Dynamic phase transition for the Taylor problem in the wide-gap case

Zhibo Hou; Tian Ma

doi:10.1186/1687-2770-2013-227

Research

Open Access

Dynamic phase transition for the Taylor problem in the wide-gap case

Zhibo Hou¹Email author and
Tian Ma¹

Boundary Value Problems20132013:227

https://doi.org/10.1186/1687-2770-2013-227

Received: 26 July 2013

Accepted: 6 September 2013

Published: 7 November 2013

Abstract

The main objective of this paper is to study the stability and the type of transition of the Taylor problem in the wide-gap case by using the averaging method, and we conclude that the stability of the Taylor problem in the wide-gap case is essentially the same with that in the case of the narrow-gap. The main technical tools are the spectral theory for linear and completely continuous fields, the dynamic bifurcation theory and the transition theory for incompressible flows, both developed by Ma and Wang (Bifurcation Theory and Applications, 2005; Stability and Bifurcation of Nonlinear Evolution Equations, 2007).

Keywords

Taylor problemspectral theorydynamic bifurcationtransition theory

1 Introduction

In 1921, Taylor [1] observed and studied the stability of laminar flow, which is known as the Couette flow. Taylor studied the case, where the gap between two cylinders is smaller in comparison with the mean radius, and both of them rotate in the same direction. He found that when the Taylor number T is smaller than a critical value $T_{c} > 0$ , called the critical Taylor number, the Couette flow is stable, and when the Taylor number crosses $T_{c}$ , the Couette flow breaks out into a cellular pattern which is radially symmetric.

Since Taylor’s work, there have been many studies on this kind of problem. Such as Couette [2] and Mallock [3] did a lot of experiments. Cloes [4] published the most comprehensive experimental results associated with the Taylor vortex and the secondary instability. Chandrasekhar [5], Walowit et al. [6], Drazin and Reid [7] studied the linear theories. Velte [8], Kirchgässner [9], Kirchgässner and Sorger [10], Yudovich [11], Ma and Wang [12–14] studied the nonlinear theories. Especially, Ma and Wang established a new notion of bifurcation, called an attractor bifurcation, which was applied to the Taylor problem and obtained a series of fine results. This paper focuses on the Taylor problem in the wide-gap case. In this case, the radius of inner cylinder is small, while the radius of the outer cylinder is big. In addition to the same direction, the two cylinders could rot in the converse direction.

The main objective is to study the stability and the type of transition of the Taylor problem in the wide-gap case by using the averaging method and to compare with the Taylor problem in the narrow-gap case.

The main technical tools are the spectral theory for linear and completely continuous fields, the dynamic bifurcation theory and the transition theory for incompressible flows. These theories are directly applied to the Taylor problem in the wide-gap case.

The main conclusion is that the stability of the Taylor problem in the wide-gap case is essentially the same with that in the case of a narrow-gap. The main theorems are presented in Section 4, through which we can give pictures depicting the Couette flow stability in the wide-gap case, and compare with the Taylor problem in the narrow-gap case. In the later research, we intend to simulate the Taylor problem in the wide-gap case by using computer.

This paper is organized as follows. Section 2 introduces the governing equations for the Taylor problem. Section 3 studies the Taylor problem in the wide-gap case by using the averaging method, and establishes its mathematical frame. All the main theorems and the proofs are presented in Section 4.

2 Governing equations for the Taylor problem

The governing equations for the Taylor problem are the Navier-Stocks equations in the cylindrical coordinates

(z, r, θ)

, which have the form

{\begin{cases} \frac{\partial u_{z}}{\partial t} + (u \cdot \nabla) u_{z} = - \frac{\partial}{\partial z} (\frac{p}{ρ}) + ν Δ u_{z}, \\ \frac{\partial u_{r}}{\partial t} + (u \cdot \nabla) u_{r} - \frac{u_{θ}^{2}}{r} = - \frac{\partial}{\partial r} (\frac{p}{ρ}) + ν (Δ u_{r} - \frac{2}{r^{2}} \frac{\partial u_{θ}}{\partial θ} - \frac{u_{r}}{r^{2}}), \\ \frac{\partial u_{θ}}{\partial t} + (u \cdot \nabla) u_{θ} + \frac{u_{r} u_{θ}}{r} = - \frac{\partial}{\partial θ} (\frac{p}{ρ}) + ν (Δ u_{θ} + \frac{2}{r^{2}} \frac{\partial u_{r}}{\partial θ} - \frac{u_{θ}}{r^{2}}), \\ \frac{\partial (r u_{r})}{\partial r} + \frac{\partial u_{θ}}{\partial θ} + \frac{\partial (r u_{z})}{\partial z} = 0, \end{cases}

(2.1)

where ν is the kinematic viscosity, ρ the density,

u = (u_{z}, u_{r}, u_{θ})

the velocity field, p the pressure function, and

\begin{matrix} u \cdot \nabla = u_{z} \frac{\partial}{\partial z} + u_{r} \frac{\partial}{\partial r} + \frac{u_{θ}}{r} \frac{\partial}{\partial θ}, \\ Δ = \frac{\partial^{2}}{\partial z^{2}} + \frac{1}{r} \frac{\partial}{\partial r} (r \frac{\partial}{\partial r}) + \frac{1}{r^{2}} \frac{\partial^{2}}{\partial θ^{2}} . \end{matrix}

The basic flow for (2.1) is the Couette flow, namely a steady state solution defined by

\begin{aligned} u_{z} = u_{r} = 0, u_{θ} = V (r), p = ρ \int \frac{1}{r} V^{2} (r) d r, \\ V (r) = a r + \frac{b}{r}, \end{aligned}

(2.2)

where

\begin{matrix} a = - Ω_{1} η^{2} \frac{1 - μ / η^{2}}{1 - η^{2}}, b = Ω_{1} \frac{r_{1}^{2} (1 - μ)}{1 - η^{2}}, \\ μ = \frac{Ω_{2}}{Ω_{1}}, η = \frac{r_{1}}{r_{2}} . \end{matrix}

In order to investigate the stability of the Couette flow defined by (2.1), we need to consider the perturbed state

u_{z}, u_{r}, u_{θ} + V (r), p + ρ \int \frac{1}{r} V^{2} (r) d r .

Assume that the perturbations are axisymmetric and independent of θ, from (2.1), we have:

{\begin{cases} \frac{\partial u_{z}}{\partial t} + (\tilde{u} \cdot \nabla) u_{z} = ν Δ u_{z} - \frac{\partial}{\partial z} (\frac{p}{ρ}), \\ \frac{\partial u_{r}}{\partial t} + (\tilde{u} \cdot \nabla) u_{r} - \frac{u_{θ}^{2}}{r} = ν (Δ u_{r} - \frac{u_{r}}{r^{2}}) - \frac{\partial}{\partial r} (\frac{p}{ρ}) + \frac{2 V}{r} u_{θ}, \\ \frac{\partial u_{θ}}{\partial t} + (\tilde{u} \cdot \nabla) u_{θ} + \frac{u_{r} u_{θ}}{r} = ν (Δ u_{θ} - \frac{u_{θ}}{r^{2}}) - (V^{'} + \frac{V}{r}) u_{r}, \\ \frac{\partial (r u_{r})}{\partial r} + \frac{\partial (r u_{z})}{\partial z} = 0, \end{cases}

(2.3)

where

\begin{matrix} \tilde{u} \cdot \nabla = u_{z} \frac{\partial}{\partial z} + u_{r} \frac{\partial}{\partial r}, \\ Δ = \frac{\partial^{2}}{\partial z^{2}} + \frac{\partial^{2}}{\partial r^{2}} + \frac{1}{r} \frac{\partial}{\partial r} . \end{matrix}

The spatial domain for (2.3) is $M = (r_{1}, r_{2}) \times (0, L) \subset R^{2}$ , where L is the height of the field between the two cylinders. There are different physically sound boundary conditions.

In the radial direction, there are two kinds of boundary conditions:

Free boundary condition:

\frac{\partial u_{z}}{\partial r} = 0, u_{r} = u_{θ} = 0, at r = r_{1}, r_{2} .

Rigid boundary condition:

u_{z} = u_{r} = u_{θ} = 0, at r = r_{1}, r_{2} .

In the z direction, there are four kinds of boundary conditions:

Free slip boundary condition:

u_{z} = 0, \frac{\partial u_{r}}{\partial z} = \frac{\partial U_{θ}}{\partial z} = 0, at z = 0, L .

Dirichlet boundary condition (or rigid condition):

u_{z} = u_{r} = u_{θ} = 0, at z = 0, L .

Free rigid boundary condition:

\begin{matrix} u_{z} = 0, \frac{\partial u_{r}}{\partial z} = \frac{\partial u_{θ}}{\partial z} = 0, at z = L, \\ u_{z} = u_{r} = u_{θ} = 0, at z = 0 . \end{matrix}

Periodic condition:

u = (u_{z}, u_{r}, u_{θ}) is periodic in the z direction .

Using the transform:

{\begin{cases} x = l x^{'}, x = (z, r, θ), \\ t = l^{2} t^{'} / ν, \\ u = ν u^{'} / l, u = (u_{z}, u_{r}, u_{θ}), \\ p = ρ ν^{2} p^{'} / l^{2}, \end{cases}

where l is a certain length unit in (2.3).

We get the dimensionless form:

{\begin{cases} \frac{\partial u_{z}}{\partial t} = Δ u_{z} - \frac{\partial p}{\partial z} - (\tilde{u} \cdot \nabla) u_{z}, \\ \frac{\partial u_{r}}{\partial t} = (Δ u_{r} - \frac{u_{r}}{r^{2}}) - \frac{\partial p}{\partial r} + 2 Ω_{1} \frac{l^{2}}{ν} (\frac{r_{1}^{2}}{r^{2}} \frac{1 - μ}{1 - η^{2}} - \frac{η^{2} - μ}{1 - η^{2}} u_{θ}) + \frac{u_{θ}^{2}}{r} - (\tilde{u} \cdot \nabla) u_{r}, \\ \frac{\partial u_{θ}}{\partial t} = (Δ u_{θ} - \frac{u_{θ}}{r^{2}}) + 2 Ω_{1} \frac{l^{2}}{ν} \frac{η^{2} - μ}{1 - η^{2}} u_{r} - \frac{u_{r} u_{θ}}{r} - (\tilde{u} \cdot \nabla) u_{θ}, \\ \frac{\partial (r u_{r})}{\partial r} + \frac{\partial (r u_{z})}{\partial z} = 0 . \end{cases}

(2.4)

3 Averaging for the Taylor problem in the wide-gap case

In the case of a wide-gap,

r_{1}

is small, while

r_{2}

is big. The averaging method is that we let

l = r_{1} = 1

, and

r_{0} = (r_{1} + r_{2}) / 2

replace variable r approximately. Since

r_{x}

is big,

r_{1}

and ν are small, we can ignore

r^{- n}

(

n \geq 1

), which are associated with the inner friction of fluid. In this paper, moreover, our analysis will be conducted using free boundary conditions in z and radial directions. Then from (2.4), we obtain the averaged governing equation for the case of the wide-gap:

{\begin{matrix} \frac{\partial u_{z}}{\partial t} = Δ u_{z} - \frac{\partial p}{\partial z} - (\tilde{u} \cdot \nabla) u_{z}, \\ \frac{\partial u_{r}}{\partial t} = Δ u_{r} - \frac{\partial p}{\partial r} + λ (\frac{1}{r_{0}^{2}} - κ) u_{θ} + \frac{u_{θ}^{2}}{r_{0}} - (\tilde{u} \cdot \nabla) u_{r}, \\ \frac{\partial u_{θ}}{\partial t} = Δ u_{θ} + λ κ u_{r} - \frac{u_{r} u_{θ}}{r_{0}} - (\tilde{u} \cdot \nabla) u_{θ}, \\ \frac{\partial u_{r}}{\partial r} + \frac{\partial u_{z}}{\partial z} = 0, \end{matrix}

(3.1)

{\begin{matrix} \frac{\partial u_{z}}{\partial r} = 0, u_{r} = u_{θ} = 0, at r = 1, r_{2}, \\ u_{z} = 0, \frac{\partial u_{r}}{\partial z} = \frac{\partial u_{θ}}{\partial z} = 0, at z = 0, L, \end{matrix}

(3.2)

where

λ = \sqrt{T}

, T is the Taylor number, T and κ are defined by

\begin{matrix} T = \frac{4 Ω_{1}^{2} (1 - μ^{2})}{ν^{2} (1 - η^{2})}, κ = \frac{η^{2} - μ}{1 - μ}, \\ \tilde{u} \cdot \nabla = u_{z} \frac{\partial}{\partial z} + u_{r} \frac{\partial}{\partial r}, \\ Δ = \frac{\partial^{2}}{\partial z^{2}} + \frac{\partial^{2}}{\partial r^{2}}, \end{matrix}

spatial domain for (3.1) is $M = (1, r_{2}) \times (0, L) \subset R^{2}$ .

Let

\tilde{u} = (u_{z}, u_{r})

,

\begin{matrix} H = {u = (\tilde{u}, u_{θ}) \in L^{2} (M, R^{3}) {| div \tilde{u} = 0, \tilde{u} \cdot n |}_{\partial M} = 0}, \\ H_{1} = {u = (\tilde{u}, u_{θ}) \in H^{2} (M, R^{3}) \cap H | u satisfies (3.2)} . \end{matrix}

By the Hodge decomposition theorem,

L^{2} (M, R^{3})

can be decomposed as follows.

\begin{matrix} L^{2} (M, R^{3}) = H \oplus G, \\ H = {u = (\tilde{u}, u_{θ}) \in L^{2} (M, R^{3}) {| div \tilde{u} = 0, \tilde{u} \cdot n |}_{\partial M} = 0}, \\ G = {\nabla_{φ} \in L^{2} (M, R^{3}) | φ \in H^{1} (M)}, \\ G ⊥ H . \end{matrix}

Here, $H^{1} (M)$ and $H^{2} (M, R^{3})$ are the usual Sobolev spaces.

Then we can define an orthogonal projection, called Leray projection

P : L^{2} (M, R^{3}) \to H .

Let

\begin{matrix} L_{λ} = - A + λ B : H_{1} \to H, \\ G : H_{1} \to H \end{matrix}

be a mapping defined by

\begin{aligned} A (u) = - P (Δ u_{z}, Δ u_{r}, Δ u_{θ}), \\ B (u) = P (0, (\frac{1}{r_{0}^{2}} - κ) u_{θ}, κ u_{r}), \\ G (u) = - P ((\tilde{u} \cdot \nabla) u_{z}, (\tilde{u} \cdot \nabla) u_{r} - \frac{u_{θ}^{2}}{r_{0}}, (\tilde{u} \cdot \nabla) u_{θ} + \frac{u_{θ} u_{r}}{r_{0}}), \end{aligned}

(3.3)

where $u = (u_{z}, u_{r}, u_{θ}) \in H_{1}$ , and the mapping P is the Leray projection.

Thus, (3.1) can be written in the abstract form:

\frac{d u}{d t} = L_{λ} u + G u, u \in H_{1} .

(3.4)

So far, the stability of the Couette flows of equation (2.1) equals to the stability of $(u_{z}, u_{r}, u_{θ}) = 0$ of equation (3.4).

4 Main results and proofs

Let

X_{1}

, X be two Banach spaces,

X_{1} \subset X

be a dense and compact inclusion. Consider the following nonlinear evolution equations

{\begin{matrix} \frac{d u}{d t} = L u + G u, \\ u (0) = φ, \end{matrix}

(4.1)

where $L : X_{1} \to X$ is bounded linear operator, $G : X_{1} \to X$ is $C^{r}$ ( $r \geq 1$ ) mapping.

Definition 4.1 [14]

Let $Ω \subset X$ be a bounded open set, we say that (4.1) is Lyapunov stable in Ω, if the solution of (4.1) $u (t, φ) \in Ω$ , $\forall t \geq 0$ , for any initial point $φ \in Ω$ .

Definition 4.2 [14]

We say (4.1) is asymptotically Lyapunov stable in Ω, or

u_{0}

is an asymptotically stable equilibrium point of (4.1) in Ω, if there exists a steady solution

u_{0} \in Ω

, satisfying

lim_{t \to \infty} {∥ u (t, φ) - u_{0} ∥}_{X} = 0 for any φ \in Ω .

Lemma 4.3 [14]

Let $L : X_{1} \to x$ be a sectorial operator, $G : X_{α} \to X$ be a $C^{r}$ ( $r \geq 1$ ) mapping for certain $0 \leq α < 1$ , v be a steady solution of (4.1). If the spectral $β (λ)$ of $L + D G (v)$ satisfy $Re β (λ) < - ε$ , for certain $ε > 0$ , then v is a locally asymptotically stable equilibrium point of (4.1), and it decays exponentially. Namely, there exist $M > 0$ , $σ > 0$ , $δ > 0$ , for the solution of (4.1) $u (t, φ)$ , ${∥ u (t, φ) - v ∥}_{X} \leq M e^{- σ t}$ ( $\forall t \geq 0$ ) hold true, as $∥ φ - v ∥ < δ$ .

Let

δ = \frac{3 r_{2}^{2} - 2 r_{2} - 1}{r_{2}^{2} (r_{2} - 1) (r_{2} + 3)} > 0, as r_{2} > r_{1} = 1,

and as above

μ = \frac{Ω_{2}}{Ω_{1}}, η = \frac{r_{1}}{r_{2}} .

Theorem 4.4 If $μ \leq - δ$ , or $η^{2} \leq μ$ , then $(u_{z}, u_{r}, u_{θ}) = 0$ is a locally asymptotically stable equilibrium point of (3.4), and it decays exponentially.

Theorem 4.5 If

- δ < μ < η^{2}

(

μ \neq 0

), we have the following conclusions for (3.4):

(1)

Equation (3.4) happens with a continuous transition at $(u, λ) = (0, \sqrt{T_{c}})$ , namely there is an attractor bifurcation, and it bifurcates exactly into two singular points $v_{i}$ ( $i = 1, 2$ ), which attract two open subsets of U separately. U is the neighborhood of $u = 0$ .
(2)

The two singular points $v_{i}$ ( $i = 1, 2$ ) can be written by:
$v_{1, 2} (λ) = \pm {| β_{n_{1} 1} / σ |}^{\frac{1}{2}} ψ_{n_{1} 1} + o ({| β_{n_{1} 1} / σ |}^{\frac{1}{2}}),$

where $σ = \frac{- 1}{4 β_{2 n_{1} 2}} a_{1}^{4} b_{1}^{2}$ , $β_{n_{1} 1}$ is the first eigenvalue.

Remark 4.6 From the point of view of mathematics, Theorem 4.4 explains that there exists an open set $Ω \subset H_{1}$ , which guarantees that if the initial point $u_{0} \in Ω$ , then the solution of (3.4) satisfies ${∥ u (t, u_{0}) ∥}_{H} \to 0$ ( $t \to \infty$ ).

Remark 4.7 From the point of view of physics, Theorem 4.4 explains that in the wide-gap case, the Couette flow of the Taylor problem is metastable. Namely, if the initial perturbation is in a certain range, the disturbed fluid will become the Couette flow in a short time. But if the initial perturbation is beyond that range above, the disturbed fluid will become another steady flow. The explanation is the same as that for Theorem 4.5.

Remark 4.8 Theorems 4.4-4.5 give the entire results of stability of the Taylor problem in the wide-gap case; see Figure 1.

Σ_{1}

represents the unstable area,

Σ_{2}

represents the stable area,

T_{0} \neq 0

and

T_{0} = min_{(n, l)} \frac{α^{3} ν^{2} {(1 - η^{2})}^{2}}{4 a^{2} η^{2} (\frac{1}{r_{0}^{2}} - η^{2})} .

Remark 4.9 [15]

In the narrow-gap case, let

r_{1} = 3.55 cm

and

r_{2} = 4.035 cm

, the results can be seen in Figure 2.

Σ_{1}

represents the unstable area,

Σ_{2}

represents the stable area.

Remark 4.10 By Figure 1 and Figure 2, we can conclude that the results of the stability of the Taylor problem in the wide-gap case, obtained by using the averaging method, are essentially the same as those of the Taylor problem in the narrow-gap case.

Proof of Theorem 4.4 Obviously, A and B are linear operators. According to [16], A is a homeomorphism, then it is a sectorial operator. According to the Sobolev compact embedding theorem [17] and the Leray projection, P is bounded, B is compact. So,

L_{λ} = - A + λ B : H_{1} \to H

is a sectorial operator, and also a linear completely continuous field.

Now,

\begin{array}{rcl} \int_{Ω} (u \cdot \nabla) u \cdot v + u^{2} \cdot v d x & \leq & {∥ u ∥}_{C^{0}} {∥ \nabla u ∥}_{L^{2}} {∥ v ∥}_{L^{2}} + {∥ u ∥}_{C^{0}} {∥ u ∥}_{L^{2}} {∥ v ∥}_{L^{2}} \\ = & {∥ u ∥}_{C^{0}} {∥ u ∥}_{W^{1, 2}} {∥ v ∥}_{L^{2}}, \end{array}

we have

{∥ G u ∥}_{H} \leq {∥ u ∥}_{C^{0}} {∥ u ∥}_{W^{1, 2}} .

According to [18], Theorem 7.2.2, the fraction spaces of

L_{λ}, H_{α} = D (L_{λ}^{α})

, satisfy

{∥ u ∥}_{C^{0}} + {∥ u ∥}_{W^{1, 2}} \leq C {∥ u ∥}_{H_{α}},

where

α > \frac{3}{4}

. Then

G : H_{α} \to H (\frac{3}{4} < α < 1) is a bounded analytic mapping .

Consider the eigenvalue problem:

L_{λ} ψ = β (λ) ψ, ψ = (u_{z}, u_{r}, u_{θ}) \in H_{1} .

By (3.3), the abstract form can be referred to the following eigenvalue equations in

H_{1}

:

Δ u_{z} - \frac{\partial p}{\partial z} = β (λ) u_{z},

(4.2)

Δ u_{r} - \frac{\partial p}{\partial r} + λ (\frac{1}{r_{0}^{2}} - κ) u_{θ} = β (λ) u_{r},

(4.3)

Δ u_{θ} + λ κ u_{r} = β (λ) u_{θ},

(4.4)

\frac{\partial u_{r}}{\partial r} + \frac{\partial u_{z}}{\partial z} = 0,

(4.5)

\frac{\partial u_{z}}{\partial r} = 0, u_{r} = u_{θ} = 0, at r = 1, r_{2},

(4.6)

u_{z} = 0, \frac{\partial u_{r}}{\partial z} = \frac{\partial u_{θ}}{\partial z} = 0, at z = 0, L .

(4.7)

According to [5], we take the separation of variables as follows:

\begin{aligned} u_{z} = \frac{d h (z)}{d t} \frac{d R (r)}{d r}, \\ u_{r} = a^{2} h (z) R (r), \\ u_{θ} = h (z) φ (r) . \end{aligned}

(4.8)

We utilize (4.8) in (4.5) and (4.7) to obtain:

{\begin{matrix} h^{″} R^{'} + a^{2} h R^{'} = 0, \\ h^{'} (0) = h^{'} (L) = 0 \end{matrix} \Rightarrow {\begin{matrix} h = cos a z, \\ a = \frac{n π}{L}, n = 1, 2, \dots . \end{matrix}

(4.9)

We utilize (4.8), (4.9) in (4.4) to obtain:

h^{″} φ + h φ^{″} + λ κ a^{2} h R = β (λ) h φ

that equals to

(\frac{d^{2}}{d r^{2}} - a^{2}) φ = - λ κ a^{2} R + β (λ) φ .

(4.10)

We utilize (4.8), (4.9) in (4.2), then differentiate with respect to r,

h^{‴} R^{″} + h^{'} R^{''''} - \frac{\partial^{2} p}{\partial r \partial z} = β (λ) h^{'} R^{″},

and utilize (4.8), (4.9) in (4.3), then differentiate with respect to z,

a^{2} h^{‴} R + a^{2} h^{'} R^{″} + λ (\frac{1}{r_{0}^{2}} - κ) h^{'} φ - \frac{\partial^{2} p}{\partial z \partial r} = β (λ) a^{2} h^{'} R,

the two equations above subtract to obtain:

{(\frac{d^{2}}{d r^{2}} - a^{2})}^{2} R = λ (\frac{1}{r_{0}^{2}} - κ) φ + β (λ) (\frac{d^{2}}{d r^{2}} - a^{2}) R .

(4.11)

Utilizing (4.8) in (4.6), we have

φ (1) = φ (r_{2}) = R (1) = R (r_{2}) = R^{″} (1) = R^{″} (r_{2}) = 0,

then let

R = sin l π (\frac{r - 1}{r_{2} - 1}), φ = C_{n l} sin l π (\frac{r - 1}{r_{2} - 1}), l = 1, 2, \dots .

Combine (4.10), (4.11) to obtain:

{\begin{matrix} α C_{n l} = a^{2} λ κ - C_{n l} β (λ), \\ α^{2} = C_{n l} (\frac{1}{r_{0}^{2}} - κ) λ - α β (λ), \end{matrix} α = a^{2} + {(\frac{l π}{r_{2} - 1})}^{2},

(4.12)

therefore, we get

{\begin{matrix} C_{n l} = \frac{2 α λ κ a^{2}}{\pm \sqrt{Δ}}, \\ β_{n l} = - α \pm \frac{\sqrt{Δ}}{2 α}, \end{matrix}

(4.13)

where

Δ = 4 α a^{2} λ^{2} κ (\frac{1}{r_{0}^{2}} - κ) .

The corresponding eigenvectors of (4.2)-(4.7) are as follows:

ψ_{n l} = {\begin{matrix} \frac{n l π^{2}}{L (r_{2} - 1)} sin \frac{n π z}{L} cos \frac{l π (r - 1)}{(r_{2} - 1)}, \\ {(\frac{n π}{L})}^{2} cos \frac{n π z}{L} sin \frac{l π (r - 1)}{(r_{2} - 1)}, \\ C_{n l} cos \frac{n π z}{L} sin \frac{l π (r - 1)}{(r_{2} - 1)} . \end{matrix}

(4.14)

Now, as

(n, l) \neq (0, 0)

,

\begin{aligned} L_{λ} ψ_{n l} = β_{n l} (λ) ψ_{n l}, \\ E_{1} = {ψ_{n l} | ψ_{n l} satisfies (4.14), (n, l) \in N \times N}, \\ B_{1} = {β_{n l} (λ) | β_{n l} (λ) satisfies (4.13), (n, l) \in N \times N}, \end{aligned}

as

n = 0

, from (4.2)-(4.8),

L_{λ} ψ_{0 l} = β_{0 l} (λ) ψ_{0 l},

where

\begin{aligned} ψ_{0 l} \in E_{2} = {ψ_{0 l} | ψ_{0 l} = {(0, 0, sin \frac{l π (r - 1)}{(r_{2} - 1)})}^{T}, l = 1, 2, \dots}, \\ β_{0 l} \in B_{2} = {β_{0 l} | β_{0 l} = - {(\frac{l π}{r_{2} - 1})}^{2}, l = 1, 2, \dots}, \end{aligned}

as $l = 0$ , $ψ_{n 0} = 0$ .

Now, according to the Fourier expansion, $B_{1} \cup B_{2}$ are all the eigenvalues of (4.2)-(4.7), the corresponding eigenvectors $E_{1} \cup E_{2}$ form a complete basis in $H_{1}$ .

In view of the condition

μ < - δ

, by simple calculation, we know that

κ > 0

,

κ r_{0}^{2} - 1 > 0

, then

Δ \leq 0,

thus,

Re β_{n l} < - ε (0 < ε < α),

and

β_{0 l} = - {(\frac{l π}{r_{2} - 1})}^{2} < 0 .

Now, by Lemma 4.3, the proof of Theorem 4.4 is completed. □

Proof of Theorem 4.5 Utilizing the method in Theorem 4.4, we have the following results for the eigenvalue problem of $L_{λ}^{*}$ , here $L_{λ}^{*}$ is adjoint operator of $L_{λ}$ .

As

(n, l) \neq (0, 0)

,

{\begin{matrix} C_{n l}^{*} = \frac{2 α (\frac{1}{r_{0}^{2}} - κ) α^{2} λ}{\pm \sqrt{Δ}}, \\ β_{n l} (λ) = - α \pm \frac{\sqrt{Δ}}{2 α}, \end{matrix}

(4.15)

where

Δ = 4 α a^{2} λ^{2} κ (\frac{1}{r_{0}^{2}} - κ),

the corresponding eigenvectors are as follows:

ψ_{n l}^{*} = {\begin{matrix} \frac{n l π^{2}}{L (r_{2} - 1)} sin \frac{n π z}{L} cos \frac{l π (r - 1)}{(r_{2} - 1)}, \\ {(\frac{n π}{L})}^{2} cos \frac{n π z}{L} sin \frac{l π (r - 1)}{(r_{2} - 1)}, \\ C_{n l}^{*} cos \frac{n π z}{L} sin \frac{l π (r - 1)}{(r_{2} - 1)}, \end{matrix}

(4.16)

satisfying

\begin{matrix} E_{1}^{*} = {ψ_{n l}^{*} | ψ_{n l}^{*} satisfies (4.16), (n, l) \in N \times N}, \\ B_{1} = {β_{n l} (λ) | β_{n l} (λ) satisfies (4.15), (n, l) \in N \times N} . \end{matrix}

As

n = 0

,

L_{λ}^{*} ψ_{0 l}^{*} = β_{0 l} (λ) ψ_{0 l}^{*},

where

\begin{matrix} ψ_{0 l}^{*} \in E_{2}^{*} = {ψ_{0 l}^{*} | ψ_{0 l}^{*} = {(0, 0, sin \frac{l π (r - 1)}{(r_{2} - 1)})}^{T}, l = 1, 2, \dots}, \\ β_{0 l} \in B_{2} = {β_{0 l} | β_{0 l} = - {(\frac{l π}{r_{2} - 1})}^{2}, l = 1, 2, \dots} . \end{matrix}

As $l = 0$ , $ψ_{n 0}^{*} = 0$ .

Consider equation

β_{n l} (λ) = 0

, by (4.15), we have

\begin{matrix} α {(α + \frac{1}{r_{0}^{2}})}^{2} - λ^{2} a^{2} κ (\frac{1}{r_{0}^{2} - κ}) = 0, \\ λ^{2} = \frac{α^{3}}{a^{2} κ (\frac{1}{r_{0}^{2}} - κ)} . \end{matrix}

Now, we have the critical Taylor number

T_{c} = min_{(n, l) \in N^{2}} \frac{α^{3}}{a^{2} κ (\frac{1}{r_{0}^{2}} - κ)},

(4.17)

through sample calculation, as $(n, l) = (\frac{L}{\sqrt{2} (r_{2} - 1)}, 1)$ , $\frac{α^{3}}{a^{2} κ (\frac{1}{r_{0}^{2}} - κ)}$ reaches its minimal value. We assume that $n + \frac{1}{2} \neq \frac{L}{\sqrt{2} (r_{2} - 1)}$ for $\forall n > 0$ , then there exists only one couple $(n_{1}, l)$ , making $\frac{α^{3}}{a^{2} κ (\frac{1}{r_{0}^{2}} - κ)}$ reach the minimal value.

Therefore, we have

\begin{aligned} β_{n_{1} l} (λ) = {\begin{matrix} < 0, & if λ < T_{c}, \\ = 0, & if λ = T_{c}, \\ > 0, & if λ > T_{c}, \end{matrix} \\ β_{n l} (T_{c}) < 0, \forall (n, l) \neq (n_{1}, 1) . \end{aligned}

(4.18)

The reduced form of (3.4) into its center manifold in H is:

\frac{d x}{d t} = β_{n_{1} 1} x + \frac{1}{{〈 ψ_{n_{1} 1}, ψ_{n_{1} 1}^{*} 〉}_{H}} {〈 G (ψ, ψ), ψ_{n_{1} 1}^{*} 〉}_{H},

(4.19)

where

ψ \in H_{1}

can be written by

ψ = x ψ_{n_{1} 1} + Φ,

(4.20)

Φ is a center manifold function.

G (u, v)

is a bilinear operator, satisfying

G (u, v) = - P ((\tilde{u} \cdot \nabla) v_{z}, (\tilde{u} \cdot \nabla) v_{r} - \frac{u_{θ} v_{θ}}{r_{0}}, (\tilde{u} \cdot \nabla) v_{θ} + \frac{u_{θ} v_{r}}{r_{0}}) .

(4.21)

Through direct calculation, we have

{〈 G (ψ_{n_{1} 1}, ψ_{n_{1} 1}), ψ_{J}^{*} 〉}_{H} = {\begin{matrix} 0, & ψ_{J}^{*} \neq ψ_{2 n_{1} 2}^{*}, \\ - a_{1}^{2} b_{1} c_{11} c_{11}^{*} L (r_{2} - 1), & ψ_{J}^{*} = ψ_{2 n_{1} 2}^{*}, \end{matrix}

(4.22)

where,

a_{1} = \frac{n_{1} π}{L}, b_{1} = \frac{π}{r_{2} - 1}, c_{11} = C_{n_{1} 1}, c_{11}^{*} = C_{n_{1} 1}^{*} .

Here, we use the mark as follows:

o (2) = o (x^{2}) + O (| β_{n_{1} l} (λ) x^{2} |), o (3) = o (x^{3}) + O (| β_{n_{1} l} (λ) x^{3} |) .

Then the center manifold function can be written by

Φ = - \frac{x^{2} {〈 G (ψ_{n_{1} 1}, ψ_{n_{1} 1}), ψ_{2 n_{1} 2}^{*} 〉}_{H}}{β_{2 n_{1} 2} {〈 ψ_{2 n_{1} 2}, ψ_{2 n_{1} 2}^{*} 〉}_{H}} ψ_{2 n_{1} 2} + o (2) .

(4.23)

Direct computation yields that

\begin{aligned} {〈 G (ψ_{n_{1} 1}, ψ_{2 n_{1} 2}), ψ_{n_{1} 1}^{*} 〉}_{H} = 2 a_{1}^{2} b_{1} c_{11} c_{11}^{*} L (r_{2} - 1), \\ {〈 G (ψ_{2 n_{1} 2}, ψ_{n_{1} 1}), ψ_{n_{1} 1}^{*} 〉}_{H} = - a_{1}^{2} b_{1} c_{11} c_{11}^{*} L (r_{2} - 1), \\ {〈 ψ_{n_{1} 1}, ψ_{n_{1} 1}^{*} 〉}_{H} = \frac{1}{2} c_{11} c_{11}^{*} L (r_{2} - 1), {〈 ψ_{2 n_{1} 2}, ψ_{2 n_{1} 2}^{*} 〉}_{H} = 8 c_{11} c_{11}^{*} L (r_{2} - 1) . \end{aligned}

(4.24)

Finally, we utilize (4.20), (4.22)-(4.24) in (4.19) to obtain

\frac{d x}{d t} = β_{n_{1} 1} x - σ x^{3} + o (3),

(4.25)

where

σ = \frac{- 1}{4 β_{2 n_{1} 2} a_{1}^{4} b_{1}^{2}} > 0 .

Now, according to Theorem 6.9 in [14], the proof of Theorem 4.5 is completed. □

Declarations

Acknowledgements

The authors express their sincere thanks to the referees for helpful comments and suggestions which led to the improvement of the presentation and quality of the work.

Authors’ Affiliations

(1)

Department of Mathematics, Sichuan University, Chengdu, China

References

Taylor GI: Experiments with rotating fluid. Proc. R. Soc. Lond. Ser. A 1921, 100: 114-121. 10.1098/rspa.1921.0075View ArticleGoogle Scholar
Couette MFA: Études sur le Frottement des Liquides. Ann. Chim. Phys. 1890, 21: 433-510.Google Scholar
Mallock A: Experiments on fluid viscosity. Philos. Trans. R. Soc. Lond. A 1896, 187: 41-56. 10.1098/rsta.1896.0003View ArticleGoogle Scholar
Cloes D: Transition in circular Couette flow. J. Fluid Mech. 1965, 21: 385-425. 10.1017/S0022112065000241View ArticleGoogle Scholar
Chandrasekhar S: Hydrodynamic and Hydromaganetic Stability. Dover, New York; 1981.Google Scholar
Walowit J, Tsao S, DiPrima RC: Stability of flow between arbitrarily spaced concentric cylindrical surfaces including the effect of a radial temperature gradient. J. Appl. Mech. 1964, 31: 585-593. 10.1115/1.3629718MathSciNetView ArticleGoogle Scholar
Drazin P, Reid W: Hydromagnetic Stability. Cambridge University Press, Cambridge; 1981.Google Scholar
Velte W: Stabilität und Verzweigung stationärer Lösungen der Navier-Stokeschen Gleichungen beim Taylorproblem. Arch. Ration. Mech. Anal. 1966, 22: 1-14. 10.1007/BF00281240MathSciNetView ArticleGoogle Scholar
Kirchgässner K: Die Instabilität der Strömung zwischen zwei rotierenden Zylindern gegenüber Taylor-Wirbeln für beliebige Spaltbreiten. Z. Angew. Math. Phys. 1961, 12: 14-30. 10.1007/BF01601104MathSciNetView ArticleGoogle Scholar
Kirchgässner K, Sorger P: Branching analysis for the Taylor problem. Q. J. Mech. Appl. Math. 1969, 22: 183-209. 10.1093/qjmam/22.2.183View ArticleGoogle Scholar
Yudovich VI: Secondary flows and fluid instability between rotating cylinders. J. Appl. Math. Mech. 1966, 30: 1193-1199. 10.1016/0021-8928(66)90081-5View ArticleGoogle Scholar
Ma T, Wang S: Bifurcation Theory and Applications. World Scientific, Singapore; 2005.Google Scholar
Ma T, Wang S: Stability and bifurcation of the Taylor problem. Arch. Ration. Mech. Anal. 2006, 181: 149-176. 10.1007/s00205-006-0415-8MathSciNetView ArticleGoogle Scholar
Ma T, Wang S: Stability and Bifurcation of Nonlinear Evolution Equations. Science Press, Beijing; 2007. (in Chinese)Google Scholar
Taylor GI: Stability of a viscous liquid contained between two rotating cylinders. Philos. Trans. R. Soc. Lond. A 1923, 223: 289-343. 10.1098/rsta.1923.0008View ArticleGoogle Scholar
Pazy A: Semigroups of Linear Operators and Applications to Partial Differential Equations. Springer, New York; 1983.View ArticleGoogle Scholar
Adams RA: Sobolev Spaces. Academic Press, New York; 1975.Google Scholar
Wang M: Semigroups and Evolution Equations. Science Press, Beijing; 2006. (in Chinese)Google Scholar

Copyright

This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.