Abstract
The regularity for the 3-D nematic liquid crystal equations is considered in this paper, it is proved that the Leray–Hopf weak solutions (u, d) is in fact smooth, if the velocity field \(u\in L^\infty (0,T;L^{3,\infty }_x(\mathbb {R}^3))\) satisfies some addition local small condition
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
The three-dimensional incompressible liquid crystals system are the following coupled equations
in the domain \(Q_T\equiv \mathbb {R}^3\times (0, T)\). Here, the unknowns \(u=(u_1, u_2, u_3)\) is the velocity field, p is the scalar pressure and \(d=(d_1, d_2, d_3)\) is the optical molecule direction after penalization, and \(f(d)=\frac{1}{\sigma ^2}(|d|^2-1)d\), \(\nabla d \odot \nabla d\) is a symmetric tensor with its component \((\nabla d \odot \nabla d)_{ij}\) is given by \(\partial _i d \cdot \partial _j d\). And the initial conditions are
with \(|d_0|=1.\)
System (1.1) is the simplified system of the original Ericksen–Leslie system of variable length for the flow of liquid crystals, that is the Ginzburg–Landau energy \(\int _{\Omega } (\frac{1}{2} |\nabla d |^2 +\frac{(1-|d|^2)^2}{4\sigma ^2})\). For this system, Lin and Liu [19,20,21] first proved a global existence of weak solutions under \(L^2\) data and regularity result of the suitable weak solution under the C-K-N condition. Other results to liquid crystals equations refer to [8,9,10,11, 15, 16, 18, 22, 23, 28].
Let us now recall the notion of a suitable weak solution of liquid crystals equations.
Definition 1.1
([20]) a triple (u, d, p) is called a suitable weak solution of (1.1) in \(\mathbb {R}^3\times (0,T)\) if the following conditions hold:
-
(1).
the weak solution (u, d) satisfies system (1.1) in the distribution sense;
-
(2).
the solution (u, d) satisfy the energy inequality, i.e.,
$$\begin{aligned}&\Vert u\Vert _{L^{2,\infty }(\mathbb {R}^3\times (0,T))}^2+\Vert \nabla d\Vert _{L^{2,\infty }(\mathbb {R}^3\times (0,T))}^2\\&\quad +\Vert \nabla ^2d\Vert _{L^2(\mathbb {R}^3\times (0,T)))}^2 +\Vert \nabla u\Vert _{L^2(\mathbb {R}^3\times (0,T)))}^2\le c_0; \end{aligned}$$ -
(3).
the press \(p\in L^{\frac{3}{2}}_{loc}(\mathbb {R}^3\times (0,T))\);
-
(4).
the triple (u, d, p) satisfy the modified generalized local energy inequality, for a.e. \(t\in (0,T)\) and for all \(\phi \in C_{0}^{\infty }(\mathbb {R}^3\times (0,T))\) with \(\phi \ge 0,\)
$$\begin{aligned}&\int _{\mathbb {R}^3\times \{t\}}(|u|^2+|\nabla d|^2)\phi \mathrm{d}x +2\int _{0}^t\int _{\mathbb {R}^3} (|\nabla u|^2+|\nabla ^2 d|^2)\phi \\&\quad \le \int _{0}^t\int _{\mathbb {R}^3} (|u|^2+|\nabla d|^2)(\phi _t+\Delta \phi )\\&\qquad +\int _0^t\int _{\mathbb {R}^3} (|u|^2+|\nabla d|^2+2p)(u\cdot \nabla \phi )\\&\qquad +2\int _0^t\int _{\mathbb {R}^3} ((u\cdot \nabla ) d\odot \nabla d)\cdot \nabla \phi -2\int _0^t\int _{\mathbb {R}^3}\nabla _x f(d)\cdot \nabla \mathrm{d} \phi . \end{aligned}$$
Denote for \(z=(x,t)\in \mathbb {R}^{3}\times \mathbb {R}_{+}\) the standard notations
to be the Euclidean ball and parabolic cylinder. For \(z=(0,0)\), we simply write them as \(B_{r}\) and \(Q_{r}\).
Now, we mention some relevant results on regularity. For 3D Navier–Stokes system, the study of partial regularity was originated by Scheffer in a series of papers [31, 32] and [33]. The notion of suitable weak solutions was introduced in a celebrated paper [1] by Caffarelli, Kohn and Nirenberg. It is proved that, for any suitable weak solution (u, p), there is an open subset in which the velocity field u is Hölder continuous, and the complement of it has zero 1-D Hausdorff measure. Latter Lin [17] gave a simpler proof for the CKN theorem, Lin’s method was used extensively by Seregin’s papers for example in [3, 12]. These results are based on some regularity criterions when certain dimensionless quantities are small. For liquid crystal system (1.1), Lin and Liu[20] proved the following theorem.
Theorem(A). Let the triple (u, d, p) be a suitable weak solution to system (1.1). There exist a small constant \(\epsilon _{0}>0\), such that if
then u and d are smooth in \(\overline{Q_{\frac{1}{2}}}\). In particular, for any \(z_0\) if
then \(z_0\) is a regular point.
We can drop the pressure p by Wolf’ method of local suitable weak solutions, the proof to see Sect. 5.
Proposition 1.2
Let the triple (u, d, p) be a suitable weak solution to system (1.1). There exist a small constant \(\varepsilon _{0}>0\), such that if
then u and d are smooth in \(\overline{Q_{\frac{1}{2}}}\).
There is another type of regularity criterion called the Ladyzhenskaya–Prodi–Serrin condition (to see [13, 30, 36]). For system (1.1), [24] showed that if
where \(\frac{3}{p}+\frac{2}{q}=1\) and \((p,q)\ne (3,\infty )\), then the weak solution (u, d) is regular in \(Q=B\times (-1,0)\), Serrin’s method [36] and Struwe’s method [38] dealing with Navier–Stokes equations are applied. For the Borderline case where \((p,q)=(3,\infty )\), [28] showed that the weak solution (u, d) is smooth in \(\mathbb {R}^{3}\times (0,T]\) when \(u\in L^{\infty }(0,T;L^{3}(\mathbb {R}^{3}))\)(also to see [25]). The \((3,\infty )\) case requires a technique utilizing the backward uniqueness of heat operator and unique continuation through spatial boundary, which was used to deal with the Navier–Stokes equations in [3]. For general Lorenz space \(L^{3,q}(\mathbb {R}^3), 3<q<\infty \) case, we prove in [26] if \(u\in L^\infty (0,T; L^{3,q}(\mathbb {R}^3))\) then (u, d) is smooth. For N-S equations we refer to [29]. For \(L^{3,\infty }(\mathbb {R}^3)\) case, Choe, Wolf and Yang [2] prove that if \(u\in L^\infty (0,T; L^{3,\infty }(\mathbb {R}^3))\) and an additional local small condition
then weak solution u of Navier–Stokes equations is smooth on \(Q_{\epsilon r}(z_0)\). Also to see Seregin [35].
In this paper, we shall established the regularity of weak solutions of liquid crystals equations in Lorentz space \(L^{3,\infty }\), the technicality is different from Navier–Stokes equations, the main technique is to deal with the new norms. Our main results can be stated as following.
Theorem 1.3
Assume (u, d, p) is a weak solution of (1.1) with \(u\in L^\infty (0,T; L^{3,\infty }(\mathbb {R}^n))\). There exists a positive constant \(\varepsilon \) small with following property. If \(z_0=(x_0,t_0)\in Q_T\) and \(R>0\) such that \(Q_R(z_0)\subset Q_T\) and (d, p) satisfy
for some \(0< r\le R/2,\)
Then (u, d) is smooth in \(Q_{\varepsilon r}(z_0).\) Here \(C(\nabla d,R,z_0)\) and \(D(p, R,z_0)\) are dimensionless quantities in Sect. 2.
Theorem 1.4
Let (u, d) be a suitable weak solution to the liquid crystal equations in \(Q_T\) with \(u\in L^\infty (0,T; L^{3,\infty }(\mathbb {R}^n))\). Then there exist at most finite number \(\mathcal {N}\) of singular points at any singular time t.
2 Notations and preliminaries
Let us recall the scaling property of (1.1). Denote
If (u, p, d) is a solution in \(\mathbb {R}^{3}\times (0,T)\), then obviously \((u_{\lambda },p_{\lambda },d_{\lambda })\) is a solution to the following equations
Thus, the scaling dimension of corresponding quantities are \(\dim u=-1\), \(\dim p=-2\), and \(\dim d=0\) (we assign x with dimension 1 and t with 2). There are some useful dimensionless quantities and we list them here, let \(z_0=(x_0,t_0),\)
Similarly, we denote these notations for \(\nabla d\):
For simplicity when \(z_0=(0,0)\), we write \(A(u,r)=A(u,r,(0,0))\) , and write \(A(r)\equiv A(u,\nabla d,r)=A(u,r)+A(\nabla d,r)\), and the meaning of E(r), C(r), K(r) are alike.
Next, we write down several facts about Lorentz spaces. We say a locally integrable function \(f\in L^{p,q}(\Omega )\), if the quasi-norm below is bounded
where
A basic fact for such spaces is
where \(0<q_{1}<p<q_{2}<\infty \), and \(L^{p}_{w}\) is the weak-\(L^{p}\) space. If \(|\Omega |\) is finite then \(L^{p,q}(\Omega )\subset L^r(\Omega )\) for all \(0<q\le \infty \) and \(0<r<p\),
Lemma 2.1
Let (u, d, p) be a weak solution to the liquid crystal Eq. (1.1) in \(Q=\Omega \times (a,b)\). Let \(z_0=(x_0,t_0)\) and let \(\rho >0\) be such that \(Q_\rho (z_0)\subset Q.\) For every \(r<(0,\frac{\rho }{4}]\), we have
Proof
Let \(z_0=(0,0)\), we decompose p so that
where \(p_1\) satisfies in \(B_\rho \) for a.e. \(t\in [-\rho ^{2},0]\), in the weak sense,
And \(p_2\) is a harmonic function in \(B_\rho \), i.e.,
Regarding \(p_1\), by theory of Laplace operator and Calderón–Zygmund theorem, we have
For \(x\in B_{\frac{\rho }{2}}\),
i.e., for \(r\le \frac{\rho }{2},\)
Now,
so that
On the other hand
we have
For \(x\in B_{\frac{\rho }{2}}\),
we have
therefore,
\(\square \)
We use following analysis Lemma 2.2 which can be found in ([6], Lemma 6.1) to prove local estimate Lemma 2.3.
Lemma 2.2
Let I(s) be a bounded nonnegative function in the interval \([R_1,R_2].\) Suppose that for any s, \(\rho \in [R_1,R_2]\) and \(s<\rho \), the following yields
with \(\alpha>\beta>\gamma >0\), \( a_i>0 , i=1,2,3, 4 \) and \(\theta \in [0,1)\). Then,
Lemma 2.3
Let (u, d, p) be a suitable weak solution to the liquid crystal Eq. (1.1) in \(Q=\Omega \times (a,b).\) Assume that \(z_0=(x_0,t_0)\) and \(1\ge r>0\) with \(Q_r(z_0)\subset Q\). Then the following holds:
Proof
Let \(r/2\le s<\rho \le r<1,\) and \(Q_r\subset Q_1\equiv Q.\) Choosing test function \(\phi (x,t)=\eta _1(x)\eta _2(t)\) with \(\eta _1\in C^\infty _0(B_\rho (x_0)),\) \(0\le \eta _1\le 1\) in \(\mathbb {R}^3,\) \(\eta _1\equiv 1\) on \(B_s(x_0),\) and \( |\nabla ^\alpha \eta _1|\le \frac{C}{(\rho -s)^{|\alpha |}},\) for all multi-index \(\alpha ,\) with \(|\alpha |\le 3.\) And \(\eta _2\in C^\infty _0(t_0-\rho ^2,t_0+\rho ^2),\) \(0\le \eta _2\le 1\) in \(\mathbb {R},\) \(\eta _2(t)\equiv 1\) for \(t\in [t_0-s^2,t_0+s^2],\) with \( |\eta _2'(t)|\le \frac{C}{\rho ^2-s^2}\le \frac{C}{r(\rho -s)}. \) From the local energy inequality we have
Here, we rewrite the term
Denote
and
Estimate \(J_1,\) \(J_2,\) \(J_3, J_4\) and \(J_5\), respectively, as the following:
by Young’s inequality, we get
similarly, we have
For the term \(J_4\), using Hölder’s inequality and Sobolev inequality, we have
From (2.8), using estimates above with respect to \(J_1, J_2, J_3, J_4\) and \(J_5\) we get
By Lemma 2.2, we have
Finally, we get
For \(f\in L^{7/5}(B_r(x_0))\) and \(\varphi \in C^\infty _0(B_r(x_0))\), we have
where \(\mathbf {I}_1\) is the first order Riesz’s potential defined by
By using Hardy–Littlewood–Sobolev inequality, we have
Applying (2.11) with \(f=|y|^2+|\nabla d|^2,\) and \(f= |u||\nabla d|,\) we obtain
\(\square \)
We need the bounded estimates for \(C(\nabla d,r)\) and D(p, r) with the help of the bounded of C(u, r).
Lemma 2.4
Suppose that (u, d, p) is a suitable weak solution in \(Q_1(z_0)=B_1(x_0)\times (t_0-1,t_0)\). Let
for \(M>0\). Then for every \(0<r<1/4,\) we have the following estimates:
Proof
Without loss of generality, we consider \(z_0=(0,0).\) It is easy to see that
Combining with (2.7) and (2.12) we obtain
Let \(r=\theta \rho \) with \(\theta \le \frac{1}{4}.\) From (2.5) (2.13), we have
Using Young’s inequality, we have
Set \(F(r)=C(\nabla d,r)+D(p,r)^\frac{5}{6}.\) Choose \(\eta >0\) and \(\theta >0\) small enough, we have
By the standard iterating argument
So that for \(0<r\le 1/4,\)
The estimates of \(C_1(u,\nabla d,r)\) and \(D_1(p,r)\) are immediate results. \(\square \)
3 Proof of Theorem 1.3
We shall prove the following Proposition 3.1, Theorem 1.3 is an immediate result.
Proposition 3.1
Let (u, d, p) a weak solution of (1.1) with
for \(z_0=(x_0,t_0)\) and \(R>0\) such that \(Q_R(z_0)\subset Q_T\), (d, p) satisfy
there exists a positive number \(\varepsilon (M, N)<\frac{1}{4}\) such that if for some \(0<r\le R/2\),
then there exists \(\rho \in [2r\varepsilon , r]\) such that
where \(\varepsilon _0\) is the same number in Proposition 1.2.
Proof
Let (u, d, p) be a weak solution of (1.1) with \(u\in L^\infty (0,T; L^{3,\infty }(\mathbb {R}^3)\). We assume
Note that \(|d|\le 1\), we have (see [28])
Also since the real interpolation \(L^{4}=[L^{6,\infty },L^{3,\infty }]_{\frac{1}{2},4}\) holds, then
From energy inequality and estimates above we get
which yields (u, d, p) is a local suitable weak solution of (1.1) and \(u\in C([0,T]; L^2(\mathbb {R}^3))\).
We use a contradiction argument for \(z_0=(0,0)\) and \(R=1\). Fixed \(N, M>0\) if the assertion of the proposition were false, then there would exist \(\epsilon _k \downarrow 0\), and suitable weak solutions \((u_k,d_k, p_k)\) of (1.1) and \(r_k\le 1/2\) such that
and for all \(\rho \in [2r_k\epsilon _k, r_k]\),
Since for \(0<r\le 1\)
combining the estimate and (3.6) with Lemma 2.4, we get, for \(0<r\le 1/2\),
Similarly, for any \(z_0\in Q_{1/2}\) and \(0<r\le 1/2\), we have
Define, for \((x,t)\in Q_{r_k^{-1}}\),
Obviously, \((U_k, D_k, P_k)\) are weak solutions to system (1.1) with the right side of (1.1)\(_{3}\) replaced by \(r_{k}^{2}f(D_{k})\) in \(Q_{r_k^{-1}}\). Now, for \(a>0,\) and \(a r_k\le 1/2,\)
We have by Lemma 2.4 again
So that
Thus, the \(L^p\) estimate holds for \((U_k, D_k, P_k)\) in \(Q_a\), for any \(a>0\),
By Aubin–Lion’s lemma, there exists a triplet (v, e, q) such that
and
Using estimates above, the limit function (v, e, q) satisfy, in the sense of suitable weak solutions on \(\mathbb {R}^3\times (-\infty ,0)\),
and for \(\rho \in [2\epsilon _k,1]\)
Taking limit we get
and for \(\rho \in (0, 1]\)
The crucial point here is a reduction to backward uniqueness for the heat operator with lower order terms as [3]. Set
Then \((v_k, e_k, q_k)\) satisfy (3.13), and similar to (3.10) for any \(a>0\)
As before there exists a triplet \((\widetilde{v},\widetilde{e},\widetilde{q})\) is suitable weak solution of (3.13) such that
and
From (3.16) we get
from (3.17) and take \(\rho =\rho _k\)
On the other hand, for fixed \(z_0\) and \(0<R\le \frac{1}{2r_k}\), as (3.18)
By the Fubini theorem, we have,
Hence, for any \(\eta >0,\) there exists a \(B_R\) such that
Let \(Q_1(z_0)\subset (\mathbb {R}^3\backslash B_R)\times (-T,0]\), by (3.22), we have \(A(\widetilde{v},\nabla \widetilde{e},\theta ,z_0)+E(\widetilde{v},\nabla \widetilde{e},\theta ,z_0)\le C(M,N)\) for any \(0<\theta \le 1.\) Thus, by the interpolation inequality we have
Thus
For any \(\epsilon >0\) we choose \(\gamma \) and \(\eta \) such that \(C_1(\widetilde{v},1,z_0)<\epsilon \).
It is easy to see that, by [28] (to see Lemma 3.2),
Utilizing (3.23) and Hölder’s inequality we have
Thus
First we take \(\theta \) such that \(c(M,N)\theta ^{1/2}\le \varepsilon _0/2,\) then take \(\epsilon \) such that \(\theta ^{-2}\epsilon +c(M, N)\theta ^{-3/4}\epsilon ^{1/6}\le \varepsilon _0/2\), i.e.,
which implies that \(z_0\) is a regular point by Proposition 1.2, therefor \((\widetilde{v},\widetilde{e},\widetilde{q})\) are smooth and their derivatives are bounded in \((\mathbb {R}^{3}\setminus B_{2R})\times (-T/2,0)\). Next, we show
For any B(y) and \(\phi \in C_0^\infty (B(y)),\) since e is Hölder continuous (to see following Lemma 3.2), we have
The backward uniqueness theorem of parabolic equations [3], we conclude
Using unique continuation theorem of parabolic equation in the bounded domain again [3], we conclude that
Thus, \(\widetilde{v}\) satisfies Navier–Stokes equations in \(\mathbb {R}^3\times (-T/2,0)\)
Using (3.20) and backward uniqueness of heat operator again [3], we get
which is a contradiction with (3.21).\(\square \)
We need following lemma.
Lemma 3.2
For \(v\in L^\infty (0, T; L^{3,\infty }(\mathbb {R}^3))\), if e satisfies in \(\mathbb {R}^3\times (0,T)\)
Then e is Hölder continuous.
Proof
From [34], if \(v\in L^\infty (0, T; BMO^{-1}(\mathbb {R}^3))\), then e is Hölder continuous. We only prove the following inclusion relationship for \(3<p<\infty \),
where \(L^{3}_{w}(\mathbb {R}^{3})=L^{3,\infty }(\mathbb {R}^{3})\) is the weak Lebesgue space, and \(\dot{B}^{-1+\frac{3}{p}}_{p,\infty }(\mathbb {R}^{3})\) is a homogeneous Besov space.
The first inclusion is obtained through Sobolev embedding and real interpolation. To be specific, we write weak space \(L^{3}_{w}\) a real interpolation
where \(\theta =\frac{p}{3(p-2)}\). Notice we have the following two embedding relations
where we have used Littlewood–Paley Theorem to characterize \(L^{r}\) by the Triebel–Lizorkin space \(F^{0}_{r,2}\) (for \(1<r<\infty \)). Thus the identity map is bounded:
By real interpolation, and \((1-\theta )\cdot (\frac{3}{p}-\frac{3}{2})+\theta \cdot 0=-1+\frac{3}{p}\), this leads to
thus the identity map
is also bounded. The second inclusion is obtained by using the heat kernel characterization of the corresponding spaces, denote \(s=-1+\frac{3}{p}\),
Direct calculation yields:
The proof is thus finished.\(\square \)
4 Proof of Theorem 1.4
Define for any \( r>0\) such that \(Q_r(z_0)\subset Q_T,\)
By interpolation inequality we have
and
where \(\Vert u\Vert _{L^\infty (-1,0; L^{3,\infty }(\mathbb {R}^3))}\le M\). On the other hand, by Calderón-Zygmund theorem,
and
which implies
we have
Since for any \(0<r\le R,\)
by Lemma 2.4, we have for any \(0<r\le R/2\) and \(z_0\in \Omega \times (0,T), \Omega \subset \subset \mathbb {R}^3\),
The number \(\varepsilon (M,N)\) of Proposition 3.1 can be determined.
Let S be a singular points set of (u, d) at \(\left\{ (x,T): x\in \mathbb {R}^3\right\} \). Assume that it contains more than \(M^3\varepsilon ^{-4}\) elements. Letting \(P=[M^3\varepsilon ^{-4}]+1\), we can find P different singular points \(\left\{ (x_k, T): k = 1, 2,\ldots , P\right\} \) of the set S. We can choose \(R_0\le R\) such that \(B_{R_0}(x_k)\cap B_{R_0}(x_l)=\emptyset , k\ne l,\) and bounded domain \(\Omega \) such that \(\cup _{k=1}^PB_{R_0}(x_k)\subset \Omega \). According to Proposition 3.1, for all \(r\in (0,R_0/2]\), it holds true
for all \(k=1,2,\ldots , P.\) In particular, taking \(r=r_0=R_0/2,\) we have
i.e., \(P\le M^3\varepsilon ^{-4}<P,\) which is a contradiction. \(\square \)
5 Appendix: Proof of Proposition 1.2
According to the \(L^p\) theorem of Stokes system in [5], if \(\mathbf {f}\in W^{-1,q}(\Omega ;\mathbb {R}^3), 1<q<\infty \), and \(\Omega \) is a \(C^1\) bounded domain, the following Stokes equations
there exists exactly one solution \((v,p)\in W^{1,q}(\Omega )\times L^q(\Omega ),\) and
Wolf’s the local pressure projection \(\mathcal {P}_q\) tell us,
As in [39], we have following Lemma.
Lemma 5.1
Let (u, d) be a weak solution of (1.1), then for every \(C^2\) bounded sub-domain \(\Omega \), and any \(\phi \in C_0^\infty (\Omega \times (0,T))\), there holds
i.e., set \(v_\Omega =v:=u+\nabla p_h\),
where \(\mathbf {I}\) is identity matrix, and
In addition, following estimates hold for a.e. \(t\in (0,T)\)
Here \(c>0\) depends on the geometry of \(\Omega \) and in (5.5)\(_1\) on m only. In particular, if \(\Omega \) is the ball \(B_R(x_0)\) then c in (5.5)\(_1\) depends only on m, while in (5.5)\(_2\) and (5.5)\(_3\) c is an absolute constant.
Hence, we have local energy inequality, for \(\varphi \in C_0^\infty (\Omega \times (0,T))\)
Note that the suitable weak solution of (1.1) satisfies the local energy inequality (5.6).
From local energy inequality we can get the Caccioppoli-type estimates
Here, \(W=(u, \nabla d)\), and
Obviously,
Proposition 1.2 is an immediate result of following lemma and Theorem (A).
Lemma 5.2
Suppose that (u, d) is a local suitable weak solution of (1.1). Then there exist universal constants \(\varepsilon ^*>0\) and \(\theta \in (0,\frac{1}{4}]\) with following property. For any \(\varepsilon \in (0, \varepsilon ^*]\) if
then
Proof
We prove by contradiction. Let \(\theta \in (0,\frac{1}{4}]\) be a constant to be specified later. Suppose there exist a decreasing sequence \(\{\varepsilon _n\}\) converging to 0, and a sequence of pairs of local suitable weak solutions \((u_n, d_n, p_n )\) such that
and
Define \((v_n, e_n, q_n)=(\frac{u_n}{\varepsilon _n}, \frac{d_n}{\varepsilon _n},\frac{p_n}{\varepsilon _n}),\) then they satisfy
Write \(w_n=(v_n, \nabla e_n)\), then
and
Using the Caccioppoli estimate (5.7) we conclude
which implies
The coercive estimate for the Stokes system (see, for instance, [27]) with a suitable cutoff function implies
where the constant c is independent of n. Thanks to the compact embedding theorem and (5.12), there exist \(w\in L^{3}(Q_{1/3}(z_0))\) and \(q\in L^{\frac{5}{4}}(Q_{1/3}(z_0))\) such that
Thus, (v, e, q) satisfy
Moreover
By the classical estimate of the Stokes system [37], we get
which implies that for \(0<\theta \le 1/3\)
This contradicts (5.11), if we choose \(\theta \) sufficiently small. The lemma is proved.\(\square \)
References
Caffarelli, L., Kohn, R., Nirenberg, L.: Partial regularity of suitable weak solutions of the Navier–Stokes equations. Commun. Pure Appl. Math. 35(2), 771–831 (1982)
Choe, H.J., Wolf, J., Yang, M.: On Regularity and singularity for \(L^\infty (0,T;L^{3,\omega }(\mathbb{R}^3))\) solutions to the Navier–Stokes equations. arXiv:1611.0472v1, [math.AP], 15 Nov 2016
Eskauriaza, L., Seregin, G.A., S̆verák, V.: \(L_{3,\infty }\)-solutions of Navier–Stokes equations and backward uniqueness. Uspekhi Mat. Nauk,Rossi skaya Akademiya Nauk. Moskovskoe Matematicheskoe Obshchestvo. Uspekhi Matematicheskikh Nauk 58(2), 3–44 (2003)
Ericksen, J.L.: Hydrostatic theory of liquid crystal. Arch. Ration. Mech. Anal. 9, 371–378 (1962)
Galdi, G.P., Simader, C.G., Sohr, H.: On the Stokes problem in Lipschitz domains. Ann. Di Mat. Pura Ed Appl. (IV) 167, 147–163 (1994)
Giusti, E.: Direct Methods in the Calculus of Variations. World Scientific Publishing, River Edge (2003)
Giga, Y., Sohr, H.: Abstract \(L^p\)-estimates for the Cauchy problem with applications to the Navier–Stokes equations in exterior domains. J. Funct. Anal. 102, 72–94 (1991)
Huang, T., Lin, F.H., Liu, C., Wang, C.Y.: Finite time singularity of the nematic liquid crystal flow in dimension three. Arch. Ration. Mech. Anal. 221(3), 1223–1254 (2016)
Hong, M.C.: Global existence of solutions of the simplified Ericksen–Leslie system in dimension two. Calc. Var. Partial Differ. Equ. 40(1–2), 15–36 (2011)
Hong, M.C., Li, J.K., Xin, Z.P.: Blow-up criteria of strong solutions to the Ericksen–Leslie system in \(\mathbb{R}^3\). Commun. Partial Differ. Equ. 39(7), 1284–1328 (2014)
Hong, M.C., Xin, Z.P.: Global existence of solutions of the liquid crystal flow for the Oseen–Frank model in \(\mathbb{R}^2\). Adv. Math. 231(3–4), 1364–1400 (2012)
Ladyžgenskaya, O.A., Seregin, G.A.: On partial regularity of suitable weak solutions to the three-dimensional Navier–Stokes equations. J. Math. Fluid Mech. 1, 356–387 (1999)
Ladyženskaya, O.A.: Uniqueness and smoothness of generalized solutions of Navier–Stokes equations. Zap. Naučn. Sem. Leningrad. Otdel. Mat. Inst. Steklov. (LOMI) 5, 169–185 (1967)
Leslie, F.M.: Some constitutive equations for liquid crystals. Arch. Ration. Mech. Anal. 28, 265–283 (1968)
Lei, Z., Li, D., Zhang, X.Y.: Remarks of global wellposedness of liquid crystal flows and heat flows of harmonic maps in two dimensions. Proc. Am. Math. Soc. 142(11), 3801–3810 (2014)
Lin, F.H.: Nonlinear theory of defects in nematic liquid crystal: phase transition and flow phenomena. Commun. Pure Appl. Math. 42, 789–814 (1989)
Lin, F.H.: A new proof of the Caffarelli–Kohn–Nirenberg theorem. Commun. Pure. Appl. Math. 51(3), 0241–0257 (1998)
Lin, F.H., Lin, J.Y., Wang, C.Y.: Liquid crystal flows in two dimensions. Arch. Ration. Mech. Anal. 197, 297–336 (2010)
Lin, F.H., Liu, C.: Nonparabolic dissipative systems modeling the flow of liquid crystals. Commun. Pure Appl. Math. 48(5), 501–537 (1995)
Lin, F.H., Liu, C.: Partial regularity of the dynamic system modeling the flow of liquid crystals. Discrete Contin. Dyn. Syst. 2(1), 1–22 (1996)
Lin, F.H., Liu, C.: Existence of solutions for the Ericksen–Leslie system. Arch. Ration. Mech. Aanl. 154(2), 135–156 (2000)
Lin, F.H., Wang, C.Y.: Recent developments of analysis for hydrodynamic flow of nematic liquid crystals. Philos. Trans. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci. 372(2029), 20130361 (2014)
Lin, F.H., Wang, C.Y.: Global existence of weak solutions of the nematic liquid crystal flow in dimension three. Commun. Pure Appl. Math. 69(8), 1532–1571 (2016)
Liu, X., Min, J., Wang, K., Zhang, X.: Serrin’s regularity results for the incompressible liquid crystals system. Discrete Contin. Dyn. Syst.-A 36(10), 5579–5594 (2016)
Liu, X., Min, J., Zhang, X.: \(L^{3,\infty }\) solutions of the liquid crystals system. J. Differ. Equ. 267, 2643–2670 (2019)
Liu, X., Liu, Z., Min, J.: Regularity criterion for 3D liquid crystal system in critical Lorentz spaces. 2020, preprint
Maremonti, P., Solonnikov, V.A.: On the estimate of solutions of evolution Stokes problem in anisotropic Sobolev spaces with a mixed norm. Zap. Nauchn. Sem. LOMI 223, 124–150 (1994)
Yueyang, Men, Wendong, Wang, Gang, Wu: Endpoint regularity criterion for weak solutions of the 3d incompressible liquid crystals system. Math. Methods Appl. Sci. 41(10), 3672–3683 (2018)
Phuc, N.C.: The Navier–Stokes equations in nonendpoint borderline Lorentz spaces. J. Math. Fluid Mech. 17(4), 741–760 (2015)
Prodi, Q.: Un teorema di unicit per le equazioni di Navier–Stokes. Ann. Mat. Pura Appl. (4) 48, 173–182 (1959)
Scheffer, V.: Partial regularity of solutions to the Navier–Stokes equations. Pac. J. Math. 66(2), 535–552 (1976)
Scheffer, V.: Hausdorff measure and the Navier–Stokes equations. Commun. Math. Phys. 55(2), 97–112 (1977)
Scheffer, V.: The Navier–Stokes equations on a bounded domain. Commun. Math. Phys. 73, 1–42 (1980)
Seregin G., Silvestre L., S̆verák V. and Zlatos A., On divergence-free drifts. J. Differ. Equ., 252, 505–540 (2012)
Seregin, G.: A note on weak solutions to the Navier–Stokes equations that are locally in \(L^\infty (L^{3,\infty })\). arXiv:1906.06707v1 [math.AP] 16 Jun 2019
Serrin, J.: On the interior regularity of weak solutions of the Navier–Stokes equations. Arch. Rational Mech. Anal. 9, 187–195 (1962)
Solonnikov, V.A.: Estimates of solutions to the linearized system of the Navier–Stokes equations. Trudy Steklov Math. Inst. LXX, 213–317 (1964)
Struwe, M.: On partial regularity results for the Navier–Stokes equations. Commun. Pure Appl. Math. 42(4), 437–458 (1988)
Wolf, J.: On the local regularity of suitable weak solutions to the generalized Navier–Stokes equations. Ann. Univ. Ferrara 61, 149–171 (2015)
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This work was supported partly by NSFC Grant 11971113, 11631011.
Rights and permissions
About this article
Cite this article
Liu, X., Liu, Y. & Liu, Z. A remark on regularity of liquid crystal equations in critical Lorentz spaces. Annali di Matematica 200, 1709–1734 (2021). https://doi.org/10.1007/s10231-020-01056-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10231-020-01056-4