+ All Categories
Home > Documents > Self-pumped forward phase conjugator

Self-pumped forward phase conjugator

Date post: 01-Oct-2016
Category:
Upload: baruch
View: 218 times
Download: 2 times
Share this document with a friend
3

Click here to load reader

Transcript
Page 1: Self-pumped forward phase conjugator

1462 OPTICS LETTERS / Vol. 17, No. 20 / October 15, 1992

Self-pumped forward phase conjugator

Nery Strasman, Ron Daisy, and Baruch Fischer

Department of Electrical Engineering, Advanced Optoelectronics Center, Technion-Israel Institute of Technology,Haifa 32000, Israel

Received June 11, 1992

A self-pumped forward phase-conjugate mirror based on nonlinear wave mixing in photorefractive media isdemonstrated. A single input beam to a BaTiO3 crystal excites an oscillation of pump beams between surfacesof the crystal and a forward phase-conjugate beam. The phase conjugation is carried on the first-orderdiffraction from the grating induced by the input and the self-generated pump beams.

One of the most common methods for obtainingphase conjugation is based on nonlinear four-wavemixing." The built-in solution for the phase-matching requirement made this technique superiorto an earlier suggestion for phase conjugation byusing three-wave mixing.5 The attractiveness andsimplicity of the four-wave-mixing scheme becameeven more evident when the self-pumped phase con-jugators in photorefractive media were developed'and the need for external pump beams was relaxed.In these devices, the pump beams are self-generatedwhen feedback is added by external mirrors (in thelinear, semilinear, and ring conjugators3), by thecrystal surfaces (in the cat4 and linear devices3), orwithout any mirrors (in the double phase-conjugatemirror6). In all these four-wave-mixing techniquesa backward phase conjugation is generated. Thereis another way to obtain phase conjugation byusing the first-order diffraction of the gratingsin two-beam coupling.7 -9 This method gives aforward-propagating phase conjugation and needsthin gratings (Raman-Nath regime) to permit thefirst-order diffraction. In this Letter we demonstratesuch a forward phase conjugator that does not needthe external second beam for two-beam coupling.The first-order diffraction by the grating that givesthe forward phase conjugation is induced by theinterference of the input beam and a self-generatedpump beam.

The experimental configuration is schematicallyshown in Fig. 1. Beam 1, carrying spatial informa-tion and having an extraordinary polarization, entersa photorefractive crystal that has a 450 orientationof the c axis with respect to the input facets (a 45°-cut crystal). Amplified light in a broad angle (theFanning effect) occurs, and eventually an oscillationbuilds up between the opposite crystal facets. Thefeedback for the oscillation is supplied by Fresnel re-flections of the facets. In fact, this part of the processprovides a compact self-pumped linear (or semilinear)phase conjugator2'3 in which the feedback is providedby the crystal surfaces instead of by external mirrors.A similar oscillation was demonstrated with LiNbO3by Odoulov and Soskin.'0 The four-wave mixing ofthe beams produces a backward-propagating phaseconjugation (beam 1') of input beam 1. The new re-

sult in the present research is the second stage inthe beam-coupling dynamics. The grating, inducedby the input beam 1 and the self-generated pumpbeam 2, produces a first-order diffracted beam 3. Inthis three-beam coupling process beam 3 is a for-ward phase conjugation of the input beam.7 "9 Aphotograph of the beam patterns inside the crystalis shown in Fig. 2, and the forward phase-conjugateimage carried on beam 3 is shown in Fig. 3. Theangle between beams 2 and 3, which has to be ap-proximately the same as the angle between beams 1and 2, seems in Fig. 2 to be slightly smaller notonly owing to illusion that results from the shiftedbeams' intersection (the buildup of beam 3 is at theright side of the grating zone), but there is also areal angular reduction of approximately 3% owing tothe crystal anisotropy, for which the extraordinaryrefractive index is higher as the beam direction movestoward the c axis.

We now turn to a more detailed description of theeffect through the wave-mixing processes. As men-tioned above, we have two stages in the dynamics.In the first part, the pump beams 2 and 2' build up(see Fig. 4) between the two crystal surfaces, andthe linear phase conjugator starts to operate. Infact, in the analysis below we consider it a semilin-ear self-pumped conjugator having only one feedbackmirror since the Fresnel reflectivity of the surfacesis relatively low (M = 0.17), so that the front surfacecontributes only -M 2 to the pump's feedback. Nev-ertheless, this small feedback, together with the factthat it is generated by a flat surface, causes signifi-cant fluctuations in the intensities of the backwardand forward phase-conjugate beams. The inducedgrating in the buildup of the semilinear conjugator

BaTiO3 , 3

Fig. 1. Schematic of the self-pumped forward phaseconjugator.

0146-9592/92/201462-03$5.00/0 © 1992 Optical Society of America

Page 2: Self-pumped forward phase conjugator

October 15, 1992 / Vol. 17, No. 20 / OPTICS LETTERS 1463

Fig. 2. Photograph of the beams inside the crystal whenthe self-pumped forward phase conjugator operates, show-ing the input beam 1 (bottom pattern), the pump beams 2and 2' (middle pattern), and the forward phase-conjugatebeam 3 (top pattern).

beams 2 and 3 adds a new component (1/IO)A3A2 +c.c. to the grating g in relation (1), with the samegrating wave number, and another grating, which isdue to the coupling of beams 1 and 3, with a wavenumber that is larger by a factor of 2. Nevertheless,we can neglect these two contributions since theintensity of beams 3 is much smaller than that ofbeams 1 and 2, as we show below.

A simple quantitative derivation of the beams' in-tensities Ii = IA 12 can be done by using our assump-tions of treating the wave-mixing configuration asa semilinear phase conjugator with the buildup ofpumps and thin gratings from which the first-orderbeam 3 is diffracted. Then we can use the knownsolution3 for the beam intensities in the semilinearself-pumped phase conjugator and the double phase-conjugate mirror. In particular, we want to findthe grating's diffraction transmissivity that is thetransmissivity of a double phase-conjugate mirror(Eq. 4.17 of Ref. 3),

I2(1) _ II,(O)MlO) I21(1)

(3)

This can be obtained by using Eq. 4.10 in Ref. 3for the reflectivity of the semilinear phase-conjugatemirror, defined by R = I (O)/II(O) (where 0 and 1denote the input and output sides, respectively, ofthe coupling zone whose length is 1) and the surfacefeedback reflectivity M (it is defined as M2 in Ref. 3),

image carried on

is proportional to

g = I (Al*A2 + A1,A2 1) + c.c. x I Al*A2 + c.c.,

(1)

where Ai is the complex amplitude of beam i andIo = ZIAi 12. We also used the relations Al, cc Al"' andA21 °c A *. The grating turns out to be thin, in theRaman-Nath regime. The reason is the strong de-pletion of the input beam that confines the beaminteraction zone and the grating near the input sur-face. Then the Bragg condition can be violated andthe first-order diffraction of beam 3 is allowed. Itsk vector is k3 = 2k2 - k, + Ak, where Ak is the smallphase mismatch involved in such high-order diffrac-tion processes. The effective width of the grating,which is measured in terms of its period, can be re-duced by taking a small angle between the input andthe oscillating beams. This improves the diffractionefficiency of beam 3, while still maintaining a highenough coupling constant, owing to the 45° cut of thecrystal. Then

(2)

and if the self-induced pump A2 is spatially uniform,A3 is the forward phase conjugation of the inputAl. After beam 3 is generated (its experimental timeevolution is shown in Fig. 4) the coupling between

T = (R ) 2 =1 - a2

M(1 + a2) - 2M"12[a2(M + 1) - 1]12

(4)

Here a is a constant derived from the photorefractivecoupling strength (yl) by3 tanh[-(yl/2)a] = a. Thethreshold of the semilinear conjugator is given by(Eqs. 4.12 and 4.13 in Ref. 3)

NI+ )1"2 - 1](ylX (1± M)"2 In (1 + M)1"2 + 1]'

L1004

Ce S-

i a

Time [sec]

Fig. 4. Buildup behavior of the output intensities 11(curve 1), I2 (curve 2), and I3 (curve 3). In this ex-periment, the input was a 5-mW He-Ne laser beam (at633 nm) that had an angle of 10° (inside the crystal)with respect to the input surface normal.

Fig. 3.beam 3.

Forward phase-conjugate

A 3 c gA 2 ih IA 22Al*XIo

Page 3: Self-pumped forward phase conjugator

1464 OPTICS LETTERS / Vol. 17, No. 20 / October 15, 1992

which in our experiment gives (yl)t = -3.5 for M =0.17 [compared with (yI)t = -2.49 for maximum feed-back with M = 1 that can be obtained with an exter-nal mirror or by high-reflection coating of the crystal].It is important to emphasize that all the above rela-tions are derived for the steady-state behavior andneglect absorption in the crystal.

At this point, the experimental data of the outputintensities in Fig. 4 can be examined. The figuregives II(t = 0), which can be taken as I,(l = 0)-II(0), and the steady-state values of the beams inten-sities, from which we obtain T = I2(1)1I1(0) 0.18.Then we can also have a good estimate of the forwardphase-conjugate beam intensity, relying on our aboveexplanations that the same grating responsible forthe diffraction I, - I2 gives the diffraction I2 - I3,if we assume that the coupling constant does notvary significantly for the slightly different beam di-rections. Consequently,

I3(1) : T(I2) _ T12(l) (5)2

where the angle brackets indicate an averageover the interaction length. Thus we have thediffraction efficiency from the previously obtained T(for the diffraction I, -- 12), I3(1)/I2(l) T/2 - 0.09.This is in reasonable agreement with the directexperimental steady-state value from Fig. 4, which

gives 13(1)/I2(1) 0.08. More experimental data forthe beams' intensities with slightly different inputangles, and therefore different coupling constants,showed similar agreement with the expected values.

In conclusion, we have demonstrated a self-pumpedforward phase-conjugate mirror. The demonstrationwas done with a photorefractive BaTiO3 crystal. Ananalysis was shown to give a good understanding ofthe wave-mixing process.

References1. R. W. Hellwarth, J. Opt. Soc. Am. 67, 1 (1977).2. J. 0. White, M. Cronin-Golomb, B. Fischer, and A.

Yariv, Appl. Phys. Lett. 40, 450 (1982).3. M. Cronin-Golomb, B. Fischer, J. 0. White, and A.

Yariv, IEEE J. Quantum Electron. QE-20, 12 (1984).4. J. Feinberg, Opt. Lett. 7, 486 (1982).5. A. Yariv, Appl. Phys. Lett. 28, 88 (1976); J. Opt. Soc.

Am. 66, 301 (1976).6. B. Fischer, S. Sternklar, and S. Weiss, IEEE J. Quan-

tum Electron. 25, 550 (1989); Appl. Phys. Lett. 50,483 (1987).

7. C. V. Heer and N. C. Griffen, Opt. Lett. 4, 239 (1979).8. D. C. Jones, S. F. Lynksytov, and L. Solymar, Opt.

Lett. 15, 935 (1990), and references therein.9. I. C. Khoo, IEEE J. Quantum Electron. QE-22, 1268

(1986); QE-23, 2020 (1987).10. S. G. Odoulov and M. S. Soskin, Sov. JETP Lett. 37,

289 (1983).


Recommended