OSA's Digital Library

Journal of the Optical Society of America B

Journal of the Optical Society of America B


  • Vol. 16, Iss. 2 — Feb. 1, 1999
  • pp: 236–239

Soliton dynamics in a nonlocal medium

D. John Mitchell and Allan W. Snyder  »View Author Affiliations

JOSA B, Vol. 16, Issue 2, pp. 236-239 (1999)

View Full Text Article

Enhanced HTML    Acrobat PDF (187 KB)

Browse Journals / Lookup Meetings

Browse by Journal and Year


Lookup Conference Papers

Close Browse Journals / Lookup Meetings

Article Tools



The nonlinear response of various materials extends beyond the illuminating beam. We present what is to our knowledge the first analytically tractable model for the dynamics of beams in partially nonlocal media. As far as an isolated beam is concerned, propagation is qualitatively the same, independently of the radius of nonlocality.

© 1999 Optical Society of America

OCIS Codes
(190.4400) Nonlinear optics : Nonlinear optics, materials

D. John Mitchell and Allan W. Snyder, "Soliton dynamics in a nonlocal medium," J. Opt. Soc. Am. B 16, 236-239 (1999)

Sort:  Author  |  Year  |  Journal  |  Reset  


  1. A. W. Snyder and D. J. Mitchell, Science 276, 1538 (1997). [CrossRef]
  2. Y. R. Shen, Science 276, 1520 (1997). [CrossRef]
  3. A. W. Snyder and D. J. Mitchell, [Opt. Lett. 22, 16 (1997)] describe the mighty morphing spatial solitons and bullets that are characteristic of a highly saturating medium. The ln I law had been introduced previously, unbeknown to us, in a different context. See, for example, I. Bialynicki-Birula and J. Mycielski, Phys. Scr. 20, 539 (1979). But no one seems to have recognized its relevance to optical spatial solitons, possibly because of its singularity at I=0. However, Gausian beams in a ln I medium induce parabolic index waveguides, which are a standard model in linear optics. Coming at the problem, as we do, from the linear perspective shows from the outset that ln I is, like the parabolic index fiber, a good approximation. [CrossRef] [PubMed]
  4. M. Segev, B. Crosignani, A. Yariv, and B. Fisher, Phys. Rev. Lett. 68, 923 (1993). [CrossRef]
  5. P. Yeh, Introduction to Photorefractive Nonlinear Optics (Wiley, New York, 1993).
  6. F. C. Khoo, Prog. Opt. 26, 107 (1988).
  7. M. Mitchell and M. Segev, Nature (London) 387, 880 (1997). [CrossRef]
  8. V. Tikhonenko, Opt. Lett. 23, 594 (1998). [CrossRef]
  9. A. W. Snyder and A. P. Sheppard, Opt. Lett. 18, 482 (1993). A simple qualitative theory is advanced here to predict when annihilation, fusion, or birth occurs. [CrossRef] [PubMed]
  10. M. Segev, B. Crosignani, A. Yariv, and B. Fischer, Phys. Rev. Lett. 68, 923 (1992); B. Crosignani, M. Segev, D. Engin, P. DiPorto, A. Yariv, and G. Salamo, J. Opt. Soc. Am. B 10, 440 (1993); D. N. Christodoulides and M. I. Carvalho, Opt. Lett. OPLEDP 19, 1714 (1994); M. Segev, A. Yariv, B. Crosignani, P. DiPorto, G. Duree, G. Salamo, and E. Sharp, Opt. Lett. OPLEDP 19, 1296 (1994); N. Fressegeas, J. Maufoy, and G. Kugel, Phys. Rev. E PLEEE8 54, 6866 (1996). [CrossRef] [PubMed]
  11. Section 8 of Ref. 12 shows that Maxwell’s equations for transverse field E are equivalent to the Schrödinger equation, provided only that the medium is homogeneous. This is so because the maximum nonlinear induced refractive-index change is small.
  12. A. W. Snyder, D. J. Mitchell, and Yu. S. Kivshar, Mod. Phys. Lett. B 9, 1479 (1995). [CrossRef]
  13. We substitute ψ=B exp(ax2) or ψ=B exp(ar2) into the Schrödinger equation. Taking B=b−1/2 exp(iΦ) in two dimensions or B=b−1 exp(iΦ) in three dimensions, where Φ=(k/2n0)∫[n2(0, z)−n02]dz, and using Eq. (5), we find that a=ikn0(db/dz)/(2b) and 2ikn0(da/dz)+4a2− (k2Δ/ρw2)=0. Now the radius is given by 1/ρ2= −2 Re(a). Thus, after a little algebra, we obtain d2ρdz2+1n02ρ Δρ2ρw2−1k2ρ2=0. Equation (4) follows immediately.
  14. A first integration of Eq. (7) yields dρdz2+1k2n02ρ¯2 ln(ρ2M2)+ρ¯2ρ2=const. Taking dρ/dz=0 and ρ=ρmax or ρ=ρmin immediately yields Eq. (10).
  15. Following Ref. 13, we substitute ψ=B exp(axx2)exp(ayy2) into Schrödinger equation. Taking B=b−1/2 exp(iΦ), where Φ=(k/2n0)∫[n2(0, z)−n02]dz, we find that ikn0(db/dz)/(2b)=ax+ay, and ax and ay both satisfy 2ikn0(da/dz)+4a2−(k2Δ/ρw2)=0, with ρ replaced by ρx and ρy, respectively. Thus in Ref. 13 both ρx and ρy satisfy Eq. (4).
  16. M. Shih and M. Segev, Opt. Lett. 21, 1538 (1996). [CrossRef] [PubMed]
  17. W. Krolikowski and S. A. Holmstrom, Opt. Lett. 22, 369 (1997). [CrossRef]

Cited By

Alert me when this paper is cited

OSA is able to provide readers links to articles that cite this paper by participating in CrossRef's Cited-By Linking service. CrossRef includes content from more than 3000 publishers and societies. In addition to listing OSA journal articles that cite this paper, citing articles from other participating publishers will also be listed.


Fig. 1

« Previous Article  |  Next Article »

OSA is a member of CrossRef.

CrossCheck Deposited