Cell Adhesion Study Using MEMS and Digital Image Correlation

1
Cell Adhesion Study Using MEMS and Digital Image
Correlation

• Keith Gall
• Tareef Jafferi
• Matt Schwabauer

2
Presentation Outline

• Device and experiment concept
• Digital Image Correlation (DIC)
• What is DIC?
• How can it be used in MEMS and Bio applications?
• Device design considerations
• Force Capacity
• Calibration
• Static and dynamic tests
• Fabrication and Processing
• Mask design
• Surface micro-machining
• Conclusions/Questions?

3
MEMS Tensile Device Schematic
4
MEMS Tensile Device Schematic
Top View
Loaded
lt 10microns
Loading Plates
750 combs
750 combs
Side View
Protein Coating
Cell will neck
Surface Florescent Beads
Cellular Florescent Beads
5
MEMS Tensile Device Fluorescent Markers
40 microns
lt1 micron
5 microns
8 microns
30 microns

• Fluorescent beads embedded in substrate
• Commercially available fluorescent microspheres
  (FluoSpheres, Molecular Probes, Eugene, OR)
• Size 0.2 micron, well defined
• Latex beads embedded in polyacrylimide
• Substrate funtionalized with fibronectin
• Chemical labels tag specific cell substructures
• Size lt1 micron but variable
• GFP-zyxin, vinculin,
• Tagging of lamellipodia, focal adhesion sites,
  extracellular matrix
• Technique sucessfully used in studies of fish
  fin, monkey liver, bovine endothelial, rabbit and
  human corneal fibroblasts, H-ras 3T3 cancer
  cells, etc.
• Petrol, et al, Investigative Opthalmology
  Visual Science, (2004)
• Munevar, et al, Biophysical Journal, (2001)
• Yang, et al, IEEE, (2004)

6
Digital Image Correlation Optical Measurement

• Computer assisted technique for measuring
  displacements
• Compares the location of points from a reference
  image to the location of those points in a second
  (deformed) image
• Typically used to generate 2-D strain maps (or
  3-D using stereo triangulation or confocal
  microscopy)
• Why is it useful?
• Strain maps provide detailed information
• Non-intrusive measurement
• Measurement in controlled environment

Munevar, et al, Biophysical Journal, (2001)

• Large range of time scales (static to 50 KHz)
• Large range of length scales (10-9 to 1012
  meters)
• Sub-pixel accuracy obtained thru interpolation
  functions and iterative, correlation-function
  maximization

7
Digital Image Correlation Methodology
Sum of Squared Differences

• Pattern displaces 1 pixel up, and 1 pixel over
• Subset (5 x 5) in red, required to track unique
  signature in Image 2
• Use Correlation Function to score possible
  candidate displacements, minimize function to
  find perfect match

Example as presented in Introduction to Digital
Image Correlation, at the SEM X international
Congress in Costa Mesa, CA on June 6, 2004 by
Dorian Garcia
8
Digital Image Correlation Experimental
Considerations

• Spatial Resolution
• Speckle size, lt1micron
• Speckle contrast signal to noise ratio, good
• Random speckle pattern, good
• Out-of-plane deformation
• 3-D deformation maps possible
• Stereo triangulation
• Z-axis or confocal microscopy
• 3-D reconstruction by imaging at different focal
  lengths along z-axis
• Possible for static tests
• Possible for dynamic tests, but limited by camera
  exposure time
• Non-intrusive measurement
• Cells tested in suitable environment
• Measurements can be made in hydrated state
  (submerged test device?)
• Use different color fluorescent trackers
• Use filters to monitor either substrate beads or
  cell structure
• Isolate cell cytoskeleton response from
  substrate/adhesive response

in-vivo Z-axis tracking of migrating T cells in
mouse lymh nodes
M. Miller, et al, Proceedings of National Academy
of the Sciences of USA, (2003)
9
Design Parameters Comb Drive
(per adhsion site)
40 adhesion sites

• Reaction/Applied Force
• Force applied between plates Force of cell
  stretch
• Force applied thru plate reaction force of
  cell shear reaction force of deforming
  substrate
• Comb Drive Force
• Electrostatic force applied by comb drive
• Gap 2 µm, thickness1µm, voltage 20V
• Beam Bending Force
• 4 folded beams
• Modulus of elasticity for Silicon E170 GPa
• Ktotal18.36 N/m
• Maximum Displacement
• gap between testing plates travel of comb drive
  bending of beams

10
Design Parameters Calibration

• Proof test with standard material first
• Soft polymer or gel
• Calibrate using material that matches
  stiffness/response of typical cell
• Compare calibration results to literature
• Static test Modulus, ultimate tensile strength,
  adhesive failure
• Dynamic test Viscoelasticity, hysteresis,
  fatigue

11
Design Parameters Design of Experiment

• Eliminate Cell creep into gap
• Apply negative voltage to close gap
• Place cell on closed plates
• Static test on Cell
• More consistent Data
• Less error due to inertial effects
• Closely monitor cell response to external force
• Dynamic test on Cell
• Data analysis more complex
• Measure time-dependent effects
• Simulate real world conditions of cells

12
Fabrication Process of Device Part 1 (The Comb
Driver)
(a) Copper is deposited on the glass substrate by
E-beam evaporation
(b) Photoresist (PR) is coated on the copper
layer and is patterned as the etching mask of the
copper. The pole layer of copper is patterned
with wet etching techniques in FeCl3 etching
solutions (Mask 1)
(c) Polyimide (PI) is spun as sacrificial layer
and achor is defined. The PI is cured at 150 C to
endure the next process steps
(d) A seed layer of Ti/Cu (300A/500A) for
electroplating Ni is deposited on the sacrificial
layer
Comb drive fabrication method as described by
Chien, et al, Design, Test, Integration
Packaging of MEMS, (2003)
13
Fabrication Process of Device Part 1 (The Comb
Driver)
Mask 1
Mask 2
14
Fabrication Process of Device Part 1 (The Comb
Driver)
(e) Thick Photoresist is spun and comb is defined
(f) Ni is deposited for finger layer by
electroplating
(g) The seed layer of Ti/Cu (300A/500A) is etched
by wet etching techniques in thin nitric acid and
Ti etching solution after strip of the PR with
acetone. The last release step is performed by O2
plasma dry etching
15
Fabrication Process of Device Part 1 (The Comb
Driver)
Mask 3
16
Fabrication Process of Device Part 2 (The Stage
and Beams)
a) Metal deposition and patterning (using Mask 1)
b) Sacrificial layer coating and patterning
c) PR patterning Ti/Au e-beam deposition and
lift off (using Mask 2)
d) Removal of sacrificial layer by dry etching
17
Fabrication Process of Device Part 2 (The Stage
and Beams)
Mask 1
Mask 2
18
Conclusions and Questions

• Existing cellular studies lack
• Cell response to external loading
• Reliable methods for collecting data
• Digital image correlation enables accurate
  measurements through non-contacting methods
• The proposed device
• Can apply forces and accurately measure
  deformation
• Tests both static and dynamic response of cells
• Simplified design process by adopting existing
  fabrication methods
• Questions?