Download Chip Scale Light Deflector Enables Solid

ULTRAFAST OPTICS
Chip Scale Light Deflector Enables
Solid-State Streak Camera
John E. Heebner, Chris H. Sarantos
and Susan M. Haynes
Swept spatial representation
of temporal waveform
Short pulse
pump beam
Serrated
gold mask
Pump activated
prism array
Temporal
waveform
C
Gu
d
la
id
S
Input
coupling lens
di
ing
s
ub
Focusing
lens
ng
er
lay
tr a
te
High
dynamic
range
camera
Output coupling lens
Single-shot recording
10,000
Intensity [counts]
S
cientists have used the rapid deflection of beams to provide a record
of events at many time scales. Conventional, electron-based streak cameras
represent the fastest embodiment of this
concept, but they are limited by spacecharge effects that blur the focused beam
when high signal amplitudes are present.
This forces a tradeoff between temporal
resolution and dynamic range. A streak
camera that deflects a beam of photons
would eliminate this tradeoff.
Unlike electron beams that are readily
manipulated via electromagnetic forces,
the sustained deflection of an optical
beam through many picosecond-scale
resolvable spots has been historically difficult to achieve. For each resolvable spot
of deflection in the far-field, the near-field
wavefront must be rapidly tilted by one
wave. Nonlinear optical mechanisms
based on the Kerr effect are ultrafast1 but
also ultraweak, making them impractical.
Optically excited carriers have a much
stronger influence on the refractive index
of a semiconductor. Due to a long-lived
(nanosecond-scale) electron-hole recombination time, they have often been
overlooked as a means for devising ultrafast optical switches. We demonstrate a
deflector concept that achieves picosecond
response, exploiting these strong refractive
index changes and actually benefitting
from long recombination times.
The device concept is illustrated in
the figure. A signal beam carrying a temporal waveform is coupled into a planar
waveguide. When the temporal region
of interest is fully contained, a normally
incident pump beam patterned by a serrated mask imprints a one-dimensional
array of prisms in the waveguide core.
The prisms are generated via optical
nonlinearities (plasma loading, band
filling and bandgap shrinkage),2 which
turn on rapidly and remain latched for
the sweep.
Signal input
beam
1,000
100
10
0
0 10 20 3040 50
Time [ps]
The SLIDER concept is based on the optically induced deflection of an optical signal injected into a planar slab waveguide. The deflection is caused by a sequential array of prisms
that are initially nonexistent, then simultaneously created by a sub-ps pump pulse while the
signal is in transit. (Inset) A single-shot recorded trace of a ring-down test signal consisting
of 1-ps impulses separated by 10 ps.
The signal then experiences a distributed deflection that is finely discretized
over a large number of prisms. Because
the prism array is created while the signal
is in transit through it, later portions
of the signal propagate through more
prisms. The signal thus deflects in linear
proportion to its time delay. The swept
beam is focused onto a conventional camera for recording. We term this concept
serrated light illumination for deflectionencoded recording (SLIDER).3
To test this concept, we fabricated a
planar waveguide with a GaAs guiding
layer, surrounded by AlGaAs claddings.
A Ti:sapphire regenerative amplifier provided an above-bandgap (800 nm) 150-fs
pump pulse that was spatially formatted
to a uniform fluence of 65 μJ/cm2. An
optical parametric amplifier was used
to generate a below-bandgap (950 nm)
signal that was spectrally filtered to 1.4
nm (1 ps transform-limited). A ring-down
test pattern was then generated by a
Gires–Tournois cavity with a round-trip
time of 10 ps. SLIDER enabled a singleshot recording of 1-ps impulses resolved
at 2.5 ps across a record of 50 ps. The
dynamic range of the measurement was
3,000:1, limited by the camera.
The SLIDER technique is potentially
scalable to a high dynamic range (104)
across hundreds of picoseconds, making
it a credible replacement technology for
conventional streak cameras. t
John Heebner ([email protected]) and Susan Haynes
are with the Science and Technology Directorate
of the Lawrence Livermore National Laboratory in
Livermore, Calif., U.S.A. Chris Sarantos is now with
Phoebus Optoelectronics. This work was performed
under the auspices of the U.S. Department of Energy
under contract DE-AC52-07NA27344.
References
1. J. Hübner et al. Opt. Lett. 30, 3168 (2005).
2. B R. Bennett et al. IEEE J. Quantum Electron. 26, 113
(1990).
3. C.H. Sarantos and J.E. Heebner. Opt. Lett. 35, 1389
(2010).
December 2010 | 51
ULTRAFAST OPTICS
Versatile Airy–Bessel
Light Bullets
Andy Chong, William H. Renninger, Demetrios N. Christodoulides
and Frank W. Wise
A
52 | OPN Optics & Photonics News
Axicon
ug
x
Position [mm]
y
-500
XC signal [a.u.]
ny localized wave packet tends to
broaden in space and time under
the combined actions of two universal physical processes: diffraction and
dispersion. Researchers have explored
strategies to overcome the often undesirable consequences of these two effects
since the early days of lasers.
In the early 1990s, nonlinear schemes
based on three-dimensional (3-D) soliton
solutions (or soliton bullets) have been
suggested as a means of overcoming this
problem.1 However, 3-D soliton bullets
are by nature highly unstable and thus
tend to disintegrate during propagation.
In addition, they require high power levels that are on many occasions detrimental, especially in biomedical applications.
Researchers have also reported 3-D linear
wave packets that can propagate in the
form of “O,” “X,” and Bessel-X waves.
Apart from the fact that such solutions involve complex spectra and often
require diffraction/dispersion equalization, their utility is severely limited by
the need to tailor the properties of the
wave packet precisely to material properties. Ideally, such light bullets should be
versatile and able to withstand any level
of diffraction and dispersion, irrespective of whether the latter is normal or
anomalous. The only way to circumvent
this problem is to disengage space and
time. To overcome these limitations, we
have recently explored a new approach
based on the unique properties of onedimensional Airy wave packets.3
Earlier this year, we reported the
first observation of a new class of
versatile three-dimensional linear light
bullets,4 which combine Bessel beams5
in the transverse plane with temporal
Airy pulses. These Airy–Bessel waves
are impervious to both dispersion and
diffraction effects. Thus, their evolution
does not depend critically on the mate-
1.0
0.5
0
-0.5
z
500
0
-500 0 500
Position [mm]
-500
0 500 1,000
Delay [fs]
(Top) Schematic to generate Airy-Bessel wave packets. (Bottom) Propagation of an AiryBessel wave packet A initial spatial and temporal profile, B profiles after propagation
through 3.3 LR and 1.8 Ld, C 5.4 LR and 3.6 Ld, D 7.5 LR and 5.4 Ld. (LR =diffraction length of
a beam with a diameter of 180mm, Ld =dispersion length of a 100 fs pulse).
rial in which they propagate. Temporal
self-healing and free acceleration, which
are signatures of Airy packets, were
also demonstrated. These spatiotemporal waves are possible for either sign
of dispersion and do not require any
diffraction/dispersion equalization. This
family of robust and versatile waves can
be used in ultrafast probing or imaging
in media with poorly known, or even
time-varying, properties.
The figure shows the generation of
the Airy-Bessel wave packets. An Airy
pulse with a Gaussian spatial profile is
generated by impressing a large cubic
spectral phase on a femtosecond pulse
from a mode-locked laser. The spatial
profile is then simply converted from a
Gaussian to a Bessel beam by an axicon
lens. The propagation of an Airy-Bessel
light bullet in a dispersive glass is shown.
Its nonspreading nature was verified
for propagation through seven diffraction lengths and five dispersion lengths
of the glass. The peak intensity of the
corresponding Gaussian pulse and beam
would fall by a factor of 250 during
propagation through the same material.
We expect that these versatile localized
wave packets will find a wide range of
scientific applications. t
Andy Chong ([email protected]), William H. Renninger and Frank W. Wise are with the department
of applied physics at Cornell University, Ithaca,
N.Y., U.S.A. Demetrios N. Christodoulides is with
the College of Optics/CREOL, University of Central
Florida, Orlando, Fla., U.S.A.
References
1. Y. Silberberg. Opt. Lett. 15, 1282-4 (1990).
2. Sonajalg et al. Opt. Lett. 22, 310-12 (1997).
3. G.A. Silviloglou et al. Opt. Lett. 32, 979-81 (2007).
4. A. Chong et al. Nat. Photonics 4, 103-6 (2010).
5. J. Durnin et al. Phys. Rev. Lett. 58, 1499-1501
(1987).
www.osa-opn.org