Transcript Slide 1
Micro and
Nanofabrication
Products
Carbon Nanotube Sites for Neural-Network
Patterning and Recording
Tamir Gabay1, Itshak Kalifa1, Lisa Ezra1, Eshel Ben-Jacob2, Yael Hanein1
1School
of Electrical Engineering, department of physical electronics,
2School of Physics and Astronomy
Tel-Aviv University, Tel-Aviv 69978, Israel
Abstract
Extra-cellular recordings with multi electrode array (MEAs) systems have been
used for the last several decades to study the formation and behavior of invitro neuronal networks. It is widely accepted that improved MEAs, with high
resolution and better control over cell density and patterning, are expected to
be useful to expand our understanding of high brain functions and to facilitate
novel neuro-chip sensors.
This work presents a new MEA configuration, which enables the formation of
electrically viable engineered neuronal networks with high-resolution
extracellular recording. The networks are engineered according to
lithographically defined carbon nanotube (CNT) templates. These CNT
templates strongly anchor cells and enable the formation of stable networks.
Carbon nanotube coated surfaces are biocompatible, provide an excellent
surface for cell growth and offer high specific capacitance crucial in
electrochemical applications. According to the new MEA scheme molybdenum
lines electrically connect the CNT templates and external instrumentation.
Cyclic Voltammetry
Top – Optical microscope
image of regular array of
molybdenum pads (Dia.
150µm) coated by carbon
nanotube. Uncoated
molybdenum traces are shown
also. Middle – HRSEM (El-Mul
Corp.) image of one of the
molybdenum pads surfaces,
and Bottom – HRSEM (El-Mul
Corp.) image of a typical edge
of such a molybdenum pad.
CVD system
Cyclic voltammetry experiment results
of two 100 µm diameter molybdenum
electrodes. Dashed line – electrode
coated by carbon nanotube layer;
continues line – bare
Inverted microscope images of interconnected neuronal systems
formed with CNT islands. (Left) One hour after cell deposition cells
are randomly distributed (scale bar is 150 µm). (Center) After four
days neurons form clusters at the CNT sites, which form connections
between these islands (scale bar is 150 µm). Some islands are not
yet connected at this stage and several short and unconnected
dendrites and axons are apparent. (Right) Neuron cluster at a CNT
island (scale bar is 100 µm). Cell density was 4.5×10-3 µm-2.
Assembly Mechanisms
Neural network formation from
nucleation centers. (Left) In the first
step neuronal migration from low affinity
surface results in cell aggregation at
high affinity regions (large, dark disks).
(Right) In a second step high density
neuronal islands form well defined
interconnection following the shortest
link between islands.
Neural network collapse into
isolated clusters due to tension.
(Left) The first step of this process
is based mainly of self-wiring of the
neurons with only limited neuronal
migration. (Right) In a later stage
the network segregates into
separate clusters.
Inverted microscope images of interconnected network formed with CNT islands. (Left)
after 96 hours, (Center) after 128 hours and (Right) after 150 hours after plating. Cell
density was 1x10-3 cells/µm2. Arrows indicate well defined connections between
neuronal clusters. (Scale bar is 150µm)
Inverted microscope images of interconnected networks formed with CNT islands at a
cell density of 4.5×10-3 cells/µm2 at (Left) 96 hours, (Center) 128 hours and (Right)
150 hours after plating the cells on the substrate. At this high cell density mechanical
tension overpowers cell adhesion to the CNT islands. Scale bar is 150 µm.
Tel Aviv Nano-Technology Workshop, Maagan, April 2005