfaces (studies III and V).
Prior to the measurements, the samples were ultrasonicated in
ethanol and deionized water and allowed to dry. Contact angles
were measured using the sessile drop method at constant
laboratory conditions (22 °C and in 45% relative humidity) with
the aid of a custom made apparatus that consisted of a stereo
microscope SZ-PT (Olympus Corp., Tokyo, Japan) equipped
with a digital camera (Camedia C-3030-ZOOM, Olympus). In
this method, a small droplet (15 µl) was pipetted onto the
surface under study, and after 5 seconds, the droplet was
photographed. Gnu Image Manipulation program (GIMP,
www.gimp.org) was used to determine the contact angle from
the shape geometry of the droplet. One polar liquid, i.e. water
and one non-polar liquid, i.e. diiodomethane was used to access
the total SFE as well as its polar and dispersive components. The
Owens-Wendt model, described in chapter 5.2.2, was used to
determine SFEs.
7 – Materials and methods
92
7.3 ELECTROCHEMICAL CHARACTERIZATION OF SURFACES
The electrochemical properties of electrodes and other
biomaterial surfaces were investigated by electrochemical
impedance spectroscopy (EIS) and zeta potential measurements.
Descriptions of methods used are given in the following sections.
7.3.1 Electrochemical impedance spectroscopy
Electrochemical impedance spectroscopy was used (studies I
and II) to characterize the electrochemical properties of the
fabricated microelectrodes, i.e. to assess the recording
capabilities of the electrode for making neural measurements.
The measurements were performed with an LCR meter (3531 Z
HiTester, HIOKI E.E. Corp., Nagano, Japan) (study I) or with a
Solartron 1260 impedance gain/phase analyzer (Solartron
Analytical, Farnborough, UK) (study II). In studies I and II, the
MEA was immersed in physiological saline solution (0.9% NaCl)
at room temperature and a small sinusoidal perturbation signal
(100 mV without any DC offset) was applied between the
individual microelectrode and the Pt counter electrode having a
much larger surface area (Fig. 12a). The induced current and its
phase were recorded. Frequency bands of 50 Hz - 1 MHz and 1
Hz-100 kHz were used in the studies I and II, respectively.
Electrode-electrolyte interface impedance spectra of SS
electrodes with and without a nanotextured Pt layer were also
measured (unpublished data). The EIS measurements were
implemented in a custom-made miniature measurement cell
(Fig. 12b), in which a working electrode and a counter (Pt)
electrode were clamped against the opposing walls of the cell.
The holes with a diameter of 8 mm at the both walls exposed
both the working and the counter electrodes with equal areas of
50 mm2. The measurements were carried out at frequencies
between 1 Hz and 1 MHz by applying a sinusoidal voltage of
100 mV without any DC offset using the Solartron 1260
impedance analyzer.
7 – Materials and methods
93
Pt counter electrode
microelectrode
Pt counter
electrode
Sample under test
saline
solution
A
B
Figure 12: Set-ups used to measure the electrode-electrolyte interface impedances.
Electrical impedance spectroscopy was carried out with a two-electrode configuration,
in which a platinum sheet was used as a counter electrode. Working electrode, i.e.
microelectrode (a) or nanotextured platinum coating on stainless steel (b) was
immersed in saline (0.9 % NaCl) solution or exposed to contact with it through the
hole (Ø: 8 mm) drilled at the wall of polytetrafluoroethylene-based pool, respectively.
7 – Materials and methods
94
7.3.2 Zeta potential measurements
In study III, the streaming current measurement technique was
used to examine existing interface charges of the studied solid
biomaterials in a liquid. The streaming current measurements
were performed with an electrokinetic analyzer (SurPASS,
Anton Paar GmbH, Graz, Austria) equipped with an adjustable
gap cell. In each measurement, two identical test samples, i.e.
pieces of Si wafer coated with biocompatible thin film were
fixed on the sample holders with a cross section of 20 mm x 10
mm by using a double-sided adhesive tape. The sample holders
were then inserted in the adjustable gap cell and a gap of
approximately 100 µm was adjusted between the surfaces of the
samples during the measurement. An electrolyte was circulated
through the adjustable gap cell and forced under pressure to
flow through a narrow gap. This evoked a comparative
movement of charges in the electrochemical double layer which
could be detected by two Ag-AgCl electrodes placed at both
ends of the gap channels. The measurements were performed
using 1 mM KCl as an electrolyte solution at a fixed pH of 7.4 ±
0.2. The pH of KCl was adjusted with 0.1 M HCl and 0.1 M
NaOH. An electrolyte was circulated through the adjustable gap
cell and forced under pressure to flow through a narrow gap.
This evoked a comparative movement of charges in the
electrochemical double layer which could be detected by two
Ag-AgCl electrodes placed at both ends of the gap channels.
The measurements were performed using 1 mM KCl as an
electrolyte solution at a fixed pH of 7.4 ± 0.2. The pH of KCl was
adjusted with 0.1 M HCl and 0.1 M NaOH. The zeta potential (ζ)
was obtained from streaming current measurements according
to the Helmholtz-Smoluchowski equation (Lyklema 1995):
.0 A
L
dp
dI
(7)
Here dI/dp is the slope of the streaming current versus the
differential pressure, is the electrolyte viscosity, 0 is the
vacuum permittivity, is the dielectric constant of the
7 – Materials and methods
95
electrolyte and L is the length of the streaming channel, and A is
the cross-section of the streaming channel (the rectangular gap
between the planar samples).
7.4 RECORDING OF EVOKED POTENTIALS IN RATS
The suitability of the MEAs for cortical mapping of evoked
potentials was investigated in acute and chronic recordings in
Wistar rats in the A.I.V. Institute of the University of Eastern
Finland (studies I and II). All experiments were conducted in
accordance with the Council of Europe guidelines and approved
by the Institutional Animal Care and Use committee and the
State Provincial Office of Eastern Finland.
Prior to surgery, Wistar rats were anesthetized with urethane
(1.2-1.5 g/kg) for acute recordings and with medetomidine/
ketamine (0.5 mg/kg and 75 mg/kg, respectively) for implanting
the chronic electrodes, and placed in a stereotaxic apparatus. A
microelectrode array was placed on the dura over the pre- and
postparietal cortices between lambda and bregma (Fig. 13).
Figure 13: Depiction of the implanted electrode location in Wistar rat skull.
Microelectrode array was inserted on the dura over the pre- and postparietal cortices
between lambda and bregma. A stainless steel screw used as a reference/ground
electrode (R) was placed 2-3 mm in front of bregma.
A stainless steel screw inserted 2-3 mm in front of bregma
was used as a reference electrode connected to a ground
potential. In the chronic recordings, electrodes were anchored
onto the rat’s skull with screws and bone cement. After surgery
7 – Materials and methods
96
under medetomidine/ketamine anesthesia, the rat was aroused
by giving the antagonist, atimepazole, and rats were allowed to
recover for a minimum of 7 days before the first experiments. In
the chronic trials, the implanted rats were either allowed to
move freely or were gently immobilized by holding it securely
in a towel.
Both auditory and electrical current stimuli were used to
generate EPs. The auditory stimuli consisted of 25 sets of paired
pulses (pulse duration: 10 ms) with pulse intervals of 500 ms,
followed by a 9.5 s quiet interval. The square wave current
stimuli consisted of 25 sets of paired pulses (pulse duration: 1
ms) with pulse intervals of 500 ms, followed by a 9.5 s quiet
interval, delivered to the left front paw of the rat. Auditory and
somatosensory evoked field potentials were amplified with a
preamplifier (Neuralynx Inc., Bozeman, MT, USA) and with a
main amplifier (Grass Instruments, West Warwick, RI, USA)
and led to the data acquisition PC running SciWorks
(DataWave Technologies, Loveland, CO, USA). In study II, in
vivo electrode impedance measurements were performed using
a multi-task tester (Fredrich Hare Inc., Brunswick, ME, USA) in
order to clarify the long-term performance of the MEAs.
7.5 CELL CULTURE STUDIES
Studies III, V and VI concentrated on investigating cellular
responses on the surface of the biomaterial samples. The
following chapters (7.5.1 – 7.5.4) describe the methodology used
in these studies.
7.5.1 Cell lines
In study V, hMSCs (PoieticsTM, Lonza Group Ltd., Basel,
Switzerland) were cultured in Lonza Mesenchymal Stem Cell
Growth Medium (MSCGM) at 37 °C in 5% CO2 in air. The cell
monolayer was washed twice with 140 mM phosphate buffered
saline (PBS, pH 7.4) and the cells were removed by applying
0.25% trypsin in PBS-EDTA (ethylenediamine-tetraacetic acid)
7 – Materials and methods
97
solution for 5 min at room temperature. The suspension was
then transferred to a Falcon tube and trypsin was removed by
centrifugation. The cells were resuspended in culture medium,
seeded onto the biomaterial surfaces at a density of 0.52 x 104
cells/cm2 and cultured for 7.5 h or 5 days. In the osteogenic
induction experiments, the hMSC cells were seeded at a density
of 0.31 x 104 cells/cm2. Half of the samples were induced 24 h
after seeding by replacing MSCGM with Osteogenesis Induction
Medium (Lonza). Induced hMSCs were supplied every 3–4 days
with fresh medium. Noninduced hMSCs (control samples) were
fed with MSCGM using the same schedule.
In study III, baby hamster kidney fibroblasts (BHK 21, clone
13, HPA Culture Collections, Salisbury, UK) were cultured in
Dulbecco’s Modified Eagle’s Medium (DMEM, EuroClone
S.p.A., Pero, Italy) supplemented with 10% fetal calf serum (FCS,
PAA Laboratories GmbH, Linz, Austria), 50 µg/ml ascorbic acid
(Sigma-Aldrich), 2mM l-glutamine (PAA Laboratories), 20
IU/ml penicillin (EuroClone), and 200 µg/ml streptomycin
sulfate (EuroClone). The cells were seeded onto the surface of
the samples (7mm x 7mm) by adding a medium which
contained 5.0 × 104 cells/ml. The cells were cultured for 24 h at
37 °C in 5% CO2 in air.
In study VI, human primary osteogenic sarcoma SaOS-2
(ECACC 890500205) cells were cultured in McCoy’s 5A culture
medium containing GlutaMAXTM (Gibco BRL/Life Technologies
Inc., Gaithersburg, MD, USA) and supplemented with 10% v/v
FCS, 100 IU/ml of penicillin and 100 µg/ml streptomycin. Cells
were seeded onto the biomaterial surfaces (10 mm x 10 mm) at a
density of 15-25 x 103 cells/cm2, and cultured for 48 h or 120 h at
37 °C in 5% CO2 in air.
Tissue culture-treated polystyrene 12-well (studies V and VI)
and 24-well cell culture plates (study III) supplied from Corning
Inc. (NY, USA) and Greiner Bio-One GmbH (Frickenhausen,
Germany), respectively, were used in all experiments.
7 – Materials and methods
98
7.5.2 MTS assay
The MTS (3-(4,5-dimethylthiazol-2-yl)-5-(3 carboxymethoxy-
phenyl)-2-(4-sulfophenyl)-2H-tetrazolium)-based proliferation
assay (CellTiter 96® Aqueous One Solution Reagent, Promega
Corp., Madison, WI, USA) was used as a colorimetric method to
assess the cytotoxic effects of different PI grades (study III) and
also other potential material candidates for neural sensors
(additional data). The MTS assay, a modification from the more
conventional MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyl-
tetrasodium bromide) assay, is based on the reduction of MTS
salt in the presence of phenazine methosulfate to a water-
soluble formazan product in actively functioning mitochondria.
Since it is water-soluble, no solubilization step is required with
MTS as is the essential with the MTT assay. The quantity of
formazan formed was determined by measuring the absorbance
at 490 nm, which is directly proportional to the number of living
cells in the culture. After 24-h cultivation of the BHK-21 cells,
the samples were transferred into unused wells, and a fresh
DMEM medium plus MTS reagent (200 µl) was added. The
following incubation period lasted 1 h, after which the media
were removed, the wells were washed with PBS and the
absorbances were measured at 490 nm with an ELISA reader
(Molecular Devices, Inc., Sunnyvale, CA, USA). The relative
numbers of viable cells were determined by normalizing the
optical density values to the highest measured values.
7.5.3 Scanning electron microscopy of cultured cells
Scanning electron microscopy was used to analyze the cellular
behaviour on biomaterial surfaces. In studies V and VI, after
incubation period, samples were washed twice with PBS and
fixed overnight in 2.5 % glutaraldehyde (Sigma-Aldrich) at +4°C.
After this, the samples were washed with PBS and dehydrated
in a graded ethanol series (from 50 % solution to absolute
ethanol). The dehydration was completed using a Bal-Tec CPD
030 Critical point drying unit (Bal-Tec AG, Balzers,
Liechtenstein). Samples were mounted on SEM stubs, coated
with a Pt layer with an Agar sputter device (Agar Scientific Ltd.,
7 – Materials and methods
99
Stansted, UK) and examined using a Zeiss DSM 962 SEM (Carl
Zeiss GmbH, Oberkochen, Germany) at an accelerating voltage
ranging from 8 kV to 10 kV.
In study III, the cells were fixed with 2.5% (w/v)
glutaraldehyde (Sigma-Aldrich) in sodium cacodylate buffer
(pH 7.4) and dehydrated with ethanol gradient and HMDS
(Sigma-Aldrich). These samples were examined with an
environmental SEM (Philips XL30 ESEM-TMP, Fei Company) at
an accelerating voltage of 8–15 kV after being coated with a thin
layer of Au using a Sputter Coater E 5100 (Polaron Equipment).
7.5.4 Immunofluorescence and confocal laser scanning
microscopy
After either a 7.5 or 120-hour incubation period with the hMSCs
(study V) or after a 48-hour incubation period with the SaOS-2
cells (study VI), the specimens were triple stained with
fluorescence dyes to reveal cytoskeletal (actin) and focal
adhesion (vinculin) proteins and cell nuclei. A detailed
description of the fixation, staining and viewing of the cells is
given in original articles V and VI. Briefly, the cells grown on Si
pieces were washed with PBS solution, fixed with
paraformaldehyde solution, rinsed in PBS and permeabilized in
Triton X-100 solution. After immersing in normal goat serum,
the cells were incubated with the primary antibodies (anti-
vinculin) and washed. Cells were then incubated
simultaneously with fluorochrome-conjugated stains containing
Alexa fluor 488 (Molecular Probes Inc., Eugene, OR, USA) and
Alexa fluor 568 (Invitrogen Corp., Carlsbad, CA, USA) to detect
focal adhesion (vinculin) and actin cytoskeleton. Cell nuclei
were stained with DAPI (study V) or with TO-PRO-3 (study VI)
probes and then the samples were washed again, and mounted
on objective slides. Stained cells were viewed in a fluorescence
microscope (Olympus AX70) and in more detail using a Leica
TCS SP2 confocal laser scanning microscope (Leica
Microsystems GmbH, Mannheim, Germany) equipped with
appropriate filter sets.
7 – Materials and methods
100
In study V, the development of the osteoblastic phenotype
was followed using alkaline phosphatase (ALP), osteocalcin (OC)
and mineral staining (MS) as differentiation markers at 14, 17
and 21 days, respectively. A detailed description of the staining
of hMSCs is given in original article V. After staining, the
samples were fixed onto objective slides and coverslipped. The
cells were observed using an Olympus AX70 fluorescence
microscope in ALP and OC assays and a Leitz Diaplan
microscope (Leica Microsystems GmbH, Wetzlar, Germany) in a
mineralization assay.
In study III, specimens were stained using a combination of 2
fluorescent probes, i.e. a cell-impermeable DNA-binding dye,
propidium iodide (Sigma-Aldrich), and a cell-permeable
fluorescein diacetate (Fluka Chemie GmbH, Buchs, Switzerland)
in PBS to assess the viability of the BHK-21 fibroblasts cultivated
for 24 hours on the samples. Detailed information on staining
procedure can be found in original article III. Fluorescein
diacetate is a non-fluorescent molecule that penetrates through
the cell membranes. Inside the cell, fluorescein diacetate is
hydrolyzed by non-specific esterases resulting in fluorescein
release. Fluorescein emits green fluorescence when irradiated by
laser light. Propidium iodide is a common fluorescent molecule
used to stain DNA and to identify dead cells in a cell population.
Since propidium iodide is membrane impermeable, it only
stains dead cells which have leaking membranes. Propidium
iodide emits a red fluorescence when excited. Since only viable
cells are able to produce fluorescein, these two stains can be
used to separate the living and dead cells in a cell populations.
Cells are either stained with propidium iodide or fluorescein
diacetate, but not with both. The samples were viewed in a
confocal scanner (PerkinElmer Life Sciences, Wallac-LSR,
Oxford, UK) on a Nikon Eclipse TE300 microscope (Nikon Corp.,
Tokyo, Japan), using the wavelengths 488/10 nm (excitation) and
525/50 nm (emission) for the fluorescein, and 568/10 nm
(excitation) and 607/45 nm (emission) for the propidium iodide.
Finally, the numbers of dead (red-staining) cells (nuclei/mm2)
and the surface area of live (green-staining) cells were counted
7 – Materials and methods
101
using Image J software 1.37c (National Institute of Health,
Bethesda, MD) (Abramoff et al. 2004).
7.6 STATISTICAL ANALYSES
The numerical results are expressed as mean ± standard
deviation of the mean (SD) or as mean ± standard error of the
mean (SEM). In studies III, V and VI, statistical analyses were
conducted with the SPSS software (versions 14.0 and 16.0, SPSS
Inc., Chicago, IL, USA). One-way analysis of variance (ANOVA)
followed by Tukey Post-Hoc Tests were applied to determine
the statistical significance of the differences observed between
material groups. Values of p < 0.05 were considered statistically
significant.
7 – Materials and methods
102
103
8 Results
8.1 EVALUATION OF MICROELECTRODE ARRAYS
A number of MEA prototypes were developed during the 2-year
trial period by optimizing the substrate material, substrate
thickness, electrode material, sensor layout and insulation
material (Table 6). It was demonstrated that the magnetron
sputter deposited and photolithographically patterned Pt thin
films between two layers of PI represented a very powerful and
reproducible way to produce miniaturized neural interfaces (Fig.
14). The ZIF type connector allowed the easy and reliable
connection of a thin film MEA to recording instrumentation (Fig.
14b). The use of the ZIF connector eliminated the need for
soldering, which had been observed to be very challenging to
perform on the thin Pt pads produced by the sputtering
technique on PI. Furthermore, the use of PSPI (PI-2771)
simplified the electrode fabrication considerably because this
material could be patterned directly in the same way as
photoresists without using complex multilevel processes (i.e.
lithography, etching mask formation, etching mask dissolution)
which were unavoidable steps when conventional PIs were used.
It was possible to pattern the PI-2771 layer with good alignment
to the metallization layer (Fig. 14f,g). Moreover, one could
pattern the PI-2771 layer with an almost vertical pattern profile
and without residuals (Fig. 14h).
The electrode type B, originally presented in study II, was
found to be the most suitable for recording cortical surface field
potentials in rats. The major shortcomings of the other
prototypes compared to type B are summarized in Table 6.
8 – Results
104
A B
C D
E F
G H
recording
sites
connector board
transmission
lines 5 mm
5 mm
1 mm 1 mm
200 µm 50 µm
Figure 14: The developed polyimide-based microelectrode arrays (a). The PI ribbons
are connected via a connector board consisting of a ZIF- type connector (JST Ltd.) and
a surface mount microsocket (Samtec Inc.) to the recording instrumentation (b). The
developed array is highly miniaturized as shown in comparison to a clinical subdural
grid electrode (c), and extremely flexible (d). A SEM-image of the 8-channel array
viewed at the end of the recording sites before deposition of the insulation layer (PI-
2771) (e). Patterning of this polyimide insulator layer achieved a good alignment with
the metallization layer (f,g). The 3 µm thick polyimide layer had an almost vertical
pattern profile and was virtually free of residuals (h).
8 - Results
105
Table 6: The summary of developed microelectrode array prototypes. The main
disadvantages of each array type compared to type B are presented.
Type Channel
number
Substrate Electrode
diameter (µm)
Insulator Weaknesses
A 8 PI-2525a (3 layers)
Ti/Pt, 200 PI-2771a Troublesome fabrication
B 8 Kapton HNb
+ PI-2525 Ti/Pt, 200 PI-2771 –
C 8 Kapton HN + PI-2525
Ti/Au, 200 PI-2771 Higher electrode impedances
D 16 Kapton HN
+ PI-2525
Ti/Pt, 100 PI-2771 Short-circuiting
E 16 Kapton HN
+ PI-2525
Ti/Pt, 100 SU-8c Short-circuiting
and cracking
Abbreviations: Ti: titanium; Pt: platinum; Au: gold a HD Microsystems GmbH, Bad Homburg, Germany; b Goodfellow Ltd., Cambridge, UK; c Microchem Corp., Newton, MA, USA
Due to the viscosity and chemical properties of PI-2525, the
formation of a 30 µm thick substrate layer required three spin
coating cycles (type A). This was somewhat complicated
compared to type B (study II), in which the same thickness
could be readily achieved by using Kapton HN film together
with one spin-coated PI-2525 layer. Gold electrodes (type C)
have significantly higher electrode-electrolyte interface
impedance levels compared to Pt electrodes especially at low
frequencies below 1 kHz which are the most relevant when
biological signals need to be recorded (Fig. 15). The
microelectrode surface area in high-density arrays (types D and
E, electrode diameter 100 µm) was only 25 % of the area of the
larger electrode (types A-C, electrode diameter 200 µm) causing
significantly higher electrode impedances.
8 – Results
106
Au
Pt
10 102 103 104 105
frequency (Hz)
|Z| (
Ω)
Pha
sean
gle
(°)
107
106
105
104
103
-20
-40
-60
-80-81.0
-74.1
95.5 k
25.5 k
Au
Pt
Figure 15: Impedance spectroscopy of platinum (Pt) and gold (Au) microelectrodes
(diameter 200µm), measured in 0.9 % saline solution at room temperature. The
impedance magnitude and phase are presented with error bars corresponding to
standard deviations (SD). In some instances, SD values were often smaller than the
symbol size and thus not shown. The absolute values of the impedance magnitude and
the phase are also given at 1 kHz.
Designs D and E were also susceptible to short-circuiting by
cerebrospinal fluid when the array was implanted onto the
surface of the rat cortex. Moreover, epoxy-resist SU-8 (type C)
was observed to be totally unsuitable for present electrode
application. While PI- insulated arrays were very robust and
they could be bent even into sharp angles without any cracks,
the bending/implantation did create many cracks and even
delaminations of the SU-8 layers.
The developed electrodes were tested in acute and chronic
evoked potential recordings in Wistar rats. The flexibility of PI
was particularly suitable for following the curved cortex surface,
8 - Results
107
thus ensuring good contact between electrode site and the dura
mater. This enabled reliable acquisition of cortical signals.
During the acute recording sessions, the MEA type B was
capable of capturing stable recordings as observed in the form
of the standard response parameters, such as latencies, onsets
and decays of the main components in the potential traces (Fig.
16a). The recorded signal traces were biologically meaningful
and reproducible, this being verified by comparing the
responses at the beginning of the experiments with those
obtained at the end. In chronic recordings, the electrode was
capable of yielding signals comparable with the presented acute
signals for approximately two weeks after the array
implantation (Fig. 16b). However, after 16 days, the responses
decayed. This decline in signal amplitudes was concurrent with
the observed jump in the electrode impedances which were
systematically measured in vivo at three-day intervals (Fig. 16c).
The physical appearance of the dura around the electrode
occasionally changed during chronic implantation. It became
thicker and encapsulated the electrode. However, in microscopic
examination, the surface of the brain appeared to be undamaged
with the electrode location being slightly depressed. Although
no macroscopic scar formation was seen in the examination of
the brain after the array was removed, this kind of jump in the
in vivo impedances was compatible with a thickening of the
dura, a common finding in chronic experiments, and
microscopic growth of non-conductive fibrous tissue around the
electrodes, shrinking the freely exposed surface area of the
recording electrode.
8 – Results
108
Figure 16: (a) An acute recording of somatosensory evoked potentials (SEP) in the rat
parietal cortex. Profiles of averaged SEP traces using the 8-channel microelectrode
array are shown on the right with their correspondence to the recording sites which are
numbered in schema in left side. The position of reference/ground electrode is marked
as “R”. The current stimuli consisted of 25 sets of paired pulses with pulse intervals of
500 ms, followed by a 9.5 s quiet interval, delivered to the left front paw of the rat. (b)
A chronic recording of auditory evoked potentials in the rat parietal cortex. The stimuli
consisted of 25 sets of paired pulses at intervals of 500 ms, followed by a 9.5 s quiet
interval. (c) Average electrode impedance (mean ± SD, n = 8) measured in vivo.
8 - Results
109
8.2 EVALUATION OF NANOROUGH ELECTRODE SURFACES
The USPLD technique was used to prepare the Pt coatings on
SS, PI and Si with a desired nanostructured surface topography.
The adhesion of coatings deposited upon all the tested
substrates was good without any signs of delamination. The
USPLD process parameters were adjusted to achieve different
surface topographies. Fig. 17 shows an example where the
feature size is about 30 nm. The nanostructured electrode
surface clearly increased the effective surface area compared to a
smooth Pt surface typically achieved using sputtering.
Figure 17: AFM image of nanostructured Pt-electrodes produced using the USPLD
technique. The average roughness of the surface is 18 ± 4 nm (mean ± SD)
Electrochemical impedance spectroscopy was applied in
order to study the electrode-electrolyte impedance behaviour of
different Pt-coated SS electrodes (unpublished data). A total of
six sample areas (50 mm2) of each plate were examined using
the measurement set-up described in chapter 7.3.1. The mean
impedance magnitude and phase data for Pt coating, whose
surface topography was demonstrated in Fig. 17, are shown in
Fig. 18. Pure SS electrode and sputter-coated Pt on SS are
8 – Results
110
presented as reference materials. The results clearly reveal that
the presence of the nanostructures on an electrode surface had
increased the surface area and decreased impedance. The
impedance magnitudes (mean ± SD) at 1 Hz were 7.0 kΩ ± 0.5
kΩ, 17.8 kΩ ± 1.1 kΩ and 21.7 kΩ ± 1.0 kΩ for nanostructured
USPLD-Pt, sputtered-Pt and pure SS, respectively. The
differences in impedance values between the materials were
statistically significant (p < 0.001) at low frequencies. Due to the
relative large geometrical surface area, all coatings reached the
same impedance magnitude level of approximately 180-200 Ω
by 1 kHz, which corresponds to the resistance value of
electrolyte solution. All surfaces are highly polarizable, but
nanostructuring decreases the phase difference between current
and voltage making electrodes less polarizable.
100 101 102 103 104 105102
103
104
105
Frequency (Hz)
|Z| (o
hm
)
SS.datSputtered Pt.datUSPLD-Pt.dat
-80
-60
-40
-20
0
Ph
ase
an
gle
(°)
Figure 18: Impedance magnitude and phase of a nanostructered platinum-coating
(USPLD-Pt) and sputter-coated Pt-coating deposited on stainless steel (SS). Results
for pure SS electrode are presented for reference. A clear drop in electrode impedance
can be seen for USPLD-Pt, especially at low frequencies. Note the logarithmic scale.
8 - Results
111
8.3 CYTOCOMPATIBILITY TESTING OF INSULATOR
MATERIALS
Prior fibroblastic BHK-21 cell experiments, spin-coated PIs (PI-
2525 films and PI-2771 films cured at 200 °C and at 350 °C) and
USPLD coatings (C3N4 and Al2O3) were characterized in terms of
their roughness, wettability and zeta potential (Table 7). The
surface roughness values obtained were all very low, i.e. the Ra
values were below 1 nm. However, the Ra value for the PI-2525
was significantly higher than for other coatings (p < 0.001). The
contact angles for water on both PI-2771 film types were
significantly higher than for the PI-2525 (p < 0.001). The same
tendency was also noticed when contact angles were measured
for diiodomethane (results are shown in the original article III,
Table 1) highlighting the more hydrophobic nature of the PI-
2771 with significantly lower polar and dispersive surface
energy values (p < 0.001). Somewhat surprisingly, the effect of
the chosen curing protocol (200 °C vs. 350 °C) for the PI-2771 on
roughness or on wettability was not statistically significant. The
contact angles for water on C3N4 were in the same range as on
PI-2525 film, evidence of a slightly hydrophobic thin film
coating. In contrast, the Al2O3 coating was very similar to PI-
2771 film with respect to its wettability properties. All of the
surfaces were negatively charged, and the zeta potential values
measured at a fixed pH of 7.4 were in a narrow range of -54.2
mV – -66.2 mV. The PI-2771-200 film followed by the PI-2525
film was the most negatively charged and their zeta potential
values differed significantly from the all other test materials (p <
0.001).
8 – Results
112
Table 7: Average surface roughness (Ra), water contact angle and zeta potential
values of tested polyimide and ceramic coatings. All of the materials were extremely
smooth although the roughness of PI-2525 was significantly higher than that of other
materials. The PI-2525 and carbon nitride (C3N4) were more hydrophilic compared to
alumina (Al2O3) and both PI-2771 types. Zeta potential values were all in the same
range by demonstrating the negative surface charge of these coatings.
Material Ra (nm) Contact angle (°) Zeta potential (mV)
PI-2525 0.89 ± 0.13 * 69.5 ± 1.8 ## -59.9 ± 0.3 *
PI-2771-200 0.43 ± 0.15 95.1 ± 1.3 -66.2 ± 1.6 *
PI-2771-350 0.39 ± 0.04 93.2 ± 0.8 -54.2 ± 0.1
C3N4 0.24 ± 0.05 # 70.8 ± 0.9 ## -55.1 ± 1.2
Al2O3 0.18 ± 0.03 # 94.0 ± 1.3 -56.6 ± 3.4
* p < 0.001, as compared to all other materials # p < 0.001, as compared to all polyimides ## p < 0.001, as compared to all other materials, except of C3N4/PI-2525
The values are mean ± SD. One-way ANOVA was used to determine the statistical differences between materials.
In this study, the cytotoxicity of the insulator candidate
materials was examined with in vitro tests conducted along to
the guidelines of the international standard ISO-10993-5. Latex
rubber and PE were used as positive and negative control
material, respectively. The MTS assay defined the number of
viable BHK-21 cells attached on the surfaces 24 h after cell-
seeding. The results were normalized relative to the highest
value, i.e., the value for the PE control (Fig. 19). The relative cell
numbers in the different polyimide groups and C3N4 were at a
level of 62-70 % of the PE controls. The number of cells attached
to the Al2O3 surface was lower, i.e. 50 %, but still significantly
higher than attained with the latex rubber control.
Fluorescence microscopy studies with live/dead staining
probes revealed that all test materials were highly
cytocompatible, in contrast to latex rubber, on which viable cells
were rarely found, instead dead cells had most probably
detached and were floating in the medium (illustrated in the
original article III). Results from the quantitative analysis
included the determination of the density of dead cells
8 - Results
113
(nuclei/mm2) and the surface area of live cells (µm2) is presented
in the Table 8. The numbers of dead cells were very low on each
surface, i.e., mean values ranged from 13 cells/mm2 to 48
cells/mm2. Even though the dead cell density was also minimal
on the surface of latex rubber, the cytotoxic effect could be
clearly seen when the surface area of viable cells was examined.
The total surface area of viable cells on latex rubber was
significantly lower (p < 0.001) than that of the other test
materials, being only around 2% of the corresponding values
obtained for other surfaces. The values for C3N4 and Al2O3 were
slightly but not significantly higher than that for PE control or PI
materials.
Table 8: The density of the dead cells (nuclei/mm2) and the surface area of viable cells
(µm2) on polyethylene (PE, negative control), different polyimide surfaces, alumina
(Al2O3), carbon nitride (C3N4) and latex rubber (positive control) after a 24-hour
cultivation period evaluated using a combination of two fluorescent probes and Image J
software.
Material Density of dead cells
(1/mm2)
Surface area of viable
cells (µm2)
PE (n = 10) 13.9 ± 5.4 70400 ± 2900
PI-2525 (n = 13) 13.5 ± 5.5 71900 ± 8500
PI-2771-200 (n = 8) 18.8 ± 8.2 67300 ± 13200
PI-2771-350 (n = 8) 30.8 ± 8.8 63000 ± 5800
C3N4 (n = 13) 48.0 ± 10.3 83900 ± 11600
Al2O3 (n = 15) 47.4 ± 5.8 82600 ± 12300
Latex rubber (n = 8) 24.9 ± 8.6 1400 ± 320*
The values are the mean ± SEM. *p < 0.001 (One-way ANOVA).
SEM imaging was used to monitor cell morphology,
adhesion and spreading on the test surfaces. The BHK-21 cells
which had been cultivated for 24 h on each PI surface, Al2O3 and
C3N4 exhibited a normal fibroblastic morphology, adhered well
with filamentous extensions, and had started to form clusters
(Fig. 20). Thus, all insulator materials were considered not to be
toxic and equally good for allowing the adhesion and spreading
of cultured fibroblasts.
8 – Results
114
0
20
40
60
80
100
120
PE PI2525 PI2771-200 PI2771-350 Alumina CN Latex
Re
lative
ce
ll n
um
be
r (%
)*
*
0
5
10
15
20
PE PI-2525 PI2771-200 PI2771-350 alumina CN Latex
Ce
ll co
ve
rag
e (
%)
*
PE PI-2525 PI-2771-200 PI-2771-350 Al2O3 C3N4 Latex
PE PI-2525 PI-2771-200 PI-2771-350 Al2O3 C3N4 Latex
A
B
Figure 19: (a) Relative number of BHK-21 cells on polyethylene (PE, negative control),
conventional polyimide (PI-2525), two differently cured photosensitive polyimides (PI-
2771-200 and PI-2771-350), alumina (Al2O3), carbon nitride (C3N4) and latex rubber
(positive control). The data was obtained by MTS assay and normalized to the highest
optical density value. (b) Surface area covered by BHK-21 cells at 24 h on the same test
materials. The data was acquired from confocal laser scanning microscope images of
the samples with live/dead-stained cells. All polyimide and ceramic coatings can be
considered to be non-cytotoxic and they allowed adhesion and spreading of fibroblasts.
The error bars indicate standard errors of the means. *p < 0.001 (one-way ANOVA).
8 - Results
115
Figure 20: SEM images of BHK-21 cells cultured for 24 h on PI-2771-200 at two
different magnifications. The cells exhibited a normal fibroblastic morphology, adhered
well with filamentous extensions, and had even started to form clusters. Cells cultured
on the other test materials, i.e. PI-2525, PI-2771-350, C3N4 and Al2O3, appeared to be
very similar and thus, are not shown.
8.4 MICROPATTERNED BIOMATERIAL COATINGS
The interactions of the hMSCs and SaOS-2 cells on
micropatterned biomaterial (Ti, Ta, Cr, DLC) coatings were
investigated in studies V and VI. The roughness and surface
energy data of test materials were originally published in paper
V, but these values are also valid for the samples used in study
VI. A summary of these results, supplemented with the zeta
potential values achieved for smooth (non-patterned) surfaces is
given in Table 9. All of the materials used in these studies were
extremely smooth, with Ra values of less than 2 nm leading to
the hypothesis that the observed differences in cell behaviour
were not due to the differences in surface topography but were
mediated by different chemical compositions and physical
patterns of material studied. The only relevant topographical
cues of the samples were attributable to the thickness of coating
(~ 200 nm) that forms steps between micropatterns and the
background. All coating materials also possessed very similar
wettability properties, i.e. their SFE values were in a range
between 47.3 - 52.1 mJ/m2. The Si wafer used as background
material in these experiments had a significantly higher SFE
(68.5 mJ/m2) compared to all other coating materials. The
negative zeta potential values obtained from streaming current
8 – Results
116
measurements at a fixed pH of 7.4 disclosed that the surfaces of
all coating materials were negatively charged. The differences in
zeta potential values between the surfaces were statistically
significant (p < 0.001 for all).
Table 9: Average surface roughness (Ra), water contact angle, surface free energy and
zeta potential values for the surfaces of different test materials. All surfaces were
extremely smooth (i.e. mirror finish) and their contact angle/surface energy values
were in a narrow range (except for background Si), although some statistically
significant differences existed.
Material Ra (nm) *
Contact angle (°)
Surface energy (mJ/m2)
Zeta potential (mV) *
DLC 0.3 ± 0.1 67.4 ± 2.0 49.6 ± 1.2 -70.2 ± 1.4
Ti 1.8 ± 0.2 67.6 ± 2.0 49.7 ± 1.1 -53.9 ± 0.3
Ta 1.5 ± 0.3 64.1 ± 1.6# 52.1 ± 1.1# -49.6 ± 1.1
Cr 0.8 ± 0.1 71.2 ± 1.1# 47.3 ± 0.7# -39.6 ± 2.2
Si 0.2 ± 0.1 32.4 ± 1.1# 68.5 ± 0.8# -43.5 ± 1.3
*p < 0.005 between all materials #p < 0.05, as compared to all other materials
The values are mean ± SD. One-way ANOVA was used to determine the statistical differences between materials.
8.4.1 Behaviour of human mesenchymal stem cells
An incubation time of 7.5 hours was used to investigate the
early-stage adhesion, contact guidance and morphology of
hMSC cells on the micro-patterned DLC, Ti, Ta and Cr thin films
deposited on Si (study V). Based on observations with the SEM,
hMSCs seemed to prefer the biomaterial patterns over a Si
background. This kind of guiding effect seemed to be evident
with all materials, although only Ti and DLC surfaces are
presented here (Fig. 21a,b). Quantitative analysis showed that
on Ta, Cr and DLC, the density of hMSCs was 3.0-3.5 times
higher than on the background (p < 0.0005 for all, Fig. 22). One
exception was Ti, on which the density of hMSCs on the
patterns did not differ from the density of the cells grown on the
Si background (p = 0.14). The analysis of the surface area of
patterns and Si background covered by the MSCs at 7.5 h gave
8 - Results
117
results consistent with the results from the cell density
assessment (Fig. 23). From this viewpoint, the guiding effect was
also clear on all the materials (p < 0.05) except for Ti (p = 0.46).
Figure 21: hMSC cells after 7.5 h culture (a,b) and 5 days culture (c,d) on patterned
titanium (a,c) and diamond-like carbon (b,d) surfaces. The cells seemed to prefer the
patterns over background and some cells were aligned along the edges of the patterns
(a,b). At 5 days, cells were much larger and they had grown to form a subconfluent
monolayer. The side length of the biomaterial square is 75 µm.
One interesting observation was made about the size of cells.
The mean diameter of the cell lying on patterns (68 µm) was
significantly lower (p < 0.05) than that of cells growing on a Si
background (102 µm). Moreover, it was observed that numerous
cells were aligned along the edges of the patterns. This
phenomenon was confirmed in the immunofluorescence triple
staining, which revealed that focal vinculin adhesions and actin
cytoskeleton were outgoing from the pattern edges i.e. cells were
believed to be taking on the geometric square shapes.
8 – Results
118
140
120
100
80
60
40
20
Cel
lden
sity
(cel
ls/m
m2 )
DLC Cr Ta Ti
Background (Si)
Patterns
Average
Figure 22: Density of hMSCs at 7.5 hours. On lithographically patterned diamond-
like carbon (DLC), chromium (Cr) and tantalum (Ta) samples, the cells significantly
preferred the biomaterial patterns over a silicon (Si) background. The cell density on
titanium (Ti) patterns was marginally lower (# p ≈ 0.06, one-way ANOVA) than that
on Cr and Ta patterns. The error bars indicate standard deviations of the means.
0
20
40
60
80
100
DLC (7.5 h)
DLC (5 d)
Cr (7.5 h)
Cr (5 d)
Ta (7.5 h)
Ta (5 d)
Ti (7.5 h)
Ti (5 d)
Cell
covera
ge (
%)
Background (Si)
Patterns*
**
*
+
#
Figure 23: Coverage of hMSCs at 7.5 hours and 5 days. A clear guiding effect seen at
7.5 hours was nearly lost by 5 days. * p < 0.05, as compared to DLC and Cr, # p < 0.05,
as compared to Cr and Ta, + p < 0.05, as compared to DLC. The error bars indicate
standard deviations of the means. One-way ANOVA was utilized to determine the
statistical differences.
8 - Results
119
The spreading of the hMSCs was investigated after 5 days of
incubation (Fig. 21c,d). At this stage, the individual cells had
grown considerably achieving a size which greatly exceeded
that of a singular square pattern (75 µm x 75 µm), which meant
that it was no longer possible to calculate the cell densities.
Surface coverage analysis indicated that over 65 % of the sample
surface (of each material) was covered by cells at 5 days (Fig. 23).
The guiding effect was virtually lost since the cell density was
only 1-1.9 times higher on patterns than that on Si background.
The only statistically significant difference was observed
between the DLC patterns and the Si background. Somewhat
surprisingly, the cell density on the very same Si background
varied statistically significantly, depending on the material used
in microfabricated patterns. At 7.5 h, the hMSC-covered surface
of the background Si was significantly lower on Ti than on DLC
patterns (p = 0.015), but by the 5th culture day this situation had
reversed (p = 0.006) (Fig. 23). This indicates that apart from
direct material-cell interactions, also probably indirect cell-cell
interactions were present. The cells on material islands had
remote interactions with cells growing on Si background. After
5 days, vinculin containing focal adhesion were already well
developed all over the sample surface without any preferential
localization on the square patterns as compared to the
background.
Osteogenic differentiation of hMSCs was investigated by
using ALP, OC and MS as the differentiation markers. The
results showed that the patterned surfaces, except for DLC,
allowed induced osteogenic differentiation although this was
less effective than on plain samples. It seemed to be so that cells
growing on the micropatterned surfaces did not have enough
space to organize themselves into multicellular bone tissue-type
structures. Moreover, the rectangular shape which the hMSC
cells took due to microsquare shaped islands might not be ideal
with respect to osteogenesis. The DLC coating did not support
osteogenesis, and for that reason an inert DLC seems to be a
potential coating material for applications which come into
8 – Results
120
contact with bone but which need to be removed later, e.g.
various fracture fixation devices, such as plates and screws.
8.4.2 Behaviour of osteoblast-like (SaOS-2) cells
The SEM examination of SaOS-2 cells cultured on micro-
patterned Ti and DLC coatings (on Si) for 48-hours revealed a
clear cellular preference for the biomaterial patterns over the Si
background. The cells preferentially adhered and spread well on
the large-sized (125 µm) circular (Fig. 24a) or rectangular (Fig.
24b) Ti patterns, which facilitated their occupation by several
cells. On the reverse-patterned samples containing Si features
(circles or squares) on a Ti background (Fig. 24c), the inverse
phenomenon was observed, with almost all of the cells now
attached to the Ti background, but they seemed to avoid
spreading out into the Si circles. The cytocompatibility-
enhancing and cell-guiding effects of DLC were also clear,
though much weaker than those encountered with Ti, as
demonstrated in samples containing DLC circles on Si (Fig. 24d)
and in the reverse-patterned samples containing Si squares on a
DLC background (Fig. 24e).
The influences of the pattern shape and its size on adherence
and morphology of SaOS-2 cells were studied using rectangular
and circular patterns of three different sizes (diameters of circles
and sides of squares: 5, 25 and 125 µm). Large-sized patterns
facilitated the adhesion of several cells onto one island and the
boundaries of the outermost cells followed the shape of the
edges of the patterns to which they had attached (Fig. 25a,b).
The medium pattern size (25 µm) was designed so that singular
cells could inhabit individual patterns. This pattern size seemed
to restrict the growth of cells although they no longer adapted
so strictly to the geometrical shape of the patterns (Fig. 25c).
However, neighboring patterns were so far away that cells
could not really join together to form clusters. Particularly on
circularly patterned surfaces, cells occasionally covered two or
more islands, leading to longitudinal or star-like morphologies
(Fig. 25d). On the small-sized patterns, cell bodies covered
several of the small patterns (Fig. 25e), but their thin and slender
8 - Results
121
filopodia seemed to prefer the more cell-friendly material, and
they seemed to try to avoid bare Si areas (Fig. 25f).
Figure 24: SaOS-2 cells cultured for 48 h on surfaces containing large-sized (125 µm)
features. There is a clear cellular adhesion preference for the Ti circles (a) or squares (b)
over background Si, and in the case of reverse-patterned samples for the Ti background
over the Si circles (c). The cell-guiding effect of DLC was also clear even though
weaker than that of Ti as demonstrated in samples containing DLC circles on Si (d)
and in reverse-patterned samples containing Si squares on a DLC background (e).
Scale bar is 100 µm.
Immunofluorescence triple staining disclosed that the actin
cytoskeleton was well organized in SaOS-2 cells growing on the
biomaterial islands, but poorly organized in the cells lying on Si.
8 – Results
122
Furthermore, it was observed that the formation of vinculin
containing focal adhesions was much enhanced in cells adhering
onto the biomaterial patterns compared to cells lying on the Si
backgrounds.
Figure 25: SEM images of SaOS-2 cells cultured for 48 h on micropatterned Ti (a-d)
and DLC (e,f) surfaces. Large-sized (125 µm) squared (a) and circular (b) features
facilitated the adhesion of several cells on one Ti island with the cells aligning
themselves along the edges of the cell-friendly material. The cells adhering to medium-
sized Ti islands no longer conformed strictly to the geometrical shape of the patterns (c)
but particularly on circularly patterned surfaces, star-like cellular morphologies
appeared (d). On small-sized inverse DLC samples, the cell bodies non-selectively
covered large micro-patterned areas (e), but their filopodia clearly showed a preference
for DLC trying to avoid bare Si circles (f).
8 - Results
123
Quantitative analysis of the SaOS-2 cell density, coverage and
average cell size was performed with samples containing 75 µm
squared patterns of Ti, Ta, Cr and DLC on the Si substrate. The
relative cell density at 48h was 2.9-3.7 times higher on all four
coating materials than that achieved on the Si background (p <
0.001). The relative coverage was 2.0-3.0-fold higher for Ti, Ta,
Cr and DLC compared to the corresponding value of Si (Fig. 26).
A comparison between the four patterning materials indicated
that the proportions of surface area covered by cells were higher
for Ti and Ta than for DLC and Cr (p < 0.05). Further, the Si
background was better covered with the Ti- and Cr-patterned
samples than the background of the DLC and Ta samples (p <
0.001).
DLC Cr Ta Ti
Cel
lcov
erag
e(%
)
80
60
40
20
Figure 26: Coverage of SaOS-2 cells was always higher on biomaterial patterns than
on silicon (Si) background. Titanium (Ti) and tantalum (Ta) were better covered than
diamond-like carbon (DLC) and chromium (Cr) as designated with # (p < 0.05) and
the Si background on Ti- and Cr patterned samples was better covered than the
background on DLC or Ta samples (##, p < 0.05). The error bars indicate standard
deviations of the means. One-way ANOVA was applied to determine the statistical
significance of the differences observed between material groups.
8 – Results
124
The average sizes of the cells cultured for 48-hours on each
sample type are given in Table 10. The cells growing on the
biomaterial patterns (each material) were significantly smaller
(13-41 %) than the cells growing on Si. A comparison between
different biomaterials disclosed that the cells growing on Ti
were larger than the cells on Cr (p < 0.005), and the cells on Ta
patterns were larger than the cells on DLC or Cr patterns (p <
0.05). Further, the cells growing on Ti-patterned samples were
also significantly larger on the Si background than on
backgrounds containing patterns fabricated from other coating
materials (p < 0.01). This was somewhat surprising since the
background material and the preparation techniques used in
sample fabrication were the same for all samples.
Table 10: The size of SaOS-2 cells on micropatterned samples. The SaOS-2 cells were
significantly larger on background silicon compared to the cells growing on patterns.
In addition, the cells were significantly larger on the Si background of Ti-patterned
samples than on the backgrounds of other materials.
Material Cell size on Si background (µm2)
Cell size on patterns (µm2)
DLC 1282 ± 60 962 ± 45
Cr 1404 ± 102 954 ± 25
Ta 1322 ± 86 1151 ± 65#
Ti 1786 ± 77* 1059 ± 20**
* p < 0.01, as compared to all other backgrounds
# p < 0.05, as compared to DLC and Cr
** p < 0.005, as compared to Cr
The values are mean ± SD. One-way ANOVA was used to determine the statistical
differences.
125
9 Discussion
9.1 MICROELECTRODE ARRAYS FOR INTRACRANIAL
RECORDINGS
In studies I, II and IV, the design, fabrication and experimental
use of novel flexible polymer-based microelectrode arrays was
demonstrated. Successful acute and chronic intracranial multi-
channel recordings performed on the surface of the rat cortex
clearly revealed the potential of these arrays. In particular, the
PI-Pt-PI sandwich type was proved to be capable of acquiring
stable signals with a high signal-to-noise ratio in evoked
potential recordings. The PI-based arrays demonstrated
excellent flexibility and mechanical strength during handling
and implantation onto the surface of rat cortex. The flexibility
also allowed the array to follow the movements of the rat brains
by keeping the recording points in close contact with the target
area, thus promoting reliable signal acquisition. Although SU-8
epoxy-resist has been recently utilized as an insulating material
in implantable neural electrodes (Hollenberg et al. 2006, Yeager
et al. 2008), implantation (bending) of SU-8 insulated array
developed here caused several cracks in the SU-8 layer.
Sputtering and evaporation are traditional techniques used in
producing thin film microelectrodes. The results obtained from
EIS with our magnetron sputtered Pt and Au microelectrodes
were consistent with the previously reported values (Cheung et
al. 2007, Metz et al. 2004, Stieglitz et al. 2000, Takahashi et al.
2003). The Pt electrodes displayed significantly lower electrode-
electrolyte impedance magnitudes compared to the Au
electrodes, particularly at biologically relevant frequencies, i.e.
below 1 kHz. With the help of AFM imaging, the surface of
sputtered Pt or Au was observed to be extremely smooth (Ra < 1
nm) i.e. its effective surface area was nearly the same as its
geometrical dimensions. The impedance levels for the Pt
9 – Discussion
126
electrodes used in the final array prototypes (particularly for
electrodes with diameter of 200 µm) were adequate, but if one
needs to develop smaller-sized neural electrodes for high-
resolution application, it would be advantageous to increase the
effective surface area of microelectrodes.
The fabrication process developed here enables the
fabrication of miniaturized arrays with high channel densities,
which have benefits over microwire bundles in terms of precise
control of the electrode sizes and separations between
electrodes. The lithography-based process also allows
implementation of several other neural interfaces with different
layouts, material selections or target areas either for recording or
stimulation purposes. Furthermore, these reproducible devices
can be mass produced in batches at low cost. As concluded in
paper IV, the technology developed in this thesis work also
offers several attractive applications for clinical use. The high
density arrays could possibly be utilized to achieve more
accurate mapping the brain e.g. in patients suffering from
epilepsy, Parkinson’s disease or brain tumors, enabling more
precise planning of the neurosurgery. The flexibility also eases
the insertion of the array without the need for removal of large
skull areas, in contrast to the commercially available, somewhat
bulky, designs of subdural strip and grid electrodes. In addition,
it is obvious that prosthetic systems used in electrical
stimulation of neural tissue in order to treat neuropathic pain,
replace lost sensory or motor function will also benefit from
further development of this kind of technology.
As a result of study IV, a comprehensive summary of
miniaturized flexible microelectrode arrays was obtained. Some
of these MEAs have been developed from the point of view of
human applications (Kitzmiller et al. 2006, 2007, Molina-Luna et
al. 2007, Takahashi et al. 2003, Hollenberg et al. 2006). The
technique of small-field cortical surface recordings, introduced
by Kitzmiller et al. (2006) was based on lithographically
patterned Pt electrodes on the PDMS sheet. Although being
highly miniaturized compared to present clinical electrodes, this
design did not take full advantage of all that microfabrication
9 - Discussion
127
can offer but used fragile wire bonding to create the electrical
wiring. It was realized that the transmission lines could be
formed in the same deposition step as the electrodes. Thin film
transmission lines firmly attached to the substrate surface were
demonstrated to represent a robust way to establish the signal
acquisition. By using somewhat similar technology as applied in
the present thesis, Molina-Luna et al. (2007) introduced
epidurally implantable thin-film MEA for cortical stimulation
mapping. Their approach was based on 72 round TiN contacts
(diameters 100 µm) deposited on the surface of PI foil which
was utilized for mapping the motor cortex of the rat. In
addition, a Kapton-based MEA consisting of 8 x 8 Au electrodes
has been developed for epidural recording of surface field
potential from the barrel cortex of rats (Hollenberg et al. 2006).
However, the array developed by Hollenberg et al. is thicker (75
µm) and thus stiffer than the system developed here and it is
also unsuitable for chronic recordings due to the bulky circuit-
board that the rat would have to carry on its head. A flexible PI-
based MEA with a bending structure introduced by Takahashi
et al. (2003) was successfully tested in epidural mapping of AEP
recordings. However, that system was not designed for chronic
implantation. On the contrary, the connector board with the two
miniconnectors used in our arrays was shown to be small and
light enough to allow chronic recordings.
The acute EP recordings obtained with the present MEAs
were comparable in their signal-to-noise ratios with previously
reported recordings with wire electrodes made with platinum
(Hayton et al. 1999), tungsten (Ureshi et al. 2004) or silver
(Kalliomäki et al. 1998, Jones & Barth 1999). They were also
comparable to previous recordings obtained with MEAs (Stett et
al. 2003, Hollenberg et al. 2006, Takahashi et al. 2003). The
quality of the recorded signal at the beginning of chronic EP
experiments was similar to those obtained in acute experiments
for about two weeks. Subsequently, the signal amplitude
declined due to the increase in electrode impedance resulting,
most probably from the thickening of the dura and the growth
of electrically insulating fibrous tissue around the electrodes.
9 – Discussion
128
9.2 MATERIAL SCIENCE STRATEGIES TO IMPROVE THE
PERFORMANCE OF MICROELECTRODE ARRAYS
Due to the limited long-term performance of the developed
arrays, the confirmation about biocompatibility of materials
used in those arrays was sought in study III. Moreover, different
material science strategies were investigated in attempts to
devise a solution to making the functional life span as long as
possible (unpublished findings).
Unfortunately, when implanted in neural tissue, all artificial
materials have an inconvenient tendency to induce the
formation of scar tissue, which tends to encapsulate the
electrode surface, increases electrode impedances and lowers
signal amplitudes. In present arrays, the greatest doubt with this
vigorous scar tissue formation and reduced biocompatibility
was addressed to PSPI (PI-2771, HD Microsystems) used as an
insulation material. However, the results from study III,
performed according to the ISO 10993-5 standard, revealed that
PSPI is almost as non-cytotoxic as conventional PI and PE (used
as a negative control material). The PSPI films did not adversely
affect the BHK-21 fibroblast cell function, and hence it seems to
be an appropriate material for incorporation into implantable
biomedical devices. However, studies were performed only
with one cell line and with a moderately short culture period.
Therefore, further in vitro and in vivo studies will be needed to
clarify its long-term effects.
Platinum electroplating has been shown to represent an
effective method to produce sponge-like Pt surfaces with much
higher roughness and increased effective surface areas (de Haro,
et al. 2002). Although this technique is a powerful way to
decrease electrode impedance (Gonzales & Rodriguez 1997), and
to prolong the effective life-span of microelectrode (Chen et al.
2009), it has also some serious disadvantages. Electrodeposited
Pt adheres weakly to the electrode surface, resulting in unstable
electrode impedance. Further, electrolytes commonly used in Pt
electroplating contain lead, which is known to be a highly toxic
heavy metal evoking a degree of uncertainty about the
9 - Discussion
129
biocompatibility of electrodeposited Pt surfaces (Schuettler et al.
2005). In this study, the USPLD technique (ColdabTM ) was used
to deposit Pt thin films with a nanotextured surface. Compared
to the sputter-coated smooth Pt surface, a significant reduction
in electrode-electrolyte impedances was achieved. This may be
attributable to the larger effective surface area. The USPLD
technique also conferred several other advantages in electrode
fabrication. Firstly, the USPLD films displayed high adhesion to
different substrates such as PI, SS and Si. Secondly, the samples
containing heat-sensitive materials, such as photoresists, could
be kept normally very close to room temperature, thus avoiding
hardening or flow during deposition. This meant that it was
possible to achieve both high-quality films and easy lift-off.
Thirdly, it was possible to vary deposition parameters easily to
obtain Pt films ranging from smooth (Ra < 1 nm) to relatively
rough (Ra = 100 nm) meaning that optimal results could be
obtained for each purpose. Fourthly, the biocompatibility of Pt
films is guaranteed because high purity solid source materials
are directly transferred from the target to the sample. It is
desirable that the electrode surface should also support
attachment and growth of neurons. It has been reported that
thin Pt films are non-toxic and can support the attachment and
the growth of cortical rat neurons (Thanawala et al. 2007). It has
been also shown that electrode surface morphology has a major
impact on neurocompatibility (Turner et al. 2000). Enhanced
functions of neurons have been demonstrated on biomaterials
structured at the nanoscale (Bayliss et al. 1999, Raffa et al. 2007,
Kelly et al. 2008). Furthermore, it has been reported that
astrocyte adhesion declines whereas the extension of neurons
increases on nanoporous Si surfaces (Moxon et al. 2004). There is
decreased adhesion as well as reduced proliferation and poorer
functional activation of astrocytes on carbon nanofiber materials
(diameter 60 nm) compared to conventional carbon fibres
(diameters 125-200 nm) (McKenzie et al. 2004). These results
provide clear evidence that materials structured at the nanoscale
may have promising neural applications, because they increase
the effective surface area and interact positively with neurons,
9 – Discussion
130
while they simultaneously reduce the functions of astrocytes,
leading to decreased glial scar formation (Kotov et al. 2009).
The vast majority of the tissue contact in the neural interfaces
occurs under the substrate/encapsulation material. Thus, this
material has to fulfill high demands in order to allow the stable,
long-term function of the implant. An optimal insulating
material should be biocompatible, biostable and still possess
appropriate dielectric properties. Although PI was successfully
used in the present arrays as well as in several other neural
interfaces developed by other research groups (Cheung et al.
2007, Stieglitz et al. 2000, Takahashi et al. 2003, Rousche et al.
2001), some degree of uncertainty is related to the relatively
large water uptake, which is typically in the range of 0.5 – 4%
(Stieglitz et al. 2000). In this context, one important advantage
of the USPLD method compared to several other methods is that
it is possible to create dielectric films that do not contain
microparticles or droplets, which would cause pinholes in the
films. This is due to the fact that the laser ablation pulses are so
short that the energy penetrates into only a very shallow (tens of
nanometers) surface layer, which is effectively converted to
plasma and the target surface remains smooth without any
flows or deteriorations. These coatings do not seem to contain
any microdefects which could lead to short circuiting, especially
when they come into contact with body fluids. Alumina and
carbon nitride coatings deposited by the USPLD technique were
demonstrated to be cell-friendly materials at least to fibroblastic
(BHK-21) cells. The ultrasmoothness of these depositions might
provide an additional benefit by avoiding the activation of glial
cells and the formation of scar tissue.
9.3 MICROPATTERNED BIOMATERIAL SURFACES
In studies V and VI, photolithographic micropatterning of
biomaterial surfaces was used for the production of model
surfaces with accurately defined features. These kinds of model
surfaces were demonstrated to be powerful tools for
9 - Discussion
131
investigating the effects of different surface on the adhesion,
spreading and morphology of cells and cellular communities.
Different pattern sizes (5-125 µm), shapes (rectangular or
circular) and surface chemistries (Ta, Ti, Cr, DLC) were utilized
in the creation of engineered surfaces on silicon wafers. The
optimized photolithography process with correct baking,
exposure and development parameters allowed the formation of
precisely defined photoresist features with an undercut edge
profile. The shape of profile was observed to be extremely
important for straightforward lift-off after deposition. The
surface roughness, a parameter widely known to modulate
protein adsorption and cellular responses, could be kept almost
constant with all coatings. The AFM characterization revealed
that variation in roughness between sample types was below 2
nm and the only relevant topographical cues of the samples
were the steps between the patterns and background. By precise
adjustment of the deposition time with two different types of
PVD methods, i.e. magnetron sputtering for metals and filtered
pulsed plasma arc discharge for DLC, it was possible to ensure
that the coating thickness variations were kept as low as
possible (200 ± 20 nm). This indicates that the possible clear-cut
differences observed between different materials and patterns
could not have been due to differences in surface topography
and instead suggests that the differences seen and discussed
below originated from the different chemical compositions and
physical patterns of the materials being studied.
Silicon is the most widely used substrate material in
microchips and MEMS devices. However, biocompatibility
properties of Si are often unsuitable for use in bio-MEMS
devices. In studies V and VI, the cytocompatibility enhancement
of Si was achieved using its partial coating with more
biocompatible materials such as Ti or DLC. This suggests that
even partial coating of Si-based bio-MEMS devices can be used
to enhance their cytocompatibility and to facilitate their
integration with the host during the critical initial phase of
integration. The rapid integration of host cells also reduces the
risk of implant-related infections according to the concept of the
9 – Discussion
132
‚race for the surface‛ (Gristina 1987). In cellular experiments
with both SaOS-2 and hMSC cells, a clear cellular preference for
the biomaterial patterns over background Si was observed. Both
cell types initially tended to adopt the geometrical shape of the
patterns, resulting in ‚engineered‛ square or circle shaped
cellular communities. The results from SaOS-2 cell experiments
with different pattern sizes disclosed that the size and spacing of
the patterns provide further options to guide the cells. The size
of the SaOS-2 cells growing on the same Si background varied
significantly, depending on the type of biomaterial used for
patterning. This is a very interesting observation when one takes
into account the fact that all surfaces were produced using the
same fabrication methods on the same background Si. Another
interesting observation was made in the focal adhesion staining
experiments. Preferential localization of vinculin containing
focal adhesions in SaOS-2 cells was clearly observed on the
biomaterial patterns, but hMSCs did not show any such
preference for the patterns. One could speculate that the
patterns used in the hMSC experiments (75 µm squares) were
too small to allow proper organization of the actin cytoskeleton
and subsequent preferential forming of vinculin-containing
adhesion plaques. Moreover, it has been demonstrated that the
pattern size has an effect on the differentiation of hMSCs
(McBeath et al. 2004) and in that case, relatively large islands are
needed to ensure osteogenesis.
The patterning techniques developed in this thesis provide
an effective way to enhance the biocompatibility of Si-based
microdevices. To allow proper functionality of these bio-MEMS
devices, only the areas which are not under the integrated
electronic components may be coated with biocompatible thin
films. These patterning techniques also provide an effective
means to guide the attachment, size and the shape of the cells as
well as to regulate the differentiation of MSCs into the desired
phenotypes. It is hoped that these aspects will stimulate the
design of even more advanced biomaterials and constructs for
tissue engineering and regenerative medicine.
133
10 Conclusions
In this work, novel flexible epidural microelectrode arrays were
successfully developed and tested to permit recordings of acute
and chronic evoked potentials in an animal model. In addition,
novel materials and surface textures at a micro- and nanoscales
were investigated in vitro aiming to improve the biological and
electrochemical characteristics of neural interfaces. The
knowledge about cell-biomaterial interactions obtained in this
thesis work can be used in the further development of neural
interfaces as well as in other implantable biomedical devices.
On the basis of the present series of studies, the following
detailed conclusions can be drawn:
1. The developed flexible microelectrode arrays possess
adequate mechanical and electrochemical characteristics and
they are capable of acquiring signals with high signal-to-
noise ratios in evoked potential recordings when implanted
onto the surface of the rat cortex.
2. The polyimide-platinum-polyimide sandwich structure, in
which the flexibility of PI and the superior electrochemical
properties of Pt were combined, was observed to be the most
suitable realization of the array. A novel finding was the
confirmation of the biocompatibility of the photosensitive
polyimide used in the developed arrays.
3. Nanostructures on an electrode surface can be used to
increase the effective surface area and to lower electrode
impedance as observed with nanotextured Pt surfaces
created with an ultra-short pulsed laser deposition technique.
4. Thin film dielectrics, alumina and carbon nitride produced
by the USPLD method were proved to be highly cyto-
compatible in in vitro studies with fibroblasts. However,
10 – Conclusions
134
further in vitro and in vivo studies will be needed to clarify
their long-term effects.
5. Micropatterning with biomaterials can be used to enhance
the cytocompatibility of silicon-based bio-MEMS devices.
Not only the chemical composition of the materials, but also
the shape, edges (height) and size of the features used for
surface patterning have a remarkable effect on cell guidance.
6. The morphology of cells cultured on micropatterned
surfaces seemed to be aligned along the edges of the features
e.g. forming round- or square-shaped cell morphologies
having an interesting effect of the differentiation of stem
cells into certain phenotypes. These results may be
particularly useful in applications involving regenerative
medicine and tissue engineering where cells are grown on
biomaterial scaffolds.
135
11 References
Abramoff, M.D., Magelhaes, P.J. & Ram S.J. (2004) Image processing
with ImageJ. Biophotonics Int., 11, 36-42.
Ainslie, K.M., Tao, S.L., Popat, K.C. & Desai, T.A. (2008) In vitro
immunogenicity of silicon-based micro- and nanostructured
surfaces. ACS Nano, 2, 1076-84.
Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K. & Watson, J.D.
(1994) Molecular biology of the cell, Garland Publishing, New York,
NY, USA.
Altankov, G., Grinnell, F. & Groth, T. (1996) Studies on the
biocompatibility of materials: fibroblast reorganization of
substratum-bound fibronectin on surfaces varying in wettability.
J. Biomed. Mater. Res., 30, 385-91.
Altankov, G., Richau, K. & Groth, T. (2003) The role of surface zeta
potential and substratum chemistry for regulation of dermal
fibroblasts interaction. Mat.-wiss. u. Werkstofftec, 34, 1120-8.
Altuna, A., Gabriel G., Memendez de la Prida, L., Tijero, M., Guimera,
A., Berganzo, J., Salido, R., Villa, R. & Ferdandez, L.F. (2010) SU-
8-based microneedles for in vitro neural applications. J.
Micromech. Microeng., 20, 064014.
Andersson, A.-S., Bäckhed, F., von Euler, A., Richter-Dahlfors, A.,
Sutherland, D. & Kasemo, B. (2003) Nanoscale features influence
epithelial cell morphology and cytokine production. Biomaterials,
24, 3427-36.
Amberla, T., Rekow, M., Köngäs, J., Asonen, H., Salminen, T., Viitanen,
N., Kulmala, M., Vuoristo, P., Pessa, M., Lappalainen, R. (2006)
Pulsed laser deposition with a high average power fiber laser. In: SVC -
49th Annual Technical Conference Proceedings, Washington, D.C.,
USA, April 22-27, 2006, pp. 79-82.
Anselme, K. & Bigerelle, M. (2005) Topography effects of pure
titanium substrates on human osteoblast long-term adhesion.
Acta Biomater., 1, 211-22.
11 – References
136
Ardesch, J.J., Buschman, H.P.J., Wagener-Schimmel, L.J.J.C., van der
Aa, H.E. & Hageman, G. (2007) Vagus nerve stimulation for
medically refractory epilepsy: a long-term follow-up study.
Seizure, 16, 579-85.
Badi, A.N., Kertesz, T.R., Gurgel, R.K., Shelton, C. & Normann, R.A.
(2003) Development of a novel eighth-nerve intraneural auditory
neuroprosthesis. Laryngoscope, 113, 833-42.
Bashir, R. (2004) BioMEMS: state-of-the-art in detection, opportunities
and prospects. Adv. Drug Deliv. Rev., 56, 1565-86.
Bayliss, S.C., Buckberry, L.D., Fletcher, I. & Tobin, M.J. (1999) The
culture of neurons on silicon. Sens. Act., 74, 139-42.
Ber, S., Torun Köse, G. & Hasirci, V. (2005) Bone tissue engineering on
patterned collagen films: an in vitro study. Biomaterials, 26, 1977-
86.
Bernard, A., Delamarche, E., Schmid, H., Michel, B., Bosshard, H.R. &
Biebuyck, H. (1998) Printing Patterns of Proteins. Langmuir, 14,
2225-9.
Betancourt, T. & Brannon-Peppas, L. (2006) Micro- and nanofabrication
methods in nanotechnological medical and pharmaceutical
devices. Int. J. Nanomed., 1, 483-95.
Blau, A., Ziegler, C., Heyer, M., Endres, F., Schwitzgebel, G., Matthies,
T., Stieglitz, T., Meyer, J.-U. & Gopel, W. (1997) Characterization
and optimization of microelectrode arrays for in vivo nerve signal
recording and stimulation. Biosens. Bioelectron., 12, 883-92.
Blawas, A.S. & Reichert, W.M. (1998) Protein patterning. Biomaterials,
19, 595-609.
Bodhak, S., Bose, S. & Bandyopadhyay, A. (2009) Role of surface
charge and wettability on early stage mineralization and bone
cell-materials interactions of polarized hydroxyapatite. Acta
Biomater., 5, 2178-88.
Bodhak, S., Bose, S. & Bandyopadhyay, A. (2010) Electrically polarized
HAp-coated Ti: in vitro bone cell-material interactions. Acta
Biomater., 6, 641-51.
Bogner, E., Dominizi, K., Hagl, P., Bertagnolli, E., Wirth, M., Gabor, F.,
Brezna, W. & Wanzenboeck, H.D. (2006) Bridging the gap—
Biocompatibility of microelectronic materials. Acta Biomater., 2,
229-37.
11 - References
137
Boon, P., Vonck, K., de Reuck, J. & Caemaert, J. (2001) Vagus nerve
stimulation for refractory epilepsy. Seizure, 10, 448-55.
Boppart, S.A., Wheeler, B.C. & Wallace, C.S. (1992) A flexible
perforated microelectrode array for extended neural recordings.
IEEE Trans. Biomed. Eng., 39, 37-42.
Branch, D.W., Corey, J.M., Weyhenmeyer, J.A., Brewer, G.J. & Wheeler,
B.C. (1998) Microstamp patterns of biomolecules for high-
resolution neuronal networks. Med. Biol. Eng. Comput., 36, 135-41.
Britland, S., Morgan, H., Wojiak-Stodart, B., Riehle, M., Curtis, A. &
Wilkinson, C. (1996) Synergistic and hierarchical adhesive and
topographic guidance of BHK cells. Exp. Cell Res., 228, 313-25.
Brunette, D.M., Kenner, G.S. & Gould, T.R.L. (1983) Grooved titanium
surfaces orient growth and migration of cells from human
gingival explants. J. Dent. Res., 62, 1045-8.
Burton, A.W., Phan, P.C. (2008) Spinal cord stimulation for pain
management. In: DiLorenzo, D.L. & Bronzino, J.D. (eds.)
Neuroengineering, CRC Press, Boca Raton, FL, USA, chapter 7, pp.
7.1-7.16.
Cai, K., Frant, M., Bossert, J., Hildebrand, G., Liefeith, K. & Jandt, K.D.
(2006) Surface functionalized titanium thin films: zeta-potential,
protein adsorption and cell proliferation. Colloids Surf., B, 50, 1-8.
Calzado-Martín, A., Saldaña, L., Korhonen, H., Soininen, A., Kinnari,
T.J., Gómez-Barrena, E., Tiainen, V.-M., Lappalainen, R.,
Munuera, L., Konttinen, Y.T. & Vilaboa, N. (2010) Interactions of
human bone cells with diamond-like carbon polymer hybrid
coatings. Acta Biomater., 6, 3325-38.
Campbell, P.K., Jones, K.E., Huber, R.J., Horch, K.W. & Normann, R.A.
(1991) A silicon-based, three-dimensional neural interface:
manufacturing processes for an intracortical electrode array. IEEE
Trans. Biomed. Eng., 38, 758-68.
Charest, J.L., Bryant, L.E., Garcia, A.J. & King, W.P. (2004) Hot
embossing for micropatterned cell substrates. Biomaterials, 25,
4767-75.
Chen, C.S., Mrksich, M., Huang, S., Whitesides, G.M. & Ingber, D.E.
(1997) Geometric control of cell life and death. Science, 276, 1425-8.
Chen, Y.-Y., Lai, H.-Y., Lin, S.-H., Cho, C.-W., Chao, W.-H., Liao, C.-H.,
Tsang, S., Chen, Y.-F. & Lin, S.-Y. (2009) Design and fabrication of
11 – References
138
a polyimide-based microelectrode array: Application in neural
recording and repeatable electrolytic lesion in rat brain. J.
Neurosci. Meth., 182, 6-16.
Cheney, M., Isaacson, D. & Newell, J.C. (1999) SIAM Rev., 41, 85-101.
Cheng, K., Weng, W., Wang, H. & Zhang, S. (2005) In vitro behavior of
osteoblast-like cells on fluoridated hydroxyapatite coatings.
Biomaterials, 26, 6288-95.
Cheung, K.C. (2002) Microscale and nanoscale neural interfaces, Doctoral
dissertation, University of Califormia, Berkeley, CA, USA.
Cheung, K.C. (2007) Implantable microscale neural interfaces. Biomed.
Microdev., 9, 923-38.
Cheung, K.C. & Renaud, P. (2006) BioMEMS for medicine: On-chip cell
characterization and implantable microelectrodes. Solid-State
Electron., 50, 551-7.
Cheung, K.C., Renaud, P., Tanila, H. & Djupsund, K. (2007) Flexible
polyimide microelectrode array for in vivo recordings and
current source density analysis. Biosens. Bioelectron., 22, 1783-90.
Chiappa, K.H. (ed.) (1997) Evoked potentials in clinical medicine, 3rd ed.,
Lippincott-Raven Publishers, Philadelphia, PA, USA.
Chou, L., Firth, J.D., Uitto, V.-J. & Brunette, D.M. (1995) Substratum
surface topography alters cell shape and regulates fibronectin
mRNA level, mRNA stability, secretion and assembly in human
fibroblasts. J. Cell. Sci., 108, 1563-73.
Chou, L., Firth, J.D., Uitto, V.-J. & Brunette, D.M. (1998) Effects of
titanium substratum and grooved surface topography on
metalloproteinase-2 expression in human fibroblasts. J. Biomed.
Mater. Res., 39, 437-45.
Chu, P.K., Chen, J.Y., Wang, L.P. & Huang N. (2002) Plasma-surface
modification of biomaterials. Mat. Sci. Eng. R., 36, 143-206.
Chua, P.-H., Neoh, K.-G., Kang, E.-T. & Wang, W. (2008) Surface
functionalization of titanium with hyaluronic acid/chitosan
polyelectrolyte multilayers and RGD for promoting osteoblast
functions and inhibiting bacterial adhesion. Biomaterials, 29, 1412-
21.
11 - References
139
Clark, P., Connolly, P., Curtis, A.S.G., Dow, J.A.T. & Wilkinson,
C.D.W. (1987) Topographical control of cell behaviour. I. Simple
step cues. Development, 99, 439-48.
Cogan, S.F. (2008) Neural Stimulation and Recording Electrodes. Annu.
Rev. Biomed. Eng., 10, 275-309.
Cui, F.Z. & Li, D.J. (2000) A review of investigations on
biocompatibility of diamond-like carbon and carbon nitride films.
Surf.Coat.Technol., 131, 481-7.
Cui, F.Z., Qing, X.L., Li, D.J. & Zhao, J. (2005) Biomedical
investigations on CNx coating. Surf.Coat.Technol., 200, 1009-13.
Curio, G. (2005) Ultrafast EEG activities. In: Niedermayer, E. & Lopez
da Silva, F. (eds.) Electroencephalography: basic principles, clinical
applications and related fields, 5th ed., Lippincott Williams &
Wilkins, Philadelphia, PA, USA.
Curran, J.M., Chen, R. & Hunt, J.A. (2005) Controlling the phenotype
and function of mesenchymal stem cells in vitro by adhesion to
silane-modified clean glass surfaces. Biomaterials, 26, 7057-67.
Curran, J.M., Chen, R. & Hunt, J.A. (2006) The guidance of human
mesenchymal stem cell differentiation in vitro by controlled
modifications to the cell substrate. Biomaterials, 27, 4783-93.
Curtis, A. & Wilkinson, C. (1997) Topographical control of cells.
Biomaterials, 18, 1573-83.
Dalby, M.J., McCloy, D., Robertson, M., Agheli, H., Sutherland, D.,
Affrossman, S. & Oreffo, R.O.C. (2006) Osteoprogenitor response
to semi-ordered and random nanotopographies. Biomaterials, 27,
2980-7.
Dalby, M.J., Riehle, M.O., Sutherland, D.S., Agheli, H. & Curtis, A.S.G.
(2004) Use of nanotopography to study mechanotransduction in
fibroblasts - methods and perspectives. Eur. J. Cell Biol., 83, 159-69.
Das, K., Bose, S. & Bandyopadhyay, A. (2009) TiO2 nanotubes on Ti:
Influence of nanoscale morphology on bone cell-materials
interaction. J. Biomed. Mater. Res. 90A, 225-37.
Das, M.K., Dandamudi, G. & Steiner, H.A. (2009) Modern Pacemakers:
Hope or Hype? PACE, 32, 1207-21.
de Haro, C., Mas, R., Abadal, G., Muñoz, J., Perez-Murano, F. &
Domínguez, C. (2002) Electrochemical platinum coatings for
11 – References
140
improving performance of implantable microelectrode arrays.
Biomaterials, 23, 4515-21.
den Braber, E.T., de Ruijter, J.E., Smits, H.T.J., Ginsel, L.A., von Recum,
A.F. & Jansen, J.A. (1996) Quantitative analysis of cell
proliferation and orientation on substrata with uniform parallel
surface micro-grooves. Biomaterials, 17, 1093-9.
Deng, T., Tien, J., Xu, B. & Whitesides, G.M. (1999) Using patterns in
microfiche as photomasks in 10-µm-scale microfabrication.
Langmuir, 15, 6575-81.
Deng, T., Wu, H., Brittain, S.T. & Whitesides, G.M. (2000) Prototyping
of masks, masters, and stamps/molds for soft lithography using
an office printer and photographic reduction. Anal. Chem., 72,
3176-80.
DiLorenzo, D.L. & Bronzino, J.D. (eds.) (2008) Neuroengineering, CRC
Press, Boca Raton, FL, USA.
Du, C., Su, X.W., Cui, F.Z. & Zhu, X.D. (1998) Morphological
behaviour of osteoblasts on diamond-like carbon coating and
amorphous C-N film in organ culture. Biomaterials, 19, 651-8.
Eddington, D.K. (2008) Cochlear prostheses: an introduction. In:
DiLorenzo, D.L. & Bronzino, J.D. (eds.) Neuroengineering, CRC
Press, Boca Raton, FL, USA, chapter 10, pp. 10.1-10.10.
Edell, D.J., Toi, V.V., McNeil, V.M. & Clark, L.D. (1992) Factors
influencing the biocompatibility of insertable silicon microshafts
in cerebral cortex. IEEE Trans. Biomed. Eng., 39, 635-43.
El-Ghannam, A., Hamazawy, E. & Yehia, A. (2001) Effect of thermal
treatment on bioactive glass microstructure, corrosion behavior,
zeta potential, and protein adsorption. J. Biomed. Mater. Res., 55,
387-95.
Falconnet, D., Csucs, G., Grandin, H.M. & Textor, M. (2006) Surface
engineering approaches to micropattern surfaces for cell-based
assays. Biomaterials, 27, 3044-63.
Federspil, P.A. & Plinkert, P.K. (2004) Restoring hearing with active
hearing implants. Biomed. Tech. (Berl.), 49, 78-82.
Feng, B., Weng, J., Yang, B.C., Qu, S.X. & Zhang, X.D. (2003)
Characterization of surface oxide films on titanium and adhesion
of osteoblast. Biomaterials, 24, 4663-70.
11 - References
141
Finch, D.S., Oreskovic, T., Ramadurai, K., Herrmann, C.F., George,
S.M. & Mahajan, R.L. (2008) Biocompatibility of atomic layer-
deposited alumina thin films. J. Biomed. Mater. Res. 87A, 100-6.
Flemming, R.G., Murphy, C.J., Abrams, G.A., Goodman, S.L. &
Nealey, P.F. (1999) Effects of synthetic micro- and nano-
structured surfaces on cell behavior. Biomaterials, 20, 573-88.
Folch, A., Ayon, A., Hurtado, O., Schmidt, M.A. & Toner, M. (1999)
Molding of deep polydimethylsiloxane microstructures for
microfluidics and biological applications. J. Biomech. Eng., 121, 28-
34.
Folch, A. & Toner, M. (1998) Cellular micropatterns on biocompatible
materials. Biotechnol. Prog., 14, 388-92.
Folch, A., Jo, B.-H., Hurtado, O., Beebe, D.J. & Toner, M. (2000)
Microfabricated elastomeric stencils for micropatterning cell
cultures. J. Biomed. Mater. Res., 52, 346-53.
Folch, A. & Toner, M. (2000) Microengineering of cellular interactions.
Annu. Rev. Biomed. Eng., 2, 227-56.
Franssila, S. (2004) Introduction to microfabrication, John Wiley & Sons
Ltd., Chichester, West Sussex, England.
Gale, B.K., Eddings, M.A., Sundberg, S.O., Hatch, A., Kim, J. & Ho, T.
(2008) Low-cost MEMS technologies. In: Zappe, H. (ed.)
Comprehensive Microsystems, Elsevier, Oxford, UK, volume 1, pp.
341-378.
Gallagher, J.O., McGhee, K.F., Wilkinson, C.D.W. & Riehle, M.O. (2002)
Interaction of animal cells with ordered nanotopography. IEEE
Trans. Nanobiosci., 1, 24-8.
Geddes, L.A. (1997) Historical evolution of circuit models for the
electrode-electrolyte interface. Ann. Biomed. Eng., 25, 1-14.
Geddes, L.A. & Roeder, R. (2003) Criteria for the selection of materials
for implanted electrodes. Ann. Biomed. Eng., 31, 879-90.
Geddes, L.A. & Roeder, R. (2001) Measurement of the direct-current
(faradic) resistance of the electrode-electrolyte interface for
commonly used electrode materials. Ann. Biomed. Eng., 29, 181-6.
Ghosh, M. K. & Mittal K. L. (ed.) (1996) Polyimides: fundamentals and
applications, Marcel Dekker, New York, NY, USA.
11 – References
142
Giessibl, F.J. (2003) Advances in atomic force microscopy. Rev. Mod.
Phys., 75, 949-83.
González, C. & Rodríguez, M. (1997) A flexible perforated
microelectrode array probe for action potential recording in nerve
and muscle tissues. J. Neurosci. Meth., 72, 189-95.
Grayson, A.C.R., Shawgo, R.S., Johnson, A.M., Flynn, N.T., Li, Y.,
Cima, M.J. & Langer, R. (2004) A BioMEMS review: MEMS
technology for physiologically integrated devices. Proc. IEEE, 92,
6-21.
Greenberg, R.J. (2008) Visual prostheses. In: DiLorenzo, D.L. & Bronzino,
J.D. (eds.) Neuroengineering, CRC Press, Boca Raton, FL, USA,
chapter 11, pp. 11.1-11.6.
Grill, W.M. (2008) Theory and physiology of electrical stimulation of the
central nervous system. In: DiLorenzo, D.L. & Bronzino, J.D. (eds.)
Neuroengineering, CRC Press, Boca Raton, FL, USA, chapter 20, pp.
20.1-20.15.
Gristina, A.G. (1987) Biomaterial-centered infection: microbial
adhesion versus tissue integration. Science, 237, 1588-95.
Guber, A.E., Heckele, M., Herrmann, D., Muslija, A., Saile, V.,
Eichhorn, L., Gietzelt, T., Hoffmann, W., Hauser, P.C.,
Tanyanyiwa, J., Gerlach, A., Gottschlich, N. & Knebel, G. (2004)
Microfluidic lab-on-a-chip systems based on polymers—
fabrication and application. Chem. Eng. J., 101, 447-53.
HajjHassan, M., Chodavarapu, V. & Musallam, S. (2008) NeuroMEMS:
neural probe microtechnologies. Sensors, 8, 6704-26.
Hallab, N.J., Bundy, K.J., O'Connor, K., Moses, R.L. & Jacobs, J.J. (2001)
Evaluation of metallic and polymeric biomaterial surface energy
and surface roughness characteristics for directed cell adhesion.
Tissue Eng., 7, 55-71.
Hamdan, M., Blanco, L., Khraisat, A. & Tresguerres, I.F. (2006)
Influence of titanium surface charge on fibroblast adhesion. Clin.
Implant Dent. Relat. Res., 8, 32-8.
Hamilton, D.W. & Brunette, D.M. (2005) "Gap guidance" of fibroblasts
and epithelial cells by discontinuous edged surfaces. Exp.Cell Res.,
309, 429-37.
Hamilton, D.W., Wong, K.S. & Brunette, D.M. (2006) Microfabricated
discontinuous-edge surface topographies influence osteoblast
11 - References
143
adhesion, migration, cytoskeletal organization, and proliferation
and enhance matrix and mineral deposition in vitro. Calcif. Tissue
Int., 78, 314-25.
Hao, L. & Lawrence, J. (2007) Wettability modification and the
subsequent manipulation of protein adsorption on a Ti6Al4V
alloy by means of CO2 laser surface treatment. J. Mater. Sci. Mater.
Med., 18, 807-17.
Hasirci, V. & Hasirci, N. (2005) Control of polymeric material surfaces. In:
Vadgama, P. (ed.) Surfaces and interfaces for biomaterials,
Woodhead Publishing Limited, Cambridge, UK, pp. 29-59.
Hayton, S.M., Kriss, A. & Muller, D.P.R. (1999) Comparison of the
effects of four anaesthetic agents on somatosensory evoked
potentials in the rat. Lab. Animal, 33, 243-51.
Heiduschka, P. & Thanos, S. (1998) Implantable bioelectronic interfaces
for lost nerve functions. Prog. Neurobiol., 55, 433-61.
Heimer, L. (1995) The human brain and spinal cord, Springer-Verlag,
New York, NY, USA.
Heimke, G., Griss, P., Jentschra, G. & Warner, E. (1998) Direct
anchorage of Al2O3-ceramic hip components: three years of
clinical experience and results of further animal studies. J. Biomed.
Mater. Res., 41, 519-26.
Hench, L.L. & Polak, J.M. (2002) Third-generation biomedical
materials. Science, 295, 1014-7.
Hersel, U., Dahmen, C. & Kessler, H. (2003) RGD modified polymers:
biomaterials for stimulated cell adhesion and beyond.
Biomaterials, 24, 4385-415.
Hille, B. (1992) Ionic channels of excitable membranes, 2nd ed., Sinauer
Associates, Sunderland, MA, USA.
Hollenberg, B.A., Richards, C.D., Richards, R., Bahr, D.F. & Rector,
D.M. (2006) A MEMS fabricated flexible electrode array for
recording surface field potentials. J. Neurosci. Meth., 153, 147-53.
Hormuzdi, S.G., Filippov, M.A., Mitropoulou, G., Monyer, H. &
Bruzzone, R. (2004) Electrical synapses: a dynamic signaling
system that shapes the activity of neuronal networks. Biochim.
Biophys. Acta, 1662, 113-37.
11 – References
144
Hosp, J.A., Molina-Luna, K., Hertler, B., Atiemo, C.O., Stett, A. & Luft,
A.R. (2008) Thin-film epidural microelectrode arrays for
somatosensory and motor cortex mapping in rat. J. Neurosci.
Meth., 172, 255-62.
Hsu, T.R. (2008) MEMS and microsystems: design, manufacture and
nanoscale engineering, John Wiley & Sons, Hoboken, NJ, USA.
Huang, X.-J., O'Mahony, A.M. & Compton, R.G. (2009) Microelectrode
Arrays for Electrochemistry: Approaches to Fabrication. Small, 5,
776-88.
Humayun, M.S., Weiland, J.D., Fujii, G.Y., Greenberg, R., Williamson,
R., Little, J., Mech, B., Cimmarusti, V., Van Boemel, G., Dagnelie,
G. & de Juan, E. Jr. (2003) Visual perception in a blind subject
with a chronic microelectronic retinal prosthesis. Vision Res., 43,
2573-81.
Ito, Y. (1999) Surface micropatterning to regulate cell functions.
Biomaterials, 20, 2333-42.
Iwaya, Y., Machigashira, M., Kanbara, K., Miyamoto, M., Noguchi, K.,
Izumi, Y. & Ban, S. (2008) Surface properties and biocompatibility
of acid-etched titanium. Dent. Mater. J., 27, 415-21.
James, C.D., Davis, R.C., Kam, L., Craighead, H.G., Isaacson, M.,
Turner, J.N. & Shain, W. (1998) Patterned protein layers on solid
substrates by thin stamp microcontact printing. Langmuir, 14, 741-
4.
Jinno, T., Goldberg, V.M., Davy, D. & Stevenson, S. (1998)
Osseointegration of surface-blasted implants made of titanium
alloy and cobalt-chromium alloy in a rabbit intramedullary
model. J. Biomed. Mater. Res., 42, 20-9.
Jones, M.S. & Barth, D.S. (1999) Spatiotemporal organization of fast
(>200 Hz) electrical oscillations in rat vibrissa/barrel cortex. J.
Neurophysiol., 82, 1599-609.
Judy, J.W. (2001) Microelectromechanical systems (MEMS): fabrication,
design and applications. Smart Mater. Struct., 10, 1115-34.
Kalliomäki, J., Luo, X.-L., Yu, Y.-B. & Schouenborg, J. (1998)
Intrathecally applied morphine inhibits nociceptive C fiber input
to the primary somatosensory cortex (SI) of the rat. Pain, 77, 323-9.
Kandel, E.R., Schwartz, J.H. & Jessell, T.M. (ed.) (2000) Principles of
neural science, McGraw-Hill Companies, New York, NY, USA.
11 - References
145
Kane, R.S., Takayama, S., Ostuni, E., Ingber, D.E. & Whitesides, G.M.
(1999) Patterning proteins and cells using soft lithography.
Biomaterials, 20, 2363-76.
Kanno, M., Kawakami, H., Nagaoka, S. & Kubota, S. (2002)
Biocompatibility of fluorinated polyimide. J. Biomed. Mater. Res.,
60, 53-60.
Kasemo, B. (1998) Biological surface science. Curr. Opin. Solid State
Mater. Sci., 3, 451-9.
Katiyar, P., Jin, C. & Narayan, R.J. (2005) Electrical properties of
amorphous aluminum oxide thin films. Acta Materialia, 53, 2617-
22.
Kelly, S., Regan, E.M., Uney, J.B., Dick, A.D., McGeehan, J.P., The
Bristol Biochip Group, Mayer, E.J. & Claeyssens, F. (2008)
Patterned growth of neuronal cells on modified diamond-like
carbon substrates. Biomaterials, 29, 2573-80.
Kennedy, S.B., Washburn, N.R., Simon, C.G.Jr. & Amis, E.J. (2006)
Combinatorial screen of the effect of surface energy on
fibronectin-mediated osteoblast adhesion, spreading and
proliferation. Biomaterials, 27, 3817-24.
Kern, D. S. & Kumar, R. (2007) Deep brain stimulation. The Neurologist
13: 237-52.
Khang, D., Lu, J., Yao, C., Haberstroh, K.M. & Webster, T.J. (2008) The
role of nanometer and sub-micron surface features on vascular
and bone cell adhesion on titanium. Biomaterials, 29, 970-83.
Khorasani, M.T., Moemenbellah, S., Mirzadeh, H. & Sadatnia, B. (2006)
Effect of surface charge and hydrophobicity of polyurethanes and
silicone rubbers on L929 cells response. Colloids Surf.,B, 51, 112-9.
Kim, E., Xia, Y. & Whitesides, G.M. (1995) Polymer microstructures
formed by moulding in capillaries. Nature, 376, 581-4.
Kim, E., Xia, Y. & Whitesides, G.M. (1996) Micromolding in capillaries:
applications in materials science. J. Am. Chem. Soc., 118, 5722-31.
Kim, H.J., Kim, S.H., Kim, M.S., Lee, E.J., Oh, H.G., Oh, W.M., Park,
S.W., Kim, W.J., Lee, G.J., Choi, N.G., Koh, J.T., Dinh, D.B.,
Hardin, R.R., Johnson, K., Sylvia, V.L., Schmitz, J.P. & Dean, D.D.
(2005) Varying Ti-6Al-4V surface roughness induces different
early morphologic and molecular responses in MG63 osteoblast-
like cells. J.Biomed.Mater.Res. 74A, 366-73.
11 – References
146
Kipke, D.R., Vetter, R.J., Williams, J.C. & Hetke, J.F. (2003) Silicon-
substrate intracortical microelectrode arrays for long-term
recording of neuronal spike activity in cerebral cortex. IEEE
Trans. Neural Syst. Rehabil. Eng., 11, 151-5.
Kitzmiller, J.P., Hansford, D.J., Fortin, L.D., Obrietan, K.H., Bergdall,
V.K. & Beversdorf, D.Q. (2007) Micro-field evoked potentials
recorded from the porcine sub-dural cortical surface utilizing a
microelectrode array. J. Neurosci. Meth., 162, 155-61.
Kitzmiller, J., Beversdorf, D. & Hansford, D. (2006) Fabrication and
testing of microelectrodes for small-field cortical surface
recordings. Biomed. Microdev., 8, 81-5.
Könönen, M., Hormia, M., Kivilahti, J., Hautaniemi, J. & Thesleff, I.
(1992) Effect of surface processing on the attachment, orientation,
and proliferation of human gingival fibroblasts on titanium. J.
Biomed. Mater. Res., 26, 1325-41.
Kotov, N.A., Winter, J.O., Clements, I.P., Jan, E., Timko, B.P.,
Campidelli, S., Pathak, S., Mazzatenta, A., Lieber, C.M., Prato, M.,
Bellamkonda, R.V., Silva, G.A., Kam, N.W.S., Patolsky, F. &
Ballerini, L. (2009) Nanomaterials for neural interfaces. Adv.
Mater., 21, 3970-4004.
Kotzar, G., Freas, M., Abel, P., Fleischman, A., Roy, S., Zorman, C.,
Moran, J.M. & Melzak, J. (2002) Evaluation of MEMS materials of
construction for implantable medical devices. Biomaterials, 23,
2737-50.
Kovacs, G.T.A., Maluf, N.I. & Petersen, K.E. (1998) Bulk
micromachining of silicon. P. IEEE, 86, 1536-51.
Krajewski, A., Malavolti, R. & Piancastelli, A. (1996) Albumin adhesion
on some biological and non-biological glasses and connection
with their Z-potentials. Biomaterials, 17, 53-60.
Krajewski, A., Piancastelli, A. & Malavolti, R. (1998) Albumin adhesion
on ceramics and correlation with their Z-potential. Biomaterials,
19, 637-41.
Kumar, A. & Whitesides, G.M. (1993) Features of gold having
micrometer to centimeter dimensions can be formed through a
combination of stamping with an elastomeric stamp and an
alkanethiol ``ink'' followed by chemical etching. Appl. Phys. Lett.,
63, 2002-4.
11 - References
147
Kumar, K., Toth, C. & Nath, R.K. (1997) Deep brain stimulation for
intractable pain: A 15-year experience. Neurosurgery, 40, 736-47.
Kunzler, T.P., Drobek, T., Schuler, M. & Spencer, N.D. (2007)
Systematic study of osteoblast and fibroblast response to
roughness by means of surface-morphology gradients.
Biomaterials, 28, 2175-82.
Lachaux, J.P., Rudrauf, D. & Kahane, P. (2003) Intracranial EEG and
human brain mapping. J. Physiol. Paris, 97, 613-28.
Laermer, F. & Schilp, A. (2006) Methods for anisotropic etching of silicon,
US Patent #5,501,893.
Lago, N., Yoshida, K., Koch, K.P. & Navarro, X. (2007) Assessment of
biocompatibility of chronically implanted polyimide and
platinum intrafascicular electrodes. IEEE Trans. Biomed. Eng., 54,
281-90.
Lampin, M., Warocquier-Clerout, R., Legris, C., Degrange, M. & Sigot-
Luizard, M.F. (1997) Correlation between substratum roughness
and wettability, cell adhesion, and cell migration. J. Biomed. Mater.
Res., 36, 99-108.
LeBaron, R.G. & Athanasiou, K.A. (2000) Extracellular matrix cell
adhesion peptides: functional applications in orthopedic
materials. Tissue Eng., 6, 85-103.
Lee, K., Singh, A., He, J., Massia, S., Kim, B. & Raupp, G. (2004)
Polyimide based neural implants with stiffness improvement.
Sens. Act. B: Chem., 102, 67-72.
Leuthardt, E.C., Ojemann, J.G., Schalk, G. & Moran, D.W. (2008)
General clinical issues relevant to brain-computer interfaces. In:
DiLorenzo, D.L. & Bronzino, J.D. (eds.) Neuroengineering, CRC
Press, Boca Raton, FL, USA, chapter 12, pp. 12.1-12.16.
Levon, J., Myllymaa, K., Kouri, V., Rautemaa, R., Kinnari, T.,
Myllymaa, S., Konttinen, Y.T. & Lappalainen, R. (2009) Patterned
macroarray plates in comparison of bacterial adhesion inhibition
of tantalum, titanium, and chromium compared with diamond-
like carbon. J. Biomed. Mater. Res. 92A, 1606-13.
Li, N., Tourovskaia, A. & Folch, A. (2003) Biology on a chip:
microfabrication for studying the behavior of cultured cells. Crit.
Rev. Biomed. Eng., 31, 423-88.
11 – References
148
Lim, J.Y., Shaughnessy, M.C., Zhou, Z., Noh, H., Vogler, E.A. &
Donahue, H.J. (2008) Surface energy effects on osteoblast spatial
growth and mineralization. Biomaterials, 29, 1776-84.
Liu, X., Fu, R.K.Y., Ding, C. & Chu, P.K. (2007) Hydrogen plasma
surface activation of silicon for biomedical applications. Biomol.
Eng., 24, 113-7.
Loesberg, W.A., te Riet, J., van Delft, F.C.M.J.M., Schön, P., Figdor,
C.G., Speller, S., van Loon, J.J.W.A., Walboomers, X.F. & Jansen,
J.A. (2007) The threshold at which substrate nanogroove
dimensions may influence fibroblast alignment and adhesion.
Biomaterials, 28, 3944-51.
Loizou, P.C. (1999) Introduction to cochlear implants. IEEE Eng. Med.
Biol. Mag., 18, 32-42.
Lorenz, H., Despont, M., Fahrni, N., Brugger, J., Vettiger, P. & Renaud,
P. (1998) High-aspect-ratio, ultrathick, negative-tone near-UV
photoresist and its applications for MEMS. Sens. Act. A: Phys., 64,
33-9.
Lorenz, H., Despont, M., Fahrni, N., LaBianca, N., Renaud, P. &
Vettiger, P. (1997) SU-8: a low-cost negative resist for MEMS. J.
Micromech. Microeng., 7, 121-4.
Lyklema, J. (1995) Fundamentals of interface and colloid science, vol. II,
Solid–liquid interface, Academic Press, London, UK.
Madou, M.J. (2002) Fundamentals of microfabrication: The science of
miniaturization, 2nd ed., CRC Press, Boca Raton, FL, USA.
Madou, M. & Tierney, M.J. (1993) Required technology breakthroughs
to assume widely accepted biosensors. Appl. Biochem. Biotechnol.,
41, 109-28.
Mailis-Gagnon, A., Furlan, A.D., Sandoval, J.A. & Taylor R.S. (2004)
Spinal cord stimulation for chronic pain. Cochrane Database Syst.
Rev., 3, CD003783.
Malmivuo, J. & Plonsey R. (1995) Bioelectromagnetism: principles and
applications of bioelectric and biomagnetic fields, Oxford University
Press, New York, NY, USA.
Marletta, G., Ciapetti, G., Satriano, C., Perut, F., Salerno, M. & Baldini,
N. (2007) Improved osteogenic differentiation of human marrow
stromal cells cultured on ion-induced chemically structured poly-
ε-caprolactone. Biomaterials, 28, 1132-40.
11 - References
149
Martínez, E., Engel, E., Planell, J.A. & Samitier, J. (2009) Effects of
artificial micro- and nano-structured surfaces on cell behaviour.
Ann. Anat., 191, 126-35.
Mata, A., Boehm, C., Fleischman, A.J., Muschler, G. & Roy, S. (2002)
Analysis of connective tissue progenitor cell behavior on
polydimethylsiloxane smooth and channel micro-textures.
Biomed. Microdev., 4, 267-75.
Mata, A., Boehm, C., Fleischman, A.J., Muschler, G. & Roy, S. (2002)
Growth of connective tissue progenitor cells on microtextured
polydimethylsiloxane surfaces. J. Biomed. Mater. Res., 62, 499-506.
Mata, A., Fleischman, A.J. & Roy, S. (2005) Characterization of
polydimethylsiloxane (PDMS) properties for biomedical
micro/nanosystems. Biomed. Microdev., 7, 281-93.
Matsuda, T. & Sugawara, T. (1995) Development of surface
photochemical modification method for micropatterning of
cultured cells. J. Biomed. Mater. Res., 29, 749-56.
Matsuzaka, K., Walboomers, X.F., Yoshinari, M., Inoue, T. & Jansen,
J.A. (2003) The attachment and growth behavior of osteoblast-like
cells on microtextured surfaces. Biomaterials, 24, 2711-9.
McAdams, E. (2006) Bioelectrodes. In: Webster, J.G. (ed.) Encyclopedia of
medical devices and instrumentation, 2.nd edition., John Wiley &
Sons, New York, NY, USA.
McBeath, R., Pirone, D.M., Nelson, C.M., Bhadriraju, K. & Chen, C.S.
(2004) Cell Shape, Cytoskeletal Tension, and RhoA Regulate Stem
Cell Lineage Commitment. Dev. Cell, 6, 483-95.
McDonald, J.C. & Whitesides, G.M. (2002) Poly(dimethylsiloxane) as a
material for fabricating microfluidic devices. Acc. Chem. Res., 35,
491-9.
McKenzie, J.L., Waid, M.C., Shi, R. & Webster, T.J. (2004) Decreased
functions of astrocytes on carbon nanofiber materials.
Biomaterials, 25, 1309-17.
Mendonça, G., Mendonça, D.B.S., Aragão, F.J.L. & Cooper, L.F. (2010)
The combination of micron and nanotopography by H2SO4/H2O2
treatment and its effects on osteoblast-specific gene expression of
hMSCs. J.Biomed.Mat.Res., 94A, 169-79.
Mendonça, G., Mendonça, D.B.S., Simões, L.G.P., Araújo, A.L., Leite,
E.R., Duarte, W.R., Aragão, F.J.L. & Cooper, L.F. (2009) The effects
11 – References
150
of implant surface nanoscale features on osteoblast-specific gene
expression. Biomaterials, 30, 4053-62.
Mercanzini, A., Cheung, K., Buhl, D.L., Boers, M., Maillard, A., Colin,
P., Bensadoun, J.-C., Bertsch, A. & Renaud, P. (2008)
Demonstration of cortical recording using novel flexible polymer
neural probes. Sens. Act. A: Phys., 143, 90-6.
Metz, S., Bertsch, A., Bertrand, D. & Renaud, P. (2004) Flexible
polyimide probes with microelectrodes and embedded
microfluidic channels for simultaneous drug delivery and multi-
channel monitoring of bioelectric activity. Biosens. Bioelectron., 19,
1309-18.
Michiardi, A., Aparicio, C., Ratner, B.D., Planell, J.A. & Gil, J. (2007)
The influence of surface energy on competitive protein
adsorption on oxidized NiTi surfaces. Biomaterials, 28, 586-94.
Molina-Luna, K., Buitrago, M.M., Hertler, B., Schubring, M., Haiss, F.,
Nisch, W., Schulz, J.B. & Luft, A.R. (2007) Cortical stimulation
mapping using epidurally implanted thin-film microelectrode
arrays. J. Neurosci. Meth., 161, 118-25.
Moxon, K.A., Kalkhoran, N.M., Markert, M., Sambito, M.A., McKenzie,
J.L. & Webster, J.T. (2004) Nanostructured surface modification of
ceramic-based microelectrodes to enhance biocompatibility for a
direct brain-machine interface. IEEE Trans. Biomed. Eng., 51, 881-9.
Mwenifumbo, S., Li, M., Chen, J., Beye, A. & Soboyejo, W. (2007)
Cell/surface interactions on laser micro-textured titanium-coated
silicon surfaces. J. Mater. Sci. Mater. Med., 18, 9-23.
Myllymaa, K., Myllymaa, S., Korhonen, H., Lammi, M.J., Tiitu, V. &
Lappalainen, R. (2009a) Interactions between Saos-2 cells and
microtextured amorphous diamond or amorphous diamond
hybrid coated surfaces with different wettability properties. Diam.
Relat. Mater., 18, 1294-300.
Myllymaa, K., Myllymaa, S., Korhonen, H., Lammi, M.J., Saarenpää,
H., Suvanto, M., Pakkanen, T.A., Tiitu, V. & Lappalainen, R
(2009b) Improved adherence and spreading of Saos-2 cells on
polypropylene surfaces achieved by surface texturing and carbon
nitride coating. J. Mater. Sci. Mater. Med., 20, 2337-47.
Myllymaa, S., Myllymaa, K., Korhonen, H., Djupsund, K., Tanila, H. &
Lappalainen, R. (2008) Development of flexible thin film
microelectrode arrays for neural recordings. In: IFMBE Proc. 14th
11 - References
151
Nordic-Baltic Conference on Biomedical Engineering and Medical
Physics, Riga, Latvia, June 16-20, 2008, pp. 286-289.
Nair, D.R., Burgess, R., McIntyre, C.C. & Lüders, H. (2008) Chronic
subdural electrodes in the management of epilepsy. Clin.
Neurophysiol., 119, 11-28.
Najafi, K. & Wise, K.D. (1986) Implantable multielectrode array with
on-chip signal processing. IEEE J. Solid-State Circuits, 21, 1035-44.
Navarro, M., Michiardi, A., Castaño, O. & Planell, J.A. (2008)
Biomaterials in orthopaedics. J. R. Soc. Interface, 5, 1137-58.
Nebe, J.G., Luethen, F., Lange, R. & Beck, U. (2007) Interface
interactions of osteoblasts with structured titanium and the
correlation between physicochemical characteristics and cell
biological parameters. Macromol. Biosci., 7, 567-78.
Neuman, M.R. (1978) Biopotential electrodes. In Webster, J.G. (ed.)
Medical instrumentation: application and design, Houghton Mifflin
Company, Boston, MA, USA.
Ng, J.M.K., Gitlin, I., Stroock, A.D. & Whitesides, G.M. (2002)
Components for integrated poly(dimethylsiloxane) microfluidic
systems. Electrophoresis, 23, 3461-73.
Nicolelis, M.A.L., Dimitrov, D., Carmena, J.M., Crist, R., Lehew, G.,
Kralik, J.D. & Wise, S.P. (2003) Chronic, multisite, multielectrode
recordings in macaque monkeys. Proc. Natl. Acad. Sci. U.S.A., 100,
11041-6.
Nicolelis, M.A.L., Ghazanfar, A.A., Faggin, B.M., Votaw, S. & Oliveira,
L.M.O. (1997) Reconstructing the engram: simultaneous,
multisite, many single neuron recordings. Neuron, 18, 529-37.
Niedermayer, E. & Lopez da Silva, F. (eds.) (2005) Electroencephalo-
graphy: basic principles, clinical applications and related fields, 5th ed.,
Lippincott Williams & Wilkins, Philadelphia, PA, USA.
Normann, R.A., Maynard, E.M., Rousche, P.J. & Warren, D.J. (1999) A
neural interface for a cortical vision prosthesis. Vision Res., 39,
2577-87.
Normann, R.A. (2007) Technology Insight: future neuroprosthetic
therapies for disorders of the nervous system. Nat. Clin. Pract.
Neurol, 3, 444-52.
11 – References
152
Norton, W.T., Aquino, D.A., Hozumi, I., Chiu, F.C. & Brosnan, C.F.
(1992) Quantitative aspects of reactive gliosis: a review.
Neurochem. Res., 17, 877-85.
Nunez, P.L. & Srinivasan, R. (2006) Electric fields of the brain: the
neurophysics of EEG, 2nd ed., Oxford University Press, New York,
NY, USA.
Nuxoll, E. & Siegel, R. (2009) BioMEMS devices for drug delivery. IEEE
Eng. Med. Biol. Mag., 28, 31-9.
Orsini, G., Assenza, B., Scarano, A., Piattelli, M. & Piattelli, A. (2000)
Surface analysis of machined versus sandblasted and acid-etched
titanium implants. Int. J. Oral Maxillofac. Implants, 15, 779-84.
Owens, A.L., Denison, T.J., Versnel, H., Rebbert, M., Peckerar, M. &
Shamma, S.A. (1995) Multi-electrode array for measuring evoked
potentials from surface of ferret primary auditory cortex. J.
Neurosci. Meth., 58, 209-20.
Owens, D.K. & Wendt, R.C. (1969) Estimation of the surface free
energy of polymers. J. Appl. Polym. Sci., 13, 1741-7.
Patrick, E., Sankar, V., Rowe, W., Yen, S.F., Sanchez, J.C. & Nishida, T.
(2008) Flexible polymer substrate and tungsten microelectrode
array for an implantable neural recording system. In: Proc. 30th
Annual Internat. Conf. of the IEEE-EMBS, Vancouver, British
Columbia, Canada, 20-24 August, 2008, pp. 3158-61.
Paunovic, M. & Schlesinger, M. (2006) Fundamentals of electrochemical
deposition, 2nd ed., John Wiley & Sons, Hoboken, NJ, USA.
Pearce, A.I., Richards, R.G., Milz, S., Schneider, E. & Pearce, S.G. (2007)
Animal models for implant biomaterial research in bone: a
review. Eur. Cell. Mater., 13, 1-10.
Pearce, C.W. (1988) Grystal growth and wafer preparation. In: Sze, S.M.
(ed.) VLSI technology, McGraw-Hill, New York, NY, USA.
Perlmutter, J.S. & Mink, J.W. (2006) Deep brain stimulation. Annu. Rev.
Neurosci., 29, 229-57.
Petersen, K.E. (1982) Silicon as a mechanical material. P. IEEE, 70, 420-
57.
Pittenger, M.F., Mackay, A.M., Beck, S.C., Jaiswal, R.K., Douglas, R.,
Mosca, J.D., Moorman, M.A., Simonetti, D.W., Craig, S. &
11 - References
153
Marshak, D.R. (1999) Multilineage potential of adult human
mesenchymal stem cells. Science, 284, 143-7.
Pizzoferrato, A., Ciapetti, G., Stea, S., Cenni, E., Arciola, C.R., Granchi,
D. & Savarino, L. (1994) Cell culture methods for testing
biocompatibility. Clin. Mater., 15, 173-90.
Polikov, V.S., Tresco, P.A. & Reichert, W.M. (2005) Response of brain
tissue to chronically implanted neural electrodes. J. Neurosci.
Meth., 148, 1-18.
Ponsonnet, L., Reybier, K., Jaffrezic, N., Comte, V., Lagneau, C., Lissac,
M. & Martelet, C. (2003) Relationship between surface properties
(roughness, wettability) of titanium and titanium alloys and cell
behaviour. Mater. Sci. Eng.C, 23, 551-60.
Puckett, S., Pareta, R. & Webster, T.J. (2008) Nano rough micron
patterned titanium for directing osteoblast morphology and
adhesion. Int. J. Nanomed., 3, 229-41.
Puleo, D.A. & Nanci, A. (1999) Understanding and controlling the
bone–implant interface. Biomaterials, 20, 2311-21.
Qin, D., Xia, Y. & Whitesides, G.M. (2010) Soft lithography for micro-
and nanoscale patterning. Nat. Protoc., 5, 491-502.
Raffa, V., Pensabene, V., Menciassi, A. & Dario, P. (2007) Design
criteria of neuron/electrode interface. The focused ion beam
technology as an analytical method to investigate the effect of
electrode surface morphology on neurocompatibility. Biomed.
Microdev., 9, 371-83.
Ratner, B.D. & Bryant, S.J. (2004) Biomaterials: where we have been
and where we are going. Annu. Rev. Biomed. Eng., 6, 41-75.
Ratner, B.D., Hoffman, A.S., Schoen, F.J. & Lemons, J.E. (eds.) (1996)
Biomaterial science: an introduction to materials in medicine,
Academic Press, San Diego, CA, USA.
Richardson, D.E. (2008) Deep brain stimulation for pain management. In:
DiLorenzo, D.L. & Bronzino, J.D. (eds.) Neuroengineering, CRC
Press, Boca Raton, FL, USA, chapter 6, pp. 6.1-6.18.
Richardson, R.R. Jr., Miller, J.A. & Reichert, W.M. (1993) Polyimides as
biomaterials: preliminary biocompatibility testing. Biomaterials,
14, 627-35.
11 – References
154
Roach, P., Eglin, D., Rohde, K. & Perry, C.C. (2007) Modern
biomaterials: a review - bulk properties and implications of
surface modifications. J. Mater. Sci. Mater. Med., 18, 1263-77.
Rodríguez, F.J., Ceballos, D., Schüttler, M., Valero, A., Valderrama, E.,
Stieglitz, T. & Navarro, X. (2000) Polyimide cuff electrodes for
peripheral nerve stimulation. J. Neurosci. Meth., 98, 105-18.
Roessler, S., Zimmermann, R., Scharnweber, D., Werner, C. & Worch,
H. (2002) Characterization of oxide layers on Ti6Al4V and
titanium by streaming potential and streaming current
measurements. Colloids Surf., B, 26, 387-95.
Rousche, P.J., Pellinen, D.S., Pivin, D.P. Jr., Williams, J.C., Vetter, R.J.,
& Kipke, D.R. (2001) Flexible polyimide-based intracortical
electrode arrays with bioactive capability. IEEE Trans. Biomed.
Eng., 48, 361-71.
Rousche, P.J. & Normann, R.A. (1998) Chronic recording capability of
the Utah intracortical electrode array in cat sensory cortex. J.
Neurosci. Meth., 82, 1-15.
Rubehn, B. & Stieglitz, T. (2010) In vitro evaluation of the long-term
stability of polyimide as a material for neural implants.
Biomaterials, 31, 3449-58.
Rush, S. & Driscoll, D.A. (1968) Current distribution in the brain from
surface electrodes. Anesth. Analg., 47, 717–23.
Rutten, W.L.C. (2002) Selective electrical interfaces with the nervous
system. Annu. Rev. Biomed. Eng., 4, 407-52.
Saldana, L., Gonzalez-Carrasco, J.L., Rodriguez, M., Munuera, L. &
Vilaboa, N. (2006) Osteoblast response to plasma-spray porous
Ti6Al4V coating on substrates of identical alloy. J. Biomed. Mater.
Res. 77A, 608-17.
Saliterman, S.S. (2006) Fundamentals of bioMEMS and medical micro-
devices, SPIE Press, Bellingham, WA, USA.
Santini, J.T.,Jr., Cima, M.J. & Langer, R. (1999) A controlled-release
microchip. Nature, 397, 335-8.
Santini, J.T., Jr., Richards, A.C., Scheidt, R., Cima, M.J. & Langer, R.
(2000) Microchips as controlled drug-delivery devices. Angew.
Chem. Int. Ed. Engl., 39, 2396-407.
11 - References
155
Sawase, T., Jimbo, R., Baba, K., Shibata, Y., Ikeda, T. & Atsuta, M.
(2008) Photo-induced hydrophilicity enhances initial cell behavior
and early bone apposition. Clin. Oral Implants Res., 19, 491-6.
Schadé, J.P. & Ford, D.H. (1973) Basic neuroscience, 2nd ed., Elsevier
Scientific Publishing, Amsterdam, The Netherlands.
Schlesinger, M & Paunovic, M. (eds.) (2000) Modern electroplating, John
Wiley & Sons, Hoboken, NJ, USA.
Schmidt, S., Horch, K. & Normann, R. (1993) Biocompatibility of
silicon-based electrode arrays implanted in feline cortical tissue. J.
Biomed. Mater. Res., 27, 1393-9.
Schüttler, M., Doerge, T., Wien, S.L., Becker, S., Staiger, A., Hanauer,
M., Kammer, S. & Stieglitz, T. (2005) Cytotoxicity of platinum
black. In: Proc. 10th Annual Conf. of the Internat. FES Society, July
2005, Montreal, Canada, pp. 343-345.
Scotchford, C.A., Ball, M., Winkelmann, M., Vörös, J., Csucs, C.,
Brunette, D.M., Danuser, G. & Textor, M. (2003) Chemically
patterned, metal-oxide-based surfaces produced by
photolithographic techniques for studying protein- and cell-
interactions. II: Protein adsorption and early cell interactions.
Biomaterials, 24, 1147-58.
Selvakumaran, J., Keddie, J.L., Ewins, D.J. & Hughes, M.P. (2008)
Protein adsorption on materials for recording sites on implantable
microelectrodes. J. Mater. Sci. Mater. Med., 19, 143-51.
Seo, J.-M., Kim, S.J., Chung, H., Kim, E.T., Yu, H.G. & Yu, Y.S. (2004)
Biocompatibility of polyimide microelectrode array for retinal
stimulation. Mat. Sci. Eng. C, 24, 185-9.
Seymour, J.P., Elkasabi, Y.M., Chen, H.Y., Lahann, J. & Kipke, D.R.
(2009) The insulation performance of reactive parylene films in
implantable electronic devices. Biomaterials, 30, 6158-67.
Shi, Z., Neoh, K.G., Kang, E.T., Poh, C. & Wang, W. (2008) Bacterial
adhesion and osteoblast function on titanium with surface-
grafted chitosan and immobilized RGD peptide. J. Biomed. Mater.
Res. 86A, 865-72.
Smith, D.L. (1995) Thin-film deposition: principles and practice, McGraw-
Hill, McGraw–Hill, New York, NY, USA.
11 – References
156
Soekarno, A., Lom, B. & Hockberger, P.E. (1993) Pathfinding by
neuroblastoma cells in culture is directed by preferential adhesion
to positively charged surfaces. NeuroImage, 1, 129-44.
Spence, A.J., Neeves, K.B., Murphy, D., Sponberg, S., Land, B.R., Hoy,
R.R. & Isaacson, M.S. (2007) Flexible multielectrodes can resolve
multiple muscles in an insect appendage. J. Neurosci. Meth., 159,
116-24.
St. John, P.M., Kam, L., Turner, S.W., Craighead, H.G., Issacson, M.,
Turner, J.N. & Shain, W. (1997) Preferential glial cell attachment
to microcontact printed surfaces. J. Neurosci. Meth., 75, 171-7.
Stensaas, S.S. & Stensaas, L.J. (1978) Histopathological evaluation of
materials implanted in the cerebral cortex. Acta Neuropathol., 41,
145-55.
Stett, A., Egert, U., Guenther, E., Hofmann, F., Meyer, T., Nisch, W. &
Haemmerle, H. (2003) Biological application of microelectrode
arrays in drug discovery and basic research. Anal.Bioanal. Chem.,
377, 486-95.
Stieglitz, T., Beutel, H. & Meyer, J.-U. (1997) A flexible, light-weight
multichannel sieve electrode with integrated cables for interfacing
regenerating peripheral nerves. Sens. Act. A: Phys., 60, 240-3.
Stieglitz, T. (2009) Development of a micromachined epiretinal vision
prosthesis. J. Neural Eng., 6, 065005.
Stieglitz, T. (2001) Flexible biomedical microdevices with double-sided
electrode arrangements for neural applications. Sens. Act. A:
Phys., 90, 203-11.
Stieglitz, T., Beutel, H., Schuettler, M. & Meyer, J.-U. (2000)
Micromachined, polyimide-based devices for flexible neural
interfaces. Biomed. Microdev., 2, 283-94.
Stieglitz, T. & Meyer, J.-U. (2006) Neural implants in clinical practice. In:
Urban, G.A. (ed.) BioMEMS, Springer, Dordrecht, The
Netherlands, pp. 41-70.
Stieglitz, T. & Meyer, J.-U. (2006) Biomedical microdevices for neural
implants. In: Urban, G.A. (ed.) BioMEMS, Springer, Dordrecht,
The Netherlands, pp. 71-137.
Stoddart, M.J., Grad, S., Eglin, D. & Alini, M. (2009) Cells and
biomaterials in cartilage tissue engineering. Regen. Med., 4, 81-98.
11 - References
157
Sun, Y., Lacour, S.P., Brooks, R.A., Rushton, N., Fawcett, J. & Cameron,
R.E. (2009) Assessment of the biocompatibility of photosensitive
polyimide for implantable medical device use. J. Biomed. Mater.
Res. 90A, 648-55.
Sweitzer, R., Scholz, C., Montezume, S. & Rizzo, J.F. (2006) Evaluation
of subretinal implants coated with amorphous aluminum oxide
and diamond-like carbon. Journal of bioactive and compatible
polymers, 21, 5-22.
Szarowski, D.H., Andersen, M.D., Retterer, S., Spence, A.J., Isaacson,
M., Craighead, H.G., Turner, J.N. & Shain, W. (2003) Brain
responses to micro-machined silicon devices. Brain Res., 983, 23-
35.
Takahashi, H., Ejiri, T., Nakao, M., Nakamura, N., Kaga, K. & Hervé, T.
(2003) Microelectrode array on folding polyimide ribbon for
epidural mapping of functional evoked potentials. IEEE Trans.
Biomed. Eng. 50, 510-6.
Takami, Y., Nakazawa, T., Makinouchi, K., Glueck, J. & Nose, Y. (1997)
Biocompatibility of alumina, ceramic and polyethylene as
materials for pivot bearings of a centrifugal blood pump. J.
Biomed. Mater. Res., 36, 381-6.
Tan, W. & Desai, T.A. (2003) Microfluidic patterning of cells in
extracellular matrix biopolymers: effects of channel size, cell type,
and matrix composition on pattern integrity. Tissue Eng., 9, 255-
67.
Teixeira, A.I., Abrams, G.A., Bertics, P.J., Murphy, C.J. & Nealey, P.F.
(2003) Epithelial contact guidance on well-defined micro- and
nanostructured substrates. J. Cell. Sci., 116, 1881-92.
Teixeira, S., Monteiro, F.J., Ferraz, M.P., Vilar, R. & Eugenio, S. (2007)
Laser surface treatment of hydroxyapatite for enhanced tissue
integration: surface characterization and osteoblastic interaction
studies. J.Biomed.Mater.Res. 81A, 920-9.
Thanawala, S., Palyvoda, O., Georgiev, D.G., Khan, S.P., Al-Homoudi,
I.A., Newaz, G. & Auner, G. (2007) A neural cell culture study on
thin film electrode materials. J. Mater. Sci. Mater. Med., 18, 1745-52.
Thian, E.S., Ahmad, Z., Huang, J., Edirisinghe, M.J., Jayasinghe, S.N.,
Ireland, D.C., Brooks, R.A., Rushton, N., Bonfield, W. & Best, S.M.
(2010) The role of surface wettability and surface charge of
11 – References
158
electrosprayed nanoapatites on the behaviour of osteoblasts. Acta
Biomater., 6, 750-5.
Tijero, M., Gabriel, G., Caro, J., Altuna, A., Hernández, R., Villa, R.,
Berganzo, J., Blanco, F.J., Salido, R. & Fernández, L.J. (2009) SU-8
microprobe with microelectrodes for monitoring electrical
impedance in living tissues. Biosens. Bioelectron., 24, 2410-6.
Tomlins, P.E., Leach, R., Vagdama, P., Mikhalovsky, S. & James, S.
(2005) On the topographical characterization of biomaterial surfaces. In:
Vadgama, P. (ed.) Surfaces and interfaces for biomaterials,
Woodhead Publishing Limited, Cambridge, UK, pp. 693-718.
Tortora, G.J. & Grabowski, S.R. (2000) Principles in anatomy and
physiology, 9th ed., John Wiley & Sons, New York, NY, USA.
Tuomikoski, S. (2007) Fabrication of SU-8 microstructures for analytical
microfluidic applications, Doctoral Dissertation, Helsinki University
of Technology, Espoo, Finland.
Turner, A.M.P., Dowell, N., Turner, S.W.P., Kam, L., Isaacson, M.,
Turner, J.N., Craighead, H.G. & Shain, W. (2000) Attachment of
astroglial cells to microfabricated pillar arrays of different
geometries. J. Biomed. Mater. Res., 51, 430-41.
Turner, J.N., Shain, W., Szarowski, D.H., Andersen, M., Martins, S.,
Isaacson, M. & Craighead, H. (1999) Cerebral astrocyte response
to micromachined silicon implants. Exp. Neurol., 156, 33-49.
Urban, G.A. (ed.) (2006) BioMEMS, Springer, Dordrecht, The
Netherlands.
Urban, G.A., Prohaska, O. & Olcaytug, F. (2006) Early bioMEMS multi-
sensor neuroprobes. In: Urban, G.A. (ed.) BioMEMS, Springer,
Dordrecht, The Netherlands, pp. 1-13.
Ureshi, M., Matsuura, T. & Kanno, I. (2004) Stimulus frequency
dependence of the linear relationship between local cerebral
blood flow and field potential evoked by activation of rat
somatosensory cortex. Neurosci. Res., 48, 147-53.
Vadgama P. (2005) (ed.) Surfaces and interfaces for biomaterials,
Woodhead Publishing Limited, Cambridge, UK.
Väisänen, O. (2008) Multichannel EEG methods to improve the spatial
resolution of cortical potential distribution and the signal quality of deep
brain sources, Doctoral dissertation, Tampere University of
Technology, Tampere, Finland.
11 - References
159
van Kooten, T.G., Schakenraad, J.M., van der Mei, H.C. & Busscher,
H.J. (1992) Influence of substratum wettability on the strength of
adhesion of human fibroblasts. Biomaterials, 13, 897-904.
van Oss, C.J., Good, R.J. & Chaudhury, M.K. (1988) Additive and
nonadditive surface tension components and the interpretation of
contact angles. Langmuir, 4, 884-91.
Vanhatalo, S., Holmes, M.D., Tallgren, P., Voipio, J., Kaila, K. & Miller,
J.W. (2003) Very slow EEG responses lateralize temporal lobe
seizures: An evaluation of non-invasive DC-EEG. Neurology, 60,
1098-104.
Vetter, R.J., Williams, J.C., Hetke, J.F., Nunamaker, E.A. & Kipke, D.R.
(2004) Chronic neural recording using silicon-substrate
microelectrode arrays implanted in cerebral cortex. IEEE Trans.
Biomed. Eng., 51, 896-904.
Visser, S.A., Hergenrother, R.W. & Cooper, S.L. (1996) Polymers. In:
Ratner, B.D., Hoffman, A.S., Schoen, F.J. & Lemons, J.E. (eds.)
Biomaterial science: an introduction to materials in medicine,
Academic Press, San Diego, CA, USA.
Vogler, E.A. (1998) Structure and reactivity of water at biomaterial
surfaces. Adv. Colloid Interface Sci., 74, 69-117.
Voldman, J., Gray, M.L. & Schmidt, M.A. (1999) Microfabrication in
biology and medicine. Annu. Rev. Biomed. Eng., 1, 401-25.
von Recum, A.F., Shannon, C.E., Cannon, C.E., Long, K.J., Van Kooten,
T.G. & Meyle, J. (1996) Surface Roughness, Porosity, and Texture
as Modifiers of Cellular Adhesion. Tissue Eng., 2, 241-53.
Voskerician, G., Shive, M.S., Shawgo, R.S., von Recum, H., Anderson,
J.M., Cima, M.J. & Langer, R. (2003) Biocompatibility and
biofouling of MEMS drug delivery devices. Biomaterials, 24, 1959-
67.
Waltz, J.M. (1997) Spinal cord stimulation: A quarter century of
development and investigation: A review of its development and
effectiveness in 1,336 cases. Stereot. Funct. Neuros., 69, 288-99.
Wang, X., Ohlin, C.A., Lu, Q. & Hu, J. (2006) Influence of
physicochemical properties of laser-modified polystyrene on
bovine serum albumin adsorption and rat C6 glioma cell
behavior. J. Biomed. Mater. Res. 78A, 746-54.
11 – References
160
Wang, W. & Soper, S.A. (ed.) (2007) BioMEMS: technologies and
applications, CRC Press, Boca Raton, FL, USA.
Wei, J., Yoshinari, M., Takemoto, S., Hattori, M., Kawada, E., Liu, B. &
Oda, Y. (2007) Adhesion of mouse fibroblasts on
hexamethyldisiloxane surfaces with wide range of wettability. J.
Biomed. Mater. Res. 81B, 66-75.
Whitesides, G.M., Ostuni, E., Takayama, S., Jiang, X. & Ingber, D.E.
(2001) Soft lithography in biology and biochemistry. Annu. Rev.
Biomed. Eng., 3, 335-73.
Williams, D.F. (1999) The Williams dictionary of biomaterials, Liverpool
University Press, Liverpool, UK.
Williams, D.F. (2008) On the mechanisms of biocompatibility.
Biomaterials, 29, 2941-53.
Williams, J.C., Rennaker, R.L. & Kipke, D.R. (1999) Long-term neural
recording characteristics of wire microelectrode arrays implanted
in cerebral cortex. Brain Res. Protoc., 4, 303-13.
Williams, K.R., Gupta, K. & Wasilik, M. (2003) Etch rates for
micromachining processing - Part II. J. Microelectromech. Syst., 12,
761-78.
Williams, K.R. & Muller, R.S. (1996) Etch rates for micromachining
processing. J. Microelectromech. Syst., 5, 256-69.
Wilson, B.S. & Dorman, M.F. (2008) Cochlear implants: current designs
and future possibilities. J. Rehabil. Res. Dev., 45, 695-730.
Wilson, B.S. & Dorman, M.F. (2008) Cochlear implants: A remarkable
past and a brilliant future. Hear. Res., 242, 3-21.
Wilson, S.A., Jourdain, R.P.J., Zhang, Q., Dorey, R.A., Bowen, C.R.,
Willander, M., Wahab, Q.U., Willander, M., Al-hilli, S., Nur, O.,
Quandt, E., Johansson, C., Pagounis, E., Kohl, M., Matovic, J.,
Samel, B., van der Wijngaart, W., Jager, E.W.H., Carlsson, D.,
Djinovic, Z., Wegener, M., Moldovan, C., Iosub, R., Abad, E.,
Wendlandt, M., Rusu, C. & Persson, K. (2007) New materials for
micro-scale sensors and actuators: An engineering review. Mater.
Sci. Eng. R, 56, 1-129.
Winkelmann, M., Gold, J., Hauert, R., Kasemo, B., Spencer, N.D.,
Brunette, D.M. & Textor, M. (2003) Chemically patterned, metal
oxide based surfaces produced by photolithographic techniques
11 - References
161
for studying protein– and cell–surface interactions I: Micro-
fabrication and surface characterization. Biomaterials, 24, 1133-45.
Wirth, C., Comte, V., Lagneau, C., Exbrayat, P., Lissac, M., Jaffrezic-
Renault, N. & Ponsonnet, L. (2005) Nitinol surface roughness
modulates in vitro cell response: a comparison between
fibroblasts and osteoblasts. Mater. Sci. Eng. C, 25, 51-60.
Wise, K.D. (2005) Silicon microsystems for neuroscience and neural
prostheses. IEEE Eng. Med. Biol. Mag., 24, 22-9.
Wise, K.D. & Angell, J.B. (1975) A low-capacitance multielectrode
probe for use in extracellular neurophysiology. IEEE Trans.
Biomed. Eng., 22, 212-9.
Wu, H., Odom, T.W. & Whitesides, G.M. (2002) Generation of Chrome
Masks with Micrometer-Scale Features Using Microlens
Lithography. Adv. Mater., 14, 1213-6.
Xia, Y. & Whitesides, G.M. (1998) Soft Lithography. Annu.Rev.Mater.
Sci., 28, 153-84.
Yeager, J.D., Phillips, D.J., Rector, D.M. & Bahr, D.F. (2008)
Characterization of flexible ECoG electrode arrays for chronic
recording in awake rats. J. Neurosci. Meth., 173, 279-85.
Yuen, T.G., Agnew, W.F. & Bullara, L.A. (1987) Tissue response to
potential neuroprosthetic materials implanted subdurally.
Biomaterials, 8, 138-41.
Yuen, T.G. & Agnew, W.F. (1995) Histological evaluation of
polyesterimide-insulated gold wires in brain. Biomaterials, 16, 951-
6.
Zippel, N., Schulze, M. & Tobiasch, E. (2010) Biomaterials and
mesenchymal stem cells for regenerative medicine. Recent. Pat.
Biotechnol., 4, 1-22.
Publications of the University of Eastern FinlandDissertations in Forestry and Natural Sciences
Publications of the University of Eastern Finland
Dissertations in Forestry and Natural Sciences
isbn 978-�952-�61-�0216-�0
Sami Myllymaa
Novel Micro- and Nano-technological Approaches for Improving the Performance of Implantable Biomedical Devices
Recent advances in micro- and
nanotechnology offer a great oppor-
tunity to develop intelligent bioma-
terials and the next generation of
implantable devices for diagnostics,
therapeutics, and tissue engineering.
This dissertation is focusing on the
development of novel polymer-based
microelectrode arrays suitable for
use in intracranial electroencepha-
lographic recordings. Moreover,
the performances of novel thin film
materials and their surface modifica-
tions at micro- and nanoscales were
studied with physicochemical and
cellular experiments in order to de-
vise new solutions for further devel-
opment of biomedical microdevices.
dissertatio
ns | 014 | S
am
i Myllym
aa | N
ovel M
icro- a
nd
Na
no
-techn
ologica
l Ap
pro
ach
es for Im
pro
ving th
e Perfo
rma
nce of...
Sami MyllymaaNovel Micro- and
Nano-technological Approaches for Improving
the Performance of Implantable
Biomedical Devices