microlens array Antony Orth and Kenneth Crozier 8 May CLEO 2012 Microscopy with lens arrays What is high thoughput microscopy Experimental setup confocal system Lens array characteristics resolution ID: 245828
Download Presentation The PPT/PDF document "High throughput microscopy with a" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
High throughput microscopy with a microlens array
Antony Orth and Kenneth Crozier8 MayCLEO 2012Slide2
Microscopy with lens arrays
What is high thoughput microscopy?Experimental setup – confocal system
Lens array characteristics, resolutionEffect of
confocal
filtering
Large scale imagingWhat’s next?
1Slide3
High Throughput Microscopy
Microscope field of view (FOV) << sample size.Sub-fields of large sample imaged sequentially.Sub-fields stitched together to form large continuous image.
Histological slide scanningHigh content screening (HCS)
N
2
: # of sub-fields >103 for a microscope slide > 104
for a microwell plate
With a 20x objective:
Stage translation
Autofocusing
~1-2 sec / FOV*
*
http
://www.highthroughputimaging.com/screening/imagexpress_micro.html#apps
~1-10 cm
100s of
μmSlide4
A High Throughput Microscope
- 4.26 Mpx / second (4.66 Mpx sensor)- 1.85 hrs / plate / color @ 70% coverage!
(Molecular
Devices
ImageXpress
Micro)http://www.highthroughputimaging.com/screening/imagexpress_micro.html#appsSlide5
What limits high throughput microscopy?
Specs sheet for typical systems advertise ~1s per image.Camera sensors are ~1-5Mpx, so throughput is ~1-5Mpx/s, far below the throughput available with digital cameras.1,2Limiting factors:
Motorized stages have small bandwidth.Scanning procedures (focusing, moving FOV) become temporally expensive.
Motion blur/lighting.
Can we alter optics to alleviate these problems?
Break up imaging into small, parallelized fields of view.
4
1http://www.olympus.co.uk/microscopy/22_scan_R_Specifications.htm
2 http://www.highthroughputimaging.com/screening/imagexpress_micro.html#appsSlide6
Experimental Setup
5
Piezo
scan
Movie of
microlens
apertures as sample is scanned
Microlens
focal length
Bright spots in movie = fluorescence captured by individual
micolenses
Each
microlens
= individual scanning
confocal
microscope
Stitch together
microlens
subimages
to form large image
(532nm, 38
mW
)Slide7
Reflow Molded Microlens Arrays
1.3mm100
x 100 microlens
array
6
Molded in optical adhesive (NOA 61,
n=1.56)
Pitch: 100
μm
Lens Diameter: 93
μm
Lens Height: 14
μm
1 mm
Pitch: 55
μm
Lens Diameter: 37
μm
Lens Height: 15
μm
NA: 0.41
NA: 0.31
100
x
100
microlens
arraySlide8
Imaging resolution
7
1
μm
FWHM
781 nm
37
μm
diameter lenses
Focal spot size sets resolution when
iris open
Bead FWHM =
787 nm
+/- 39 nm ~ Focal spot FWHM
200 nm beads
5
μm
Microlens
focal spotSlide9
Confocal filtering
8
Real images formed by
microlenses
.
Iris acts as
confocal filter for ALL microlenses!
Stopping down iris improves resolution via
confocal effect.Slide10
Confocal filtering
9
Iris open
Iris diameter 2 mm
(0.52 Airy diameter)
5
μm
5
μm
Confocal
ability adds another level of control:
Can trade off signal for resolutionSlide11
2 mm
50 μm
25
μm
0.85
GPx image
Raw pixel throughput 4Mpx/s
Uses only 0.124
Mpx
sensor!
Full frame sensor higher through
putSlide12
Rat Femur Slice (Cy3)
1 mmSample courtesy of Mooney lab, HarvardSlide13
Rat Femur Slice (Zoom-in)
1 mmCortical Bone
Medullary Canal
Periosteum
80
μm
80
μm
80
μmSlide14
Summary & Outlook
13
Built a parallelized scanning microscope using refractive μlensesFabricated 10,000 element
μlens
arrays. NA: 0.41 (37μm diameter),
NA: 0.31 (93 μm diameter).Constructed a 0.85 Gpx image with <790nm resolution.Resolution of <700nm can be achieved using
confocal filtering.Demonstrated imaging of microspheres, rat femur section.Throughputs up to 4Mpx/s using 352
x 352 px sensor. Lots of room for scaling.
Have recently achieved imaging through a coverslip.
Next step: image
microwell
plate, multiple wells at once.
20
μm
5
μm
“spheres”
100
μm
diam.
lensesSlide15
PDMS Reflow Molding Fabrication
Pattern posts of
photoresist
(AZ-40XT) on silicon
Place wafer on hot plate @125
o
C for 1 min. Resist melts, surface tension provides smooth lens surface
Inverse mold in PDMS
PDMS
Microscope slide
Replicate melted
photoresist
in optical adhesive (NOA 61) with UV cure
NOA 61
Peel off PDMS,
microlens
array ready for use!
14