Welcome to Scribd. Sign in or start your free trial to enjoy unlimited e-books, audiobooks & documents.Find out more
Download
Standard view
Full view
of .
Look up keyword
Like this
1Activity
0 of .
Results for:
No results containing your search query
P. 1
Wei-Ping Zhong and Milivoj Belić- Three-dimensional optical vortex and necklace solitons in highly nonlocal nonlinear media

Wei-Ping Zhong and Milivoj Belić- Three-dimensional optical vortex and necklace solitons in highly nonlocal nonlinear media

Ratings: (0)|Views: 5|Likes:
Published by 23213m

More info:

Published by: 23213m on Feb 14, 2012
Copyright:Attribution Non-commercial

Availability:

Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less

04/01/2014

pdf

text

original

 
Three-dimensional optical vortex and necklace solitons in highly nonlocal nonlinear media
Wei-Ping Zhong
1
and Milivoj Beli
ć
2
1
 Department of Electronic Engineering, Shunde College, Guangdong Province, Shunde 528300, China
2
Texas A & M University at Qatar, 23874 Doha, Qatar 
Received 21 September 2008; published 2 February 2009
We demonstrate the existence of localized optical vortex and necklace solitons in three-dimensional
3D
highly nonlocal nonlinear media, both analytically and numerically. The 3D solitons are constructed with thehelp of Kummer’s functions in spherical coordinates and their unique properties are discussed. The procedurewe follow offers ways for generation, control, and manipulation of spatial solitons.DOI:10.1103/PhysRevA.79.023804PACS number
s
: 42.65.Tg
I. INTRODUCTION
Spatial optical solitons are self-trapped beams of finitespatial cross section that travel without divergence associatedwith the freely diffracting beams. Owing to their novel phys-ics, as well as potential applications, spatial solitons havebeen under intensive study in the past decade
1
and richdynamics associated with them have been discovered. Vortexsolitons are optical beams that have phase singularitiesmixed within the wave front curvature, and frequently appearin the study of optical tweezers
2
, trapping and guiding of cold atoms
3
, and entanglement states of photons
4
.Necklace solitons are a special class of self-trapped beams,which look like necklaces with the intensity and phasemodulated periodically along the azimuthal angle
5
. Sol- jacic
et al.
predicted quasistable necklace beams in a Kerrnonlinear
NL
medium theoretically
6
, and recently suchbeams were observed experimentally
7
. Necklace vectorsolitons also exhibit quasistable evolution in a saturable NLmedium, although with a slowly expanding rate
8
. It wasshown that the presence of topological charges can slow theexpansion of such beams. Metastable necklace solitons wereobserved in lead glasses with a thermal nonlocal nonlinearity
9
. Robust spatiotemporal necklace solitons in the cubic-quintic Ginzburg-Landau system were reported in
. Dark hollow beams have also attracted attention, thanks to theirpotential for applications in the turbulent atmosphere
.However, almost all necklace solitons reported so far inevi-tably experience expansion during propagation, due to a netoutward force exerted on each “pearlby all other pearlsforming the necklace
6
.Nonlocality has also been a phenomenon of intense re-search over recent years in various NL physical systems
1
.Basically, nonlocality extends the effect of localized excita-tions in a medium, allowing a degree of interconnectionamong different regions of the medium in question
. Ithas been shown recently that the stability of localized wavesis greatly enhanced in nonlocal nonlinear
NN
media. Insuch media the NL response at a particular spatial location isdetermined by the wave intensity in the neighborhood of thislocation. Nonlocality often results from certain transport pro-cesses, such as atomic diffusion
. It can also be a signa-ture of a long-range interparticle interaction, e.g., in nematicliquid crystals
. A spatially nonlocal
NLO
response isalso naturally present in atomic condensates, where it de-scribes a noncontact bosonic interaction
. Extensivestudies of beam propagation in NN media revealed a range of interesting features. In particular, it has been shown that non-locality may affect modulational instability of plane waves
and prevent catastrophic collapse of finite beams
,as well as stabilize complex one-, two-, and three-dimensional beams, including vortices
1820
. Recently, ithas been shown that NLO media can support stable propa-gation of rotating solitons, the so-called azimuthons
.Recent theoretical studies demonstrated both stable and un-stable evolution of azimuthons
. In the latter case ithas been shown that in a highly NLO regime the azimuthonscan undergo structural transformation resulting from the en-ergetic coexistence of solitons of different symmetries. Nu-merical and analytical studies revealed that the angular ve-locity of azimuthons is governed by two contributions. Thereis the linear component, determined solely by the spatialstructure of the beam
akin to the rotation of complex wavestructures resulting from the beating of their constituentmodes
, as well as the NL component, which is broughtabout by the nonlinearity
.In this work, we investigate in detail three-dimensional
3D
highly NN solitons. We consider models for highly NNmedia where the spatial solitons have already been observed,such as lead glasses exhibiting self-focusing thermal nonlin-earity
, photorefractive media
, and nematic liquidcrystals
. We study NL waves in a generic, highly NLOoptical system by solving the nonlocal nonlinear Schrödingerequation
NNSE
, using the self-similar method. We demon-strate breathers whose evolution is periodic and discuss soli-tons in highly NN systems. We demonstrate that, by choos-ing a suitable soliton parameter, the solitons can beconveniently controlled and manipulated.The paper is arranged as follows. In Sec.IIwe introducethe highly NN model and obtain the self-similar breather andsoliton solutions in spherical coordinates. In Sec.IIIwe com-pare our exact analytical results with the numerical solutions.One of our main findings, that the azimuthons emerge fromthe internal modes of stationary highly NN soliton solutions,like vortices and necklaces, is detailed in Sec.IV.SectionV presents conclusions.
PHYSICAL REVIEW A
79
, 023804
2009
1050-2947/2009/79
2
 /023804
6
©2009 The American Physical Society023804-1
 
II. THE GENERAL NN MODEL AND SELF-SIMILARSOLITON SOLUTIONSA. Highly NN model
We consider an optical beam propagating along the
z
axisof a NL self-focusing material with the scalar amplitude of the electric field
   
,
=
   
,
exp
ik 
   
·
   
i
+c.c. Here
   
=
 x
,
 y
,
 z
,
   
is the wave number vector,
is the optical fre-quency, and
   
,
is the slowly varying amplitude. We as-sume that the refractive index change
 I 
, induced by thebeam intensity
   
,
=
 
2
, can be described by the NLOmodel
 N 
 I 

   
,
=
 R
   
   
 I 
   
,
dr 
   
.
1
The response function
R
   
is assumed to be a real, positivedefinite, localized, and rotationally symmetric function
i.e.,
 R
   
=
 R

that satisfies the normalization condition
 R
   
dr 
   
=1. The width of the response function
R
deter-mines the degree of nonlocality. For a singular response
 R
   
=
 
   
, the refractive index becomes a local function of the light intensity,
 I 
=
 I 
   
,
, i.e., the refractive indexchange at a given point is solely determined by the lightintensity at that very point. With increasing width of 
R
   
thelight intensity in the vicinity of the point
   
also contributes tothe index change at that point. In the opposite limit, when theresponse function is much broader than the intensity distri-bution, the NL term becomes proportional to the responsefunction,
 I 
=
PR
, where
P
is the beam power. The NLOresponse
1
leads to the following NNSE governing the evo-lution of an optical beam in appropriately chosen dimension-less coordinates:
i
 
 
 
+12
2
 
+
 I 
 
= 0.
2
It has been shown that as long as the response function ismonotonically decaying, the physical properties of solutionsto Eq.
2
do not depend strongly on its shape
1620
. Forconvenience in our calculations we choose to work with theGaussian response function
 R
   
   
=1
 
2
e
   
   
2
/
 
2
.
3
Assuming that the intensity distribution is peaked at the ori-gin, one can expand the response function at the origin, toobtain
 I 
P
 R
0
+
 R
2
2
. In this case the highly NNSEbecomes the
linear 
SE, leading to a NL optical model inwhich the change in the NL term is proportional to a NLfunction of the power,
 N 
 I 
 
2
P
2
. Although linear in
 
, the model still describes a highly NL phenomenon of solitons through the dependence of the coefficient
 
on thebeam power
P
. For this reason the model is referred toas the
highly nonlocal nonlinea
SE. It has been used in
,for example, to explain the experimental observation of op-tical spatial solitons in nematic liquid crystals.In the highly NLO limit, the wave equation governing thebeam propagation in 3D NN media is reduced to
i
 
 
 
+12
2
 
sr 
2
 
= 0,
4
where
s
is the parameter proportional to
 
2
P
, containingthe influence of the beam power. Note that
P
is constant,equal to the total input power
P
0
. The beam intensity doesnot explicitly enter the evolution equation any longer, andthe model becomes linear. Hence, in solving Eq.
4
we willalso be solving a linear quantum mechanical problem,namely, the 3D quantum harmonic oscillator
QHO
. Al-though many different solutions to the
time-independent 
QHO in different coordinate systems are known
, wecould not locate any analytical solutions to the
time-dependent 
3D QHO. Hence, we will be looking for self-similar time-dependent solutions of Eq.
4
in the form of localized 3D wave packets. Such solutions will naturally im-pose certain conditions on the input parameters and the pa-rameters describing these solutions. It should also be notedthat beam collapse cannot occur in Eq.
4
.
B. Self-similar solution method
The second term in Eq.
4
represents the diffraction andthe third term originates from the optical nonlinearity. Wetreat Eq.
4
in spherical coordinates, by the method of sepa-ration of variables. Note that
s
0. The separation of vari-
-202x-202y-101z20-202x-2
 
02y-101z20-202x-2
 
02y-101z20-202x-2
 
02y-101z20-2
 
02x-202y-101z2
 
0-202x-202y-101z0
a
-202x-2
 
02y-101z20-202x-2
 
02y-101z20-20
 
2x-202y-101z20-202x-2
 
02y-101z20-202x-2
 
02y-101z20-202x-2
 
02y-101z20
b
FIG. 1.
Color online
Comparison of analytical solutions withthe numerical simulations for the
 
022
wave packet, for differentparameters
=1.6
top row
and 0.6
bottom row
.
a
Analyticalsolution for intensity from Eq.
.
b
Numerical simulation of Eq.
2
. The propagation times are 2
sw
02
=0,
 
/
4,
 
/
2, from leftto right.WEI-PING ZHONG AND MILIVOJ BELI
Ć
PHYSICAL REVIEW A
79
, 023804
2009
023804-2
 
ables
 
,
,
 
,
 
=
,
 
,
 
leads to the following twoequations:1
1sin
 
  
 
sin
 
 
 
 
+1sin
2
 
 
2
 
 
2
=
l
l
+ 1
,
5
2
2
i
 
 
+12
2
  
2
 
 
sr 
2
=
l
l
+ 1
,
6
where
l
is a non-negative integer. Equation
5
, of course, hasspherical harmonics for the solution,
lm
 
,
 
=
l
+
m
!
m
1 +
q
2

l
m
!
cos
m
 
+
iq
sin
m
 

P
lm
cos
 
,
7
where the parameter
q
0,1
determines the modulationdepth of the beam intensity
5
, and
m
is a real number.Following Refs.
, we define the complex field as
,
=
 A
,
e
iB
,
, where
A
and
B
are real functions of 
and
. Substituting
,
into Eq.
4
, we find the followingcoupled equations for the phase
B
,
and the amplitude
 A
,
:
A
 
 B
 
+1
 
 A
 
+12
 
2
 A
 
2
12
 A
 
 B
 
2
sr 
2
 A
l
l
+ 1
2
2
A
= 0,
8
 
 A
 
+
 
 A
 
 
 B
 
+1
 A
 
 B
 
+12
 A
 
2
 B
 
2
= 0.
9
To find a self-similar solution of Eqs.
8
and
9
, we assumethat the amplitude and the phase have the forms
 A
=
w
3
 f 
,
10
 B
=
a
+
b
+
c
2
,
11
where
is the normalization constant,
w
is the beamwidth,
 f 
is a real function to be determined,
,
   
is a self-similar variable,
a
is the phase offset, and
c
is the chirp.The functions
,
   
,
b
, and
c
, defined by the form of 
A
and
B
, are found from Eq.
9
:
=
/
w
,
b
=0,
c
=
1
/
2
w
 
w
/
 
. The amplitude
A
,
is found from Eq.
8
,by solving the following differential equation for
:
2
 f 
2
+2
df 
+
2
w
3
2
wdt 
2
+ 2
sw
4
l
l
+ 1
2
− 2
w
2
dadt 
 f 
= 0.
12
Inserting a variable transformation
=
l
e
1
/
2−
2
/
2
g
,from Eq.
we obtain
2
g
2
+2
l
+ 1 −
2
dg
+
2
w
3
2
wdt 
2
+ 2
sw
4
− 1
2
l
+ 3
− 2
w
2
dadt 
g
= 0.
13
To further simplify Eq.
, we introduce another variabletransformation
2
=
 
, and after some algebra arrive at
2
g
 
2
+
l
+32
 
dg
 
+
ng
= 0,
14
2
wdt 
2
+ 2
sw
1
w
3
= 0,
15
14
2
l
+ 3
− 2
w
2
dadt 
=
n
,
16
where
n
=0,1,2,...
is a non-negative integer. Equation
is the well-known Kummer’s equation, whose solutionsare known as Kummer’s hypergeometric functions
,namely,
g
=
1
1
n
,
l
+
3
2
,
 
.Taking
w

=
w
0
and
dw
/
dt 

=0
=0 and integratingEq.
yields
w
2
=
w
02
1 +
− 1
sin
2
2
sw
02

,
17
where
=1
/
2
sw
04
. Hence, from Eq.
and from the defi-nition of 
c
, we obtain
a
=
a
0
2
n
+
l
+32
tan
−1
tan
2
sw
02

2
s
w
04
,
18
c
=
sw
02
− 1
sin
4
sw
02
1 +
− 1
cos
4
sw
02
.
19
Using Eqs.
and
, we finally get the exact self-similarbreather solution of Eq.
4
:
 
nlm
=
cos
m
 
+
iq
sin
m
 

w
3
P
lm
cos
 
w
l
e
1
/
2−
2
/
2
w
2
1
1
n
,
l
+32,
2
w
2
e
i
a
+
c
2
,
20
where
w
,
a
, and
c
are determined by Eqs.
and
=
2
l
+2−
n
2
l
+ 2
n
+ 1
!!
 
n
!

2
l
+ 1
!!
2
l
+
m
!
m
1 +
q
2

l
m
!.Hence, when
=1 the beam diffraction is exactly bal-anced by the nonlinearity. Since
w
=
w
0
for
=1, the beamwidth is independent of the propagation distance. The param-eters are given by
w
=
w
0
,
c
=0, and
a
=
a
0
2
n
+
l
+
3
2
/
w
02
. Thus, the exact self-similar soliton solution of Eq.
4
can be written as
THREE-DIMENSIONAL OPTICAL VORTEX AND NECKLACE
PHYSICAL REVIEW A
79
, 023804
2009
023804-3

You're Reading a Free Preview

Download
scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->