Download Holographic storage and high background imaging

Holographic storage and high background imaging using photorefractive
multiple quantum wells
R. Jones,a) S. C. W. Hyde, M. J. Lynn, N. P. Barry, J. C. Dainty, and P. M. W. Frenchb)
Femtosecond Optics Group, Physics Dept., Imperial College, London SW7-2BZ, United Kingdom
K. M. Kwolek and D. D. Nolte
Department of Physics, Purdue University, West Lafayette, Indiana 47907-1396
M. R. Melloch
School of Electrical and Computer Engineering, Purdue University, West Lafayette, Indiana 47907-1396
~Received 23 May 1996; accepted for publication 16 July 1996!
We report holographic, real time, depth-resolved image acquisition, storage, and reconstruction in
photorefractive GaAs/AlGaAs multiple quantum wells under high background radiation conditions.
Reconstructed images of 50 mm transverse and depth resolution have been achieved using this
device as a coherence gate to image through 9 mean free paths of turbid scattering medial © 1996
American Institute of Physics. @S0003-6951~96!04239-8#
Multiple quantum wells ~MQW! have many advantageous nonlinear optical properties. Compared to bulk semiconductors, the strong, narrow excitonic absorption features
lead to a large photorefractive nonlinearity.1 The nonlinearity
is due to the Franz–Keldysh ~or field broadening! effect2
when an electric field is applied parallel to the wells, or the
quantum confined stark effect,3 when a perpendicular electric
field is applied. Typically semiconductors have shown sensitivities many orders of magnitude greater than oxide
materials,4 while their high carrier mobilities lead to very
fast response times.5 Due to the small thickness of MQWs,
wave-mixing is done in the Raman–Nath regime which relaxes the Bragg constraints that are present for volume holograms. Low dark conductivity is typically achieved by proton bombardment which allows the construction of twodimensional ~2-D! imaging devices without the need for
pixellation. In this letter, we describe the use of GaAs/
AlGaAs MQW devices as a fast nonlinear medium in which
holograms have been written using femtosecond pulse in the
near infra-red ~NIR!. We apply these MQW devices to coherence gating for imaging through turbid media which has
potential applications for medical imaging.
The transverse field photorefractive MQW devices were
grown at Purdue University in a Varian GEN-II molecular
beam epitaxial chamber on a GaAs substrate. The 10% Al
barrier structure, grown at 600 °C, consists of an
Al0.5Ga0.5As stop-etch layer beneath a 1.5 mm 100 Å
Al0.1Ga0.9As/75 Å GaAs MQW region sandwiched between
Al0.1Ga0.9As buffer layers. Deep defects were introduced by
proton implantation at a double dose of 1012/cm2 at 160 keV
and 531011/cm2 at 80 keV to make the device semiinsulating throughout the MQW region. The defects provide
traps for the formation of photoreactive space-charge gratings. The device growth and processing was identical to Ref.
6 which optimized the diffraction efficiency using microcavity resonant enhancement.
The MQW was used in the transverse Franz–Keldysh
geometry.1,2 This requires a large applied voltage, of the or-
der of 1 kV, to be applied in the plane of the quantum wells,
parallel to the optical grating vector being written. Photocarriers are preferentially excited in the bright regions and drift
due to the applied field. Carriers are retrapped predominantly
in the dark regions where they cause a screening field, which
opposes the applied field. Charge generation and retrapping
occurs until the screening field produced by the trapped
charge exactly balances the applied field and charge equilibrium ensues. A large applied field will produce a broadening
of the exciton peak via field-ionization. Hence the optical
interference pattern is mapped into a refractive index/
absorption coefficient distribution.
The setup used to record/reconstruct holograms is shown
in Fig. 1. The holograms were written using a mode-locked
Ti:sapphire laser ~Spectra-Physics Tsunami!, which produced 100 fs pulses with a coherence length of '85 mm at a
wavelength of 830 nm. The holography setup consisted of a
Mach–Zender interferometer, with the object in one arm,
and an adjustable delay in the reference arm. A polarizing
beam splitter and two wave plates were used to separate the
incident beam from the image beam reflected from the object. The polarization used to record holograms was perpendicular to the plane of the writing beams. The object was
imaged in reflection onto the MQW device through a demagnifying telescope. The reference and probe beam were both
a!
FIG. 1. Showing the experiment setup used to record and reconstruct holograms in a MQW device.
Electronic mail: [email protected]
Electronic mail: [email protected]
b!
Appl. Phys. Lett. 69 (13), 23 September 1996
0003-6951/96/69(13)/1837/3/$10.00
© 1996 American Institute of Physics
1837
Downloaded¬17¬Nov¬2004¬to¬155.198.9.127.¬Redistribution¬subject¬to¬AIP¬license¬or¬copyright,¬see¬http://apl.aip.org/apl/copyright.jsp
FIG. 4. Showing the frequency response of the MQW device.
FIG. 2. Showing ~a! image of test chart as viewed directly through 9 MFP
double pass; ~b! 52 mm transverse resolution as viewed through 9 MFP of
turbid media; ~c! speckle reduction of ~b! by vibrating the reference beam.
weakly focused using 500 mm focal length lenses. The writing beams were separated by an angle of 1.5°, corresponding
to an average grating spacing of 32 mm.
A diode-pumped Cr:LiSAF continuous wave ~cw! laser,
similar to that described in Ref. 8, was used to provide a
probe beam to reconstruct the holograms. This laser was
tuned to the exciton peak wavelength ~;850 nm! to maximize diffraction efficiency. The angle between the probe
beam and the normal to the MQW device was 32°. The hologram was reconstructed in a transmission geometry, diffracting into multiple orders as dictated by the Raman–Nath
regime. The zeroth order was blocked using a spatial filter,
and the 11 or 21 order imaged onto a standard CCD camera
~Pulnix PE30! using a 200 mm focal length lens.
In all the following experiments, the illumination incident upon the MQW device was kept roughly constant with
intensities of 1 mW/cm2 in the signal beam and 1.25
mW/cm2 in the reference beam used to write the hologram.
The probe beam had an intensity of 1.25 mW/cm2 and an
FIG. 3. Showing ~a! photograph of 3D test object; ~b! image of object
viewed directly through 8 MFP of turbid media; ~c!–~e! depth resolved
images of test object, showing 100 mm depth resolution; ~f! 3D reconstruction of object.
applied voltage of either 1.5 or 2 kV was used.
The transverse resolution was measured by imaging a
USAF test chart onto the MQW device in a 1:1 imaging
geometry. The reconstructed image showed transverse resolution of 50 mm. In order to increase the detectability of low
intensity images, the constant background field was subtracted from the reconstructed CCD image. This allowed the
removal of unwanted light scattered from fixed inhomogeneities in the MQW device.
The depth resolution of the reconstructed image was
measured by imaging a 3D object, which consisted of a set
of cylindrical steps with diameters ranging from 1 to 5 mm
in 1 mm increments, and a step size of 100 mm. This object
was unpolished and gave no significant specular reflection,
but scattered back ;8% of the incident light. Since the writing laser had only a short coherence length ~;85 mm! the
arms of the interferometer are only matched at a specific
depth within the object, and so depth-resolved imaging is
possible. This technique is known as light in flight
holography.9 The depth resolution is determined by the coherence length and the angle between the writing beams.10
Due to the limiting aperture ~1 mm! of the MQW device,
large objects were imaged into the MQW device using a 3:1
demagnifying telescope. By altering the delay in the reference arm, different steps of the object could be isolated in the
reconstruction. The depth resolution was accurately measured to be 52 mm by determining the range of delays of the
reference arm over which a single step of the object remained visible.
One application of real-time holographic recording is to
image through turbid media via coherence gating.10–12 In
general, scattered photons are delayed relative to unscattered,
ballistic photons due to increased propagation path, and
therefore arrive outside the coherence gate of the reference
beam. The hologram is written only with unscattered ballistic
light, thereby providing image information in spite of the
large scattered light background. In these experiments a
weak spatial filter, 1 mm in radius, was employed in the
Fourier plane of the demagnifying telescope. This also
helped to reject some of the scattered light.
A 1 mm thick scattering cell, filled with a suspension of
0.46 mm polystyrene spheres, used as the turbid medium,
was placed before the test object. Incident light passed
through the scattering cell before reflecting off the object and
again passing through the cell to be imaged onto the MQW
device. There was about 2.8 W/cm2 power incident upon the
1838
Appl. Phys. Lett., Vol. 69, No. 13, 23 September 1996
Jones et al.
Downloaded¬17¬Nov¬2004¬to¬155.198.9.127.¬Redistribution¬subject¬to¬AIP¬license¬or¬copyright,¬see¬http://apl.aip.org/apl/copyright.jsp
scattered cell which is within the ANSI safety limit dictated
for exposure of biological tissue. The transverse resolution
was measured using a USAF test chart. With a 0.4% by
volume suspension of scattering spheres, images of 52 mm
were clearly reconstructed as shown in Fig. 2~b!. Using Mie
theory13 this concentration of spheres was calculated to have
a scattering coefficient ( m s ) of 4.46 mm21 and an anisotropy factor (g) of 0.72. Absorption is assumed to be negligible. The 0.4% scattering solution therefore corresponds to
nine scattering mean free paths ~MFP! in the double pass.
Speckle presented a significant problem for these experiments. By vibrating one of the mirror mounts in the reference arm, some of this speckle was averaged out during the
holographic exposure and higher quality images were obtained as shown in Fig. 2~b!.
The USAF test chart was replaced by the threedimensional ~3-D! test object @see Fig. 3~a!#, and again layers
separated by 100 mm could be resolved. Figures 3~c!–3~e!
show images of the object located behind a scattering solution of 3.9 MFP thickness. The limited field of view was
again due to the small ~1 mm! aperture between electrodes.
The scattering solution in this case comprised a suspension
of 0.35%, 0.46 mm polystyrene spheres, which Mie theory
predicts to have m s 53.9 mm21, and g50.72. By altering
the delay in the reference arm, the depth resolution was measured more precisely to be 53 mm. A computer generated
reconstruction of the 3D object is shown in Fig. 3~f!.
The response time of the MQW was estimated by chopping the reference beam and synchronously detecting at the
chopping frequency in the reconstructed image plane using a
photo-diode and a lock-in amplifier. When the chopping frequency becomes comparable to the response time, the grating formation in the MQW device will no longer be able to
follow the modulation and the lock-in signal will decrease.
This experiment was done imaging the 3-D object through
the scattering medium with incident powers as stated previously. As can be seen in Fig. 4, grating formation in the
MQW device continued to follow the modulation for frequencies of up to 2.5 kHz which corresponds to a response
time ,0.4 ms. This is well above normal video rate ~30
frames/s! and so provides real time image acquisition.
In conclusion, we have demonstrated real time holographic image recording in a GaAs/AlGaAs semi-insulating
MQW device with transverse and depth resolution of ;50
mm. We have used this device to acquire images through
turbid media by means of coherence gating. A 50 mm transverse resolution has been obtained through 9 MFP of scattering medium, and depth resolutions of 53 mm achieved
through 8 MFP of scattering medium. The response time of
the MQW devices has been measured to be less than 0.4 ms.
To the best of our knowledge, this technique represents the
fastest method, yet reported, of whole field, depth-resolved
image acquisition through turbid media.
Funding for this research was provided by the UK Engineering and Physical Sciences Research Council ~EPSRC!.
R. Jones acknowledges an EPSRC studentship. S. C. W.
Hyde acknowledges and EPSRC CASE studentship supported by Kodak, Ltd. M. J. Lynn acknowledges an EPSRC
CASE studentship supported by Scientific Generics, Ltd.
1
D. D. Nolte and M. R. Melloch, in Photorefractive Effects and Materials,
edited by D. D. Nolte ~Kluwer, Dordrecht, 1995!, Chap. 7.
2
Q. Wang, R. M. Brubaker, D. D. Nolte, and M. R. Melloch, J. Opt. Soc.
Am. B 9, 1626 ~1992!.
3
D. A. B. Miller, D. S. Chemla, and T. C. Damen, Phys. Rev. Lett. 53,
2173 ~1984!.
4
D. D. Nolte, D. H. Olson, G. E. Doran, W. H. Knox, and A. M. Glass, J.
Opt. Soc. Am. B 7, 2217 ~1990!.
5
A. Partovi, A. M. Glass, T. H. Chiu, and D. T. H. Liu, Opt. Lett. 18, 906
~1993!.
6
K. M. Kwolek, M. R. Melloch, and D. D. Nolte, Appl. Phys. Lett. 65, 385
~1994!.
7
D. D. Nolte and K. M. Kwolek, Opt. Commun. 115, 606 ~1995!.
8
R. Mellish, N. P. Barry, S. C. W. Hyde, R. Jones, P. M. W. French, J. R.
Taylor, C. J. van der Poel, and A. Valster, Opt. Lett. 20, 2312 ~1995!.
9
N. Abramson, Opt. Lett. 3, 121 ~1978!.
10
S. C. W. Hyde, N. P. Barry, R. Jones, J. C. Dainty, and P. M. W. French,
Opt. Lett. 20, 2330 ~1996!.
11
A. Rebane and J. Feinberg, Nature ~London! 351, 378 ~1991!.
12
M. R. Hee, J. A. Izatt, J. M. Jacobson, J. G. Fujimoto, and E. A. Swanson,
Opt. Lett. 18, 950 ~1993!.
13
A. Ishimaru, Wave Propagation in Random Media ~Academic, New York,
1978!.
Appl. Phys. Lett., Vol. 69, No. 13, 23 September 1996
Jones et al.
1839
Downloaded¬17¬Nov¬2004¬to¬155.198.9.127.¬Redistribution¬subject¬to¬AIP¬license¬or¬copyright,¬see¬http://apl.aip.org/apl/copyright.jsp