`Abstract and Applied AnalysisVolumeΒ 2012Β (2012), Article IDΒ 739156, 17 pageshttp://dx.doi.org/10.1155/2012/739156`
Research Article

On the Nonlinear Instability of Traveling Waves for a Sixth-Order Parabolic Equation

Department of Mathematics, Jilin University, Changchun 130012, China

Received 4 June 2012; Accepted 2 August 2012

Copyright Β© 2012 Zhenbang Li and Changchun Liu. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

We study the instability of the traveling waves of a sixth-order parabolic equation which arises naturally as a continuum model for the formation of quantum dots and their faceting. We prove that some traveling wave solutions are nonlinear unstable under perturbations. These traveling wave solutions converge to a constant as .

1. Introduction

In this paper, we consider the following sixth-order parabolic equation where .

Equation (1.1) arises naturally as a continuum model for the formation of quantum dots and their faceting; see [1]. Here denotes the surface slope. The high-order derivatives are a result of the additional regularization energy which is required to form an edge between two-plane surfaces with different orientations.

During the past years, only a few works have been devoted to the sixth-order parabolic equation [2β7]. Barrett et al. [2] considered the above equation with . A finite element method is presented which proves to be well posed and convergent. Numerical experiments illustrate the theory.

Recently, JΓΌngel and MiliΕ‘iΔ [5] studied the sixth-order nonlinear parabolic equation They proved the global-in-time existence of weak nonnegative solutions in one space dimension with periodic boundary conditions.

Evans et al. [3, 4] considered the sixth-order thin film equation containing an unstable (backward parabolic) second-order term By a formal matched expansion technique, they show that, for the first critical exponent for , where is the space dimension, the free-boundary problem with zero-height, zero-contact-angle, zero-moment, and zero-flux conditions at the interface admits a countable set of continuous branches of radially symmetric self-similar blow-up solutions , , where is the blow-up time.

Korzec et al. [8] considered the sixth-order equation New type of stationary solutions is derived by an extension of the method of matched asymptotic expansion.

In this paper, we study instability of the traveling waves of (1.1). Our main result is as follows.

Theorem 1.1. All the traveling waves of (1.1) satisfying , are nonlinearly unstable in the space , where denotes derivative of .

The stability and instability of special solutions for the higher-order parabolic equation are very important in the applied fields. Carlen et al. [9] proved the nonlinear stability of fronts for the Cahn-Hilliard, under perturbations. Gao and Liu [10] prove that it is nonlinearly unstable under perturbations, for some traveling wave solution of the convective Cahn-Hilliard equation. The relevant equations have also been studied in [11, 12]. The main difficulties for treating (1.1) are caused by the principal part and the lack of the Lyapunov functional. Our proof is based on the principle of linearization. We invoke a general theorem that asserts that linearized instability implies nonlinear instability.

This paper is organized as follows. In the next section, we find an exact traveling wave solution for (1.1). In Section 3, we give the proof of our main result.

2. Exact Traveling Wave Solutions

In this section, we construct an exact traveling wave which satisfies all conditions of Theorem 1.1.

If is a traveling wave solution of (1.1), then satisfies the ordinary differential equation Let . Then Substituting the above equations into (2.1), we have Then comparing the order of , we obtain A simple calculation shows that , . Hence, we get that is, that is, We easily proved that and satisfies the conditions of the Theorem 1.1.

3. Proof of The Result

To prove the Theorem 1.1, we first consider an evolution equation where is a linear operator that generates a strongly continuous semigroup on a Banach space , and is a strongly continuous operator such that . In [13], authors considered the whole problem only on space X, that is, the nonlinear operator maps to . However, many equations posses nonlinear terms that include derivatives and therefore, maps into a large Banach space . Hence, they again got the following lemma.

Lemma 3.1 (see [14]). Assume the following.(i)X, Z are two Banach spaces with and for .(ii)L generates a strongly continuous semigroup on the space Z, and the semigroup maps Z into X for and .(iii)The spectrum of on meets the right half-plane, .(iv) is continuous and ββ such that , for .
Then the zero solution of (3.1) is nonlinearly unstable in the space .

In this paper, we are going to use Lemma 3.1 for the proof of Theorem 1.1.

Definition 3.2. A traveling wave solution of (1.1) is said to be nonlinearly unstable in the space X, if there exist positive and , a sequence of solutions of (1.1) and a sequence of time such that but .

If is a traveling wave solution of (1.1), then letting , we have that is, that is, where with the initial value So the stability of traveling wave solutions of (1.1) is translated into the stability of the zero solution of (3.4). In order to prove Theorem 1.1, taking , , we need to prove that the four conditions of Lemma 3.1 are satisfied by the associated equation (3.4). The condition (i) is satisfied by our choice of and .

Denote the linear partial differential operator in (3.4) by = with . Then (3.4) may be rewritten in the form of (3.1) Note the F maps into , using the Sobolev embedding theorem, we have So, the condition (iv) is satisfied.

To prove condition (ii) in Lemma 3.1, we need the following two lemmas.

Lemma 3.3. Let . Then

Proof. We write . By Fourier transformation Hence, On the other hand, letting , we have with , . Elementary computation shows that Thus, since . Thus, Lemma 3.3 has been proved.

Lemma 3.4. Let + with , . Then

Proof. Consider the initial value problem Then , , , thus Denote , , , , and Then, we have where we use to denote .
By iteration, Let . Then Multiplying both sides of the above inequality by , we have Integrating the above inequality with respect to over , we obtain that is, Observing that is bounded and substituting the above inequality into (3.22), we get thus (3.17) has been proven.
Next, we prove the inequality (3.16). Clearly, we have where is defined in Lemma 3.3, and we use to denote .
By iteration, The second term on the right of (3.29) is where . By exchanging the order of integration, we get from the third term on the right side of (3.29), then Therefore (3.28)β(3.32) imply From (3.17), we know . Then Therefore, there exists a , such that So, we proved the inequality (3.16).
Hence (3.17) is proven and proof of Lemma 3.4 is finished.

By Lemma 3.4, the condition (ii) is proved.

We now proceed to verify condition (iii) of Lemma 3.1. Observing that if satisfies then also satisfies the above equation. By uniqueness of solution, we know that generates a strongly continuous semigroup on the Banach space (see [15] p.344). By Fourier transformation, the essential spectrum of on is The curve meets the vertical lines for because .

We now prove that the same curve belongs to the essential spectrum of .

Lemma 3.5. The essential spectrum of on contains that of .

Proof. Let and let . Following Schechter [16], if there exists a sequence with and does not have a strongly convergent subsequence in . Here we use the definition if and only if is Fredholm with index zero. Now let be a function with compact support in . Define where is chosen so that . In fact, for some positive constant . Hence . Since but is bounded away from zero, can have no convergent subsequence in .
It remains to show that . We write A simple calculation shows that Thus, Moreover, for any positive integer , , as , we have From the assumptions on , we obtain Similarly, we have In addition, Thus, So from the estimates above, The proof of Lemma 3.5 is completed.

Therefore all the four conditions of Lemma 3.1 are satisfied by the linearized equation (3.4) and Theorem 1.1 has been proved.

References

1. T. V. Savina, A. A. Golovin, S. H. Davis, A. A. Nepomnyashchy, and P. W. Voorhees, βFaceting of a growing crystal surface by surface difusion,β Physical Review E, vol. 67, no. 2, part 1, Article ID 021606, 2003.
2. J. W. Barrett, S. Langdon, and R. Nürnberg, βFinite element approximation of a sixth order nonlinear degenerate parabolic equation,β Numerische Mathematik, vol. 96, no. 3, pp. 401β434, 2004.
3. J. D. Evans, V. A. Galaktionov, and J. R. King, βUnstable sixth-order thin film equation. I. Blow-up similarity solutions,β Nonlinearity, vol. 20, no. 8, pp. 1799β1841, 2007.
4. J. D. Evans, V. A. Galaktionov, and J. R. King, βUnstable sixth-order thin film equation. II. Global similarity patterns,β Nonlinearity, vol. 20, no. 8, pp. 1843β1881, 2007.
5. A. Jüngel and J.-P. Milišić, βA sixth-order nonlinear parabolic equation for quantum systems,β SIAM Journal on Mathematical Analysis, vol. 41, no. 4, pp. 1472β1490, 2009.
6. C. Liu, βQualitative properties for a sixth-order thin film equation,β Mathematical Modelling and Analysis, vol. 15, no. 4, pp. 457β471, 2010.
7. C. Liu, βA sixth order degenerate equation with the higher order P-Laplacian operator,β Mathematica Slovaca, vol. 60, no. 6, pp. 847β864, 2010.
8. M. D. Korzec, P. L. Evans, A. Münch, and B. Wagner, βStationary solutions of driven fourth- and sixth-order Cahn-Hilliard-type equations,β SIAM Journal on Applied Mathematics, vol. 69, no. 2, pp. 348β374, 2008.
9. E. A. Carlen, M. C. Carvalho, and E. Orlandi, βA simple proof of stability of fronts for the Cahn-Hilliard equation,β Communications in Mathematical Physics, vol. 224, no. 1, pp. 323β340, 2001.
10. H. Gao and C. Liu, βInstability of traveling waves of the convective-diffusive Cahn-Hilliard equation,β Chaos, Solitons and Fractals, vol. 20, no. 2, pp. 253β258, 2004.
11. S. A. Gourley and N. F. Britton, βInstability of travelling wave solutions of a population model with nonlocal effects,β IMA Journal of Applied Mathematics, vol. 51, no. 3, pp. 299β310, 1993.
12. C. Liu, βInstability of traveling waves for a generalized diffusion model in population problems,β Electronic Journal of Qualitative Theory of Differential Equations, no. 18, pp. 1β10, 2004.
13. J. Shatah and W. Strauss, βSpectral condition for instability,β in Contemporary Mathematics, vol. 255, pp. 189β198, 2000.
14. W. Strauss and G. Wang, βInstability of traveling waves of the Kuramoto-Sivashinsky equation,β Chinese Annals of Mathematics B, vol. 23, no. 2, pp. 267β276, 2002.
15. Q. Ye and Z. Li, Theory of Reaction-Diffusion Equations, Science Press, Beijing, China, 1994.
16. M. Schechter, Spectra of Partial Differential Operators, American Elsevier, New York, NY, USA, 1971.