Transcript Adaptive optical biocompact disk for molecular recognition

Adaptive optical biocompact
disk for molecular recognition
Leilei Peng, Manoj M. Varma, Fred E. Regnier,
David D. Nolte
Applied Physics Letters, May 2005
Purdue University
Journal Club – 11-03-05
Emre Özkumur
Electrical and Computer Engineering Department
Boston University
outline


Background for phase sensitive interferometric detection
Details of how the method works




Phase modulation with a BioCD
Conversion of phase modulation to amplitude modulation
Fabrication of the BioCD
Results



Sensitivity
Background for bio-detection; specific, non-specific binding
Molecular data
Phase sensitive interferometry
ε1
E1e
ε2
E2 e
j
2

jn
d
2

d
Measure
d(n-1)
d
I  E1e
2
j d

 E2 e
2 2
jn d

 2d

 I 0  I cos
(n  1) 
 

Phase sensitive interferometry_2
I
Most sensitive
region:
quadrature
Satisfied when
d(n-1)=λ/4
or…
0
π/2
Ø
2π
by having a timeindependent π/2
phase difference
between reference
and signal beams
Phase modulation with a BioCD
• 4” diameter
• 1024 wedgedspoke pattern of
biomolecules
• spun at 50Hz,
pattern is
transferred to
probe beam as
phase modulation
at 51.2kHz
Phase modulation -> Amplitude modulation
Phase modulated probe and reference beams are sent into a
Photorefractive Quantum Well device
PRQW
E 2’
E 1”
E1
E2
E 1’
E 2”
- Fringes change the effective refractive index of the
material
- Because of the periodic change in “n”, device acts
as a diffraction grating
Why use PRQW?


2
 

E1 ( E2 , L)  e  E1 (0)   ( ) E2 (0) expi  P   ( ) 
d (t )(n  1)    

2 


i 0
Photorefractive
phase shift
Excitonic spectral
shift
- Quadrature condition is satisfied by adjusting the excitonic spectral shift
by changing the applied voltage to PRQW and wavelength of the beam
 P   ( )
- Grating can adapt to slowly varying phases, but not to rapid changes
=> low frequency variations (environmental & mechanical
disturbances) of the phase compansated, high frequency pass without
changing
- Resultant intensity amplitude modulated with 51.2kHz; total intensity is
measured and sent to a spectrum analyzer or lock-in amplifier
Fabrication of the BioCD
- Ink gel stamping
method
Sensitivity of the system


To test the system sensitivity, a phase modulator was
used instead of BioCD
Phase modulator was used to calibrate the data, to
extract the height information
Sensitivity of the system

Phase modulator created Øpp=2.1*10-2 rad,
corresponding to:
 pp 
d (n  1) 
 2.8nm
2
(optical thickness for λ=830nm)

Patterned Bovine serum albumin (BSA) optical
thickness measured to be:
d(n-1)=0.6nm
for which AFM measured 2.5±0.7nm for the height
difference, confirming the refractive index of BSA to
be n=1.2±0.1nm
Signal-to-Noise
NEOT: noise equivalent optical thickness
NL: Noise contribution of the local oscillator
NP: Phase noise from the disk
S
d
 d 
 2


2
N N L  N P  NEOT 
2
2
At 51 kHz modulation frequency NP dominates,
NEOT=0.3nm
Higher sensitivity can be achieved by narrowing
the detection bandwidth (3kHz)
Background for bio-detection data


BioCD was implemented for an antigen-antibody
molecular recognition experiment
For simplicity:





Label
Label
Label
Label
“mouse IgG antigen” = M
“anti-mouse IgG antibody” = m
“rabbit IgG antigen” = R
“anti-rabbit IgG antibody” = r
System should be able to detect the bindings of
interest (specific bindings) which are M-m, R-r,
and should ignore non-specific bindings such as
M-r or R-m
Preperation of the samples on CD



First, M-antigen 1024spoke pattern was printed
on the disk
Free surfaces (unprinted
areas) were covered with
BSA
Then the disk was divided
4 annular tracks (A-D)
Molecular data_1


Non-specific binding was tested by incubating
track C with r-antibody
Specific binding was tested by incubating both
tracks B and C with m-antibody
Molecular data_2


This time, after M-antigen 1024 spoke pattern was printed on
the cd, R-antigen was used as free-surface saturating reagent
instead of BSA
Tracks B and C exposed to R-antibody, whereas tracks C and D
exposed to M-antibody
 When no antibody agents present, the homodyne signal is
the same as noise
 When both antibodies bind in track C, homodyne signal
diminishes, meaning that the heights cancel each other
Conclusion
 First adaptive interferometric BioCD for
molecular recognition was demonstrated
 Sensitivity = 0.3nm
 Higher sensitivity can be achieved by
 Narrowing the detection bandwidth
 Shifting to a reflection geometry that
will not be sensitive to density
fluctuations in the glass