December 1, 1997 / Vol. 22, No.

23 / OPTICS LETTERS

1751

Steady-state vortex-screening solitons formed in biased photorefractive media

Zhigang Chen, Ming-feng Shih, and Mordechai Segev
Department of Electrical Engineering and Center for Photonics and Optoelectronic Materials, Princeton University, Princeton, New Jersey 08544

Daniel W. Wilson, Richard E. Muller, and Paul D. Maker

Center for Space Microelectronics Technology, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California 91109 Received August 22, 1997 We report the observation of steady-state photorefractive vortex-screening solitons. As a singly charged circular vortex nested on a broad beam propagates through a biased strontium barium niobate crystal, it self-traps in both transverse dimensions despite the inherent anisotropy of the photorefractive nonlinearity. When the vortex beam is a doughnut-shaped narrow beam, it breaks up into two elongated slices (with a selfdefocusing nonlinearity) or into two focused filaments (with a self-focusing nonlinearity). We demonstrate the optical guidance of a probe beam in a circular waveguide induced by the self-trapped vortex. 1997 Optical Society of America

Optical vortex solitons form when self-defocusing balances diffraction, leading to a vortexlike phase structure that undergoes stationary propagation in nonlinear media. Vortex solitons were predicted1 and observed2 previously in Kerr-type media and, more recently, in saturable self-defocusing nonlinear media.3 In all these studies, the nonlinearity that gives rise to the solitons is isotropic and therefore supports dark soliton stripes in any direction in the transverse plane. Thus it was possible to generate isotropic structures such as dark soliton crosses and grids4 as well as circular vortex solitons.2,3 Following the first prediction of photorefractive spatial solitons,5 self-trapping of an optical vortex was also demonstrated in these inherently anisotropic nonlinear media, including in biased strontium barium niobate6 (SBN) and in unbiased photovoltaic LiNbO3 .7 In a biased SBN crystal a vortex soliton was observed only in a transient state, corresponding to quasi-steady-state photorefractive solitons that arise from the nonlocal nature of the photorefractive effect and exist in a time window characterized by the slow screening process of an external field applied to the crystal.6 In an unbiased LiNbO3 crystal the steady-state self-trapping of the optical vortex was supported by the bulk photovoltaic effect, which gives rise to photovoltaic solitons.8 Vortex solitons observed in the photorefractive materials6,7 differ in their physical origin from those observed in the research reported in Refs. 2 and 3. A typical property of photorefractive nonlinearity is the inherent anisotropy with respect to the transverse dimensions of beam trapping, which permits formation of a one-dimensional (1D) dark soliton stripe only in specific directions in the transverse plane.6 8 Besides quasi-steady-state and photovoltaic solitons, another important type of photorefractive soliton is the screening soliton.9 Screening solitons are also generated in biased photorefractive media, but unlike quasi-steady-state solitons they exist in steady
0146-9592/97/231751-03$10.00/0

state when the applied f ield is completely screened nonuniformly. In fact, most of the recent research on photorefractive solitons has focused on screening solitons. There is substantial experimental evidence that two-dimensional (2D) circular bright screening solitons indeed exist in biased photorefractive media,10 12 despite the anisotropy of the photorefractive nonlinearity. However, the existence of steady-state 2D dark(vortex-) screening solitons has been an intriguing topic that poses a challenge for both theoretical and experimental investigations. In fact, it is a controversial issue whether a singly charged vortex soliton can exist in an anisotropic photorefractive medium.13 In this Letter we report the observation of steadystate photorefractive vortex-screening solitons. We study self-trapping of a singly charged vortexscreening soliton and the waveguide induced by it in a biased SBN crystal under steady-state conditions. We show that the vortex soliton can be generated only when it is borne upon an inf inite beam and that it breaks up when it is nested upon a f inite doughnutshaped beam. Our experimental setup is similar to the one used in Ref. 14, which was designed for observation of 1D dark-screening solitons, except that here the l 4 step mirror is replaced by a helicoidal phase mask. The phase mask was fabricated by e-beam microlithography, by which the phase material poly(methyl methacrylate) was spun onto a SiO2 substrate to yield a 1-cm-diameter specimen whose thickness increased azimuthally by l 2. Such a mask is used to generate an optical vortex borne upon a beam of inf inite extent, which we f ind is necessary for observing steady-state vortex solitons. When a broad and collimated beam l 514 nm is launched onto the mask, a vortex with a helical 2p phase is produced in the ref lected beam. In our earlier experiments7 we confirmed that the vortex generated from our mask is of unit topological charge. The vortex-nesting beam
1997 Optical Society of America

1752

OPTICS LETTERS / Vol. 22, No. 23 / December 1, 1997

(otherwise uniform) is much larger than the center vortex core and is broad enough to cover the entire input face of the SBN crystal when it propagates along one of the crystalline a axes. (The crystal is 5.0, 11.7, and 5.3 mm along the a, a, and c axes, respectively.) The c axis is oriented parallel to the beam polarization and the applied field; thus the change in the refractive index is dominated by the r33 electro-optic coeff icient. Because the strength of the space-charge field (and thus of the refractive-index change) depends on the ratio between the optical intensity and the dark irradiance,9 it is desirable to control this ratio and to fine tune the nonlinearity. Therefore we add artificial dark irradiance by uniformly illuminating the crystal with an ordinarily polarized background beam, as is commonly done with screening-soliton experiments.10 12,14 To test the waveguide properties of the vortex soliton, we use a HeNe beam as a probe. Typical experimental results are presented in Fig. 1, which shows photographs and beam prof iles of the vortex beam at the input output faces of the crystal. The vortex is 18 mm (FWHM) at the input [Fig. 1(a)], and it diffracts to roughly 43 mm after propagating 11.7 mm through the crystal [Fig. 1(b)]. When a negative voltage (relative to the crystal c axis) is applied across the crystal, the medium turns into the self-defocusing type.9 Figure 1(c) is obtained when the applied voltage (between electrodes separated by 5.3 mm) is set to 2450 V and the intensity ratio (the ratio between the peak intensity of the vortex beam and the sum of the background illumination and the dark irradiance9) is set to 0.95. It is apparent from Fig. 1 that the vortex is self-trapped in both transverse dimensions and maintains a circular shape. Because the vortex is self-trapped to its initial size and because this self-trapping persists in steady state, we interpret this self-trapped vortex as a vortex-screening soliton. The experimental condition is critical for generating circular-vortex solitons. If the values of applied voltage or intensity ratio, both of which control the nonlinearity, are not appropriate (i.e., if they deviate considerably from those that support a circular-vortex soliton), the vortex seems to self-trap into an elliptical rather than into a circular shape, in a manner similar to the cases of 2D bright-screening solitons11 and of photovoltaic vortex solitons.7 Large deviations in these parameters lead to regular diffraction if the resulting nonlinearity is not high enough, or to modulation instability if it is too high, similar to effects observed with 1D dark-screening solitons.14 As we pointed out above; the photorefractive nonlinearity is inherently anisotropic. To illustrate this we launch into the crystal as broad beam containing a dark cross, which is generated simply by insertion of a wire cross as an amplitude mask in the input beam. Figure 2 shows that, with the application of a bias field (parallel to the c axis as in Fig. 1), the dark stripe evolves into a Y-junction structure in the direction parallel to the c axis, owing to the even initial condition,14 whereas in the perpendicular direction the dark stripe diffracts normally as if no nonlinearity were present. Such a preference in the direction of self-trapping does not occur in isotropic nonlinear me-

dia.4 Because of this unique feature, 1D dark soliton stripes can exist in photorefractive media only when the direction of intensity variation, which is also the direction of the internal space-charge field, employs a sufficiently large term from the electro-optic tensor and thus results in the refractive-index change that is necessary for self-trapping.5 9 However, when a 2D circular beam (bright or dark) illuminates the crystal, the electric f ield bends, i.e., its vector direction varies, in the regions of transition between dark and bright. As a result, each vector component in these regions varies in both transverse directions. Thus the c component of the space-charge field varies in both transverse directions and gives rise to an index change that induces a 2D waveguide that accounts for the 2D self-trapping. The experimental results shown in Fig. 1 imply that the c component of the f ield has a roughly circular symmetry at the vicinity of the vortex core. Next we show the difference between a vortex borne upon a finite beam and that borne upon an inf inite beam. In our experiments we found that a singly charged vortex carried by a f inite beam of narrow extent cannot evolve into a steady-state vortex soliton when the crystal is biased; rather, the vortex stretches asymmetrically as its carrying beam breaks up into two pieces. This result occurs for a wide range of parameters (intensity ratio and voltage) and is a particularly

Fig. 1. Propagation of a vortex nested upon an inf inite beam through an 11.7-mm long SBN crystal: (a) input, (b) output at V 0, (c) output at V 2450 V.

Fig. 2. Propagation of a dark cross formed from an amplitude mask: (a) input, (b) output at V 0, (c) output at V 2200 V.

December 1, 1997 / Vol. 22, No. 23 / OPTICS LETTERS

1753

Fig. 3. Propagation of a vortex nested upon a f inite beam: (a) input, ( b) output at V 0, (c) output at V 2750 V, (d) output at V 1750 V.

havior. Figure 4 shows typical results. In the absence of the applied voltage, the vortex diffracts from 18 to 30 mm, and the probe beam diffracts from 20 to 38 mm after propagating 5 mm in the crystal. As the nonlinearity (bias field) is turned on and a steady state is established, the vortex self-traps to its initial size, and the probe beam is well guided by the waveguide induced by the vortex soliton. We note that the alignment for guiding a beam by a dark soliton is highly sensitive because a waveguide induced by a dark-screening soliton is a single-mode waveguide. In conclusion, we have observed steady-state vortexscreening solitons and the waveguides that they induce in biased photorefractive media. This research was supported by the National Science Foundation and the U.S. Army Research Off ice. References

Fig. 4. Waveguiding of a probe beam by a vortex-screening soliton: (a) input, ( b) diffraction output after 5 mm of propagation, (c) output of the vortex soliton and the guided probe beam.

strong effect at high bias f ields, as shown in Fig. 3. The vortex is now nested upon a focused beam, which gives a doughnut-like structure (the entire beam is roughly 27 mm at input face of crystal). When a negative voltage 2750 V is applied, the doughnut breaks up into two self-defocused bright pieces, whereas the vortex itself, now experiencing a nonsymmetric base, stretches along the c axis [Fig. 3(c)]. Slight rotation of the vortex is observed in the temporal transient state. These results resemble those reported in previous experiments13 that led the authors to conclude that a single charge-one photorefractive vortex soliton cannot exist. Evidently our results shown in Fig. 1 prove that a circular photorefractive vortex soliton can exist in steady state, provided that it is borne upon a beam of inf inite extent, much as for the case of 1D dark screening solitons.14 Interestingly enough, as we reverse the polarity of the applied voltage and turn the medium into a self-focusing type (keeping all other experimental conditions unchanged), the vortex beam disintegrates into two bright spots that seem to selftrap into two solitonlike beams rotating temporally as seen from the crystal output face [Fig. 3(d)]. These results are similar to those reported in previous experiments on vortex propagation in isotropic saturable nonlinear media.15 A feature common to all saturable self-focusing nonlinear materials seems to be that a vortex borne upon a f inite beam evolves into a twosolitonlike structure. As is well known, an optical vortex soliton induces a circular waveguide akin to an optical fiber that can guide another beam.1,2 To test the waveguide properties of the vortex-screening soliton we launched an extraordinarily polarized probe beam l 633 nm into the vortex and monitored the waveguiding be-

1. A. W. Snyder, L. Poladian, and D. J. Mitchell, Opt. Lett. 17, 789 (1992). 2. G. A. Swartzlander and C. T. Law, Phys. Rev. Lett. 69, 2503 (1992). 3. B. Luther-Davies, R. Powles, and V. Tikhonenko, Opt. Lett. 19, 1816 (1994); J. Christou, V. Tikhonenko, and Y. S. Kivshar, Opt. Lett. 21, 1649 (1996). 4. G. A. Swartzlander, D. R. Andersen, J. J. Regan, H. Yin, and A. E. Kaplan, Phys. Rev. Lett. 66, 1583 (1991). 5. M. Segev, B. Crosignani, A. Yariv, and B. Fischer, Phys. Rev. Lett. 68, 923 (1992). 6. G. Duree, M. Morin, G. Salamo, M. Segev, A. Yariv, B. Crosignani, P. DiPorto, and E. Sharp, Phys. Rev. Lett. 74, 1978 (1995). 7. Z. Chen, M. Segev, D. W. Wilson, R. E. Muller, and P. D. Maker, Phys. Rev. Lett. 78, 2948 (1997). 8. G. C. Valley, M. Segev, B. Crosignani, A. Yariv, M. M. Fejer, and M. Bashaw, Phys. Rev. A 50, R4457 (1994); M. Taya, M. Bashaw, M. M. Fejer, M. Segev, and G. C. Valley, Phys. Rev. A 52, 3095 (1995). 9. M. Segev, G. C. Valley, B. Crosignani, P. DiPorto, and A. Yariv, Phys. Rev. Lett. 73, 3211 (1994); D. N. Christodoulides and M. I. Carvalho, J. Opt. Soc. Am. B 12, 1628 (1995). 10. Steady-state self-focusing in biased photorefractive media was f irst observed by M. Iturbe-Castillo, P. Marquez-Aguilar, J. Sanchez-Mondrag n, S. Stepanov, o and V. Vysloukh, Appl. Phys. Lett. 64, 408 (1994). 11. Photorefractive screening solitons were first observed by M. Shih, M. Segev, G. C. Valley, G. Salamo, B. Crosignani, and P. DiPorto, Electron. Lett. 31, 826 (1995); M. Shih, P. Leach, M. Segev, M. H. Garrett, G. Salamo, and G. C. Valley, Opt. Lett. 21, 324 (1996); M. Shih, M. Segev, and G. Salamo, Opt. Lett. 21, 931 (1996). 12. W. Krolikowski and S. Holmstrom, Opt. Lett. 22, 369 (1997). 13. A. V. Mamaev, M. Saffman, and A. A. Zozulya, Phys. Rev. Lett. 77, 4544 (1996). 14. Z. Chen, M. Mitchell, M. Shih, M. Segev, M. Garrett, and G. C. Valley, Opt. Lett. 21, 629 (1996); Z. Chen, M. Mitchell, and M. Segev, Opt. Lett. 21, 716 (1996). 15. V. Tikhonenko, J. Christou, and B. Luther-Davies, J. Opt. Soc. Am. B 12, 2046 (1995).