Abstract
This paper is concerned with the existence and stability of time periodic traveling curved fronts for reaction–diffusion equations with bistable nonlinearity in \({\mathbb {R}}^3\). We first study the existence and other qualitative properties of time periodic traveling fronts of polyhedral shape. Furthermore, for any given \(g\in C^{\infty }(S^1)\) with \(\min \nolimits _{0\le \theta \le 2\pi }g(\theta )=0\) that gives a convex bounded domain with smooth boundary of positive curvature everywhere, which is included in a sequence of convex polygons, we show that there exists a three-dimensional time periodic traveling front by taking the limit of the solutions corresponding to the convex polyhedrons as the number of the lateral surfaces goes to infinity.
We’re sorry, something doesn't seem to be working properly.
Please try refreshing the page. If that doesn't work, please contact support so we can address the problem.
1 Introduction
Traveling fronts have been extensively studied since the last decade, one can refer to [2, 3, 12, 40] and references therein for the study of planar traveling fronts to the following autonomous reaction–diffusion equation
in one or multidimensional space. Recently, the study on nonplanar traveling fronts among mathematicians has attracted an increasing attention and many new types of nonplanar traveling fronts have been observed of (1.1). For instance, Brazhnik and Tyson [6], Hamel and Nadirashsili [15], Hamel and Roquejoffre [16], El Smaily et al. [11] considered nonplanar traveling fronts of (1.1) with monostable nonlinearity for \(N\ge 2\) (i.e., \(f(0)=f(1)=0\), \(f^{\prime }(0)>0\), \(f^{\prime }(1)<0\)); Bonnet and Hamel [5], Hamel et al. [17] examined the V-shaped traveling front of (1.1) with the combustion nonlinearity (i.e., \(f(s)=0\) for \(s\in [0,\theta ]\cup \{1\}\), \(f(s)>0\) in \((\theta ,1)\)); Chen et al. [9], Gui [13] studied the existence and qualitative properties of cylindrically symmetric traveling fronts of (1.1) with the balanced bistable nonlinearity (i.e., \(f(0)=f(a)=f(1)=0\), \(f^{\prime }(0)<0\), \(f^{\prime }(1)<0\), \(f^{\prime }(a)>0\) and \(\int _0^1f(s)\hbox {d}s=0\)); Hamel et al. [18, 19], Ninomiya and Taniguchi [28, 29] studied V-shaped traveling fronts of (1.1) for the unbalanced bistable case (i.e., \(f(0)=f(a)=f(1)=0\), \(f^{\prime }(0)<0\), \(f^{\prime }(1)<0\), \(f^{\prime }(a)>0\) and \(\int _0^1f(s)\hbox {d}s\ne 0\)). Additionally, Hamel et al. [18, 19] considered cylindrically symmetric traveling fronts of (1.1) for \(N\ge 3\); Taniguchi [35, 36], Kurokawa and Taniguchi [23] studied pyramidal-shaped traveling fronts of (1.1) for \(N\ge 3\). More recently, Wang [41] and Wang et al. [45] developed the arguments of [28, 29, 35, 36] to reaction–diffusion systems. See also Haragus and Scheel [20, 21] for the study of almost planar traveling fronts. Very recently, Taniguchi [37] studied multidimensional traveling fronts of (1.1) for \(N=3\). For the study on the nonconvex and nonconnected traveling fronts, we refer to del Pino et al. [10]. Other related works on traveling fronts for autonomous reaction–diffusion equations can be referred to [7, 8, 14, 26, 27, 32, 34, 38, 39, 43].
It is well known that in population dynamics interactive species may live in a fluctuating environment, for instance, physical environment conditions such as temperature, humidity and the available of food, water and other resources usually varies in time with seasonal or daily changes [46]. Therefore, in nature, another more realistic model might be of the following form
When the data of (1.2) are functions with commensurate time period, we call (1.2) a periodic equation. Recently, there are a lot of works devote to the study of traveling fronts of (1.2). One can refer to [1, 4, 22, 25, 30, 31, 47] for the study of time almost periodic and time periodic planar traveling fronts. Nevertheless, a very little attention has been paid to the study of nonplanar traveling fronts for nonautonomous reaction–diffusion equations, even for the time periodic case. As far as we know, Wang and Wu [44] proved that there exists a two-dimensional time periodic V-shaped traveling front. Moreover, they showed that such a traveling curved front is asymptotically stable. Sheng et al. [33] addressed the existence and asymptotic stability of time periodic pyramidal-shaped traveling fronts. Very recently, Wang [42] showed the existence of time periodic cylindrically symmetric traveling fronts by appealing to the method of comparison principle and the asymptotic speed of propagation.
However, the issue of the existence and stability of multidimensional time periodic traveling curved fronts for reaction–diffusion equation with bistable nonlinearity is still open. The main contribution of the current study is to give an affirmative answer to this issue. Actually, motivated by [33, 37, 42, 44], we first consider the existence, uniqueness and stability of three-dimensional time periodic traveling fronts of polyhedral shape. Then, for any given \(g\in C^{\infty }(S^1)\) with \(\min \nolimits _{0\le \theta \le 2\pi } g(\theta ) = 0\) that defines a convex domain with smooth boundary of positive curvature everywhere, which is included in a sequence of convex polygons, we show that there exists a three-dimensional time periodic traveling front by taking the limit of the solutions associated with the convex polyhedrons as the number of the lateral surfaces goes to infinity.
In this paper, we study traveling curved fronts of the following reaction–diffusion equation
under the following hypotheses:
- (H1) :
-
There exists \(T > 0\) such that \(f (u,t)=f(u,t+T)\) for all \((u, t) \in {\mathbb {R}}^2\).
- (H2) :
-
The period map \(P(\alpha ):= \omega (\alpha , T )\) has exactly three fixed points \(\alpha ^{-},{\alpha }^0,\alpha ^+ \) such that \(\alpha ^{-}<\alpha ^0<\alpha ^+ \), where \(\omega (\alpha ,t)\) is the solution of
$$\begin{aligned} \omega _t=f(\omega ,t),\quad t\in {\mathbb {R}},\quad \omega (\alpha ,0) = \alpha \in {\mathbb {R}}. \end{aligned}$$Furthermore, they are nondegenerate and \(\alpha ^{\pm }\) are stable, i.e.,
$$\begin{aligned} \frac{d}{\hbox {d}\alpha } P(\alpha ^{\pm })<1<\frac{d}{\hbox {d}\alpha } P(\alpha ^0). \end{aligned}$$ - (H3) :
-
There exists \(\nu _0>0\) such that \(\nu ^{+}+\nu ^{-}+f_u(W^{\pm }(t),t)>\nu _0\) for any \(t \in [0, T ]\), where
$$\begin{aligned} \nu ^{\pm }:=-\frac{1}{T}\int _0^T f_u(W^{\pm }(\lambda ),\lambda )\hbox {d}\lambda , \end{aligned}$$and
$$\begin{aligned} W^{\pm }(t):=\omega (\alpha ^{\pm },t),\quad W^0(t):=\omega (\alpha ^0,t). \end{aligned}$$ - (H4) :
-
There exist constants \(r_0 > 0\) and \(\epsilon \in \left( 0,\min \nolimits _{t\in [0,T ]}(W^0(t)-W^-(t))\right) \) such that \(\overline{f}(u,t) \ge r_0 u(\epsilon -u)\) for any \(u \in (0,\epsilon )\) and \(t\in [0, T]\), where \(\overline{f }(u,t):=f(W^0(t), t)-f(W^0(t)-u, t)\).
A typical example of f satisfying (H1)–(H3) is the cubic potential \(f=(1 -u^2)(2u-\rho (t))\), where \(\rho (t)\in (-2,2)\) is T-periodic, which is the particular case of the following more general example (see Alikakos et al. [1])
where \(\rho \in C^1\) and \(p\in C^3\) satisfy \(\rho (\cdot +T) =\rho (\cdot )\), and \(p(\pm 1) = 0\), \(p(\cdot )>0\) in \((-1,1)\). Moreover, by taking \(|\rho (t)|\le 2\sqrt{5}/5\), then such a function f satisfies (H4) (see Wang [42]).
It is known from [1] that when \(f (u, t)\in C^{2,1}({\mathbb {R}}\times {\mathbb {R}})\) satisfies (H1) and (H2), there exists a unique solution pair (c, U) of (1.3) such that
where the function \(U(\cdot ,\cdot ):{\mathbb {R}}\times {\mathbb {R}}\rightarrow \mathbb {R}\) is the wave profile and the constant \(c \in {\mathbb {R}}\) is the wave speed. In addition, (c, U) enjoys the following properties:
- (i) :
-
\(U(\xi ,t)\) is monotone increasing with respect to the moving coordinate for each t. Namely, \(U_{\xi }(\xi ,t)>0\) in \({\mathbb {R}}\times {\mathbb {R}}\).
- (ii) :
-
There exist positive constants \(C_1\) and \(\beta _1\) satisfying
$$\begin{aligned} |U(\pm \xi ,t)-W^{\pm } (t)|+|U_{\xi }(\pm \xi ,t)|+|U_{\xi \xi }(\pm \xi ,t)| \le C_1 \hbox {e}^{-\beta _1 \xi },\quad \xi \ge 0, t\in {\mathbb {R}}. \end{aligned}$$That is, U exponentially approaches its limits as \(\xi \rightarrow \pm \infty \).
Without loss of generality, we assume that the solutions travel toward z-direction. Set
For simplicity, we denote \(v(x,y,z_1,t)\) by v(x, y, z, t). Substituting v into (1.3), we have
Hereafter, we always assume \(l>c>0\). The objective of this paper is to seek for the solution V (x, y, z, t) of
Let
Given \(n \ge 3\) be an integer. Assume that \(\{\theta _j\}_{1\le j\le n}\) satisfy
where \(\theta _{n+1}=\theta _1+2\pi \). Given, \(l_j\) with
Then,
is the unit normal vector of a surface \(\{z = \tau (x\cos \theta _j + y\sin \theta _j)\}\). Putting
then \(\{(x,y,z)\in {\mathbb {R}}^3|-z \ge h(x, y)\}\) is a convex polyhedron. If \((l_1,l_2,\ldots ,l_n) = (0,0,\ldots ,0)\), the polyhedron becomes a pyramid in \({\mathbb {R}}^3\).
Denote
For \(j=1,2\ldots ,n\), define
We note that \(\Omega _j\ne {\emptyset }\) for all \(1\le j \le n\). Here \(\Omega _1,\Omega _2,\ldots ,\Omega _n\) are located counterclockwise. Set
Let
be a part of an edge of a polyhedron \(\{(x,y,z)\in {\mathbb {R}}^3|-z \ge h(x, y)\}\). If \((l_1,l_2,\ldots ,l_n) = (0,0,\ldots ,0)\) and \(\Theta =0\), then \(\Gamma _j\) and \(\bigcup _{j=1}^n \Gamma _j\) stand for an edge and the set of all edges of a pyramid, respectively. For each \(\gamma > 0\), we define
Theorem 1.1
Let \(l>c>0\) and h(x, y) be given by (1.8). Under the assumptions (H1)–(H3), there exists a solution V(x, y, z, t) of (1.6)–(1.7) such that
Moreover, if
then, the solution \(v(x,y,z,t;v_0)\) of (1.5) satisfies
where \(\hat{l}:=\max \nolimits _{1\le j\le n} l_j\ge 0\), \(E_j\) is the two-dimensional V-shaped traveling front defined in (2.6), \(\widetilde{V}\) is the pyramidal traveling front given in Theorem 2.2, \(X_j(-\hat{l})\), \(Y_j(-\hat{l})\) and \(X_j(\rho )\), \(Y_j(\rho )\) satisfy \(h(X_j(-\hat{l}), Y_j(-\hat{l}))=-\tau \hat{l}\) and \(h(X_j(\rho ), Y_j(\rho ))=\tau \rho \), respectively. Furthermore, V enjoys the following properties:
- (i) :
-
Let h(x, y) be defined in (1.8), \(\overline{h}(x,y):=\tau \max _{1\le j\le n}(x\cos \theta _j + y\sin \theta _j-\overline{l}_j)\) with \(\min \nolimits _{1\le j\le n}\overline{l}_j\ge 0\) and \(V({\mathbf {x}},t)\) be given in Theorem 1.1, \(\overline{V}({\mathbf {x}},t)\) be the traveling fronts of polyhedral-shape associated with \(\overline{h}\). If \(\overline{h}(x,y)\ge h(x,y)\) for any \((x,y)\in {\mathbb {R}}^2\), then it holds \(\overline{V}({\mathbf {x}},t)\ge V({\mathbf {x}},t)\) for all \(({\mathbf {x}},t)\in {\mathbb {R}}^4\).
- (ii) :
-
There holds
$$\begin{aligned} \frac{\partial V}{\partial \nu }>0\quad {\mathrm {in}}\,\,{\mathbb {R}}^4 \end{aligned}$$for
$$\begin{aligned} \nu =\frac{1}{\sqrt{1+t_1^2+t_2^2}}\left( \begin{array}{c} t_1 \\ t_2\\ 1 \\ \end{array}\right) \quad {\mathrm {with}}\,\sqrt{t_1^2+t_2^2}\le \frac{1}{\tau }. \end{aligned}$$ - (iii) :
-
If \( h(x,y)=h(|x|,|y|)\), then one has
$$\begin{aligned}&V(x,y,z,t)=V(|x|,|y|,z,t),\quad (x,y,z,t) \in {\mathbb {R}}^4,\\&V_x(x,y,z,t)>0\quad {\mathrm {for}}\,(x,y,z,t) \in (0,\infty )\times {\mathbb {R}}^3,\\&V_x(0,y,z,t)=0\quad {\mathrm {for}}\,(y,z,t) \in {\mathbb {R}}^3,\\&V_y(x,y,z,t)>0\quad {\mathrm {for}}\,(x,y,z,t) \in {\mathbb {R}}\times (0,\infty )\times {\mathbb {R}}^2,\\&V_y(x,0,z,t)=0\quad {\mathrm {for}}\,(x,z,t) \in {\mathbb {R}}^3. \end{aligned}$$
An immediate consequence of this theorem is the following corollary.
Corollary 1.2
Let \(V({\mathbf {x}},t)\) be the time periodic traveling curved front defined in Theorem 1.1. If there is a time periodic solution \(w({\mathbf {x}},t)\) of (1.6) and (1.7) satisfying (1.10), then it holds
In what follows, we treat the three-dimensional traveling fronts of (2.1) for any given \(g\in C^\infty (S^1)\), where
for any \(\theta \in {\mathbb {R}}\). We identify \(S^1\) with \({\mathbb {R}}/2\pi {\mathbb {Z}}\). Let \(g \in C^\infty (S^1)\) be any given function with \(\min _{0 \le \theta \le 2\pi } g(\theta )=0\) and \(r_{*}\ge 1\) be large enough such that
Setting
then,
is a smooth convex closed curve such that
where \(\kappa (\theta )\) is the curvature of \({\mathcal {C}}\) given by
Define
Then, \({\mathcal {C}}\) is the corresponding boundary to \({\mathcal {D}}\). Let
Then, we have \(0<\kappa _{\mathrm{min}}\le \kappa _{\mathrm{max}}<\infty \). Set
Let \(R_{*}\in \left( \kappa _{\mathrm{min}}^{-1},\infty \right) \) be large enough such that, for each \(\theta \in [0,2\pi )\), there is a circle of radius \(R_{*}\) that circumscribes \({\mathcal {C}}\) at \((R(\theta )\cos \theta ,R(\theta )\sin \theta )\). Let \((\xi _{*}(\theta ),\eta _{*}(\theta ))\) and \(B(\theta )\) be the center and the interior of this circle for each \(\theta \in [0,2\pi )\). It is obvious that \(\mathcal {\overline{D}}\subset \overline{B}(\theta )\) for all \(\theta \in [0,2\pi )\).
Theorem 1.3
Assume that (H1)–(H4) hold. Let \(g\in C^\infty (S^1)\) be any given function such that \(\min \nolimits _{0\le \theta \le 2\pi }g(\theta )=0\) and \(R(\theta )=r_{*}+g(\theta )\). Then, there exits a solution \(\widetilde{W}({\mathbf {x}},t)\in C^{2,1}({\mathbb {R}}^3\times {\mathbb {R}})\) of (1.6)–(1.7) satisfying
for all \((x,y,z,t)\in {\mathbb {R}}^4\). Morevoer, one has
Furthermore, such \(\widetilde{W}(x,y,z,t)\) is uniquely determined by (1.6) and (1.19). Moreover, if
for all \((x,y,z)\in {\mathbb {R}}^3\), then the solution \(v({\mathbf {x}},t;v_0)\) of (1.5) with initial value \(v_0\) satisfies
or equivalently
where \(\Psi \) is the cylindrically symmetric traveling front defined in Theorem 2.4.
The rest of this paper is organized as follows. In Sect. 2, we state some preliminaries including two-dimensional time periodic V-shaped traveling fronts, three-dimensional time periodic pyramidal-shaped traveling fronts and cylindrically symmetric time periodic traveling fronts. Section 3 is devoted to the existence and stability of time periodic traveling curved fronts with polyhedral shape, that is, we prove Theorem 1.1. In Sect. 4, we show Theorem 1.3.
2 Preliminaries
In this section, we recall some results established by Wang and Wu [44], Sheng et al. [33] and Wang [42] about two-dimensional time periodic V-shaped taveling fronts, three-dimensional time periodic pyramidal traveling fronts and time periodic cylindrically symmetric in \({\mathbb {R}}^3\), respectively. In the sequel, we write (c, U) be the planar traveling front defined by (1.4).
2.1 Two-dimensional V-shaped fronts
Let \(\hat{v}(\xi ,\eta ,t;\hat{v}_0)\) be the solution of the following equation
Then, from [44, Theorem 1.1], we have the following theorem.
Theorem 2.1
Assume that (H1)–(H3) hold and \(l> c>0\). Then, there exists a unique \(\widehat{V}(\xi ,\eta ,t)\) such that
and
Moreover, there holds
One also has
2.2 Three-dimensional pyramidal traveling fronts
Consider the following problem:
Set
and
Define
Substituting \(E_j(x,y,z,t)\) into (2.4), we obtain that every \(E_j(x,y,z,t)\) is a time periodic V-shaped traveling front with speed \(\frac{l}{\sqrt{1+\tau ^2k_j^2}}>c\). By Sheng et al. [33, Theorems 1.1–1.2 and Lemma 4.6], we get the following theorem.
Theorem 2.2
Assume that \(l>c>0\) and (H1)–(H3) hold. Let p(x, y) be given by (2.5). Then, there exists a solution \(\widetilde{V}(x, y, z, t)\) of (2.4) such that
and
Lemma 2.3
There exists positive constants \(\delta _0\), \(\sigma \) and \(\beta \) \((\beta <\frac{\nu _0}{4})\) such that, for any \(\delta \in (0, \delta _0)\), the functions \(w^\pm \) defined by
are a supersolution and a subsolution of (2.4) on \({\mathbf {x}}\in {\mathbb {R}}^3\) and \(t\in [0,\infty )\), respectively, where
and
with the constants \(\nu _0\), \(\nu ^+\) and \(\nu ^-\) are defined as in (H3).
2.3 Cylindrically symmetric traveling fronts
Let
Clearly,
is tangent to
for any \(m\in {\mathbb {N}}\) and \(1\le j\le 2^{m}\). Replacing p(x, y) by \(p^{(m)}(x,y)\) in Theorem 2.2, we obtain a sequence of time periodic pyramidal traveling fronts of (2.4), namely,
where
Denote the edge of the pyramid \(-z=p^{(m)}(x,y)\) by \(\Gamma ^m\) and put
Owing to \(\widetilde{v}_0^{m,-}({\mathbf {x}},0)\) is nondecreasing on \(x \in (0,\infty ) \) and \(y\in (0,\infty )\) and is even on \(x\in {\mathbb {R}}\) and \(y\in {\mathbb {R}}\), respectively. It then follows from Theorem 2.2 that
where \(\nu =\frac{1}{\sqrt{1+\nu _1^2+\nu _2^2}} \left( \begin{array}{c} \nu _1 \\ \nu _2 \\ 1 \\ \end{array} \right) \) with \(\sqrt{\nu _1^2+\nu _2^2}\le \frac{1}{\tau }\). Thanks to
one infers
where
Let \(z^m \in {\mathbb {R}}\) be such that
for a given constant \(\theta _0 \in (\alpha ^-,\alpha ^0)\). Denote
It then deduces from the parabolic estimate [24] and Theorem 2.2 that there exists a solution \(W({\mathbf {x}},t)\in C^{2,1}({\mathbb {R}}^3\times {\mathbb {R}})\) of (2.4) (even if up to an extraction of some subsequence) such that
Define
for any \((x,y,z,t)\in {\mathbb {R}}^4\). Then, one has
For a given constant \(\theta _0 \in (\alpha ^-,\alpha ^0)\), we define \(\phi (r)\in {\mathbb {R}}\) by
By a shift, one can assume \(U(0,0)=\theta _0\) without loss of generality. Then, Wang [42] established the following results.
Theorem 2.4
Assume that (H1)–(H4) hold. Suppose \(c > 0\). Then, \(\Psi (r,z,t)\) defined by (2.13) satisfies (2.14) and \(\Psi (r,z,t+T)=\Psi (r,z,t)\) for all \((r,z,t)\in {\mathbb {R}}^3\). Moreover, there hold
- (i) :
-
\(\frac{\partial }{\partial r} \Psi (r,z,t)>0\) and \(\frac{\partial }{\partial z} \Psi (r,z,t)>0 ~\text {for any}~ (r,z,t)\in (0,\infty )\times {\mathbb {R}}^2.\)
- (ii) :
-
\( \lim _{z\rightarrow +\infty }\Vert \Psi (\cdot ,z,t)-W^{+}(t)\Vert _{C({\mathbb {R}}^2)}=0\), \(\lim _{z\rightarrow -\infty }\Vert \Psi (\cdot ,z,t)-W^{-}(t)\Vert _{C_{loc}({\mathbb {R}}^2)}=0\).
- (iii) :
-
\(\frac{\partial }{\partial \nu } \Psi (r,z,t)>0\) for any \(r>0\), \(z>0\), \(t\in {\mathbb {R}}\), where
$$\begin{aligned} \nu =\frac{1}{\sqrt{1+{\nu ^{\prime }}^2}} \left( \begin{array}{c} \nu ^{\prime } \\ 1 \\ \end{array}\right) \quad {\mathrm {with}}\, \nu ^{\prime }\ge -\frac{1}{\tau }. \end{aligned}$$ - (iv) :
-
\(\lim _{r\rightarrow \infty }\phi ^{\prime }(r)=-\tau \).
- (v) :
-
\( \lim _{r\rightarrow \infty }\Vert \Psi (x+r,z+\phi (r),t)-U(\frac{c}{l}(z+\tau x),t)\Vert _{C^{2,1}_{loc}({\mathbb {R}}^2\times {\mathbb {R}})}=0\).
3 Proof of Theorem 1.1
In this section, we study the existence and asymptotic stability of traveling fronts with convex polyhedral shapes, that is, we prove Theorem 1.1.
Firstly, note that \(\{(x,y,z)\in {\mathbb {R}}^3| -z \ge h(x,y)\}\) is a convex polyhedron. Indeed, if \(-z_i\ge h(x_i,y_i)\) for \(i = 1,2\), then \(-z_i\ge h_j(x_i,y_i)\) for all \(1\le j \le n\) and \(i = 1,2\). It then follows from (1.8) that
for all \(1\le j \le n\) and any \(a \in (0,1)\). Then, one has
Hence, \(\{(x,y,z)\in {\mathbb {R}}^3| -z \ge h(x,y)\}\) is a convex polyhedron.
On the other hand, for any \(\zeta \in {\mathbb {R}}\) and \(1\le j \le n\), let \((X_j(\zeta ),Y_j(\zeta ))\) be such that
Direct computations give
Here, we would like to point out that, for every \(\zeta \in {\mathbb {R}}\), the set \(\{(x,y)\in {\mathbb {R}}^2|h(x,y)\le \zeta \}\) is either an empty set or a nonempty convex closed set in \({\mathbb {R}}^2\). Indeed, if \((x_i,y_i)\) satisfies \(h(x_i,y_i)\le \zeta \) for \(i=1,2\), then \(h_j(x_i,y_i)\le \zeta \) for all \(1\le j\le n\) and \(i=1,2\) . In view of (1.8), we have
for all \(1\le j\le n\) and any \(a\in (0,1)\), whence
If \(\zeta <-\max _{1\le j\le n} l_j\) with \(l_j\) given in (1.8), then \(\{(x,y)\in {\mathbb {R}}^2|h(x,y)\le \zeta \}\) is an empty set. Moreover, it derives from [37, Lemma 3.1] that the set \(\{(x,y)\in {\mathbb {R}}^2|h(x,y)\le \tau \rho \}\) is a convex n-polygon in the x-y plane with vertices \(\{(X_j(\rho ),Y_j(\rho ))\}_{1\le j\le n}\) for any fixed number \(\rho \in (\Theta ,\infty )\).
Proof of Theorem 1.1
Thanks to \(h(X_j(\rho ),Y_j(\rho ))=\tau \rho \) for all \(1\le j\le n\), one infers that
where h and p are defined in (1.8) and (2.5), respectively. Set
Write the solution of (1.5) with \(v_0(x,y,z)=v^-(x,y,z,0)\) by \(v({\mathbf {x}},t;v^-(x,y,z,0))\). Call
Then, the function \(V({\mathbf {x}},t)\in C^{2,1}({\mathbb {R}}^4)\) is a solution of (1.6). Since
then, we have
Letting \(k\rightarrow \infty \), we arrive at
Moreover, a similar discussion to Sheng [33, Lemma 3.1] yields that there is a supersolution \(v^+\) that converges to \(v^-\) far away from the set of edges of the given polyhedron. Thus, (1.10) follows.
Notice that the function \(v^-\) defined in (3.1) is a subsolution of (1.6) and the pyramidal traveling front \(\widetilde{V}\) defined in Theorem 2.2 is solution of (1.6). As a result of the comparison principle, we have
for all \((x,y,z,t)\in {\mathbb {R}}^4\) and \(1\le j\le n\). This shows that
is a supersolution of (1.6) for any \((x,y,z)\in {\mathbb {R}}^3\) and \(t\in [0,T]\). It then follows from the comparison principle that
for all \((x,y,z)\in {\mathbb {R}}^3\) and \(t\in [0,T]\).
On the other hand, since
for all \(1\le j\le n\), then
We consider the left-hand side and the right-hand side as an initial value of (1.5), respectively. Then the comparison principle yields that
for all \(1\le j\le n\). Notice that
Sending \(k\rightarrow \infty \) in (3.4), it then follows from Theorem 2.1, (2.10) and (2.6) that
This together with (3.3), we arrive at
for all \((x,y,z)\in {\mathbb {R}}^3\) and \(t\in [0,T]\). As a consequence, (1.11) and (1.12) follow from (2.8), (2.2) and (3.5).
By (2.3) and (1.11) and applying the Schauder interior estimate to the following equation:
we get
that is, (1.13) holds.
Notice that \(|z+h(x,y)|\rightarrow \infty \) implies \(\mathrm{dist}({\mathbf {x}},\Gamma _j)\rightarrow \infty \) for \(1\le j \le n\). In light of (1.10), we have
By the interpolation \(\Vert \cdot \Vert _{C^1}\le 2\sqrt{\Vert \cdot \Vert _{C^0}\Vert \cdot \Vert _{C^2}}\), we get (1.14) due to
Now, we show that the time periodic curved front V is asymptotically stable. Set
Then, there holds
whence
Considering the left-hand side and the right-hand side of (3.6) as initial values of (1.5), we have
from the comparison principle. Passing \(k\rightarrow \infty \), one derives that
for \(1\le j\le n\). Combining (3.5) and (3.7), we have
for all \((x,y,z)\in {\mathbb {R}}^3\) and \( t\in [0,T]\).
In view of (3.8), we have
For all \((x,y,z)\in {\mathbb {R}}^3\), and \(t\in [0,T]\), set
Then, the comparison principle gives that
and
for all \((x,y,z)\in {\mathbb {R}}^3\), and \(t\in [0,T]\). By (1.11) and (2.8), we infer
On the other hand, owing to (1.13), there exists a positive constant \(\sigma \) such that
where \(0<\delta <\delta _0\) is small enough, \(\delta _0\), \(\beta \) and \(K_0\) are given in Lemma 2.3, \(f_u(u,t)\) denotes the partial derivative of f with respect to u. For \(0<\delta <\delta _0\), it follows from Lemma 2.3 that
is a supersolution to (1.6) on \({\mathbb {R}}^3\times [0,\infty )\). In light of the boundedness of \(V^{*}(x,y,z,t)\) and the monotonicity of V(x, y, z, t) with respect to z, we can take \(\lambda >0\) large enough such that
Then, the comparison principle implies
for \((x,y,z)\in {\mathbb {R}}^3\), \(t\in [0,T]\) and \(k\in {\mathbb {N}}\). Sending \(k\rightarrow \infty \), we get
for all \((x,y,z)\in {\mathbb {R}}^3\) and \(t\in [0,T]\).
Define
It is evident that \(\Lambda \ge 0\) from (3.9). If \(\Lambda = 0\), then \(V^{*}\equiv V\) in \({\mathbb {R}}^4\). We prove \(\Lambda = 0\) by a contradiction argument. Assume that \(\Lambda > 0\). Then, we have
It then follows from the strong maximum principle that
By (1.14), there exists a constant \(R_0>0\) large enough such that
Let
be small enough. If \(|z+h(x,y)|\le R_0-\Lambda -1\), we have
If \(|z+h(x,y)|\ge R_0-\Lambda -1\), we have
Combining (3.10) and (3.11), we get
for all \((x,y,z)\in {\mathbb {R}}^3\). It then follows from the comparison principle and Lemma 2.3 that
for all \((x,y,z)\in {\mathbb {R}}^3\), \(t\in [0,T]\) and \(k\in {\mathbb {N}}\). Letting \(k\rightarrow \infty \), we have
This contradicts with the definition of \(\Lambda \). Thus, \( V\equiv V^{*}\) in \({\mathbb {R}}^4\). Namely,
By a similar argument to
we have
Note that for any fixed \({\mathbf {x}}\in {\mathbb {R}}^3\) and \(t\ge 0\), \(v({\mathbf {x}},t;\cdot )\) is a continuous mapping in \(BU({\mathbb {R}}^3)\) with \(BU({\mathbb {R}}^3)\) standing for the set of bounded and continuous functions. By this continuity, Theorem 2.2 and the comparison principle, we obtain
In other words, the desired result (1.16) holds. The proof is complete.
4 Proof of Theorem 1.3
Set
and
for every \(m\ge 2\), where \(\mathcal {\overline{D}}\) and \(p^{(m)}\) are defined as in (1.17) and (2.11), respectively. Since \((\xi _{*}(\theta ),\eta _{*}(\theta ))\in \mathcal {A}_0\) for all \(\theta \in [0,2\pi )\), and \(\partial B(0;R_{*})\) is the inscribed circle of a polygon \(\{(x,y)\in {\mathbb {R}}^2|p^{(m)}(x,y)=\tau R_{*}\}\), then we have
For any \(\varepsilon >0\) there holds
namely,
as m large enough. It then follows that
for m sufficiently large.
Proof of Theorem 1.3
Define
Since \(\partial B(0;R_{*})\) is the inscribed circle of a polygon \(\{(x,y)\in {\mathbb {R}}^2|p^{(m)}(x,y)=\tau R_{*}\}\), then one arrives at
Moreover, it holds
for all \((\xi ,\eta )\in \mathcal {A}_m\) and \(\theta \in [0,2\pi ]\), whence
It then follows that
for all \((\xi ,\eta )\in \mathcal {A}_m\), \(\theta \in [0,2\pi ]\) and \((x,y)\in {\mathbb {R}}^2\). As a result, one infers that
for all \((\xi ,\eta )\in \mathcal {A}_m\) and \(\theta \in [0,2\pi ]\), where \(z^m\) is defined as in (2.12) and \(\widetilde{V}^m\) is the pyramidal-shaped traveling fronts corresponding to \( p^{(m)}\). Combining this inequality with parabolic estimates [24] and Sobolev imbedding theorem, we obtain that the function \(\widetilde{W}({\mathbf {x}},t)\) defined by
is a solution of (1.6). Moreover, it holds \(\widetilde{W}({\mathbf {x}},t)=\widetilde{W}({\mathbf {x}},t+T)\). On the other hand, by letting \(m\rightarrow \infty \) in (4.1), one derives from Theorem 2.4 that
for all \((\xi ,\eta )\in \mathcal {A}_m\) and \(\theta \in [0,2\pi ]\). Thus (1.18) follows. Furthermore, we have
and
from the definition of \( \widetilde{W}(x,y,z,t)\).
In view of Theorem 2.4 (ii), (iv) and (v), we get
for all \(\theta \in [0,2\pi ]\). It then follows that
Thus, (1.19) follows.
It remains to prove that such \(\widetilde{W}\) is the unique solution of (1.6) under (1.19) which is also stable. We first show that
for \(\delta >0\) small enough. We only sketch the proof here, for details one can refer to [33, Lemma 4.6]. Indeed, we have \(\widetilde{W}_z>0\) in \({\mathbb {R}}^4\). Hence, \(\widetilde{W}_z\) has a positive minimum on any compact subset of \({\mathbb {R}}^4\). Thus, we need only to study \(\widetilde{W}_z({\mathbf {x}}, t)\) as \(|{\mathbf {x}}|\rightarrow \infty \). Assume that \({\mathbf {x}}_i=(x_i,y_i,z_i)\) satisfies \(\lim _{i\rightarrow \infty }|{\mathbf {x}}_i|=\infty \) and \(W^-(t)+\delta \le \widetilde{W}({\mathbf {x}}_i, t)\le W^+(t)-\delta \) for all \(t\in [0,T]\). It suffices to prove \(\lim \inf _{i\rightarrow \infty ,t\in [0,T]}\widetilde{W}_z({\mathbf {x}}_i,t)>0\). It then follows from (1.19) that
Namely,
By the interpolation \(\Vert \cdot \Vert _{C^1}\le 2\sqrt{\Vert \cdot \Vert _{C^0}\Vert \cdot \Vert _{C^2}}\), we have
This together with (2.9) and Theorem 2.4 yields
Consequently, (4.3) holds. For \(0<\delta <\delta _0\) with \(\delta _0\) given in Lemma 2.3, it follows from (4.3) and Lemma 2.3 that
is a supersolution of (1.6), where \(\beta >0\) sufficiently small and \(\sigma >0\) large enough such that
with \(K_0\) given in Lemma 2.3.
Suppose that \(\widehat{W}({\mathbf {x}},t)\) satisfies (1.6) and (1.19). We now prove that \(\widehat{W}({\mathbf {x}},t)\equiv \widetilde{W}({\mathbf {x}},t)\) in \({\mathbb {R}}^4\). Assume that this is not true. Without loss of generality, we assume that \(\widehat{W}({\mathbf {x}},0)\le \widetilde{W}({\mathbf {x}},0)\) but \(\widehat{W}({\mathbf {x}},0)\not \equiv \widetilde{W}({\mathbf {x}},0)\). We can choose a constant \(\lambda >0\) large enough such that
Indeed, (4.4) is obviously true if \(|{\mathbf {x}}|^2<A^2\) for some constant \(A>0\) sufficiently large. If \(|{\mathbf {x}}|^2\ge A^2\), (4.4) follows from (1.19) and \(W_z(x,y,z,t)>0\) for all \((x,y,z,t)\in {\mathbb {R}}^4\). It then follows from Lemma 2.3 and the comparison principle that
for all \((x,y,z)\in {\mathbb {R}}^3\), \(t\in [0,T]\) and \(k\in {\mathbb {N}}\). Sending \(k \rightarrow \infty \), we have
for all \((x,y,z)\in {\mathbb {R}}^3\) and \(t\in [0,T]\).
Define
If \(\Lambda _1 = 0\), we get \(\widehat{W}({\mathbf {x}},0)\le \widetilde{W}({\mathbf {x}},0)\). Similarly, we can obtain \(\widehat{W}({\mathbf {x}},0)\ge \widetilde{W}({\mathbf {x}},0)\). Thus, \(\widehat{W}({\mathbf {x}},t)\equiv \widetilde{W}({\mathbf {x}},t)\) for all \((x,y,z)\in {\mathbb {R}}^3\) and \(t\in [0,T]\). We prove \(\Lambda _1 = 0\) by a contradiction argument. Assume that \(\Lambda _1 \ne 0\). Without loss of generality, we assume that \(\Lambda _1 >0\). It then follows from the strong maximum principle that
Fix \(R_1>0\) sufficiently large such that
where \(\phi \) is defined in (2.15). Define
For \((x,y,z)\in {\mathcal {O}}\), we have
where \(\alpha ^\pm \) are given in the assumption (H2). Then, we have
If \({\mathbf {x}}\in {\mathcal {O}}\) and \(|{\mathbf {x}}|\) is large enough, say \(|{\mathbf {x}}|\ge R_0\) for some \(R_0>0\), by (1.20), we get
for \(\varepsilon >0\) small enough. If \({\mathbf {x}} \in {\mathcal {O}}\cap B(0;R_0)\), we have
for sufficiently small \(0<\varepsilon <\delta _0\) due to the compactness of the set \({\mathcal {O}}\cap B(0;R_0)\) in \({\mathbb {R}}^3\). Thus, we obtain that
In \({\mathbb {R}}^3\setminus {\mathcal {O}}\), we have
This yields
Combining(4.5) and (4.6), we have
It then follows from Lemma 2.3 that
is a supersolution of (1.6) on \({\mathbb {R}}^3\times [0,\infty )\). Hence, we have
for all \({\mathbf {x}} \in {\mathbb {R}}^3\), \(t\in [0,T]\) and \(k\in {\mathbb {N}}\). Letting \(k\rightarrow \infty \), we have
for all \((x,y,z)\in {\mathbb {R}}^3\) and \(t\in [0,T]\). This contradicts the definition of \(\Lambda _1\). Thus, \(\Lambda _1=0\) and \(\widehat{W}({\mathbf {x}},t)\equiv \widetilde{W}({\mathbf {x}},t)\) in \({\mathbb {R}}^4\) follows.
Now we prove that (1.21) holds. Define
and
for all \((x,y,z,t)\in {\mathbb {R}}^4\). Then, \(\underline{W}(x,y,z,t)\) and \(\overline{W}(x,y,z,t)\) are a subsolution and a supersolution of (1.6), respectively. Thus, \(\lim _{t\rightarrow \infty }v({\mathbf {x}},t;\underline{W}(x,y,z,0))\) and \(\lim _{t\rightarrow \infty }v({\mathbf {x}},t;\overline{W}(x,y,z,0))\) are solutions of (1.6) between \(\underline{W}(x,y,z,t)\) and \(\overline{W}(x,y,z,t)\). By the uniqueness of the solution bounded between \(\underline{W}\) and \(\overline{W}\), we have \(\lim _{t\rightarrow \infty }v({\mathbf {x}},t;\underline{W}(x,y,z,0))=\lim _{t\rightarrow \infty } v({\mathbf {x}},t;\overline{W}(x,y,z,0))=\widetilde{W}({\mathbf {x}},t)\). Thanks to the continuity of \(v({\mathbf {x}},t;v_0)\) with respect to \(v_0\), we get the desired result from the assumption (1.20). The proof is complete. \(\square \)
References
Alikakos, N., Bates, P.W., Chen, X.: Traveling waves and locating oscillating patterns in multidimensional domains. Trans. Am. Math. Soc. 351, 2777–2805 (1999)
Aronson, D.G., Weinberger, H.F.: Nonlinear diffusion in population genetics, combustion, and nerve pulse propagation. In: J.A. Goldstein (eds.) Partial Differential Equations and Related Topics, Lecture Notes in Mathematics, vol. 446, pp. 5–49 (1975)
Aronson, D.G., Weinberger, H.F.: Multidimensional nonlinear diffusions arising in population genetics. Adv. Math. 30, 33–76 (1978)
Bao, X.X., Wang, Z.C.: Existence and stability of time periodic traveling waves for a periodic bistable Lotka–Volterra competition system. J. Differ. Equ. 255, 2402–2435 (2013)
Bonnet, A., Hamel, F.: Existence of non-planar solutions of a simple model of premixed Bunsen flames. SIAM J. Math. Anal. 31, 80–118 (1999)
Brazhnik, P.K., Tyson, J.J.: On traveling wave solutions of Fisher’s equation in two spatial dimensions. SIAM J. Appl. Math. 60, 371–391 (1999)
Cao, M.L., Sheng, W.J.: Traveling curved fronts of bistable Lotka–Volterra competition-diffusion systems in \({\mathbb{R}}^3\). Comput. Math. Appl. 71, 1270–1286 (2016)
Chapuisat, G.: Existence and nonexistence of curved front solution of a biological equation. J. Differ. Equ. 236, 237–279 (2007)
Chen, X., Guo, J.S., Hamel, F., Ninomiya, H., Roquejoffre, J.M.: Traveling waves with paraboloid like interfaces for balanced bistable dynamics. Ann. Inst. H. Poincaré Analyse Non Linéaire 24, 369–393 (2007)
del Pino, M., Kowalczyk, M., Wei, J.C.: Traveling waves with multiple and non-convex fronts for a bistable semilinear parabolic equation. Commun. Pure Appl. Math. 66, 481–547 (2013)
El Smaily, M., Hamel, F., Huang, R.: Two-dimensional curved fronts in a periodic shear flow. Nonlinear Anal. Theory Methods Appl 74, 6469–6486 (2011)
Fife, P.C., McLeod, J.B.: The approach of solutions of nonlinear diffusion equations to travelling front solutions. Arch. Ration. Mech. Anal. 65, 335–361 (1977)
Gui, C.: Symmetry of traveling wave solutions to the Allen–Cahn equation in \({\mathbb{R}}^2\). Arch. Ration. Mech. Anal. 203, 1037–1065 (2012)
Hamel, F., Monneau, R.: Solutions of semilinear elliptic equations in \({\mathbb{R}}^N\) with conical-shaped level sets. Commun. Partial Differ. Equ. 25, 769–819 (2000)
Hamel, F., Nadirashvili, N.: Traveling fronts and entire solutions of the Fisher-KPP equation in \({\mathbb{R}}^N\). Arch. Ration. Mech. Anal. 157, 91–163 (2001)
Hamel, F., Roquejoffre, J.M.: Heteroclinic connections for multidimensional bistable reaction-diffusion equations. Discrete Contin. Dyn. Syst. 4, 101–123 (2011)
Hamel, F., Monneau, R., Roquejoffre, J.M.: Stability of travelling waves in a model for conical flames in two space dimensions. Ann. Scient. Ec. Norm. Sup. 37, 469–506 (2004)
Hamel, F., Monneau, R., Roquejoffre, J.M.: Existence and qualitative properties of multidimensional conical bistable fronts. Discrete Contin. Dyn. Syst. 13, 1069–1096 (2005)
Hamel, F., Monneau, R., Roquejoffre, J.M.: Asymptotic properties and classification of bistable fronts with Lipschitz level sets. Discrete Contin. Dyn. Syst. 14, 75–92 (2006)
Haragus, M., Scheel, A.: Almost planar waves in anisotropic media. Commun. Partial Differ. Equ. 31, 791–815 (2006)
Haragus, M., Scheel, A.: Corner defects in almost planar interface propagation. Ann. Inst. H. Poincaré Anal. Non Linéaire 23, 283–329 (2006)
Huang, J., Shen, W.: Speeds of spread and propagation for KPP models in time almost and space periodic media. SIAM J. Appl. Dyn. Syst. 8, 790–821 (2009)
Kurokawa, Y., Taniguchi, M.: Multi-dimensional pyramidal traveling fronts in the Allen–Cahn equations. Proc. R. Soc. Edinb. Sect. 141A, 1–24 (2011)
Lieberman, G.M.: Second Order Parabolic Differential Equations. World Scientific, Singapore (2005)
Liang, X., Yi, Y., Zhao, X.Q.: Spreading speeds and traveling waves for periodic evolution systems. J. Differ. Equ. 231, 57–77 (2006)
Morita, Y., Ninomiya, H.: Monostable-type traveling waves of bistable reaction-diffusion equations in the multi-dimensional space. Bull. Inst. Math. Acad. Sin.(NS) 3, 567–584 (2008)
Ni, W.M., Taniguchi, M.: Traveling fronts of pyramidal shapes in competition-diffusion systems. Netw. Heterog. Media 8, 379–395 (2013)
Ninomiya, H., Taniguchi, M.: Existence and global stability of traveling curved fronts in the Allen–Cahn equations. J. Differ. Equ. 213, 204–233 (2005)
Ninomiya, H., Taniguchi, M.: Global stability of traveling curved fronts in the Allen–Cahn equations. Discrete Contin. Dyn. Syst. 15, 819–832 (2006)
Shen, W.: Traveling waves in time almost periodic structures governed by bistable nonlinearities. I. Stability and uniqueness. J. Differ. Equ. 159, 1–54 (1999)
Shen, W.: Traveling waves in time almost periodic structures governed by bistable nonlinearities. II. Existence. J. Differ. Equ. 159, 55–101 (1999)
Sheng, W.J.: Time periodic traveling curved fronts of bistable reaction-diffusion equations in \({\mathbb{R}}^N\). Appl. Math. Lett. 54, 22–30 (2016)
Sheng, W.J., Li, W.T., Wang, Z.C.: Periodic pyramidal traveling fronts of bistable reaction-diffusion equations with time-periodic nonlinearity. J. Differ. Equ. 252, 2388–2424 (2012)
Sheng, W.J., Li, W.T., Wang, Z.C.: Multidimensional stability of V-shaped traveling fronts in the Allen–Cahn equation. Sci. China Math. 56, 1969–1982 (2013)
Taniguchi, M.: Traveling fronts of pyramidal shapes in the Allen–Cahn equations. SIAM J. Math. Anal. 39, 319–344 (2007)
Taniguchi, M.: The uniqueness and asymptotic stability of pyramidal traveling fronts in the Allen–Cahn equations. J. Differ. Equ. 246, 2103–2130 (2009)
Taniguchi, M.: Multi-dimensional traveling fronts in bistable reaction-diffusion equations. Discrete Contin. Dyn. Syst. 32, 1011–1046 (2012)
Taniguchi, M.: An (\(N-1\))-dimensional convex compact set gives an \(N\)-dimensional traveling front in the Allen–Cahn equation. SIAM J. Math. Anal. 47, 455–476 (2015)
Taniguchi, M.: Convex compact sets in \({\mathbb{R}}^{N-1}\) give traveling fronts of cooperation-diffusion systems in \({\mathbb{R}}^{N}\). J. Differ. Equ. 260, 4301–4338 (2016)
Volpert, A.I., Volpert, V.A., Volpert, V.A.: Travelling wave solutions of parabolic systems. In: Translations of Mathematical Monographs, vol. 140. American Mathematical Society, Providence, RI (1994)
Wang, Z.C.: Traveling curved fronts in monotone bistable systems. Discrete Contin. Dyn. Syst. 32, 2339–2374 (2012)
Wang, Z.C.: Cylindrically symmetric traveling fronts in periodic reaction-diffusion equation with bistable nonlinearity. Proc. R. Soc. Edinb. Sect. 145, 1053–1090 (2015)
Wang, Z.C., Bu, Z.H.: Nonplanar traveling fronts in reaction-diffusion equations with combustion and degenerate Fisher-KPP nonlinearities. J. Differ. Equ. 260, 6405–6450 (2016)
Wang, Z.C., Wu, J.: Periodic traveling curved fronts in reaction-diffusion equation with bistable time-periodic nonlinearity. J. Differ. Equ. 250, 3196–3229 (2011)
Wang, Z.C., Li, W.T., Ruan, S.G.: Existence, uniqueness and stability of pyramidal traveling fronts in reaction-diffusion systems. Sci. China Math. (2016). doi:10.1007/s11425-000-0000-0
Zhao, X.Q.: Dynamical systems in population biology. In: CMS Books in Mathematics/Ouvrages de Mathématiques de la SMC, vol. 16. Springer, New York (2003)
Zhao, G.Y., Ruan, S.G.: Existence, uniqueness and asymptotic stability of time periodic traveling waves for a periodic Lotka–Volterra competition system with diffusion. J. Math. Pures Appl. 95, 627–671 (2011)
Acknowledgments
The author is very grateful to the referees and the editors for their valuable comments and suggestions that helped to improve the original manuscript. This work was supported by NSF of China (11401134), China Postdoctoral Science Foundation Funded project 2012M520716, and by China Scholarship Council for a one year visit of Aix Marseille Université.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Sheng, WJ. Time periodic traveling curved fronts of bistable reaction–diffusion equations in \({\mathbb {R}}^3\) . Annali di Matematica 196, 617–639 (2017). https://doi.org/10.1007/s10231-016-0589-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10231-016-0589-0