Abstract
It is proved that if the solution of the Navier–Stokes system satisfies
or
then the solution is smooth on (0, T]. These two improve many previous results.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
In this paper, we continue our study [31, 35, 36] of the regularity criteria of the following Navier–Stokes equations (NSE):
where \(\varvec{u}=(u_1,u_2,u_3)\) is the fluid velocity field, \(\pi \) is a scalar pressure, \(\varvec{u}_0\) is the prescribed initial velocity field satisfying the compatibility condition \(\nabla \cdot \varvec{u}_0=0\), and
Leray [18] and Hopf [13] have established a global weak solution to (1); however, it remains an open problem of its regularity and uniqueness. Serrin [25] first showed that if
then the solution is regular on (0, T]. See also [8, 22]. The so-called Serrin-type regularity criterion (2) was generalized by Beir\(\tilde{\text {a}}\)o da Veiga [1] by adding integrability conditions on the velocity gradient,
In view of the divergence-free condition \(\nabla \cdot \varvec{u}=0\), it is natural and important to consider components reduction improvements of (2) and (3), that is, whether or not integrability conditions on partial components of the velocity or velocity gradient could still ensure the smoothness of the solution. There are so many studies devoted to this refinement, and without no intention to be complete, we recommend [2, 3, 5, 6, 12, 15, 16, 19,20,21, 26, 28, 31, 32, 34, 37,38,39,40].
In this paper, we would like to investigate the regularity criterion of (1) based on one directional derivative of the velocity field, say \(\partial _3\varvec{u}\), or one diagonal entry of the velocity gradient, say \(\partial _3u_3\). Let us first review what have happened in the last decades. In [20, Theorem 4 (i)], Penel–Pokorný showed that if
then the solution is smooth. This is based on a regularity criterion in terms of \(u_3\), \(\partial _3u_1\) and \(\partial _3u_2\) [20, Theorem 1 (a)]:
Then, Kukavica–Ziane [16] established a fine property of the horizontal convective terms (denoting by \(\displaystyle {\Delta _h=\partial _1\partial _1+\partial _2\partial _2}\) the horizontal Laplacian)
and refined (4) to be critical, but with limited range of space integrability indices,
Later on, Cao [2] employed multiplicative Sobolev inequalities
and
to get the following extended regularity condition
It should be remarked that Cao [2] claimed the range of q in (10) is \(\displaystyle {q\ge \frac{27}{16}}\), but it is indeed (10) which is actually proved. See the footnote of [31, p. 35] for more information.
In a recent paper, Zhang [31] generalized (9) as
and employed general \(L^{2\lambda }\) estimate (instead of \(L^4\) estimate as in [2]) to improve (7) and (10) simultaneously. Precisely, he showed the following regularity criterion,
Notice in establishing (12), Zhang have missed a condition (say, in [31, (31)], we need \(1\le c\le \infty \)), which was noticed by Yuliya–Skalak [30]. Skalak [27] then covered all of the range \(\left( \frac{3}{2},3\right] \),
Finally, Zhang–Yuan–Zhou [36] showed two new refinements of (4),
and
Whence, the state of the art is the following smoothness condition
The first purpose of this paper is to improve (16) in the range \(3<q<4\).
As far as regularity criterion \(\partial _3u_3\) is concerned, Zhou–Pokorný [39] first established a regularity condition based on \(u_3\) and then showed that if
then the solution is smooth. The equality in (17) was verified by Jia–Zhou [14]:
Later, Cao–Titi [4] established a bilateral multiplicative Sobolev inequality (see [32, Remark 8] for more information, and [35] for a more efficient form)
With (19) in hand, Cao–Titi showed the following two smoothness conditions,
and
Then Fang–Qian [9, Theorems 1.1 and 1.2] dominated \(u_3\) by \(\partial _1u_3\), employed some tricks in [4] and improved (21) as (after rationalizing the denominator of [9, Equation (1.10)])
Finally, Zhang–Zhong–Huang [35] found a more effective substitute of (19),
Invoking (23), Zhang–Zhong–Huang [35] were able to improve (21) and (22) as
but could not refine (20).
As for (20), Fang–Qian [9] made a contribution by invoking a regularity criterion of Zhang [33]. Fang–Qian [10, Theorem 1.8] then used an integration by parts technique in estimating \(u_3\) by \(\partial _3u_3\) and obtained the finest result up to now,
For later developments in anisotropic Lebesgue spaces, see [11, 24]. The second aim of the present paper is to make (25) better.
Before stating the precise result, let us recall the weak formulation of (1), see [7, 17, 23, 29] for instance.
Definition 1
Let \(\varvec{u}_0\in L^2(\mathbb {R}^3)\) with \(\nabla \cdot \varvec{u}_0=0\), \(T>0\). A measurable \(\mathbb {R}^3\)-valued function \(\varvec{u}\) defined in \([0,T]\times \mathbb {R}^3\) is said to be a weak solution to (1) if
-
(1)
\(\varvec{u}\in L^\infty (0,T;L^2(\mathbb {R}^3)\cap L^2(0,T;H^1(\mathbb {R}^3))\);
-
(2)
(1)\(_1\) and (1)\(_2\) hold in the sense of distributions, i.e.,
$$\begin{aligned} \int \limits _0^t\int \limits _{\mathbb {R}^3}\varvec{u}\cdot \left[ \partial _t\varvec{\phi }+\left( \varvec{u}\cdot \nabla \right) \varvec{\phi }\right] \text { d}x\text { d}s+\int \limits _{\mathbb {R}^3}\varvec{u}_0\cdot \varvec{\phi }(0)\text { d}x =\int \limits _0^T\int \limits _{\mathbb {R}^3} \nabla \varvec{u}:\nabla \varvec{\phi }\text { d}x\text { d}t, \end{aligned}$$for each \(\varvec{\phi }\in C_c^\infty ([0,T)\times \mathbb {R}^3)\) with \(\nabla \cdot \varvec{\phi }=0\), where \(\displaystyle {A:B=\sum \nolimits _{i,j=1}^3 a_{ij}b_{ij}}\) for \(3\times 3\) matrices \(A=(a_{ij})\), \(B=(b_{ij})\), and
$$\begin{aligned} \int \limits _0^T \int \limits _{\mathbb {R}^3}\varvec{u}\cdot \nabla \psi \text { d}x\text { d}t=0, \end{aligned}$$for each \(\psi \in C_c^\infty (\mathbb {R}^3\times [0,T))\);
-
(3)
the strong energy inequality, that is,
$$\begin{aligned} \left\| \varvec{u}(t)\right\| _{L^2}^2+2\int \limits _s^t\left\| \nabla \varvec{u}(s)\right\| _{L^2}^2\text { d}s \le \left\| \varvec{u}(s)\right\| _{L^2}^2,\quad s\le t\le T, \end{aligned}$$holds for \(s=0\) and almost all times \(s\in (0,T)\).
Now, our main result reads
Theorem 2
Let \(\varvec{u}_0\in L^2(\mathbb {R}^3)\) with \(\nabla \cdot \varvec{u}_0=0\), \(T>0\). Assume that \(\varvec{u}\) is a weak solution to (1) on [0, T] with initial data \(\varvec{u}_0\). If one of the following two conditions holds,
then the solution \(\varvec{u}\) is smooth in \((0,T]\times \mathbb {R}^3\).
Remark 3
-
(1)
Our regularity criterion (26) is better than (16) in case \(3<\alpha <4\). See Fig. 1, where “Skalak” refers to (13), “one Zhang-Zhou” (the upper one) means (14), “two Zhang-Zhou (the lower one)” demonstrates (15), “Penel-Pokorny” reveals (4), and “this” reflects (26).
-
(2)
Our regularity criterion (27) is better than (17), (20) and (25). See Fig. 2, where “Zhou-Pokorný” refers to (17); “Cao–Titi” means (20); “Fang-Qian” demonstrates (25); and “this” reflects our result (27).
-
(3)
It is not so hard to deduce that the scaling dimension \(\displaystyle {\frac{3\left( \sqrt{65\alpha ^2-78\alpha +49}+7-\alpha \right) }{16\alpha }}\) in (27) is strictly decreasing with respect to \(\displaystyle {\frac{3+\sqrt{17}}{4}\le \alpha \le \infty }\). Notice that
$$\begin{aligned} \begin{aligned}&\lim _{\alpha \rightarrow \frac{3+\sqrt{17}}{4}}\frac{3\left( \sqrt{65\alpha ^2-78\alpha +49}+7-\alpha \right) }{16\alpha } =\frac{3(\sqrt{17}-3)}{2}\approx 1.68466,\\&\lim _{\alpha \rightarrow \infty } \frac{3\left( \sqrt{65\alpha ^2-78\alpha +49}+7-\alpha \right) }{16\alpha } =\frac{3(\sqrt{65}-1)}{16}\approx 1.32417, \end{aligned} \end{aligned}$$we have the following rough, but maybe more elegant regularity criterion in terms of \(\partial _3u_3\),
$$\begin{aligned} \begin{aligned}&\partial _3u_3\in L^\beta (0,T;L^\alpha (\mathbb {R}^3)),\\&\frac{2}{\beta }+\frac{3}{\alpha }= \frac{3(\sqrt{65}-1)}{16}\approx 1.32417,\quad 1.78078\approx \frac{3+\sqrt{17}}{4}\le \alpha \le \infty . \end{aligned} \end{aligned}$$(28)
2 Proof of Theorem 2
In this section, we shall prove Theorem 2.
Case I (26) holds. For any \(\varepsilon \in (0,T)\), due to the fact that \(\nabla \varvec{u}\in L^2(0,T;L^2(\mathbb {R}^3))\), we may find a \(\delta \in (0,\varepsilon )\), such that \(\nabla \varvec{u}(\delta )\in L^2(\mathbb {R}^3)\). Take this \(\varvec{u}(\delta )\) as initial data, there exists an \({\tilde{\varvec{u}}}\in C([\delta ,\varGamma ^*),H^1(\mathbb {R}^3)) \cap L^2(\delta ,\varGamma ^*;H^2(\mathbb {R}^3))\), where \([\delta , \varGamma ^*)\) is the life span of the unique strong solution, see [29]. Moreover, \({\tilde{\varvec{u}}}\in C^\infty (\mathbb {R}^3\times (\delta ,\varGamma ^*))\). According to the uniqueness result, \({\tilde{\varvec{u}}}=\varvec{u}\) on \([\delta ,\varGamma ^*)\). If \(\varGamma ^*\ge T\), we have already that \(\varvec{u}\in C^\infty (\mathbb {R}^3\times (0,T))\), due to the arbitrariness of \(\varepsilon \in (0,T)\). In case \(\varGamma ^*<T\), our strategy is to show that \(u_3\in L^3(\delta ,\varGamma ^*;L^9(\mathbb {R}^3))\). Then, by the fact that
we may conclude the proof by invoking (5).
For this purpose, we multiply the equation of \(u_3\) in (1) by \(|u_3|u_3\) and integrate over \(\mathbb {R}^3\),
By the Hölder inequality,
To dominate \(\partial _3\pi \), we apply the divergence of (1)\(_1\) to obtain
and thus by the interpolation inequality,
Employing the multiplicative Sobolev inequalities (8) and (11) yields
After collection, we deduce by the Young inequality,
Putting this above inequality into (29) and applying the Gronwall inequality give
as desired.
Case II (27) holds. Argue as in Case I, it suffices to show that \(\left\| \nabla \varvec{u}(t)\right\| _{L^2}\) is uniformly bounded as \(t\nearrow \varGamma ^*\). By the absolute continuity property of the Lebesgue integrable function, for \(\delta _2\in (0,1)\) to be determined, we can choose a \(\delta _1\in [\delta ,\varGamma ^*)\) such that
We first establish the \(L^q\) bound of \(u_3\) in terms of \(\partial _3u_3\), which have been used in [9, 10]. Multiplying the third component of (1)\(_1\):
by \(|u_3|^{q-2}u_3\) with
integrating over \(\mathbb {R}^3\) and applying integration by parts, we obtain
By the Hölder inequality with
we have
Thanks to (30) and (11), it follows that
provided that
Employing the Gagliardo–Nirenberg inequality with
gives
By the fact that \(\varvec{u}\in L^\infty (0,T;L^2(\mathbb {R}^3))\) from Definition 1 and the Young inequality, we deduce
if
Plugging (37) into (33), absorbing the last term into the left-hand side and integrating with respect to the time, we find
Then, we establish the bound of \(\left\| \nabla _h\varvec{u}\right\| _{L^2}^2\) with \(\nabla _h=(\partial _1,\partial _2)\) the horizontal gradient operator. Testing (1)\(_1\) by \(-\Delta _h\varvec{u}\), it follows from [3, 4, 39] that
By the Hölder inequality, the Minkowski inequality and the Gagliardo–Nirenberg inequality,
Here, the exponents appeared above should satisfy
Putting (41) into (40) and integrating with respect to the time give
Finally, we obtain the \(H^1\) estimate of the solution. Taking the inner product of (1)\(_1\) by \(-\Delta \varvec{u}\) in \(L^2(\mathbb {R}^3)\), it follows from [39] that
Invoking the Hölder inequality (8) and the Young inequality, we get
Gathering (45) into (44) and integrating with respect to the time provide
Thanks to (31) and the obtained estimates (43) and (39), we have
Now, if
then the last integral in (47) is finite, and once we choose \(\delta _2\) sufficiently small, then we can absorb the last term in (47) into the left-hand side to deduce
as desired. The proof of Theorem 2 is thus completed.
Now, we calculate all the parameters above. Denote by \(\displaystyle {{\tilde{\beta }}=\frac{1}{\beta },\ {\tilde{\alpha }}=\frac{1}{\alpha }}\), then
On the other hand,
Solving this quadratic equation gives
Hence,
Now, the main restriction of \(\alpha \) comes from (32) and (38). After some calculations, we find (38) reduces to \(\displaystyle {\frac{3+\sqrt{17}}{4}\le \alpha \le \infty }\) (in (38), \(\beta =\infty \) corresponds to \(\displaystyle {\alpha =\frac{3+\sqrt{17}}{4}}\)), and all the assumptions, say (32), (34)-(36), (38), (42), (48), are all valid.
Remark 4
If we apply the same method in the proof of Theorem 2 to [9, Theorem 1.2], that is, in showing [9, Lemma 2.1], we use the generalized multiplicative Sobolev inequality (11), we get better result than (22), but no better result than (24).
References
Beirão da Veiga, H.: A new regularity class for the Navier–Stokes equations in \(\mathbb{R}^n\). Chin. Ann. Math. Ser. B 16, 407–412 (1995)
Cao, C.S.: Sufficient conditions for the regularity to the 3D Navier–Stokes equations. Discrete Contin. Dyn. Syst. 26, 1141–1151 (2010)
Cao, C.S., Titi, E.S.: Regularity criteria for the three-dimensional Navier–Stokes equations. Indiana Univ. Math. J. 57, 2643–2661 (2008)
Cao, C.S., Titi, E.S.: Global regularity criterion for the \(3\)D Navier–Stokes equations involving one entry of the velocity gradient tensor. Arch. Ration. Mech. Anal. 202, 919–932 (2011)
Chemin, J.Y., Zhang, P.: On the critical one component regularity for \(3\)-D Navier–Stokes systems. Ann. Sci. Éc. Norm. Supér. 49, 131–167 (2016)
Chemin, J.Y., Zhang, P., Zhang, Z.F.: On the critical one component regularity for \(3\)-D Navier–Stokes system: general case. Arch. Ration. Mech. Anal. 224, 871–905 (2017)
Constantin, P., Fioas, C.: Navier–Stokes equations, Chicago Lectures in Mathematics Series (1988)
Eskauriaza, L., Serëgin, G.A., Šverák, V.: \(L_{3,\infty }\)-solutions of Navier–Stokes equations and backward uniqueness. Russ. Math. Surv. 58, 211–250 (2003)
Fang, D.Y., Qian, C.Y.: The regularity criterion for the \(3\)D Navier–Stokes equations involving one velocity gradient component. Nonlinear Anal. 78, 86–103 (2013)
Fang, D.Y., Qian, C.Y.: Some new regularity criteria for the 3D Navier–Stokes equations (2012). arXiv preprint arXiv:1212.2335
Guo, Z.G., Li, Y.F., Skalak, Z.: Regularity criteria of the incompressible Navier–Stokes equations via only one entry of velocity gradient. J. Math. Fluid Mech. 21, Art. 35 (2019)
Guo, Z.G., Caggio, M., Skalak, Z.: Regularity criteria for the Navier–Stokes equations based on one component of velocity. Nonlinear Anal. Real World Appl. 35, 379–396 (2017)
Hopf, E.: Über die Anfangwertaufgaben für die hydromischen Grundgleichungen. Math. Nachr. 4, 213–321 (1951)
Jia, X.J., Zhou, Y.: Remarks on regularity criteria for the Navier–Stokes equations via one velocity component. Nonlinear Anal. Real World Appl. 15, 239–245 (2014)
Kukavica, I., Ziane, M.: One component regularity for the Navier–Stokes equations. Nonlinearity 19, 453–469 (2006)
Kukavica, I., Ziane, M.: Navier–Stokes equations with regularity in one direction. J. Math. Phys. 48, 065203 (2007)
Ladyzhenskaya, O.A.: The mathematical theory of viscous incompressible flow. Moscow (1970)
Leray, J.: Sur le mouvement d’un liquide visqueux emplissant l’espace. Acta Math. 63, 193–248 (1934)
Neustupa, J., Novotný, A., Penel, P.: An interior regularity of a weak solution to the Navier–Stokes equations in dependence on one component of velocity, Topics in mathematical fluid mechanics. Quad. Mat. 10, 163–183 (2002)
Penel, P., Pokorný, M.: Some new regularity criteria for the Navier–Stokes equations containing gradient of the velocity. Appl. Math. 49, 483–493 (2004)
Pokorný, M.: On the result of He concerning the smoothness of solutions to the Navier–Stokes equations. Electron. J. Differ. Equ. 2003, 1–8 (2003)
Prodi, G.: Un teorema di unicitá per le equazioni di Navier–Stokes. Ann. Mat. Pura Appl. 48, 173–182 (1959)
Lemarié-Rieusset, P.G.: Recent Developments in the Navier–Stokes Problem. Chapman and Hall, London (2002)
Qian, C.Y.: The anisotropic regularity criteria for 3D Navier–Stokes equations involving one velocity component. Nonlinear Anal. Real World Appl. 54, 103094 (2020)
Serrin, J.: On the interior regularity of weak solutions of the Navier–Stokes equations. Arch. Ration. Mech. Anal. 9, 187–195 (1962)
Skalak, Z.: On the regularity of the solutions to the Navier–Stokes equations via the gradient of one velocity component. Nonlinear Anal. 104, 84–89 (2014)
Skalak, Z.: The optimal regularity criterion for the Navier–Stokes equations in terms of \(\partial _3u\) (2018). Preprint https://mat.fsv.cvut.cz/nales/preprints/preprinty/2018/ulozeneclanky/preprint22018.pdf
Skalak, Z.: A regularity criterion for the Navier–Stokes equations based on the gradient of one velocity component. J. Math. Anal. Appl. 437(2016), 474–484 (2016)
Temam, R.: Navier–Stokes Equations, Theory and Numercial Analysis. AMS Chelsea Publishing, New York City (2001)
Yuliya, N., Skalak, Z.: The optimal regularity criterion for the Navier-Stokes equations in terms of one directional derivative of the velocity. ZAMM-J. Appl. Math. Mech. 100(1), e201800114 (2020)
Zhang, Z.J.: An improved regularity criterion for the Navier–Stokes equations in terms of one directional derivative of the velocity field. Bull. Math. Sci. 8, 33–47 (2018)
Zhang, Z.J.: An almost Serrin-type regularity criterion for the Navier–Stokes equations involving the gradient of one velocity component. Z. Angew. Math. Phys. 66, 1707–1715 (2015)
Zhang, Z.J.: A Serrin-type regularity criterion for the Navier–Stokes equations via one velocity component. Commun. Pure Appl. Anal. 12, 117–124 (2013)
Zhang, Z.J.: Serrin-type regularity criterion for the Navier–Stokes equations involving one velocity and one vorticity component. Czechoslov. Math. J. 68, 219–225 (2018)
Zhang, Z.J., Zhong, D.X., Huang, X.T.: A refined regularity criterion for the Navier–Stokes equations involving one non-diagonal entry of the velocity gradient. J. Math. Anal. Appl. 453, 1145–1150 (2017)
Zhang, Z.J., Yuan, W.J., Zhou, Y.: Some remarks on the Navier–Stokes equations with regularity in one direction. Appl. Math. 64, 301–308 (2019)
Zhou, Y.: A new regularity criterion for the Navier–Stokes equations in terms of the gradient of one velocity component. Methods Appl. Anal. 9(4), 563–578 (2002)
Zhou, Y.: A new regularity criterion for weak solutions to the Navier–Stokes equations. J. Math. Pures Appl. (9) 84(11), 1496–1514 (2005)
Zhou, Y., Pokorný, M.: On the regularity of the solutions of the Navier–Stokes equations via one velocity component. Nonlinearity 23, 1097–1107 (2010)
Zhou, Y., Pokorný, M.: On a regularity criterion for the Navier–Stokes equations involving gradient of one velocity component. J. Math. Phys. 50, 123514 (2009)
Acknowledgements
This work is partially supported by the National Natural Science Foundation of China (Grant No. 11761009) and the Natural Science Foundation of Jiangxi Province (Grant No. 20202BABL201008).
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.
Rights and permissions
About this article
Cite this article
Zhang, Z., Zhang, Y. On regularity criteria for the Navier–Stokes equations based on one directional derivative of the velocity or one diagonal entry of the velocity gradient. Z. Angew. Math. Phys. 72, 24 (2021). https://doi.org/10.1007/s00033-020-01442-1
Received:
Revised:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00033-020-01442-1