28.01.2015 Views

Three - University of Arkansas Physics Department

Three - University of Arkansas Physics Department

Three - University of Arkansas Physics Department

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

September 1,1996 I Vol. 21, No. 17 I OPTICS LETTERS 1333<br />

Self-trapping <strong>of</strong> planar optical beams by use<br />

<strong>of</strong> the photorefractive effect in 1nP:Fe<br />

M. Chauvet, S. A. Hawkins, and G. J. Salamo<br />

<strong>Department</strong> <strong>of</strong> <strong>Physics</strong>, <strong>University</strong> <strong>of</strong> <strong>Arkansas</strong>, Fayetteville, <strong>Arkansas</strong> 72701<br />

M. Segev<br />

<strong>Department</strong> <strong>of</strong> Electrical Engineering and Advanced Center for Photonics and Optoelectronic Materials,<br />

Princeton <strong>University</strong>, Princeton, New Iersey 08544<br />

D. F. Bliss and G. Bryant<br />

U.S. Air Force, Rome Laboratory, Hanscom Air Force Rase, Massachusetts 01 731<br />

Received January 24, 1996<br />

We demonstrate what we believe to be the first experimental observation <strong>of</strong> self-trapping and self-deflection <strong>of</strong><br />

a planar optical beam by the photorefractive effect in a semiconductor. The semiconductor material is indium<br />

phosphide doped with iron. We show that the observed focusing and defocusing effects follow the component<br />

<strong>of</strong> the two-wave-mixing space charge field that is in phase with the intensity pattern, whereas the spatial beam<br />

deflection effects follow the 90"-shifted component. O 1996 Optical Society <strong>of</strong> America<br />

Spatial solitons in photorefractive materials have<br />

been the subject <strong>of</strong> recent intere~tl-~ Compared<br />

with Kerr spatial198 solitons, the most distinctive<br />

features <strong>of</strong> spatial solitons are that they are observed<br />

at low light intensitie~~.~~%nd that trapping<br />

occurs in both transverse dimension^.^,^ Until now,<br />

photorefractive spatial solitons have been observed<br />

in the tungsten bronze ferroelectric o~ides~'~ and<br />

in the nonferroelectric sillenite oxide^.^ Several<br />

reasons have led us to carry out similar experiments<br />

in the photorefractive semiconductor crystal<br />

1nP:Fe. First, this material is sensitive in the range<br />

<strong>of</strong> the near-infrared wavelengths used in optical<br />

telecommunications. Second, the photorefractive<br />

effect in semiconductors has the advantage <strong>of</strong> a<br />

faster response time than that observed in either the<br />

tungsten bronze or the sillenite crystal materials.<br />

Third, the possibility <strong>of</strong> monolithic integration with<br />

other optoelectronic components (lasers, detectors) is<br />

also attractive.<br />

We experimentally demonstrate in this Letter what<br />

we believe to be the first self-trapping <strong>of</strong> optical beams<br />

in a semiconductor, using the photorefractive effect.<br />

The self-trapping is observed at steady state with a low<br />

light intensity (50 mW/cm2) and with a moderate d.c.<br />

applied field (9 kV/cm) in bulk 1nP:Fe. The output<br />

diameter is observed to be trapped at the input value<br />

after propagation <strong>of</strong> both 5 and 10 mm.<br />

To observe the self-trapping effects, we focus a<br />

continuous beam at 1.04 or at 1.3 pm with a 70-mm<br />

focal-length cylindrical lens onto the entrance face <strong>of</strong> a<br />

1nP:Fe crystal. A beam pr<strong>of</strong>ile system and an imaging<br />

lens are used to analyze the beam shape. The light<br />

beam is linearly polarized along the (110) direction and<br />

propagates along the (110) direction. Absorption is<br />

-1.5 cm-' at the 1.04-pm wavelength and0.15 cm-' at<br />

1.3 pm. A d.c. field (Eo) is applied between the (001)<br />

faces. The temperature <strong>of</strong> the crystal is stabilized<br />

at 297 K.<br />

The resonance behavior <strong>of</strong> the two-wave-mixing<br />

(TWM) gain in 1nP:Fe has been characterized both<br />

experimentally and theoreti~ally.~*l~ In this model<br />

electrons are dominantly thermally excited from Fe2+<br />

to the conduction band, whereas holes are dominantly<br />

optically excited from Fe3+ to the valence band. The<br />

resonance in the gain as a function <strong>of</strong> intensity occurs<br />

at an intensity I,,, for which the optical excitation <strong>of</strong><br />

holes is equal to the thermal excitation <strong>of</strong> electrons.<br />

In addition, the relative phase between the intensity<br />

spatial pattern and the space-charge-field pattern is<br />

also intensity and temperature dependent. The reso-<br />

\. , - I<br />

0 40 80 120 160 200<br />

Intensity (mWlcm2)<br />

Fig. 1. Real and imaginary parts <strong>of</strong> the normalized TWM<br />

space charge field in 1nP:Fe at 1.06 pm calculated with<br />

formula 18 from Ref. 10. m is the modulation rate <strong>of</strong><br />

the intensity grating. Eo = 10 kV/cm; T = 297 K;<br />

grating spacing = 10 pm; [Fez+] nTo = 0.5 x 1016 ~rn-~;<br />

[Fe3+] p~o = 6 X 1016 ~rn-~; and hole and electron<br />

photoionization cross sections an0 = 6 X 10-l8 cm2,<br />

upo = 1 X lo-'" cm2. I,,, = 38 mW/cm2.<br />

O 1996 Optical Society <strong>of</strong> America

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!