Commun. Theor. Phys. (Beijing, China) 52 (2009) pp.

10041012 c Chinese Physical Society and IOP Publishing Ltd

Vol. 52, No. 6, December 15, 2009

Quasi-Periodic Structures Based on Symmetrical Lucas Function of (2+1)-Dimensional Modied Dispersive Water-Wave System
Emad A-B. ABDEL-SALAM
Assiut University, Department of Mathematics, New Valley Faculty of Education, El-Kharga, New Valley, Egypt

(Received December 24, 2008)

Abstract By introducing the LucasRiccati method and a linear variable separation method, new variable separation solutions with arbitrary functions are derived for a (2+1)-dimensional modied dispersive water-wave system. The main idea of this method is to express the solutions of this system as polynomials in the solution of the Riccati equation that the symmetrical Lucas functions satisfy. From the variable separation solution and by selecting appropriate functions, some novel Jacobian elliptic wave structure with variable modulus and their interactions with dromions and peakons are investigated.
PACS numbers: 02.30.Jr, 05.45.Yv, 03.65.Ge

Key words: Lucas functions, quasi-periodic structure, variable separation excitations, modied dispersive water-wave system

1 Introduction
Due to the wide applications of soliton theory in mathematics, physics, chemistry, biology, communications, astrophysics, and geophysics, the study of integrable models has attracted much attention of many mathematicians and physicists. To nd some exact explicit soliton solutions for integrable models is one of the most important and signicant tasks. There has been a great amount of activities aiming to nd methods for exact solution of nonlinear partial dierential equations (NLPDEs). Such include the Bcklund transformation, Darboux transformation,[1] a various tanh methods,[26] various Jacobi elliptic function methods,[78] multi-linear variable separation approach,[9] Painlev property, homogeneous balance method, similare ity reduction method and so on.[4143] For a given NPLDE with independent variables x = (x0 = t, x1 , x2 , x3 , . . . , xn ) and dependent variable u, P ( u, ut , uxi , uxi xj , . . .) = 0 , (1)

where P is in general a polynomial function of its argument, and the subscripts denote the partial derivatives, by using the traveling wave transformation, Eq. (1) possesses the following anstz, a
m

u = u(),

=
i=0

ki xi ,

(2)

where ki , i = 0, 1, 2, . . . , m are all arbitrary constants. Substituting Eq. (2) into Eq. (1) yields an ordinary dierential equation (ODE): O(u(), u() , u() , . . .) = 0. Then u() is expanded into a polynomial in g()
n

u() = F (g()) =
i=0
E-mail:

ai g i () ,

(3)

where ai are constants to be determined and n is xed by balancing the linear term of the highest order with the nonlinear term in Eq. (1). If we suppose g() = tanh , g() = sech , and g() = sn or g() = cn respectively, then the corresponding approach is usually called the tanh-function method, the sech-function method, and the Jacobian-function method. Although the Jacobian elliptic function method is more improved than the tanhfunction method and the sech-function method, the repeated calculations are often tedious since the dierent function g() should be treated in a repeated way. The main idea of the mapping approach is that, g() is not assumed to be a specic function, such as tanh, sech, sn, and cn, etc., but a solution of a mapping equation such as the Riccati equation (g = g 2 + a0 ), or a solution of the cubic 2 non-linear Klein Gordon (g = a4 g 4 + a2 g 2 + a0 ), or a 4 2 solution of the general elliptic equation ( g = i=0 ai g i ), where ai (i = 0, 1, . . . , 4) are all arbitrary constants. Using the mapping relations and the solutions of these mapping equations, one can obtain many explicit and exact traveling wave solutions of Eq. (1). Now an interesting or important question is that whether the localized excitations based on the former multilinear approach[9] can be derived by the latter mapping approach, which is usually used to search for traveling wave solutions. The crucial technic is how to obtain some solutions of Eq. (1) with certain arbitrary functions, also the quite rich localized excitations, such as lumps, dromions, peakons, compactons, foldons, ring solitons, fractal solitons, chaotic solitons and so on[914,1731,39] are obtained, and the novel interactive behavior among the same types and various types of soliton excitations are revealed. There is a well-known fact that two mathematical constants of nature, the - and e-numbers, play a great role

emad abdelsalam@yahoo.com

No. 6

Quasi-Periodic Structures Based on Symmetrical Lucas Function of (2+1)-Dimensional

1005

in mathematics and physics. Their importance consists in the fact that they generate the main classes of socalled elementary functions: sin, cosine (the -number), exponential, logarithmic and hyperbolic functions (the enumber). It is impossible to imagine mathematics and physics without these functions. For example, there is the well-known greatest role of the classical hyperbolic functions in geometry and in cosmological researches. However, there is the one more mathematical constant playing a great role in modeling of processes in living nature termed the Golden Section, Golden Proportion, Golden Ratio, Golden Mean.[3236] However, we should certify that a role of this mathematical constant is sometimes undeservedly humiliated in modern mathematics and mathematical education. There is the well-known fact that the basic symbols of esoteric (pentagram, pentagonal star, platonic solids etc.) are connected to the Golden Section closely. Moreover, the materialistic science together with its materialistic education had decided to throw out the Golden Section. However, in modern science, an attitude towards the Golden Section and connected to its Fibonacci and Lucas numbers is changing very quickly. The outstanding discoveries of modern science based on the Golden Section have a revolutionary importance for development of modern science. These are enough convincing conrmation of the fact that human science approaches to uncovering one of the most complicated scientic notions, namely, the notion of Harmony, which is based on the Golden Section, Harmony was opposed to Chaos and meant the organization of the universe. In Euclids the elements we nd a geometric problem called the problem of division of a line segment in the extreme and middle ratio. Often this problem is called the golden section problem.[3336] Solution of the golden section problem reduces to the following. algebraic equation x2 = x + 1 this equation has two roots. We call the positive root, = (1 + 5)/2, the golden proportion, golden mean, or golden ratio. El Naschies works[3438] develop the golden mean applications into modern physics. In the paper[38] devoted to the role of the Golden Mean in quantum physics El Naschie concludes the following: In our opinion it is very worthwhile enterprise to follow the idea of Cantorian space-time with all its mathematical and physical ramications. The nal version may well be a synthesis between the results of quantum topology, quantum geometry and may be also Rosslers endorphysics, which like Nottales latest work makes extensive use of the ideas of Nelsons stochastic mechanism. Thus, in the Shechtmans, Butusovs, Mauldin and Williams, ElNaschies, Vladimirovs works, the Golden Section occupied a rm place in modern physics and it is impossible to imagine the future progress in physical researches without the Golden Section. In our present paper, we review symmetrical Lucas functions[39] and we nd new solutions of the Riccati equa-

tion by using these functions. Also, we devise an algorithm called LucasRiccati method to obtain new exact solutions of NLPDEs. Along with the above line, i.e., in order to derive some new solutions with certain arbitrary functions, we assume that its solutions in the form,
n

u(x) =
i=0

ai (x) F i (x) ,

(4)

with F = A + B F2 , (5) where x = (x0 = t, x1 , x2 , x3 , . . . , xn ) and A, B are constants and the prime denotes dierentiation with respect to . To determine u explicitly, one may take the following steps: First, similar to the usual mapping approach, determine n by balancing the highest non-linear terms and the highest-order partial terms in the given NLPDE. Second, substituting (4) and (5) into the given NLPDE and collecting coecients of polynomials of F , then eliminating each coecient to derive a set of partial dierential equations of ai (i = 0, 1, 2, . . . , n) and . Third, solving the system of partial dierential equations to obtain ai and . Substituting these results into (4), then a general formula of solutions of equation (1) can be obtained. Choose properly A and B in ODE (5) such that the corresponding solution F () is one of the symmetrical Lucas function given bellow. Some denitions and properties of the symmetrical Lucas function are given in Appendix. Case 1 If A = ln and B = ln , then (5) possesses solutions tLs(), cotLs() . Case 2 If A = ln /2 and B = ln /2, then (5) possesses a solution tLs() . 1 secLs() Case 3 If A = ln and B = 4 ln , then (5) possesses a solution tLs() . 1 tLs2 () In the following section we apply the LucasRiccati method to obtain new localized excitations. Also, we pay our attention to some novel Jacobian elliptic wave structure with variable modulus and their interactions with dromions and peakons.

2 New Variable Separation Solutions of (2+1) -Dimensional Modied Dispersive WaterWave System
We consider here the (2+1)-dimensional modied dispersive water-wave (MDWW) system uyt + uxxy 2 vxx (u2 )xy = 0 , vt vxx 2(vu)x = 0 . (6)

1006

Emad A-B. ABDEL-SALAM

Vol. 52

The (2+1)-dimensional MDWW system was used to model nonlinear and dispersive long gravity waves traveling in two horizontal directions on shallow waters of uniform depth, and can also be derived from the well-known KadomtsevPetviashvili (KP) equation using the symmetry constraint.[1516] Abundant propagating localized excitations were derived by Tang et al.[9] with the help of PainlevBcklund transformation and a multilinear varie a able separation approach. It is worth mentioning that this system has been widely applied in many branches of physics, such as plasma physics, uid dynamics, nonlinear optics, etc. So, a good understanding of more solutions of the (2+1)-dimensional MDWW system (6) is very helpful, especially for coastal and civil engineers in applying the nonlinear water model in harbor and coastal design. Meanwhile, nding more types of solutions to system (6) is of fundamental interest in uid dynamics. Now we apply the LucasRiccati method to Eqs. (6). First, let us make a transformation of the system (6): v = uy . Substituting this transformation into system (6), yields uty uxxy (u2 )xy = 0 . (7) Balancing the highest order derivative term with the nonlinear term in Eq. (7), gives n = 1, we have the anstz a u(x, y, t) = a0 (x, y, t) + a1 (x, y, t)F ((x, y, t)) , (8)

fxx ft + fx cotLs(f + g) ln , 2fx v2 = fx gy ln 2 fx gy cotLs2 (f + g) ln 2 , fxx ft fx tLs(f + g) ln u3 = + , 2fx 2[1 secLs(f + g)] 2 fx gy ln 2 fx gy ln 2 tLs(f + g) v3 = , 4 4 1 secLs(f + g) fxx ft 4fx tLs(f + g) ln u4 = + , 2fx 1 + tLs2 (f + g) 2 tLs(f + g) v4 = 4fx gy ln 2 16fx gy ln 2 , 2 (f + g) 1 + tLs u2 =

(16) (17) (18) (19) (20) (21)

where f (x, t) and g(y) are two arbitrary variable separation functions. Especially, for the potential U = u1y has the following form U = 4 fx gy sec Ls2 (f + g) ln 2 . (22)

3 Periodic and Quasi-Periodic and Waves of (2+1)-Dimensional MDWW System

All rich localized coherent structures, such as nonpropagating solitons, dromions, peakons, compactons, foldons, instantons, ghostons, ring solitons, and the interactions between these solitons,[914,1731,39] can be derived by the quantity U expressed by (22). It is known that for the (2+1)-dimensional integrable models, there are many more abundant localized structures than in (1+1)-dimensional case because some types of arbitrary functions can be included in the explicit solution expression.[9] Moreover, the periodic waves also have been studied by some authors. However, the quasi- and non-periodic wave solutions of the integrable systems have been given limited attention with a fewer publications in the literature. In the present paper, we pay our attention to some periodic, quasi-periodic, and non-periodic wave evolutional behaviors for the eld U in (2+1) dimensions. 3.1 Doubly Periodic and Line Periodic Solitary Waves It is known that for the nonlinear system, the periodic (and doubly periodic) wave solutions can usually be expressed by means of the Jacobi elliptic functions with constant modulus. If we take f = cn(k1 x + 1 t + x01 ; m1 ) , g = cn(K1 y + y01 ; n1 ) , (23)

where a0 (x, y, t) a0 , a1 (x, y, t) a1 , and (x, y, t) are arbitrary functions of x, y, t to be determined. Substituting (8) with (5) into (6), and equating each of the coecients of F () to zero, we obtain system of PDEs. Solving this system of PDEs, with the help of Maple, we obtain the following solution: a0 (x, y, t) = xx (x, y, t) t (x, y, t) , 2 x (x, y, t) a1 (x, y, t) = B x (x, y, t) , (x, y, t) = f (x, t) + g(y) , (9) (10) (11)

where f (x, t) f and g(y) = g are two arbitrary functions of x, t, and y, respectively. Now based on the solutions of (5), one can obtain new types of localized excitations of the (2+1)-dimensional MDWW system. We obtain the general formulae of the solutions of the (2+1)-dimensional MDWW system fxx ft Bfx F (f + g) , 2fx v = ABfx gy B 2 fx gy F 2 (f + g) . u= (12) (13)

By selecting the special values of the A, B and the corresponding function F we have the following solutions of (2+1)-dimensional MDWW system: u1 = fxx ft + fx tLs(f + g) ln , 2fx v1 = fx gy ln 2 fx gy tLs2 ( f + g) ln 2 , (14) (15)

of (22) leads to a periodic solution for the potential U . In (23), cn(k1 x+ 1 t+ x01 ; m1 ), cn(K1 y + y01 ; n1 ) are the Jacobian elliptic cn functions with the modulus m1 , n1 , and x01 , y01 , k1 , K1 are arbitrary constants. Figure 1 shows the detailed structures of (22) with (23) and k1 = 1 = K1 = 1, x01 = y01 = 12 , (24)

No. 6

Quasi-Periodic Structures Based on Symmetrical Lucas Function of (2+1)-Dimensional

1007

when t = 0. Figure 1(a) is related to the modulus of the Jacobian cn functions being taken as m1 = n1 = 0.6. Figure 1(b) shows the density plot of 1(a). If we select the functions f and g such that one of them possesses a localized structure while the other has a periodic structure, then the wave solution (22) becomes a line-periodic wave excitation. For example, the selection f = cn(k1 x + 1 t + x01 ; m1 ) ,

g = tanh(K1 y + y01 ) ,

(25)

makes the wave solution (22) to be line-periodic solitary wave. Figure 2 displays the structure of equation (22) with the condition (25) and the parameter selections as k1 = K1 = 0.6, 1 = 1, x01 = y01 = 8 , (26)

at time t = 0. Figure 2(a) is related to the modulus of the Jacobian cn functions being taken as m1 = 0.4.

Fig. 1 The structures of doubly-periodic solution (22) with (23) and (24): (a) m1 = n1 = 0.6; (b) the density plot of (a) when t = 0.

Fig. 2 t = 0.

The structures of line-periodic solution (22) with (23) and (24): (a) m1 = 0.4; (b) the density plot of (a) when

3.2 Quasi-Periodic Wave In this subsection, we give doubly quasi-periodic (quasi-periodic in both x and y directions) wave solutions by selecting the arbitrary functions as Jacobi elliptic functions with variable modulus. To see more concretely, we take some concrete selections for the arbitrary functions f and g for the (2+1)-dimensional MDWW system. The rst example, we take f and g as f = cn(1 ; m(2 )) , m = 0.3 + 0.2 tanh(2 ) + 0.4 tanh (2 ) ,
2

g = cn(1 ; n(2 )) , n = 0.3 + 0.2 tanh(2 ) + 0.4 tanh2 (2 ) , i = ki x + i t + x0i , i = Ki y + y0i , i = 1, 2 , (27) then Eq. (22) denotes a doubly quasi-periodic wave solution. Figure 3 displays the structure of Eq. (22) with the condition (27) and the parameter selection as k1 = k2 = 0.2, at time t = 0. K1 = K2 = 0.4, 1 = 2 = 1 , (28) x01 = y01 = x02 = y02 = 6 ,

1008

Emad A-B. ABDEL-SALAM

Vol. 52

Fig. 3

(a) The structures of quasi-periodic solution (22) with (27) and (28) (b) the density plot of (a) when t = 0.

Fig. 4

(a) The structures of quasi-periodic solution (22) with (29) and (30); (b) the density plot of (a) when t = 0.

Fig. 5

(a) The structures of quasi-periodic solution (22) with (31) and (32); (b) the density plot of (a) when t = 0.

The second selection of the functions f and g reads f = sn(1 ; m(2 )) , m = 0.3 + 0.2 sech (2 ) + 0.4 sech 2 (2 ) , g = sn(1 ; n(2 )) , n = 0.3 + 0.2 sech (2 ) + 0.4 sech 2 (2 ) , i = ki x + i t + x0i , i = Ki y + y0i , i = 1, 2 , (29)

then Eq. (22) denotes another doubly quasi-periodic wave solution. Figure 4 displays the structure of Eq. (22) with the condition (29) and the parameter selections as k1 = k2 = 0.1, K1 = K2 = 0.2, 1 = 2 = 1 , (30) x01 = y01 = x02 = y02 = 4,

at time t = 0. The last selection of the functions f and g reads f = cn(1 ; m(2 )) ,

No. 6

Quasi-Periodic Structures Based on Symmetrical Lucas Function of (2+1)-Dimensional

1009

m = 0.6 + 0.3 tanh(2 ) , g = cn(1 ; n(2 )) , n = 0.6 + 0.3 tanh(2 ) , i = ki x + i t + x0i , i = Ki y + y0i , i = 1, 2 , (31)

the condition (31) and the parameter selections as k1 = k2 = 0.2, K1 = K2 = 0.3, 1 = 2 = 1 , (32) x01 = y01 = x02 = y02 = 3 ,

then Eq. (20) denotes another doubly quasi-periodic wave solution. Figure 5 displays the structure of Eq. (22) with

at time t = 0. The idea to introduce the variable modulus into the conoidal wave solutions is simple but important and never been bethought. The introduction of the variable modulus in the Jacobi elliptic functions is essential and it results completely dierent types of waves solutions.

3.3 Interaction of Quasi-Periodic Waves and Dromions

Fig. 6 (a) The interaction of dromion and quasi-periodic solution (22) with (33) and (34); (b) the density plot of (a) when t = 0.

Fig. 7 (a) The interaction of dromion and line quasi-periodic solution (22) with (33) and (34) when m = 1; (b) the density plot of (a) when t = 0.

The dromion solutions which are localized in all directions are driven by multiple straight-line ghost solitons with some suitable dispersion relation. Also, multiple dromion solutions are driven by curved line and straight line solitons. Here, we study the interaction between this localized structure in the background of Jacobian elliptic wave and the Jacobian elliptic waves with variable modulus. If we take f = cn(k1 x + 1 t + x01 ; m) + tanh(k1 x + 1 t + x01 ) ,

g = sn(1 ; n(2 )), i = Ki y + y0i ,

n = 0.3 + 0.2 tanh(2 ) , i = 1, 2 , (33)

from (22) we can obtain the interaction between the Jacobian elliptic waves with variable modulus and dromions. In order to understand the properties of the interaction between the periodic and quasi-periodic waves, we illustrate it by several gures. Figure 5 shows the detail on the interaction properties of the quasi-periodic wave and

1010

Emad A-B. ABDEL-SALAM

Vol. 52

dromions with the parameter values K1 = k1 = 1, K2 = k2 = 2, 1 = 3 , (34) x01 = y01 = y02 = 0, m = 0.3 ,

and t = 0, respectively. It is very interesting to see the

interaction properties of the Jacobian elliptic waves and dromions in the limit case. As m is close to 1, from (22) we can obtain the interaction of dromions and line-quasi periodic solution depicted in Fig. 7 with the same parameter values as Fig. 6.

3.4 Interaction of Quasi-Periodic Waves and Peakons

Fig. 8 (a) The interaction of peakon and quasi-periodic solution (22) with (35) and (36) when m = 0.4; (b) the density plot of (a) when t = 0.

Fig. 9 (a) The interaction of peakon and quasi-line periodic solution (22) with (35) and (36) when m = 1; (b) the density plot of (a) when t = 0.

The celebrated (1+1)-dimensional CamassaHolm equation[40] ut + 2kux uxxt + 3uux = 2uxuxx + uuxxx , possesses a special type of novel solution, namely the peakon solution (weak continuous solution) which is discontinuous at its crest. u(x, t) = k + c e(|xct|), k 0 . Moreover, we can also discuss the interaction between Jacobian elliptic waves with variable modulus and peakon with the background of Jacobian elliptic wave. When we consider M Fi (ki x + i t), ki x + i t 0 , i=1 f = cn(kx + t + x0 ; m) + M Fi (ki x + i t) + 2Fi (0), ki x + i t > 0 ,
i=1

g = cn(1 ; n(2 )),

n = 0.3 + 0.2 tanh(2 ),

i = Ki y + y0i ,

i = 1, 2 ,

(35)

No. 6

Quasi-Periodic Structures Based on Symmetrical Lucas Function of (2+1)-Dimensional

1011

the interaction between the Jacobian elliptic cn wave with variable modulus and peakon with the background of Jacobian elliptic wave can be constructed. A simple example of this interaction is depicted in Fig. 8 when choosing F1 = 0.3 ex t , K1 = K2 = 0.2, 1 = 1, M = 1, m = 0.5, k = 0.3, = 0, (36) k1 = 1 ,

x0 = y01 = y02 = 0 ,

and t = 0, respectively. Figure 8 shows the interaction between the Jacobian elliptic waves with variable modulus and peakon. It is very interesting to see the interaction properties of the Jacobian elliptic waves and peakon in the limit case. As m is close to 1, from (22) we can obtain the interaction of peakon and line quasi-periodic solution depicted in Fig. 9 with the same parameter values as Fig. 8.

4 Summary and Discussion

In conclusion, the Lucas Riccati method is applied to obtain variable separation solutions of (2+1)-dimensional MDWW system. Based on the quantity (22), some novel Jacobian elliptic wave with variable modulus, quasiperiodic wave evolutional, and their interaction behaviors are found. We hope that in future experimental studies these novel structures and quasi-periodic wave evolutional behaviors obtained here can be realized in some elds. Actually, our present short paper is merely a beginning work, more application to other nonlinear physical systems should be concerned and deserve further investigation. In our future work, on the one hand, we devote to generalizing this method to other (2+1)-dimensional nonlinear systems such as the ANNV system and BKK system, BoitiLeonPempinelle system, etc. On the other hand, we will look for more interesting localized excitations. The solutions obtained here be useful to some physical problems in uid dynamics.

and Lucas numbers. Taking into consideration a great role played by the hyperbolic functions in geometry and physics, (Lobatchevskis hyperbolic geometry, Fourdimensional Minkowskis world, etc.), it is possible to expect that the new theory of the hyperbolic functions will bring to new results and interpretations on mathematics, biology, physics, and cosmology. In particular, the result is vital for understanding the relation between transnitness i.e. fractal geometry and the hyperbolic symmetrical character of the disintegration of the neural vacuum, as pointed out by El Naschie. The denition and properties of the symmetrical Lucas functions, the symmetrical Lucas sine function (sLs), the symmetrical Lucas cosine function (cLs) and the symmetrical Lucas tangent function (tLs) are dened[3235] as sLs(x) = x x , cLs(x) = x + x , x x . (A1) tLs(x) = x + x They are introduced to consider so-called symmetrical representation of the hyperbolic Lucas functions and they may present a certain interest for modern theoretical physics taking into consideration a great role played by the Golden Section, Golden Proportion, Golden ratio, Golden Mean in modern physical researches.[3233] The symmetrical Lucas cotangent function (cotLs) is cotLs(x) = 1/tLs(x), the symmetrical Lucas secant function (secLs) is secLs(x) = 1/cLs(x), the symmetrical Lucas cosecant function (cscLs) is cscLs(x) = 1/sLs(x). These functions satisfy the following relations[3233] cLs2 (x)sLs2 (x) = 4 , 1tLs2 (x) = 4secLs2 (x) , (A2) cotLs2 (x) 1 = 4cscLs2 (x) .

Appendix
Stakhov and Rozin in [32] introduced a new class of hyperbolic functions that unite the characteristics of the classical hyperbolic functions and the recurring Fibonacci and Lucas series. The hyperbolic Fibonacci and Lucas functions, which are the being extension of Binets formulas for the Fibonacci and Lucas numbers in continuous domain, transform the Fibonacci numbers theory into continuous theory because every identity for the hyperbolic Fibonacci and Lucas functions has its discrete analogy in the framework of the Fibonacci

Also, from the above denition, we give the derivative formulas of the symmetrical Lucas functions as follows: d cLs(x) d sLs(x) = cLs(x) ln , = sLs(x) ln , dx dx d tLs(x) = 4secLs2 (x) ln . (A3) dx The above symmetrical hyperbolic Lucas functions are connected with the classical hyperbolic functions by the following simple correlations: sLs(x) = 2 sinh(x ln ) , tLs(x) = tanh(x ln ) . cFs(x) = 2 cosh(x ln ) , (A4)

References
[1] M.J. Ablowitz and H. Segur, Solitons and the Inverse Scattering Transform, SIAM, Philadelphia (1981); V.B. Matveev and M.A. Salle, Darboux Transformation and Solitons, Springer-Verlag, Berlin (1991).

[2] W. Maliet, Am. J. Phys. 60 (1992) 650. [3] E.G. Fan, Phys. Lett. A 277 (2000) 212. [4] D.S. Li, F. Gao, and H.Q. Zhang, Chaos, Solitons & Fractals 20 (2004) 1021.

1012

Emad A-B. ABDEL-SALAM

Vol. 52

[5] C.L. Bai and H. Zhao, Chaos, Solitons & Fractals 27 (2006) 1026. [6] H.A. Abdusalam, Int. J. Nonlinear Sci. Numer. Simul. 6 (2005) 99. [7] S.K. Liu, Z.T. Fu, S.D. Liu, and Q. Zhao, Phys. Lett. A 289 (2001) 69. [8] Z.T. Fu, S.K. Liu, S.D. Liu, and Q. Zhao, Phys. Lett. A 290 (2001) 72. [9] X.Y. Tang, S.Y. Lou, and Y. Zhang, Phys. Rev. E 66 (2002) 046601. [10] X.Y. Tang and S.Y. Lou, J. Math. Phys. 44 (2003) 4000. [11] C.L. Zheng and L.Q. Chen, J. Phys. Soc. Jpn. 73 (2004) 293. [12] C.L. Zheng and Z.M. Sheng, Inter. J. Mod. Phys. B 17 (2003) 4407. [13] C.L. Zheng and J.F. Zhang, Chin. Phys. Lett. 19 (2002) 1399. [14] C.L. Zheng, J.F. Zhang, and Z.M. Sheng, Chin. Phys. Lett. 20 (2003) 331. [15] S.Y. Lou and X.B. Hu, J. Math. Phys. 38 (1997) 6401. [16] S.Y. Lou and X.B. Hu, Commun. Theor. Phys. 29 (1998) 145. [17] J.F. Zhang and C.L. Zheng, Chin. J. Phys. 41 (2003) 242. [18] X.Y. Tang and S.Y. Lou, J. Math. Phys. 44 (2003) 4000. [19] C.Q. Dai and J.F. Zhang, J. Math. Phys. 47 (2006) 43501. [20] C.Q. Dai and R.P. Chen, Z. Naturf. A 61 (2006) 307. [21] C.Q. Dai and Y.Z. Ni, Phys. Scr. 74 (2006) 584. [22] C.Q. Dai, Phys. Scr. 75 (2007) 310. [23] C.L. Bai and H. Zhao, Eur. Phys. J. B 42 (2004) 581. [24] C.L. Bai and H. Zhao, Chin. Phys. 14 (2005) 285.

[25] C.L. Bai and H. Zhao, Z. Naturf. A 59 (2004) 729. [26] C.L. Bai and H. Zhao, Z. Naturf. A 59 (2004) 412. [27] C.L. Bai and H. Zhao, Chaos, Solitons and Fractals 25 (2005) 481. [28] R. Radha and M. Lakshmanan, J. Math. Phys. 35 (1994) 4746; J. Phys. A: Math. Gen. 29 (1997) 1551; J. Phys. A: Math. Gen. 30 (1997) 3229. [29] C.L. Bai and H. Zhao, J. Phys. A: Math. Gen. 39 (2006) 3283. [30] C.L. Bai, H. Zhao, and X.Y. Wang, Nonlinearity 19 (2006) 1697. [31] C.L. Bai and H. Zhao, Chin. J. Phys. 44 (2006) 94. [32] A.P. Stakhov and B. Rozin, Chaos, Solitons & Fractels 23 (2005) 379. [33] A.P. Stakhov, Comput. Math. Appl. 17 (1989) 613. [34] M.S. El Naschie, Chaos, Solitons & Fractals 1 (1992) 485. [35] M.S. El Naschie, Chaos, Solitons & Fractals 4 (1994) 177. [36] M.S. El Naschie, Chaos, Solitons & Fractals 14 (2002) 649. [37] M.S. El Naschie, Chaos, Solitons & Fractals 9 (1998) 975. [38] M.S. El Naschie, Chaos, Solitons & Fractals 17 (2003) 631. [39] E.A-B. Abdel-Salam, Z. Naturf. 63a (2008) 671. [40] R. Camassa and D.D. Holm, Phys. Rev. Lett. 71 (1993) 1661. [41] E.A-B. Abdel-Salam and D. Kaya, Z. Naturforsch. 64a (2009) 1. [42] M.F. El-Sabbagh, M.M. Hassan, and E.A-B. AbdelSalam, Physica Scripta 80 (2009) 015006. [43] E. A-B. Abdel-Salam, Z. Naturforsch. 64a (2009) 639.