Home Gradient estimates for a class of higher-order elliptic equations of p-growth over a nonsmooth domain
Article Open Access

Gradient estimates for a class of higher-order elliptic equations of p-growth over a nonsmooth domain

  • Hong Tian and Shenzhou Zheng EMAIL logo
Published/Copyright: February 23, 2024

Abstract

This article is devoted to a global Calderón-Zygmund estimate in the framework of Lorentz spaces for the m -order gradients of weak solution to a higher-order elliptic equation with p -growth. We prove the main result based on a proper power decay estimation of the upper-level set by the principle of layer cake representation for the L γ , q -estimate of D m u , while the coefficient satisfies a small BMO semi-norm and the boundary of underlying domain is flat in the sense of Reifenberg. In particular, a tricky ingredient is to establish the normal component of higher derivatives controlled by the horizontal component of higher derivatives of solutions in the neighborhood at any boundary point, which is achieved by comparing the solution under consideration with that for some reference problems.

MSC 2010: 35J30; 35J92; 46E30

1 Introduction

Throughout the article, we assume that m 1 is a positive integer, α = ( α 1 , , α d ) is a multi-index with the length α = α 1 + + α d , and D α = D 1 α 1 D d α d with D i = x i for i = 1 , , d . Let Ω R d for d 3 be a given bounded domain with its nonsmooth boundary Ω determined later. We consider the following zero Dirichlet problem of a higher-order elliptic equation of p -growth for 1 < p < :

(1.1) α = β = m D α A ( x ) D m u , D m u p 2 2 A ( x ) D β u = α = m D α ( f p 2 f α ) in Ω , l m 1 D l u = 0 on Ω ,

where η , ξ α = β = m η β ξ α denotes a standard inner product of η , ξ R m d in the following context. Here, the variable coefficient tensor A ( x ) satisfies the uniform ellipticity and boundedness, i.e., there exist two positive constants 0 < λ Λ < such that

(1.2) λ ξ 2 α = β = m A α β ( x ) ξ α ξ β Λ ξ 2

for all ξ = { ξ α R d α = m } R m d and almost every x Ω . For a given function f = { f α : Ω R d α = m } R m d belonging to the classical Lebesgue spaces, we are concerned with the weak solution of the Dirichlet problem (1.1) in the classical Sobolev space W 0 m , p ( Ω ) , which means that

(1.3) Ω A ( x ) D m u , D m u p 2 2 A ( x ) D m u , D m φ d x = Ω f p 2 f , D m φ d x

holds for any φ W 0 m , p ( Ω ) . By using a suitable Galerkin-approximation in the L p -norm, there exists a unique weak solution u W 0 m , p ( Ω ) of problem (1.1) with the standard L p -estimate

(1.4) D m u L p ( Ω ) C f L p ( Ω ) ,

where C is a constant depending only on λ , Λ , p , and Ω , see [28, Part II, Page 94].

The main aim of this article is to prove a global regularity in the framework of Lorentz spaces for the m -order gradients of weak solution to problem (1.1) under weaker regular assumptions imposed on the given datum. In other words, we are to impose some minimal regular assumptions on the coefficient A ( x ) and on the rough geometric boundary Ω to ensure the following nonlinear Calderón-Zygmund estimate of the Dirichlet problem (1.1):

(1.5) f p L γ , q ( Ω ) D m u p L γ , q ( Ω ) ,

where the Lorentz spaces L γ , q ( Ω ) for any 1 < γ < d d 2 + ε , 0 < q , and ε > 0 as a small constant. In particular, for 1 < γ = q < d d 2 + ε , we see that equation (1.5) leads to the L γ -estimate for the Dirichlet problem (1.1), which was just done by Habermann for the elliptic equations of p ( x ) -growth in [28] while p ( x ) is a constant. Especially, it is necessary for the range of 1 < γ < d d 2 + ε to be confined even for a special coefficient A ( x ) I d and f ( x ) with a higher integrability.

It is a well-known fact that the uniform ellipticity and boundedness (1.2) is not enough to ensure the validity of the Calderón-Zygmund type estimate (1.5) if one does not impose more regularity assumptions on A ( x ) and Ω . A counterexample introduced by Meyers in [39] shows that the gradients for the solutions of linear elliptic equations involved in highly oscillatory coefficients cannot be expected to have higher integrability, irrespective of the nice regularity of the data f ( x ) . Another counterexample given by Jerison and Kenig’s work [33, Theorem A] also demonstrates that the Poisson equation over a Lipschitz domain may not be solvable in the W 1 , p -spaces for large p , unless the boundary is sufficiently smooth. Therefore, to ensure a nonlinear Calderón-Zygmund estimate of (1.1), it is necessary to find minimal regular assumptions for the coefficients A ( x ) and the boundary of underlying domain Ω . For this, let us recall some notations and basic facts. For any fixed point x R d and r > 0 , we set B r ( x ) = { y R d : x y < r } and briefly denote B r = B r ( 0 ) . The average for a measurable function g ( y ) on B r ( x ) is denoted by

( g ) B r ( x ) = B r ( x ) g ( y ) d y = 1 B r B r ( x ) g ( y ) d y ,

where B r is the d -dimensional Lebesgue measure of B r ( x ) . In the following context, we always confine δ ( 0 , 1 8 ) by a scale way.

Assumption 1.1

We say that ( A ( x ) , Ω ) is ( δ , R 0 ) -vanishing if for any x 0 Ω and r ( 0 , R 0 ] with

dist ( x 0 , Ω ) = min z Ω dist ( x 0 , z ) > 2 r ,

there exists a coordinate system depending on x 0 and r , whose variables are still denoted by x , such that in the new coordinate system, x 0 is the origin and

B r ( x 0 ) A ( x ) ( A ) B r ( x 0 ) p d x δ p

while, for any x 0 Ω and r ( 0 , R 0 ] with

dist ( x 0 , Ω ) = min z Ω dist ( x 0 , z ) = dist ( x 0 , z 0 ) 2 r

for some z 0 Ω , there exists a coordinate system depending on x 0 and r , whose variables are still denoted by x , such that in the new coordinate system, z 0 is the origin,

(1.6) B 3 r ( z 0 ) { x 1 3 δ r } B 3 r ( z 0 ) Ω B 3 r ( z 0 ) { x 1 3 δ r } ,

and

B 3 r ( z 0 ) A ( x ) ( A ) B 3 r ( z 0 ) p d x δ p ,

where A ( x ) is a zero-extension from B 3 r ( z 0 ) Ω to the ball B 3 r ( z 0 ) , and the parameters R 0 and δ > 0 will be specified later.

The domain Ω is called to be ( δ , R 0 ) -Reifenberg flat if equation (1.6) holds in the new coordinate system, see [44]. This ensures that Ω locally satisfies the so-called A -type one, which means that there exist R 0 > 0 and a small constant δ > 0 such that

(1.7) sup 0 < r R 0 B r ( x 0 ) Ω B r ( x 0 ) 2 1 δ d

for any x 0 Ω . The estimate (1.7) actually implies that a higher integrability of the m -order gradients of the solution holds for the Dirichlet problem (1.1) in a neighborhood of every boundary point. Moreover, we also note that the extension theorem of Sobolev embedding inequalities is available over the ( δ , R 0 ) -Reifenberg flat domains, see [11]. To present our main result precisely, we first introduce the definition of Lorentz spaces.

Definition 1.2

Let U be an open subset of R d . The Lorentz space L γ , q ( U ) with 1 γ <  and 0 < q <  is the set of all measurable functions g : U R such that

g L γ , q ( U ) = γ 0 ( κ γ { x U : g ( x ) > κ } ) q γ d κ κ 1 q < + .

For q = , the space L γ , ( U ) is set to be the usual Marcinkiewicz space with quasi-norm

g L γ , ( U ) = sup κ > 0 ( κ γ { x U : g ( x ) > κ } ) 1 γ < + .

We observe that if 1 γ = q < , then the Lorentz space L q , q ( U ) is nothing but the classical Lebesgue space L q ( U ) , which is equivalently endowed with the norm by

g L q ( U ) = U g ( x ) q d x 1 q < + .

Let us now state our research motivation under consideration. It is an important observation that for the linear case of p = 2 , the problem (1.1) leads to the following higher-order linear elliptic form:

(1.8) α = β = m D α ( A ( x ) D β u ) = α = m D α f α in Ω , l m 1 D l u = 0 on Ω .

Some recent studies of problem (1.8) are recalled in order. Byun and Ryu [12] established a global Orlicz estimate of higher-order elliptic and parabolic equations on nonsmooth domains, and they got that f p L ϕ ( Ω ) D m u p L ϕ ( Ω ) for any ϕ Δ 2 2 by a geometric method provided that the principle coefficients satisfy small BMO and the boundary of underlying domain is a Reifenberg flatness. Furthermore, Dong and Kim [1719] Dong and Gallarati [20] proved the L p -solvability for various complex boundary conditions of higher-order elliptic and parabolic systems by a different argument based on an estimate of the Fefferman-Stein sharp function theorem under main assumptions on the coefficients with small partially BMO quasi-norm, and the boundary of underlying domain being Reifenberg flat. Wang and Yao [51] verified the L p -regularity for a higher-order elliptic and parabolic equation in the whole space by way of the so-called large-M-inequality principle under assumption that the leading coefficients have small BMO quasi-norm. On the other hand, for a special case of m = 1 , this leads to a degenerate elliptic problem of p -growth as follows:

(1.9) div A ( x ) D u , D u p 2 2 A ( x ) D u = div ( f p 2 f ) in Ω , u = 0 on Ω .

As we know, Iwaniec [32] first proved a local nonlinear Calderón-Zygmund estimate for the equation of p -Laplacian, and he got that f L loc q D u L loc q for all q p . It was extended by DiBendetto and Manfredi in [21] to the setting of the elliptic systems of p -Laplacian. For the case of discontinuous coefficients, Caffarelli and Peral [15] used the modified Vitali covering originated from Safonov’s “crawling of ink spots” argument [45] to attain the W loc 1 , p -estimates for a class of general elliptic equations of p -growth with a small deviation of “variable coefficients” from “constant coefficients.” Kinnunen and Zhou proved the interior and boundary L q estimates of problem (1.9) in [34,35] based on the Perturbation approach, respectively, where the coefficients A ( x ) are of VMO class and the domain is C 1 , α for any 0 < α 1 . Later, Byun et al. [11] reduced the C 1 , α boundary regularity to the so-called Reifenberg flat boundary by the geometric method, and they also showed a global W 1 , q estimate of the gradients to the above problem (1.9). Recently, we also observed that a global nonlinear Calderón-Zygmund estimate is generalized to the weighted Lorentz spaces by Mengesha and Phuc [40] under the same regular assumptions on the coefficients and the Reifenberg domain. Here, we would like to mention that Acerbi and Mingione [1] first made use of the so-called large-M-inequality technique to prove a local W 1 , q -estimate of the weak solution for parabolic systems of p -growth with small BMO coefficients in all ( t , x ) variables. By way of Acerbi and Mingione’s idea, Baroni [6,7] established local Lorentz estimates of the gradients to the weak solution of parabolic equations and parabolic obstacle problems of p -growths, respectively, where the leading coefficients are allowed to be small BMO seminorms in all the spatial variables and measurable in time variable. In addition, it would be remarked that Eleuteri and Habermann [23] got the Calderón-Zygmund-type estimates for a class of obstacle problems with p ( x ) -growth. Tian and Zheng [50] also proved a global weighted Lorentz estimate for a variable power of the gradients to the weak solution of nonlinear parabolic equations with BMO nonlinearity over the quasiconvex domain by using the so-called large-M-inequality principle. For more study involving the existence and regularity of solutions to p -Laplacian elliptic and parabolic equations, we also refer to the recent articles [31,36,43,47,49].

Note that the research on regularity results of higher-order elliptic and parabolic problems is substantially rare in comparison to the second-order cases. Giaquinta and Modica in [25] showed some regularity results for higher-order nonlinear elliptic systems. Campanato and Cannarsa in [13] established a local ( m + 1 ) -order differentiability of the solutions of nonlinear elliptic systems with quadratic growth. The above result had been expanded by Campanato in [14] to nonlinear elliptic systems with the standard p -growths. Maz’ya [38] dealt with the Wiener test for the Dirichlet problem of strongly elliptic differential operators of an arbitrary even order 2 m with constant real coefficients, which is concerned with an equivalent relation between the capacitary Wiener’s-type criterion and the regularity of a boundary point. The partial Hölder regularity of the gradients of the solutions for such a higher-order homogeneous nonlinear elliptic system in divergence form had been completed by Duzaar et al. [22], while Habermann [30] also showed partial regularity for minima of higher-order functionals with p ( x ) -growth. Recently, Miyazaki [42] obtained a global L p ( Ω ) estimate when Ω is a C 1 -smooth domain and the coefficients satisfy uniform continuity. Bögelein [10] gave a higher integrability of the gradients of the solutions by way of proving a reverse Hölder inequality of the m -order gradients for quasilinear parabolic systems of higher-order. Habermann [28] showed a nonlinear Calderón-Zygmund estimate of D m u in L loc q ( Ω ) for any 1 < q < d d 2 for a higher-order nonlinear elliptic system of p ( x ) -growth, which the nonlinearities are assumed to be continuous in all spatial variables. Furthermore, Habermann [29] improved the integrable exponent of m -order gradients to L loc q ( Ω ) for any 1 < q < d d 2 + ε with a positive constant ε > 0 by proving the validity of the reverse Hölder inequality of the m -order gradients under the same regularity assumption on the nonlinearities. Choi and Kim [16] proved the weighted L p , q -estimates for divergence-type higher-order elliptic and parabolic systems with irregular coefficients on Reifenberg flat domains, where the coefficients do not have any regularity assumptions in the time-variable for the parabolic case, and the leading coefficients are permitted to have small mean oscillations in the spatial-variables for the elliptic case. On the other hand, the Lorentz regularity of gradients of solutions for elliptic and parabolic equations have been substantially studied since Talenti’s seminal work [48], for more references see [36,41,50]. It is a well-known fact that the Lorentz space is a refined version of the Lebesgue spaces. Therefore, the regularity under our consideration in the scale of Lorentz spaces for the Dirichlet problem (1.1) is inspired by the aforementioned works, which is supposed to be under weaker regular assumptions on the coefficient A ( x ) and the nonsmooth domain Ω .

We are now in a position to summarize the main result of this article.

Theorem 1.3

For 1 < p < and m 1 , let u W 0 m , p ( Ω ) be the weak solution of the Dirichlet problem (1.1) with the ellipticity and boundedness (1.2). Then, there exists a small constant δ = δ ( d , m , p , λ , Λ , Ω ) > 0 such that ( A ( x ) , Ω ) satisfies ( δ , R 0 ) -vanishing for some R 0 > 0 shown as Assumption 1.1; if f p L γ , q ( Ω ) for 1 < γ < d d 2 + ε and 0 < q with some ε = ε ( d , m , p , λ , Λ ) > 0 , then we have D m u p L γ , q ( Ω ) with the following estimate:

(1.10) D m u p L γ , q ( Ω ) C f p L γ , q ( Ω ) ,

where the constant C = C ( d , m , p , γ , q , λ , Λ , δ , R 0 , Ω ) > 0 while q < ; the constant is not dependent on q while q = .

It would be remarked that the restriction for γ < d d 2 + ε does not appear in the order-two case m = 1 because the solutions of second-order homogeneous elliptic problems with constant coefficients have local Lipschitz regularity, which implies that f p L γ , q ( Ω ) means that D u p L γ , q ( Ω ) for any 1 < γ < + and 0 < q in this setting. We refer the readers to [6,11,34] for more details. Indeed, a key point for such a higher-order problem of p -growth in comparison to the second-order case is the absence of a priori estimate in the W m , -spaces for the higher-order elliptic equations with constant coefficients. To this end, our scheme is arranged as follows. We are the first to control the normal component of higher derivatives by the horizontal component of higher derivatives of solutions on the boundary point, which is based on making comparisons among the solutions under consideration with some reference problems. Thanks to a validity of the ( m + 1 ) -order derivatives of the solution for such a higher-order elliptic equation with constant coefficients based on the difference quotient argument, then we attain the W m + 1 , p ˜ -estimate of the solution of problem (1.1) with a suitable exponent p ˜ depending only on d and p > 1 . This estimate combined with the Sobolev embedding theorem yields a higher integrability of the m -order gradients of the solution, see equation (3.14) in Section 3. Furthermore, an integrability of the m -order gradients is improved by the reverse Hölder inequality based on the definition of the weak solution of the Dirichlet problem (1.1). Here, we would also like to point out that it is infeasible to show an estimate of the ( m + 2 ) -order gradients of the solution by the same way as the difference quotient technique even for the higher-order equation of p -Laplacian. Finally, we are ascribed to study the upper-level set E ( κ ) = { x Ω : D m u p > κ } for sufficiently large κ by way of the modified Vitali covering and the large-M-inequality technique. It is essential to show a proper power decay estimate of such upper-level set so that we ensure the validity of the principle of layer cake representation for the L γ , q -estimate of D m u .

A brief outline of this article is given as follows: in Section 2, we introduce some preliminary lemmas; Section 3 is mainly devoted to local interior and boundary estimates of the m -order gradients of the weak solution to the Dirichlet problems (1.1), respectively; finally, the proof of main Theorem 1.3 is given in Section 4.

2 Some technical lemmas

In this section, we collect some preliminary results that will be used in the following context. Throughout the article, we denote by C ( d , m , p , ) a universal constant depending only on prescribed quantities and possibly varying from line to line. We begin this section with a few standard embedding relations for the Lorentz spaces. It is worth noting that the quantity L γ , q ( U ) is only a quasi-norm due to the lack of subadditivity. However, the mapping g g L γ , q ( U ) is still a lower semi-continuous, see [8, Section 3] or [40, Proposition 3.9].

Lemma 2.1

For any given U R d with a finite measure, we have the following conclusions:

  1. If 1 γ 1 γ 2 < and g L γ 2 , q ( U ) for any 0 < q , then it holds a continuous embedding L γ 2 , q ( U ) L γ 1 , q ( U ) with

    g L γ 1 , q ( U ) U 1 γ 1 1 γ 2 g L γ 2 , q ( U ) .

  2. If 0 < q 1 < q 2 and g L γ , q 1 ( U ) for any 1 γ < , then there holds a continuous embedding L γ , q 1 ( U ) L γ , q 2 ( U ) with

    g L γ , q 2 ( U ) C ( γ , q 1 , q 2 ) g L γ , q 1 ( U )

    while q 2 < ; the constant is not dependent on q 2 while q 2 = .

  3. There holds a continuous embedding L γ , q ( U ) L p ( U ) for any γ > p and all 0 < q .

Note that the boundary of Reifenberg flat domain is locally the graph of a Lipschitz continuous function with a small Lipschitz constant, which ensures the Sobolev embedding theorem and interpolation due to the validity of the Sobolev extension theorem. Let Ω r = B r Ω for r > 0 . The following interpolation inequality involved in the higher gradients on an annulus of the form Ω R \ Ω r ( B R ( x 0 ) \ B r ( x 0 ) ) Ω with 0 < r < R < will be frequently used in our main proof, see [10, Lemma 3] and [9, Chapter 1].

Lemma 2.2

Let Ω be a ( δ , R 0 ) -Reifenberg flat domain as shown in equation (1.6). If g W m , p ( Ω r 2 ) for p 1 , and m 1 , then for any 0 k m 1 , 0 < ε 1 , and 0 < r 1 < r 2 1 , there exists a positive constant C = C ( d , m , p , δ , R 0 ) such that

Ω r 2 \ Ω r 1 D k g p ( r 2 r 1 ) p ( m k ) d x ε Ω r 2 \ Ω r 1 D m g p d x + C ε k m k Ω r 2 \ Ω r 1 g p ( r 2 r 1 ) p m d x .

In the next context, we will intensively use the difference quotients to approximate the weak derivative of the solution. Therefore, it is necessary to recall the following relationship between the difference quotient and the weak derivative, see [24, Theorem 3 in Chapter 5]. For any locally summable function g ( x ) in the bounded domain U and V U , we first introduce the notation of the s -th difference quotient with size h denoted by

(2.1) D s , h g ( x ) = τ s , h g h x V ,

with τ s , h g g ( x + h e s ) g ( x ) for s = 1 , 2 , , d , where we confine 0 < h < dist ( V , U ) to ensure x + h e s U . In the following, we write

(2.2) D h g = ( D 1 , h g , D 2 , h g , , D d , h g ) .

Lemma 2.3

  1. Let g W 1 , p ( U ) with 1 p < . Then, for all V U and all 0 < h < dist ( V , U ) , there holds

    D h g L p ( V ) C D g L p ( U ) ,

    where C = C ( d , p ) is a positive constant.

  2. Let g L p ( V ) with 1 < p < . If there exists a constant C > 0 such that

    D h g L p ( V ) C

    for all 0 < h < dist ( V , U ) , then g W 1 , p ( V ) with

    D g L p ( V ) C .

Next, it is useful to recall the following two basic inequalities regarding general elliptic operator of p -growth, see [52, Lemma 2] or [34, Eq. (3.16)].

Lemma 2.4

Let the coefficients A ( x ) satisfy the uniform elliptic condition (1.2). Then, there exists a positive constant C = C ( d , m , p , λ , Λ ) such that for any 1 < p < ,

A ( x ) ξ , ξ p 2 2 A ( x ) ξ A ( x ) η , η p 2 2 A ( x ) η , ξ η C ( ξ 2 + η 2 ) p 2 2 ξ η 2

for any ξ R m d and η R m d . In particular, if p 2 , then

A ( x ) ξ , ξ p 2 2 A ( x ) ξ A ( x ) η , η p 2 2 A ( x ) η , ξ η λ p 2 ξ η p .

Lemma 2.5

Let the coefficients A ( x ) and B ( x ) satisfy the uniform ellipticity and boundedness (1.2), then for any 1 < p < , ξ = { ξ α R d α = m } R m d and η = { η α R d α = m } R m d , and there hold

α = m A ( x ) ξ , ξ p 2 2 A ( x ) ξ α B ( x ) ξ , ξ p 2 2 B ( x ) ξ α C A ( x ) B ( x ) ξ p 1

and

α = m A ( x ) ξ , ξ p 2 2 A ( x ) ξ α A ( x ) η , η p 2 2 A ( x ) η α C ( ξ + η ) p 2 ξ η ,

where C = C ( d , m , p , λ , Λ ) is positive constant.

It is necessary for the proof of main theorem to recall the following two useful technical lemmas. The first inequality is actually a variant of the classical Hardy’s inequality, whose proof can be found in [6, Lemma 3.4].

Lemma 2.6

Let g be a nonnegative measurable function defined on [ 0 , + ) satisfying

0 g ( β ) d β < ,

then for any α 1 and τ > 0 , we have

0 β τ β g ( μ ) d μ α d β β α τ α 0 β τ ( β g ( β ) ) α d β β .

The second lemma is a useful inequality to our main proof, which really encodes the inclusion property with respect to the second exponent of the Lorentz spaces, see [6, Lemma 3.5].

Lemma 2.7

Let g : [ 0 , + ) [ 0 , + ) be a nonincreasing measurable function, α 1 α 2 , and τ > 0 . Then, if α 2 < , for any ε ( 0 , 1 ] and β 0 , we have

β ( μ τ g ( μ ) ) α 2 d μ μ 1 α 2 ε β τ g ( β ) + C ε α 2 α 1 1 β ( μ τ g ( μ ) ) α 1 d μ μ 1 α 1 ,

with the constant C = C ( α 1 , α 2 , τ ) . If α 2 = , then we have

sup μ > β ( μ τ g ( μ ) ) C β τ g ( β ) + C β ( μ τ g ( μ ) ) α 1 d μ μ 1 α 1 ,

with the constant C = C ( α 1 , τ ) .

Finally, the following iterating lemma is mainly used in our main proof, which is referred to [26].

Lemma 2.8

Let ϕ be a bounded nonnegative function on ( r , R ) . Suppose that for any r 1 , r 2 with 0 < r r 1 < r 2 R < , there holds

ϕ ( r 1 ) ε ϕ ( r 2 ) + A ( r 2 r 1 ) θ + B ,

where the constants A , B 0 , 0 ε < 1 , and θ > 0 . Then, there exists a positive constant C = C ( ε , θ ) such that

ϕ ( r ) C A ( R r ) θ + B .

3 Local interior and boundary estimates

This section is devoted to some auxiliary estimates by comparing the solution with some reference problems. First, we prove a boundary a priori estimate, which is a crucial ingredient in the proof for a higher integrability of problem (1.1). Note that a similar result can be found in [10, Lemma 6] for problem (1.1) in the interior setting. Here, we only focus our L p -estimate on the boundary neighborhood.

Lemma 3.1

(Caccioppoli inequality) Let u W 0 m , p ( Ω ) , for 1 < p < and m 1 , be the weak solution of problem (1.1)with condition (1.2), and let Ω be a ( δ , R 0 ) -Reifenberg flat domain as shown in equation (1.6). If f L p ( Ω ) , then for every fixed point x 0 Ω ¯ and 0 < r < R 1 , there holds

(3.1) Ω r ( x 0 ) D m u p d x C Ω R ( x 0 ) f p d x + Ω R ( x 0 ) u p ( R r ) m p d x ,

where C = C ( d , m , p , λ , Λ , R 0 ) is a positive constant.

Proof

Without loss of generality, we let x 0 = 0 Ω ¯ . For any r and R : 0 < r r 1 < r 2 R 1 , we take η C 0 ( B r 2 ) as a cutting-off function with

(3.2) η = 1 in B r 1 , 0 η 1 in B r 2 ; D k η C ( r 2 r 1 ) k in B r 2 \ B r 1 for all 1 k m .

Now, we define η ¯ η χ Ω , where χ Ω is the characteristic function on Ω . Choosing the test function φ = u η ¯ in equation (1.3), we obtain

(3.3) Ω r 2 A ( x ) D m u , D m u p 2 2 A ( x ) D m u , D m ( u η ¯ ) d x = Ω r 2 f p 2 f , D m ( u η ¯ ) d x .

Leibniz’s rule yields that

D m ( u η ¯ ) = D m u η ¯ + k = 0 m 1 m k D k u D m k η ¯ L O T m ,

where m k = m ! k ! ( m k ) ! . Then, we put it into equation (3.3) to obtain that

(3.4) I Ω r 2 A ( x ) D m u , D m u p 2 2 A ( x ) D m u , D m u η ¯ d x = Ω r 2 A ( x ) D m u , D m u p 2 2 A ( x ) D m u , L O T m d x + Ω r 2 f p 2 f , D m u η ¯ d x + Ω r 2 f p 2 f , L O T m d x I 1 + I 2 + I 3 .

Now we are to estimate I , I 1 , I 2 , and I 3 , respectively.

Estimate of I: using the uniform ellipticity equations (1.2) and (3.2), we obtain

I λ p 2 Ω r 2 D m u p η ¯ d x λ p 2 Ω r 1 D m u p d x .

Estimates of I 1 , I 2 , and I 3 : employing uniform boundedness (1.2) and Young’s inequality, we obtain that

I 1 Λ p 2 Ω r 2 D m u p 1 L O T m d x ε 1 Ω r 2 D m u p d x + C ε 1 1 p Ω r 2 LOT m p d x

for any ε 1 > 0 and some constant C = C ( Λ , p ) > 0 ,

I 2 Ω r 2 f p 1 D m u d x ε 2 Ω r 2 D m u p d x + C ε 2 1 1 p Ω r 2 f p d x

for any ε 2 > 0 and some constant C = C ( p ) > 0 , and

I 3 Ω r 2 f p 1 LOT m d x Ω r 2 LOT m p d x + C Ω r 2 f p d x

for some constant C = C ( p ) > 0 . Putting the estimates I , I 1 , I 2 , and I 3 together into equation (3.4), we obtain

(3.5) Ω r 1 D m u p d x λ p 2 ( ε 1 + ε 2 ) Ω r 2 D m u p d x + C ( 1 + ε 2 1 1 p ) Ω r 2 f p d x + ( 1 + ε 1 1 p ) Ω r 2 LOT m p d x ,

where C = C ( p , λ , Λ ) is a positive constant. To estimate the term LOT m , we note that D k η ¯ = χ Ω D k η in B r 2 for 1 k m . Therefore, we apply the interpolation Lemma 2.2 on Ω r 2 \ Ω r 1 to obtain that

Ω r 2 LOT m p d x C k = 0 m 1 Ω r 2 \ Ω r 1 D k u p ( r 2 r 1 ) ( m k ) p d x C ε 3 Ω r 2 D m u p d x + k = 0 m 1 ε 3 k m k Ω r 2 u p ( r 2 r 1 ) p m d x

for any ε 3 > 0 , where C = C ( m , d , p ) is a positive constant. Therefore, this in combination with equation (3.5) yields

Ω r 1 D m u p d x C ( ε 1 + ε 2 + ( 1 + ε 1 1 p ) ε 3 ) Ω r 2 D m u p d x + C ( 1 + ε 2 1 1 p ) Ω r 2 f p d x + ( 1 + ε 1 1 p ) k = 0 m 1 ε 3 k m k Ω r 2 u p ( r 2 r 1 ) p m d x

for any ε i > 0 , i = 1 , 2 , 3 , where C = C ( d , m , p , λ , Λ ) is a positive constant. Finally, to obtain the estimate (3.1), we only need to use the iterating Lemma 2.8 and take ε i > 0 sufficiently small for each i = 1 , 2 , 3 .□

In what follows, we are to show a higher integrability of the D m u to problem (1.1). It merely suffices to prove a generalized reverse Hölder inequality in the neighborhood of boundary point by using the type- A condition of ( δ , R 0 ) -Reifenberg flat domain as in equation (1.7).

Lemma 3.2

(Reverse Hölder inequality) Assume that u W 0 m , p ( Ω ) for 1 < p < and m 1 is the weak solution of equation (1.1) with equation (1.2), and Ω is a ( δ , R 0 ) -Reifenberg flat domain as shown in equation (1.6). If f L loc ν ( Ω ) for some ν > p , then for every x 0 Ω ¯ and 0 < R 1 , there exists a small positive constant σ 0 = σ 0 ( d , m , p , ν , λ , Λ , Ω ) such that for all 0 < σ σ 0 , we have

(3.6) Ω R 2 ( x 0 ) D m u p ( 1 + σ ) d x C Ω R ( x 0 ) D m u p d x ( 1 + σ ) + Ω R ( x 0 ) f p ( 1 + σ ) d x ,

where C = C ( d , m , p , λ , Λ , δ , R 0 ) is a positive constant.

Proof

We also let x 0 = 0 Ω ¯ . Using Lemma 3.1 with r = R 2 , we have

(3.7) Ω R 2 D m u p d x C Ω R f p d x + R m p Ω R u p d x .

For 1 < p < and 1 q d p d + m p < p , we recall the following Sobolev imbedding inequality

u L p ( Ω R ) C R m + d p d q D m u L q ( Ω R ) ,

see [2, Theorem 5.4 in Chapter V], where C = C ( d , m , p , δ , R 0 ) is a positive constant. Here, we would like to point out that the above Sobolev imbedding inequality also holds in the neighborhood of a Reifenberg boundary point. Now, putting it into equation (3.7) deduces

Ω R 2 D m u p d x C Ω R f p d x + R m p Ω R D m u d p d + m p d x d + m p d ,

which implies that

Ω R 2 D m u p d x C Ω R f p d x + Ω R D m u d p d + m p d x d + m p d .

Then, we attain the desired inequality (3.6) by using the Gehring-Giaquinta-Modica lemma, see Proposition 1.1 in [26, p. 122].□

For B 6 Ω , we let

(3.8) B 6 A ( x ) ( A ) B 6 p d x δ p

and consider the localized problem

(3.9) α = β = m D α A ( x ) D m u , D m u p 2 2 A ( x ) D β u = α = m D α ( f p 2 f α ) in B 6

under the assumption of

(3.10) B 6 D m u p + 1 δ p f p d x 1 .

We are now to give various interior comparison estimates by comparing the local original problem with a few of reference problems. Let h W m , p ( B 5 ) be the weak solution of the following boundary problem:

(3.11) α = β = m D α A ( x ) D m h , D m h p 2 2 A ( x ) D β h = 0 in B 5 , l m 1 D l ( h u ) = 0 on B 5 ,

with u as the solution of equation (3.9). We also consider v W m , p ( B 4 ) as a weak solution of the following limiting problem:

(3.12) α = β = m D α A ¯ 4 D m v , D m v p 2 2 A ¯ 4 D β v = 0 in B 4 , l m 1 D l ( v h ) = 0 on B 4 ,

with h as the weak solution of equation (3.11) and A ¯ 4 being an average of A ( x ) denoted by A ¯ 4 = ( A ) B 4 over B 4 . Equation (3.12) is such a problem with constant coefficients satisfying the uniform ellipticity (1.2). By the standard L p -estimates for problems (3.11) and (3.12), we conclude that for some constant C = C ( d , m , p , λ , Λ ) > 0 , there hold

(3.13) B 4 D m v p d x C B 5 D m h p d x C B 6 D m u p d x + B 6 f p d x .

Unlike the p -Laplacian equation for m = 1 , one cannot obtain the L -bound for the derivative D m v to the higher-order limiting equation of p -growth. Therefore, it is necessary to recall the following interior estimates involved in the equations of p -growth with the constant coefficients as in [29, Lemma 4.1]. Here, let us recall an interior higher-order differential of D m v for higher-order p -Laplacian equations.

Lemma 3.3

Let 1 < p < and m 1 . If v W m , p ( B 4 ) is the weak solution of problem (3.12), then there exists σ 1 = σ 1 ( d , m , p , λ , Λ ) > 0 such that for any 0 < σ σ 1 , we have D D m v p 2 2 D m v L loc 2 ( 1 + σ ) ( B 4 ) with the estimate

B 2 D ( D m v p 2 2 D m v ) 2 ( 1 + σ ) d x C B 3 D m v p d x 1 + σ ,

where C = C ( d , m , p , λ , Λ , σ 1 ) is a positive constant.

In the following lemmas, we set

(3.14) p * = d p d p , if p < d , arbitrary large number > 1 , if p d ,

as the Sobolev conjugate of an arbitrary p > 1 . By the Gagliardo-Nirenberg-Sobolev inequality in [24, Theorem 6] and a higher integrability of D m v shown in Lemma 3.3, it yields the following conclusion.

Lemma 3.4

Let 1 < p < and m 1 . If v W m , p ( B 4 ) is the weak solution of problem (3.12), then there exists σ 1 = σ 1 ( d , m , p , λ , Λ ) > 0 such that for any 0 < σ σ 1 , we have

D m v p 2 L loc ( 2 ( 1 + σ ) ) * ( B 4 ) ;

moreover, there holds the following estimate:

B 2 D m v d p ( 1 + σ ) d 2 ( 1 + σ ) d x C B 3 D m v p d x d ( 1 + σ ) d 2 ( 1 + σ ) ,

where C = C ( d , m , p , λ , Λ , σ 1 ) > 0 .

We are next to give the following perturbation estimate by comparing the solutions between the original problem and the limiting problem (3.12). More precisely, we have

Lemma 3.5

Let u W m , p ( B 6 ) be the weak solution of problem (3.9), and v W m , p ( B 4 ) be the weak solution of problem (3.12). Then, for any ε > 0 , there exists a small δ = δ ( d , m , p , λ , Λ ) > 0 such that the assumptions (3.8)and (3.10)hold; moreover, we have

(3.15) B 4 D m ( u v ) p d x ε .

Proof

Now, we take h u W 0 m , p ( B 5 ) as a test function, then it follows from equations (3.9) and (3.11) that

(3.16) B 5 A ( x ) D m u , D m u p 2 2 A ( x ) D m u A ( x ) D m h , D m h p 2 2 A ( x ) D m h , D m u D m h d x = B 5 f p 2 f , D m u D m h d x .

In the following, we part the estimates of equation (3.16) into two cases of 1 < p < 2 and p 2 .

If 1 < p < 2 , by Lemma 2.4, we derive that

(3.17) B 5 D m ( u h ) p d x = B 5 ( D m u 2 + D m h 2 ) p 2 2 D m ( u h ) 2 p 2 ( D m u 2 + D m h 2 ) ( 2 p ) p 4 d x ε 1 B 5 ( D m u 2 + D m h 2 ) p 2 d x + C ε 1 p 2 p B 5 ( D m u 2 + D m h 2 ) p 2 2 D m ( u h ) 2 d x C ε 1 B 5 ( D m u p + D m h p ) d x + C ε 1 p 2 p B 5 f p 2 f , D m u D m h d x C ( ε 1 + ε 2 ) B 5 ( D m u p + D m h p ) d x + C ε 1 p 2 p 1 ε 2 1 p 1 B 5 f p d x

for any constant ε i > 0 , i = 1 , 2 , where C = C ( d , m , p , λ , Λ ) is a positive constant.

If p 2 , using a similar argument, we readily conclude

(3.18) B 5 D m ( u h ) p d x C B 5 f p 2 f , D m u D m h d x C ε 3 B 5 ( D m u p + D m h p ) d x + C ε 3 1 p 1 B 5 f p d x

for any ε 3 > 0 , where C = C ( d , m , p , λ , Λ ) is a positive constant.

Therefore, putting the two cases into equation (3.16) and using the standard L p -estimate (3.13) and the normalized assumption (3.10), we obtain

(3.19) B 5 D m ( u h ) p d x C ε 1 + ε 2 + ε 3 + ε 1 p 2 p 1 ε 2 1 p 1 + ε 3 1 p 1 δ p

for any δ > 0 and ε i > 0 with i = 1 , 2 , 3 .

On the other hand, we take v h W 0 m , p ( B 4 ) as a test function, and by making a difference to equations (3.11) and (3.12), we have

(3.20) LHS B 4 A ¯ 4 D m h , D m h p 2 2 A ¯ 4 D m h A ¯ 4 D m v , D m v p 2 2 A ¯ 4 D m v , D m h D m v d x = B 4 A ¯ 4 D m h , D m h p 2 2 A ¯ 4 D m h A ( x ) D m h , D m h p 2 2 A ( x ) D m h , D m h D m v d x RHS .

By Lemma 2.4 and Hölder’s inequality, we have

RHS B 4 A ( x ) A ¯ 4 D m h p 1 D m h D m v d x C ε 4 B 4 ( D m h p + D m v p ) d x + C ε 4 1 p 1 B 4 A ( x ) A ¯ 4 p p 1 D m h p d x C ε 4 B 4 ( D m h p + D m v p ) d x + C ε 4 1 p 1 B 4 A ( x ) A ¯ 4 p ( 1 + σ ) ( p 1 ) σ d x σ 1 + σ B 4 D m h p ( 1 + σ ) d x 1 1 + σ ,

with any ε 4 > 0 , and σ > 0 mentioned in Lemma 3.2, where C = C ( p ) is a positive constant. Therefore, using the boundedness of A ( x ) in equation (1.2), the local BMO seminorm with small δ > 0 for equation (3.8), the standard L p estimate (3.13), as well as a higher integrability of the gradients for equation (3.11) shown in Lemma 3.2, we have

RHS C B 4 ε 4 + ε 4 1 p 1 δ p .

For the estimate of LHS of equation (3.20), we use a similar argument as in the proofs of (3.17) and (3.18) to obtain

B 4 D m ( h v ) p d x C ε 5 B 4 ( D m h p + D m v p ) d x + C ε 5 p 2 p LHS

for any ε 5 > 0 , where C = C ( d , m , p , λ , Λ ) is a positive constant.

Let us put the above estimates into equation (3.20), and using the standard L p estimate (3.13) and assumption (3.10), we obtain

(3.21) B 4 D m ( h v ) p d x C ε 5 p 2 p ε 4 + ε 5 + ε 5 p 2 p ε 4 1 p 1 δ p .

Finally, putting equations (3.19) and (3.21) together shows that

B 4 D m ( u v ) p d x C B 5 D m ( u h ) p d x + B 4 D m ( h v ) p d x C i = 1 3 ε i + ε 5 + ε 5 p 2 p ε 4 + ε 1 p 2 p 1 ε 2 1 p 1 + ε 3 1 p 1 + ε 5 p 2 p ε 4 1 p 1 δ p

for any ε i > 0 , i = 1 , , 5 , and δ > 0 , where C = C ( d , m , p , λ , Λ ) is a positive constant. Thus, for any ε > 0 , there exist small constants ε i > 0 , i = 1 , , 5 , and δ > 0 , such that the estimate (3.15) holds. It completes the proof of Lemma 3.5.□

In what follows, we are devoted to the relevant estimates in the neighborhood at a point on the flat boundary. Here, we adapt an argument from Skrypnik’s monograph [46] to estimate the normal derivatives controlled by the horizontal derivatives, which is realized by the difference quotient. To this end, we choose a suitable test function consisting of the difference quotient involved the higher derivatives of the solution u to show the relevant estimates for the limiting problem in a half ball. Let

B r + = B r { x 1 > 0 } and Γ r = B r { x 1 = 0 }

for any r > 0 . We consider the following limiting problem

(3.22) α = β = m D α A ˜ 4 D m v , D m v p 2 2 A ˜ 4 D β v = 0 in B 4 + , l m 1 D l v = 0 on Γ 4 ,

where A ˜ 4 = ( A ) B 4 + is the constant coefficient satisfying the uniform ellipticity (1.2).

Lemma 3.6

Let 1 < p < and m 1 . If v W m , p ( B 4 + ) is the weak solution of problem (3.22), then we have D D m v p 2 2 D m v L loc 2 ( B 4 + ) with the following estimate:

B 2 + D ( D m v p 2 2 D m v ) 2 d x C B 4 + D m v p d x ,

where C = C ( d , m , p , λ , Λ ) is a positive constant.

Proof

We divide the proof into two aspects: one is first to estimate the horizontal derivatives D s ( D m v p 2 2 D m v ) , with s = 2 , , d . Then, we are to estimate the normal derivatives D 1 ( D m v p 2 2 D m v ) in accordance with the horizontal derivatives.

Estimate of the horizontal direction: we select a smooth cut-off function η C 0 ( B 5 2 ) satisfying 0 η 1 , η = 1 on B 2 , η = 0 outside B 5 2 , and

(3.23) D k η C for k = 1 , , m .

Let us first confine a small h in the range 0 < h < 1 2 such that x + h e s B 4 + with s = 2 , d . Then, we consider the s -th difference quotient with size h denoted by D s , h v ( x ) as equation (2.1). For s = 2 , , d , we construct a test function as follows

(3.24) φ = D s , h ( η 2 m D s , h v ) .

We use equation (3.24) as a test function, together with the integration by part for the difference quotients to obtain

(3.25) B 5 2 + D s , h A ˜ 4 D m v , D m v p 2 2 A ˜ 4 D m v , D m ( η 2 m D s , h v ) d x = 0 .

Note that

D s , h A ˜ 4 D m v , D m v p 2 2 A ˜ 4 D m v = 1 h 0 1 d d t A ˜ 4 ( D m v + t D m τ s , h v ) , D m v + t D m τ s , h v p 2 2 A ˜ 4 ( D m v + t D m τ s , h v ) d t = 0 1 D A ˜ 4 ( D m v + t D m τ s , h v ) , D m v + t D m τ s , h v p 2 2 A ˜ 4 ( D m v + t D m τ s , h v ) d t D m D s , h v .

We set

B h ( x ) 0 1 D A ˜ 4 ( D m v + t D m τ s , h v ) , D m v + t D m τ s , h v p 2 2 A ˜ 4 ( D m v + t D m τ s , h v ) d t

and then obtain

(3.26) D s , h A ˜ 4 D m v , D m v p 2 2 A ˜ 4 D m v = B h ( x ) D m D s , h v ,

with

(3.27) B h ( x ) C ( p ) Λ ( D m v 2 + D m τ s , h v 2 ) p 2 2

and

(3.28) α = β = m B h ( x ) ξ α ξ β λ ( D m v 2 + D m τ s , h v 2 ) p 2 2 ξ 2

for ξ = { ξ α R d : α = m } and almost every x B 5 2 + . For equation (3.25), using equation (3.26) and Leibniz’s rule, we obtain that

(3.29) K 1 B 5 2 + B h ( x ) D m D s , h v , η 2 m D m D s , h v d x = B 5 2 + B h ( x ) D m D s , h v , k = 1 m m k D k η 2 m D m k D s , h v d x K 2 .

Estimate of K 1 : by using equation (3.28), we obtain

K 1 λ B 5 2 + η 2 m ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x .

Estimate of K 2 : we distinguish it into the following two cases: 1 < p 2 and p > 2 .

If p > 2 , from equations (3.27) and (3.23) and Cauchy’s inequality, it follows that

K 2 C k = 1 m B 5 2 + η 2 m k B h ( x ) D m D s , h v D m k D s , h v d x C Λ k = 1 m B 5 2 + η 2 m k ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v D m k D s , h v d x ε B 5 2 + η 2 m ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x + C ε 1 k = 1 m B 5 2 + η 2 ( m k ) ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m k D s , h v 2 d x

for any ε > 0 , where C = C ( d , m , p , Λ ) is a positive constant. Subsequently, we put the above two estimates K 1 , K 2 into equation (3.29) and choose ε = λ 2 > 0 to obtain

(3.30) B 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x C k = 1 m B 5 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m k D s , h v 2 d x ,

where C = C ( d , m , p , λ , Λ ) is a positive constant. Using Young’s inequality with two exponents p ¯ p p 2 > 1 and p ¯ p 2 > 1 , it yields

k = 1 m B 5 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m k D s , h v 2 d x C B 5 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 d x + k = 1 m B 5 2 + D m k D s , h v p d x ,

where C = C ( d , m , p , λ , Λ ) is a positive constant. Note that v W m , p ( B 4 + ) with γ m 1 D γ v = 0 on Γ 4 , and by using the standard estimates for the difference quotients as in Lemma 2.3 and the Poincaré’s inequality of ( k 1 ) -order, we have

B 5 2 + D m k D s , h v p d x C B 3 + D m k + 1 v p d x C B 4 + D m v p d x

for each k = 1 , , m , where C = C ( d , m , p ) is a positive constant. A combination of the above three estimates yields

(3.31) B 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x C B 4 + D m v p d x ,

where s = 2 , , d and h > 0 is a small constant.

If 1 < p 2 , then using equation (3.26), the integration by parts for the difference quotients, the ellipticity (1.2) and Young’s inequality with exponents p ¯ p > 1 and p ¯ p p 1 > 1 , we obtain that

K 2 B 5 2 + A ˜ 4 D m v , D m v p 2 2 A ˜ 4 D m v , k = 1 m m k D s , h ( D k η 2 m D m k D s , h v ) d x C k = 1 m B 5 2 + D m v p 1 D s , h ( D k η 2 m D m k D s , h v ) d x C B 5 2 + D m v p d x + k = 1 m B 5 2 + D s , h ( D k η 2 m D m k D s , h v ) p d x .

To estimate the second term in the last inequality above, we use an estimate of the difference quotient shown in Lemma 2.3, Poincaré’s inequality, and Young’s inequality with two exponents p ¯ 2 p > 1 and p ¯ = 2 2 p > 1 to conclude that

k = 1 m B 5 2 + D s , h ( D k η 2 m D m k D s , h v ) p d x C k = 1 m B 3 + D m k + 1 v p d x + B 3 + D m D s , h v p d x C B 3 + D m v p d x + C ε 1 B 3 + ( D m v 2 + D m τ s , h v 2 ) p 2 d x + ε B 3 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x

for any ε > 0 , which implies

K 2 C B 3 + D m v p d x + ε B 3 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x .

Putting the two estimates of K 1 and K 2 into equation (3.29) and using the iterating lemma with ε > 0 small enough, we also give the same estimate as shown equation (3.31) for the case of 1 < p 2 .

Finally, using Lemma 2.5, we see that

(3.32) τ s , h D m v p 2 2 D m v 2 = D m v , D m v p 2 4 D m v ( x + h e s ) D m v , D m v p 2 4 D m v ( x ) 2 C ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m τ s , h v 2 .

Hence, we put the estimates (3.32) and (3.31) together to yield

B 5 2 + D s , h D m v p 2 2 D m v 2 d x C B 5 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x C B 4 + D m v p d x .

This means that the sequence D s , h D m v p 2 2 D m v is uniformly bounded in L 2 ( B 2 + ) . By Lemma 2.3, we obtain

(3.33) B 2 + D s D m v p 2 2 D m v 2 d x C B 4 + D m v p d x

for each s = 2 , , d .

Estimate of the normal direction: it only suffices to prove

(3.34) B 2 + D 1 D m v p 2 2 D m v 2 d x C B 4 + D m v p d x .

For this, we first need a zero-extension of v from B 4 + to B 4 , i.e.,

v ¯ = v in B 4 + , 0 in B 4 \ B 4 + .

Recalling the cut-off function η defined by equation (3.23), we conclude η 2 m D 1 , h v ¯ W 0 m , p ( B 5 2 + ) . We continue to make a zero-extension of η 2 m D 1 , h v ¯ from B 5 2 to R d , still denoted also by η 2 m D 1 , h v ¯ W 0 m , p ( R d ) with the support in B 5 2 + since v ¯ ( x + h e 1 ) moves the boundary Γ 4 upward h -step along e 1 . Next, we construct a test function by confining 0 < h < dist { Γ 4 , Ω B 4 } 2 as follows:

(3.35) φ ( x ) = D 1 , h ( η 2 m ω ( x ) ) with ω ( x ) = ϒ + m χ R + d ϒ + m ( η 2 m D 1 , h v ¯ ( x ) ) ,

where χ R + d is the characteristic function of the half space R + d , and ϒ + ± 1 is a convolution operator with constant symbol in R d , defined by the equality

(3.36) ϒ + ± 1 g = 1 ( i ξ 1 + ξ + 1 ) ± 1 g .

Here, g ( ξ ) = R d g ( x ) e i x ξ d x denotes the Fourier transform of the function g ( x ) . In such way, we employ the same argument as in the proof of Theorem 3.4 from [46, Page 197, line-9], we deduce that the function ω ( x ) vanishes in R d . By considering that the limit of h is small enough that there is still a gap between Γ 4 ( h , 0 ) and Ω B 4 , η 2 m ω ( x ) ( x h e 1 ) moves the boundary Γ 4 downward h -step. This leads to the fact that the support of φ is still in B 5 2 + , which implies φ W 0 m , p ( B 5 2 + ) . Therefore, we can make use of equation (3.35) as a test function in the limiting problem (3.22) to obtain that

B 5 2 + A ˜ 4 D m v , D m v p 2 2 A ˜ 4 D m v , D m φ d x = 0 .

Using the integration by part for the difference quotients and Leibniz’s rule, it yields that

(3.37) B 5 2 + D 1 , h A ˜ 4 D m v ¯ , D m v ¯ p 2 2 A ˜ 4 D m v ¯ , η 2 m D m ω d x = B 5 2 + D 1 , h A ˜ 4 D m v ¯ , D m v ¯ p 2 2 A ˜ 4 D m v ¯ , k = 1 m m k D k η 2 m D m k ω d x .

Let us set

B ˜ h ( x ) 0 1 D A ˜ 4 ( D m v ¯ + t D m τ 1 , h v ¯ ) , D m v ¯ + t D m τ 1 , h v ¯ p 2 2 A ˜ 4 ( D m v ¯ + t D m τ 1 , h v ¯ ) d t

and then, we obtain

(3.38) D 1 , h A ˜ 4 D m v ¯ , D m v ¯ p 2 2 A ˜ 4 D m v ¯ = B ˜ h ( x ) D m D 1 , h v ¯ .

Using the uniform ellipticity and boundedness (1.2), we obtain that

(3.39) B ˜ h ( x ) C ( p ) Λ ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2

and

(3.40) α = β = m B ˜ h ( x ) ξ α ξ β λ ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 ξ 2

for ξ = { ξ α R d : α = m } R m d and almost every x B 5 2 + . According to the definition of ϒ + ± 1 as shown in equation (3.36) and the fact that ( 1 g ) = 1 ( g ) = g for g L p ( R d ) (cf. [27, Chapter 2, Page 104]), we take g = η 2 m D 1 , h v ¯ ( x ) to conclude that

ϒ + 1 ϒ + ( η 2 m D 1 , h v ¯ ( x ) ) = 1 ( i ξ 1 + ξ + 1 ) 1 ( 1 ( i ξ 1 + ξ + 1 ) ( η 2 m D 1 , h v ¯ ( x ) ) ) = 1 ( i ξ 1 + ξ + 1 ) 1 ( i ξ 1 + ξ + 1 ) ( η 2 m D 1 , h v ¯ ( x ) ) = 1 ( η 2 m D 1 , h v ¯ ( x ) ) = η 2 m D 1 , h v ¯ ( x ) ,

which yields that for any x R + d , there holds

ω ( x ) = ϒ + m ϒ + m ( η 2 m D 1 , h v ¯ ( x ) ) = η 2 m D 1 , h v ¯ ( x ) .

Putting it into equation (3.37) and again using Leibniz’s rule, we write

(3.41) J 1 = J 2 ,

where

J 1 B 5 2 + B ˜ h ( x ) D m D 1 , h v ¯ , η 4 m D m D 1 , h v ¯ d x , J 2 B 5 2 + B ˜ h ( x ) D m D 1 , h v ¯ , η 2 m k = 1 m m k D k η 2 m D m k D 1 , h v ¯ d x B 5 2 + B ˜ h ( x ) D m D 1 , h v ¯ , k = 1 m m k D k η 2 m D m k ( η 2 m D 1 , h v ¯ ) d x .

Estimate of J 1 : using equation (3.40), we have

(3.42) J 1 λ B 5 2 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m D 1 , h v ¯ 2 η 4 m d x .

Estimate of J 2 : we estimate it in the following two cases: 1 < p 2 and p > 2 .

If p > 2 , then using equation (3.39), Cauchy’s inequality, and the definition of cut-off function (3.23), we obtain that for any ε > 0 ,

J 2 C k = 1 m B 5 2 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m D 1 , h v ¯ D m k D 1 , h v ¯ η 2 m d x + C k = 1 m B 5 2 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m D 1 , h v ¯ D m k ( η 2 m D 1 , h v ¯ ) η 2 m k d x ε B 5 2 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m D 1 , h v ¯ 2 η 4 m d x + C ε 1 k = 1 m B 5 2 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m k D 1 , h v ¯ 2 d x + k = 1 m B 5 2 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m k ( η 2 m D 1 , h v ¯ ) 2 d x .

Note that v W m , p ( B 4 + ) with γ m 1 D γ v = 0 on Γ 4 . Hence, using the standard estimate of difference quotient as in Lemma 2.3, Young’s inequality with two exponents p ¯ p p 2 > 1 and p ¯ p 2 > 1 , and the Poincaré’s inequality of ( k 1 ) -order, we obtain that

k = 1 m B 5 2 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m k ( η 2 m D 1 , h v ¯ ) 2 d x C k = 1 m B 5 2 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m k D 1 , h v ¯ 2 + i = 1 m k D m k i D 1 , h v ¯ 2 d x C B 3 + D m v p d x + C k = 1 m B 3 + D m k + 1 v p + i = 1 m k D m k i + 1 v p d x C B 4 + D m v p d x .

Putting the above two estimates together, we conclude

(3.43) J 2 ε B 3 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m D 1 , h v ¯ 2 η 4 m d x + C B 4 + D m v p d x .

If 1 < p 2 , then we see that for k 1 , the function ϒ + m χ R + d ϒ + m ( η 2 m k D m k D 1 , h v ¯ ) vanishes in R d . It implies that we can use equation (3.38) and the integration by parts for the difference quotients to obtain

J 2 = B 5 2 + A ˜ 4 D m v , D m v p 2 2 A ˜ 4 D m v , D 1 , h η 2 m k = 1 m m k D k η 2 m D m k D 1 , h v ¯ d x + B 5 2 + A ˜ 4 D m v , D m v p 2 2 A ˜ 4 D m v , D 1 , h k = 1 m m k D k η 2 m D m k ( η 2 m D 1 , h v ¯ ) d x .

Thanks to the ellipticity (1.2), the definition of cut-off function (3.23), and Young’s inequality with exponents p ¯ p > 1 and p ¯ p p 1 > 1 , we obtain that for any ε > 0 ,

J 2 C k = 1 m B 5 2 + D m v p 1 ( D m k D 1 , h D 1 , h v ¯ + D m k D 1 , h v ¯ ) d x + C k = 1 m B 5 2 + D m v p 1 ( D m k D 1 , h ( η 2 m D 1 , h v ¯ ) + D m k ( η 2 m D 1 , h v ¯ ) ) d x C B 5 2 + D m v p d x + C k = 1 m B 5 2 + ( D m k D 1 , h D 1 , h v ¯ p + D m k D 1 , h v ¯ p ) d x + C k = 1 m B 5 2 + ( D m k D 1 , h ( η 2 m D 1 , h v ¯ ) p + D m k ( η 2 m D 1 , h v ¯ ) p ) d x .

To estimate the last two terms for the above inequality, we use the standard estimate of difference quotient as in Lemma 2.3, the Poincaré’s inequality of ( k 1 ) -order, and Young’s inequality with two exponents p ¯ 2 p > 1 and p ¯ = 2 2 p > 1 to obtain that

k = 1 m B 5 2 + ( D m k D 1 , h ( η 2 m D 1 , h v ¯ ) p + D m k ( η 2 m D 1 , h v ¯ ) p ) d x C k = 1 m B 5 2 + ( D m k D 1 , h D 1 , h v ¯ p + D m k D 1 , h v ¯ p ) d x + C k = 1 m B 5 2 + i = 1 m k D m k i D 1 , h D 1 , h v ¯ p + i = 1 m k D m k i D 1 , h v ¯ p d x C B 3 + D m v p d x + C B 3 + D m D 1 , h v ¯ p d x C B 4 + D m v p d x + ε B 3 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m D 1 , h v ¯ 2 η 4 m d x + C ε 1 B 3 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 d x .

Putting the above estimate into the estimate of J 2 , we also give the estimate (3.43) for the case of 1 < p 2 .

Let us now insert the estimates of J 1 and J 2 as shown in equations (3.42) and (3.43) into equation (3.41), then we use the iterating lemma with ε > 0 small enough to obtain

(3.44) B 2 + ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m D 1 , h v ¯ 2 d x C B 4 + D m v p d x .

It is easy to check that there exists a constant C = C ( p , d ) such that

(3.45) τ 1 , h D m v ¯ p 2 2 D m v ¯ 2 C ( D m v ¯ 2 + D m τ 1 , h v ¯ 2 ) p 2 2 D m τ 1 , h v ¯ 2 .

Thus, this estimate together with equation (3.44) yields

B 2 + D 1 , h D m v ¯ p 2 2 D m v ¯ 2 d x C B 4 + D m v p d x .

We see that the sequence D 1 , h D m v p 2 2 D m v is uniformly bounded in L 2 ( B 2 + ) , which implies that D 1 D m v p 2 2 D m v has the same bound as above in L 2 ( B 2 + ) to make equation (3.34) true.

Finally let us put equations (3.33) and (3.34) together, then we obtain the desired estimate as shown in Lemma 3.6, which completes the proof.□

With Lemma 3.6 in hand, we can achieve a higher integrability of D m v by the standard technique of the reverse Hölder inequality.

Lemma 3.7

Let 1 < p < and m 1 . If v W m , p ( B 4 + ) is the weak solution of problem (3.22), then there exists σ 2 = σ 2 ( d , m , p , λ , Λ ) > 0 such that for any 0 < σ σ 2 , we have D D m v p 2 2 D m v L loc 2 ( 1 + σ ) ( B 4 + ) with the estimate

B 2 + D ( D m v p 2 2 D m v ) 2 ( 1 + σ ) d x C B 3 + D m v p d x 1 + σ ,

where C = C ( d , m , p , λ , Λ , σ 2 ) is a positive constant.

Proof

With Lemma 3.6 in hand, it suffices to establish a reverse Hölder’s inequality of D ( D m v p 2 2 D m v ) , which leads to the desired higher integrability result via the famous Gehring’s lemma.

To this end, we are first to estimate the horizontal direction, and test equation (3.22) by the following test function:

φ = D s , h ( η 2 m D s , h ( v P ) ) ,

with a suitable cutting-off function η and the polynomial P ( x ) of degree m , which will be specified later. Here, we confine 0 < h < 1 as a small constant such that x + h e s Ω for s = 2 , , d .

If p > 2 , it follows from the estimate (3.30) and Young’s inequality that

B 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x C B 5 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 d x + k = 1 m B 5 2 + D m k D s , h ( v P ) p d x .

To estimate the second term of the right-hand side above, we use Lemma 2.3 and Poincaré’s inequality of ( k 1 ) -order to obtain

(3.46) k = 1 m B 5 2 + D m k D s , h ( v P ) p d x C B 3 + D m ( v P ) p d x .

Now we choose the highest-order coefficient of polynomial P ( x ) of degree m such that

B 3 + D m ( v P ) p d x = B 3 + D m v p 2 2 D m v D m v p 2 2 D m v B 3 + 2 d x ,

which follows from the Sobolev-Poincaré inequality that

(3.47) B 3 + D m v p 2 2 D m v D m v p 2 2 D m v B 3 + 2 d x B 3 + D D m v p 2 2 D m v 2 d d + 2 d x .

Here, we note that the existence of such polynomial P is a well-known fact, see [29]. Therefore, putting the above four estimates together, we obtain

(3.48) B 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x C B 3 + D m v p d x + B 3 + D D m v p 2 2 D m v 2 d d + 2 d x .

If 1 < p 2 , by the estimates of K 1 and K 2 with equation (3.29) and the Young inequality, we obtain

B 2 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x C B 3 + D m v p d x + k = 1 m B 3 + D m k D s , h ( v P ) p d x + ε B 3 + ( D m v 2 + D m τ s , h v 2 ) p 2 2 D m D s , h v 2 d x .

To estimate the second term of the right-hand side above, we also use a similar argument as equations (3.46) and (3.47) so as to attain equation (3.48) for the case of 1 < p 2 . To estimate the left-hand side of equation (3.48), we use equation (3.32) and the difference quotient estimate of Lemma 2.3 to obtain that for each s = 2 , , d ,

(3.49) B 2 + D s D m v p 2 2 D m v 2 d x C B 3 + D m v p d x + B 3 + D D m v p 2 2 D m v 2 d d + 2 d x .

For the normal direction, we again use the estimate (3.34). Therefore, we put the horizontal estimate (3.49) and the normal estimate (3.34) together to deduce the following reverse Hölder inequality of D D m v p 2 2 D m v :

B 2 + D D m v p 2 2 D m v 2 d x C B 3 + D m v p d x + B 3 + D D m v p 2 2 D m v 2 d d + 2 d x .

Gehring-Giaquinta-Modica lemma (cf. [26, Proposition 1.1]) ensures a higher integrability of D D m v p 2 2 D m v . This combining with Lemma 3.6 leads to the desired estimate, which completes the proof.□

It is easy to check that we make use of the Gagliardo-Nirenberg-Sobolev inequality (cf. [24, Theorem 6]) and Lemma 3.7 to obtain the following conclusion.

Lemma 3.8

Let 1 < p < and m 1 . If v W m , p ( B 4 + ) is the weak solution of problem (3.22), then there exists σ 2 = σ 2 ( d , m , p , λ , Λ ) > 0 such that for any 0 < σ σ 2 , we have

D m v p 2 L loc ( 2 ( 1 + σ ) ) * ( B 4 + ) ,

with the following estimate:

B 2 + D m v d p ( 1 + σ ) d 2 ( 1 + σ ) d x C B 3 + D m v p d x d ( 1 + σ ) d 2 ( 1 + σ ) ,

where C = C ( d , m , p , λ , Λ , σ 2 ) is a positive constant.

In what follows, we note that the boundary of Reifenberg domain Ω is rough, which is not flattened by a smooth mapping. So, it is difficult to use the odd and even extension techniques to handle the boundary estimates. To consider the boundary setting, we set w Ω r = B r Ω for any point x 0 Ω and r > 0 . By a scale argument, we set

(3.50) B 6 + B 6 Ω B 6 { x : x 1 > 12 δ }

and

(3.51) Ω 6 A ( x ) ( A ) Ω 6 p d x δ p .

We consider a localized Dirichlet problem

(3.52) α = β = m D α A ( x ) D m u , D m u p 2 2 A ( x ) D β u = α = m D α ( f p 2 f α ) in Ω 6 , l m 1 D l u = 0 on w Ω 6

under the assumption of

(3.53) Ω 6 D m u p + 1 δ p f p d x 1 .

Let h W m , p ( Ω 5 ) be the weak solution of the following homogeneous equation:

(3.54) α = β = m D α A ( x ) D m h , D m h p 2 2 A ( x ) D β h = 0 in Ω 5 , l m 1 D l ( h u ) = 0 on Ω 5 ,

and let w W m , p ( Ω 4 ) be the weak solution of a limiting problem

(3.55) α = β = m D α ( A ) Ω 4 D m w , D m w p 2 2 ( A ) Ω 4 D β w = 0 in Ω 4 , l m 1 D l ( w h ) = 0 on Ω 4 .

By the standard L p -estimate for the Dirichlet problems (3.54) and (3.55), we have

(3.56) Ω 4 D m w p d x C Ω 5 D m h p d x C Ω 6 D m u p d x + Ω 6 f p d x .

Recalling that v W m , p ( B 4 + ) is the weak solution of problem (3.22), then the following approximation lemma assures that the solutions between w and v are close enough in the L p -sense.

Lemma 3.9

For any ε > 0 , there exists a small δ > 0 such that if w W m , p ( Ω 4 ) is the weak solution of equation (3.55) with boundary assumption (3.50), then there exists a corresponding weak solution v W m , p ( B 4 + ) of equation (3.22) such that

B 4 + w v p d x ε .

Proof

If m = 1 , then the proof of lemma can be found in [11, Lemma 4]. Hence, we focus on the case of m 2 . We argue it by contradiction. If not, then there would exist ε 0 > 0 , { w k } k = 1 , and { Ω 4 k } k = 1 such that w k W m , p ( Ω 4 k ) is the weak solution of

(3.57) α = β = m D α ( A ) Ω 4 k D m w k , D m w k p 2 2 ( A ) Ω 4 k D β w k = 0 in Ω 4 k , l m 1 D l w k = 0 on w Ω 4 k ,

satisfying

(3.58) Ω 4 k D m w k p d x 1

and

(3.59) B 4 + Ω 4 k B 4 x : x 1 > 8 k ,

but

(3.60) B 4 + w k v p d x > ε 0

for any weak solution v W m , p ( B 4 + ) of equation (3.22) satisfying

B 4 + D m v p d x 1 .

We note that the Reifenberg domain is locally A -type one, which means that there exists a positive constant A > 0 such that Ω r A r d . By considering l m 1 D l w k = 0 on w Ω 4 k , we make use of the generalized Poincaré’s inequality of m -order (cf. [37, Theorem 4]) and the assumption (3.58) to obtain w k W m , p ( B 4 + ) C . Therefore, this estimate implies that there exists a subsequence, which still denotes by { w k } , and w W m , p ( B 4 + ) such that

(3.61) w k w in W m , p ( B 4 + ) and D l w k D l w in L p ( B 4 + ) for all l m 1 .

Recalling that A ( x ) satisfies the uniform ellipticity and boundedness (1.2), we obtain that A ˆ k is bounded in Ω . Then, there exists a subsequence, which we denote by { A ˆ k } , such that

(3.62) A ˆ k A in Ω

for some constant coefficient A . Now, we claim that w W m , p ( B 4 + ) is still a weak solution of

(3.63) α = β = m D α A D m w , D m w p 2 2 A D β w = 0 in B 4 + , l m 1 D l w = 0 on Γ 4 .

Indeed, by the trace theorem

D l w k D l w L γ ( Ω 4 k ) D l w k D l w L γ ( Ω 4 k )

for l m 1 , we see from equation (3.61) that D l w k D l w a.e. on Ω 4 . We use the boundary conditions (3.59) and (3.57) to obtain that l m 1 D l w = 0 on Γ 4 immediately. On the other hand, we note that A k satisfies the uniform boundedness as in equation (1.2) so that A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k is uniformly bounded in L p p 1 ( Ω 4 k ) . This leads to that there exists a vector-valued function ξ L p p 1 ( B 4 + ) and a subsequence of A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k , which we still denote by the same form, such that

(3.64) A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k ξ in L p p 1 ( B 4 + ) .

Therefore, to attain our claim (3.63), it suffices only to show that

(3.65) ξ = A D m w , D m w p 2 2 A D m w a.e in B 4 + .

Using Lemma 2.4, we have

Ω 4 k A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k A ˆ k D m l , D m l p 2 2 A ˆ k D m l , D m ( w k l ) ϕ d x C Ω 4 k ( D m w k 2 + D m l 2 ) p 2 2 D m ( w k l ) 2 ϕ d x 0

for every l W m , p ( Ω k ) and every 0 ϕ C 0 ( Ω 4 k ) , or alternatively,

(3.66) Ω 4 k A ˆ k D m w k , D m w k p 2 ϕ d x Ω 4 σ A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k , D m l ϕ d x Ω 4 k A ˆ k D m l , D m l p 2 2 A ˆ k D m l , D m w k ϕ d x + Ω 4 k A ˆ k D m l , D m l p 2 ϕ d x 0 .

Taking w k ϕ as a test function in equation (3.57), we have

Ω 4 k A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k , D m ( w k ϕ ) d x = 0 ,

which implies that

Ω 4 k A ˆ k D m w k , D m w k p 2 ϕ d x = ı = 1 m m ı Ω 4 k A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k , D ı ϕ D m ı w k d x .

Therefore, this combined with equation (3.66) yields

ı = 1 m m ı Ω 4 k A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k , D ı ϕ D m ı w k d x Ω 4 σ A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k , D m l ϕ d x Ω 4 k A ˆ k D m l , D m l p 2 2 A ˆ k D m l , D m w k ϕ d x + Ω 4 k A ˆ k D m l , D m l p 2 ϕ d x 0 .

Using equations (3.59), (3.61), (3.62), and (3.64) and taking the limit k in the above inequality, we obtain

(3.67) ı = 1 m m ı B 4 + ξ , D ı ϕ D m ı w d x B 4 + ξ , D m l ϕ d x B 4 + A ˆ D m l , D m l p 2 2 A ˆ D m l , D m w ϕ d x + B 4 + A ˆ D m l , D m l p 2 ϕ d x 0 .

Moreover, we also choose w ϕ as a test function in equation (3.57),

Ω 4 k A ˆ k D m w k , D m w k p 2 2 A ˆ k D m w k , D m ( w ϕ ) d x = 0 ,

which implies that

B 4 + ξ , D m w ϕ d x = ı = 1 m m ı B 4 + ξ , D ı ϕ D m ı w d x .

Putting the above equality into equation (3.67), we have

B 4 + ξ , D m w ϕ d x B 4 + ξ , D m l ϕ d x B 4 + A ˆ D m l , D m l p 2 2 A ˆ D m l , D m w ϕ d x + B 4 + A ˆ D m l , D m l p 2 ϕ d x 0 .

Therefore, it yields

(3.68) B 4 + ξ A ˆ D m l , D m l p 2 2 A ˆ D m l , D m ( w l ) ϕ d x 0 .

Next, we choose l = w τ g for any τ > 0 , g W 1 , p ( B 4 + ) in the above inequality, to obtain

B 4 + ξ A ˆ D m ( w τ g ) , D m ( w τ g ) p 2 2 A ˆ D m ( w τ g ) , D m g ϕ d x 0 .

Taking τ 0 and using the Lebesgue-dominated convergence theorem, we obtain

B 4 + ξ A ˆ D m w , D m w p 2 2 A ˆ D m w , D m g ϕ d x 0

for every g W 1 , p ( B 4 + ) and every 0 ϕ C 0 ( Ω 4 k ) . The above inequality also holds for g W 1 , p ( B 4 + ) . Therefore, we have

B 4 + ξ A ˆ D m w , D m w p 2 2 A ˆ D m w , D m g ϕ d x = 0

for every g W 1 , p ( B 4 + ) and every 0 ϕ C 0 ( Ω 4 k ) . Therefore, we obtain the desired formula (3.65) to prove the claim (3.63).

Finally, it follows from the weak lower semi-continuity of the semi-norm for D m w k that

(3.69) B 4 + D m w p d x lim inf k Ω 4 k D m w k p d x 1 .

The claim (3.63) combined with equation (3.69) leads to a contradiction to equation (3.60) by taking A = ( A ) B 4 + , v = w , and k large enough. This completes the proof.□

To make a comparison of the solution u with the limiting problem (3.22), we need a zero-extension of v from B 4 + to B 4 . For this, we let

(3.70) v ¯ = v in B 4 + , 0 in B 4 \ B 4 + .

Lemma 3.10

Let u W m , p ( Ω 8 ) be the weak solution of problem (3.52). Then, for any ε > 0 , there exists a small δ > 0 such that the assumptions (3.50), (3.51), and (3.53) hold, then there exists a corresponding weak solution v W m , p ( B 4 + ) of equation (3.22) such that

(3.71) Ω 2 D m ( u v ¯ ) p d x ε

where v ¯ is the extension of v defined by (3.70).

Proof

First, we select a smooth cutting-off function χ defined on R such that

χ ( x 1 ) = 0 for x 1 0 , χ ( x 1 ) = 1 for x 1 4 δ , D k χ C ( 4 δ ) k for k = 0 , 1 , , m .

Recall that w W m , p ( Ω 4 ) is the weak solution of limiting problem (3.55). Let w ˆ = χ w , then w ˆ together with all its derivative vanish on B 4 { x 1 0 } , and w ˆ satisfies

(3.72) α = β = m D α ( A ) Ω 4 D m w ˆ , D m w ˆ p 2 2 ( A ) Ω 4 D β w ˆ = α = β = m D α ( A ) Ω 4 D m w ˆ , D m w ˆ p 2 2 ( A ) Ω 4 D m w , D m w p 2 2 ( A ) Ω 4 D β w ˆ + α = β = m D α ( A ) Ω 4 D m w , D m w p 2 2 ( A ) Ω 4 D β ( ( χ 1 ) w )

in B 4 + . In what follows, we let η C 0 ( B 3 + ) to be cutting-off function with

(3.73) η = 1 in B 2 + , 0 η 1 in B 3 + , D k η C in B 3 + \ B 2 + for all 1 k m ,

and take φ = ( w ˆ v ) η as a test function of equation (3.72). Then, we combine equation (3.22) with equation (3.72) and A ˆ 4 = ( A ) Ω 4 to obtain

(3.74) LHS B 3 + A ˆ 4 D m w ˆ , D m w ˆ p 2 2 A ˆ 4 D m w ˆ A ˆ 4 D m v , D m v p 2 2 A ˆ 4 D m v , D m ( ( w ˆ v ) η ) d x = B 3 + A ˆ 4 D m w ˆ , D m w ˆ p 2 2 A ˆ 4 D m w , D m w p 2 2 A ˆ 4 D m w ˆ , D m ( ( w ˆ v ) η ) d x + B 3 + A ˆ 4 D m w , D m w p 2 2 A ˆ 4 D m ( ( χ 1 ) w ) , D m ( ( w ˆ v ) η ) d x RHS .

We are now in a position to estimate the terms LHS and RHS of equation (3.74), respectively. Using Leibniz’s rule, it yields

D m ( ( w ˆ v ) η ) = η D m ( w ˆ v ) + k = 0 m 1 m k D k ( w ˆ v ) D m k η LOT ¯ m ,

where m k = m ! k ! ( m k ) ! .

To estimate LHS: we first note that

LHS = B 3 + A ˆ 4 D m w ˆ , D m w ˆ p 2 2 A ˆ 4 D m w ˆ A ˆ 4 D m v , D m v p 2 2 A ˆ 4 D m v , η D m ( w ˆ v ) d x + B 3 + A ˆ 4 D m w ˆ , D m w ˆ p 2 2 A ˆ 4 D m w ˆ A ˆ 4 D m v , D m v p 2 2 A ˆ 4 D m v , LOT ¯ m d x .

Using a similar argument to the proofs of equations (3.17) and (3.18), we obtain

B 3 + D m ( w ˆ v ) p η d x ε 1 B 3 + ( D m w ˆ p + D m v p ) d x + C ( ε 1 ) LHS B 3 + A ˆ 4 D m w ˆ , D m w ˆ p 2 2 A ˆ 4 D m w ˆ A ˆ 4 D m v , D m v p 2 2 A ˆ 4 D m v , LOT ¯ m d x

for any ε 1 > 0 . Hence, it follows from Lemma 2.5 and Young’s inequality that

(3.75) B 3 + D m ( w ˆ v ) p η d x ( ε 1 + ε 2 ) B 3 + ( D m w ˆ p + D m v p ) d x + C B 3 + LOT ¯ m p d x + LHS .

To estimate RHS : by using Lemma 2.4, one has

RHS C B 3 + ( D w ˆ + D w ) p 2 D m ( w ˆ w ) D m ( w ˆ v ) d x + C B 3 + ( D w ˆ + D w ) p 2 D m ( w ˆ w ) LOT ¯ m d x + C B 3 + D m w p 2 D m ( ( χ 1 ) w ) D m ( w ˆ v ) d x + C B 3 + D m w p 2 D m ( ( χ 1 ) w ) LOT ¯ m d x .

Employing Young’s inequality and the coefficient assumptions (1.2), we obtain

(3.76) RHS C ε 3 B 3 + ( D m w ˆ p + D m v p ) d x + ε 4 B 3 + D m w p d x + C B 3 + LOT ¯ m p d x + C B 3 + D m ( ( χ 1 ) w ) p d x

for any ε i > 0 with i = 3 , 4 . Consequently, we put the estimates (3.75) and (3.76) with equation (3.74) together to show

(3.77) B 3 + D m ( w ˆ v ) p η d x C i = 1 3 ε i B 3 + ( D m w ˆ p + D m v p ) d x + ε 4 B 3 + D m w p d x + C B 3 + LOT ¯ m p d x + C B 3 + D m ( ( χ 1 ) w ) p d x .

Now, we extend v to zero in B 3 \ B 3 + as presented in equation (3.70) and recall the fact that w ˆ = w + ( χ 1 ) w , which implies that w ˆ = 0 in B 3 \ B 3 + and w ˆ W m , p ( Ω 4 ) . By equation (3.77), it yields

(3.78) Ω 3 D m ( w v ¯ ) p η d x C B 3 + D m ( w ˆ v ) p η d x + C B 3 + D m ( ( χ 1 ) w ) p d x C i = 1 4 ε i B 3 + ( D m w p + D m v p ) d x + C B 3 + LOT ¯ m p d x + C B 3 + D m ( ( χ 1 ) w ) p d x .

To estimate the low-order term involving LOT ¯ m , we make use of the interpolation of Lemma 2.2 to obtain

B 3 + LOT ¯ m p d x C k = 0 m 1 B 3 + \ B 2 + D k ( w ˆ v ) p d x ε 5 B 3 + D m ( w ˆ v ) p d x + C B 3 + w ˆ v p d x

for any ε 5 > 0 . Considering the fact that w ˆ = w + ( χ 1 ) w and Poincaré’s inequality, we have

(3.79) B 3 + LOT ¯ m p d x ε 5 B 3 + D m ( w v ) p d x + C B 3 + w v p d x + C B 3 + D m ( ( χ 1 ) w ) p d x .

For the term Ω 3 D m ( ( χ 1 ) w ) p d x shown in equation (3.78) and (3.79), we note that χ 1 = 0 for x 1 4 δ , D m k ( χ 1 ) = 0 in ( , 0 ) ( 4 δ , + ) and

D m k ( χ 1 ) C ( 4 δ ) m k

in ( 0 , 4 δ ) , where k = 0 , 1 , , m . For any fixed point y = ( y 1 , y ) Ω 2 , let y ˆ 1 = y ˆ 1 ( y ) be the nearest point with ( y ˆ 1 , y ) Ω . Therefore, the weak solution w , as a function of y ˆ 1 , vanishes along with its derivatives up to ( m 1 ) -th order at ( y ˆ 1 , y ) . By Leibniz’s rule and the Hardy-type inequality (cf. [18, Lemma 7.9]), we obtain

Ω 3 D m ( ( χ 1 ) w ) p d x C k = 0 m Ω 3 D m k ( χ 1 ) D k w p d x C k = 0 m Ω 3 { 0 < x 1 < 4 δ } ( 4 δ ) k m D k w p d x C Ω 3 { 0 < x 1 < 4 δ } D m w p d x .

This inequality in combination with the Hölder inequality and the reverse Hölder inequality for the Dirichlet problem (3.55) yields

(3.80) Ω 3 D m ( ( χ 1 ) w ) p d x C Ω 3 { 0 < x 1 < 4 δ } σ 1 + σ Ω 3 D m w p ( 1 + σ ) d x 1 1 + σ C δ σ 1 + σ Ω 4 D m w p d x

for some σ > 0 as in Lemma 3.2. Inserting the estimates (3.79) and (3.80) into (3.78), we obtain

Ω 3 D m ( w v ¯ ) p η d x C i = 1 5 ε i + δ σ 1 + σ Ω 4 ( D m w p + D m v ¯ p ) d x + B 4 + w v p d x

for any ε i > 0 , i = 1 , , 5 and δ > 0 . Subsequently, we use the standard L p estimate (3.56) for the boundary setting and Lemma 3.9 with ε = ε 6 to obtain

Ω 2 D m ( w v ¯ ) p d x C i = 1 6 ε i + δ σ 1 + σ .

On the other hand, for the Dirichlet problems (3.52), (3.54), and (3.55), we use the same argument as the proof of Lemma 3.5 to obtain

Ω 4 D m ( u w ) p d x ε 7 .

Putting the above two estimates together, we obtain that

Ω 2 D m ( u v ¯ ) p d x C Ω 4 D m ( u w ) p d x + Ω 2 D m ( w v ¯ ) p d x C i = 1 7 ε i + δ σ 1 + σ ,

which implies the desired result (3.71) by taking enough small constants ε i > 0 , i = 1 , , 7 , and δ > 0 . It completes the proof.□

4 Proof of main theorem

In this section, we are to prove the global Lorentz estimate of the m -order gradients to the Dirichlet problem (1.1) of higher-order elliptic equations with p -growth. For 1 < p < , we assume that ( A ( x ) , Ω ) satisfies ( δ , R 0 ) -vanishing for some R 0 > 0 shown as Assumption 1.1, where 0 < δ 1 8 is temporarily confined which is determined later. In the following context, we set R ( 0 , R 0 ] and

(4.1) κ 0 = Ω D m u p d x + 1 δ p Ω ( f p + 1 ) η d x 1 η ,

where η > 1 will be specified later. We also take sufficiently large κ satisfying

(4.2) κ κ 1 2 1 δ d Ω B 1 ( y ) 128 R d κ 0

and consider the upper-level set

(4.3) E ( κ ) = { x Ω : D m u p > κ } .

Fix any point y E ( κ ) , we consider a continuous function Φ r ( y ) in r defined by

(4.4) Φ r ( y ) = Ω r ( y ) D m u p d x + 1 δ p Ω r ( y ) f p η d x 1 η

for any r > 0 . The Lebesgue differentiation theorem implies that for almost every y E ( κ ) , it holds

lim r 0 Φ r ( y ) D m u p ( y ) > κ .

On the other hand, if we confine the range: R 128 < r < R 2 , by using equation (4.2), the measure density condition (1.7) near the Reifenberg boundary and the fact of η > 1 , then we obtain that

Φ r ( y ) Ω Ω r ( y ) Ω D m u p d x + 1 δ p Ω ( f p + 1 ) η d x 1 η B r ( y ) Ω B r ( y ) Ω B r ( y ) κ 0 2 1 δ d Ω B 1 ( y ) 128 R d κ 0 κ .

Consequently, we conclude that for almost every y E ( κ ) , there exists a maximum r y = r ( y ) 0 , R 128 such that it holds

Φ r y ( y ) = κ and Φ r ( y ) < κ for any r r y , R 2 .

Then, we infer the following lemma from the well-known Vitali covering lemma due to the property of r y .

Lemma 4.1

Let κ satisfy equation (4.2). Then, there exists a disjoint family { Ω r i ( y i ) } i = 1 with y i E ( κ ) and r i = r y i 0 , R 128 such that

Φ r i ( y i ) = κ and Φ r ( y i ) < κ for a n y r r i , R 2

and

(4.5) E ( κ ) i = 1 Ω 8 r i ( y i ) .

We are now in a position to show a suitable decay estimate to each member Ω r i ( y i ) of the aforementioned covering { Ω r i ( y i ) } i = 1 .

Lemma 4.2

Under the same hypotheses as Lemma 4.1, we have

Ω r i ( y i ) C Ω r i ( y i ) E ( κ 4 ) + 1 ( A κ ) η A κ λ η { x Ω r i ( y i ) : f p > λ } d λ λ ,

where A = δ p 4 , and C = C ( d , m , p , λ , Λ ) is a positive constant.

Proof

By Lemma 4.1, we immediately conclude one of the following alternatives:

Ω r i ( y i ) D m u p d x κ 2 or Ω r i ( y i ) f p η d x κ δ p 2 η .

For the first case, we split it into two parts and use the Hölder’s inequality to obtain that

κ 2 Ω r i ( y i ) Ω r i ( y i ) D m u p d x Ω r i ( y i ) E ( κ 4 ) D m u p d x + Ω r i ( y i ) \ E ( κ 4 ) D m u p d x Ω r i ( y i ) E ( κ 4 ) σ 1 + σ Ω r i ( y i ) E ( κ 4 ) D m u p ( 1 + σ ) d x 1 1 + σ + κ 4 Ω r i ( y i ) ,

which implies

(4.6) κ 4 Ω r i ( y i ) σ 1 + σ Ω r i ( y i ) E ( κ 4 ) σ 1 + σ Ω r i ( y i ) D m u p ( 1 + σ ) d x 1 1 + σ .

Here, the constant 0 < σ σ 0 is determined by Lemma 3.2. Therefore, using Lemma 3.2, we obtain

Ω r i ( y i ) D m u p ( 1 + σ ) d x 1 1 + σ C Ω 2 r i ( y i ) D m u p d x + Ω 2 r i ( y i ) f p ( 1 + σ ) d x 1 1 + σ .

Taking η = 1 + σ and using the covering Lemma 4.1, it yields

Ω r i ( y i ) D m u p ( 1 + σ ) d x 1 1 + σ C κ .

Putting it into equation (4.6) and dividing by κ , we derive

(4.7) Ω r i ( y i ) C Ω r i ( y i ) E ( κ 4 ) .

For the second case, by the classical measure theory, we estimate the upper level set as follows:

κ δ p 2 η Ω r i ( y i ) Ω r i ( y i ) f p η d x = η 0 λ η { x Ω r i ( y i ) : f p > λ } d λ λ ( A κ ) η Ω r i ( y i ) + η A κ λ η { x Ω r i ( y i ) : f p > λ } d λ λ .

By taking A = δ p 4 , it yields

(4.8) Ω r i ( y i ) C ( A κ ) η A κ λ η { x Ω r i ( y i ) : f p > λ } d λ λ .

Putting equations (4.7) and (4.8) together, it completes the proof of Lemma 4.2.□

With Lemma 4.1 in hand, we can construct a disjoint family { Ω r i ( y i ) } i = 1 with y i E ( κ ) and r i 0 , R 128 . Now, by fixing each member Ω r i ( y i ) for i = 1 , 2 , , we consider the following two possible cases: one is for the interior case that B 36 r i ( y i ) Ω , and the other is for the boundary case that B 36 r i ( y i ) Ω .

We first focus on the estimates concerning the interior case for B 36 r i ( y i ) Ω . Since A ( x ) satisfies ( δ , R 0 ) -vanishing for some R 0 > 0 as shown in Assumption 1.1, we suppose that in a new coordinate system ( x 1 , x 2 , , x d ) , the origin is y i and

(4.9) B 36 r i ( y i ) A ( x ) ( A ) B 36 r i ( y i ) p d x δ p .

By Lemma 4.1 and the definition of Φ r ( y ) , we have

(4.10) B 36 r i ( y i ) D m u p d x κ and B 36 r i ( y i ) f p η d x 1 η κ δ p .

Recalling that η > 1 and using the Hölder’s inequality, one deduces that

(4.11) B 36 r i ( y i ) f p d x B 36 r i ( y i ) f p η d x 1 η κ δ p .

For the sake of simplicity, let us define the following scaling functions in B 6 ( y i ) :

(4.12) A ˜ i ( x ) = A ( 6 r i x ) κ p , u ˜ i ( x ) = u ( 6 r i x ) κ p , f ˜ i ( x ) = f ( 6 r i x ) κ p .

Then, it is easy to check that u ˜ i is the weak solution of

(4.13) α = β = m D α A ˜ i ( x ) D m u ˜ i , D m u ˜ i p 2 2 A ˜ i ( x ) D β u ˜ i = α = m D α ( f ˜ i p 2 f i ˜ α ) in B 6 ( y i ) ,

and it follows from equations (4.9)–(4.11) that

(4.14) B 6 ( y i ) A ˜ i ( x ) ( A ˜ i ) B 6 ( y i ) p d x δ p

and

(4.15) B 6 ( y i ) D m u ˜ i p d x + 1 δ p B 6 ( y i ) f ˜ i p d x 1 .

We readily check that the Dirichlet problem (4.13) still satisfies the hypotheses of Lemmas 3.4 and 3.5, which implies that there exist small δ > 0 and function v ˜ i defined in B 4 such that

B 2 ( y i ) D m v ˜ i d p ( 1 + σ ) d 2 ( 1 + σ ) d x N 1 and B 4 ( y i ) D m ( u ˜ i v ˜ i ) p d x ε ,

where N 1 = N 1 ( d , m , p , λ , Λ , σ 1 ) is a position constant, and 0 < σ σ 1 as in Lemma 3.4. Scaling these functions back and denoting v i by the translated function of

v ˜ i ( x ) = v ( 6 r i x ) κ p ,

we conclude that

(4.16) B 12 r i ( y i ) D m v d p ( 1 + σ ) d 2 ( 1 + σ ) d x N 1 κ d ( 1 + σ ) d 2 ( 1 + σ ) and B 24 r i ( y i ) D m ( u v ) p d x κ ε .

We now consider the estimates in the boundary case for B 36 r i ( y i ) Ω . This implies that dist { y i , Ω } = y i y 0 36 r i for y 0 Ω . We recall the geometry of Reifenberg flat domain, which shows that for every point y 0 on the boundary of Ω , there exists a coordinate system { x 1 , , x d } such that

B 64 r i ( y 0 ) { x 1 > 64 r i δ } B 64 r i ( y 0 ) Ω B 64 r i ( y 0 ) { x 1 > 64 r i δ } .

Let us confine 0 < δ 1 8 as equation (4.1), which yields B 56 r i ( y 0 + 64 r i δ e 1 ) B 64 r i ( y 0 ) , where e 1 = ( 1 , 0 , , 0 ) . We now translate the spatial coordinate system along the x 1 -direction by 64 r i δ , which implies that in the new coordinate system, the origin is 0 = y 0 + 64 r i δ e 1 . Without loss of generality, we still denote x = { x 1 , x 2 , , x d } -coordinate in this new coordinate system with

(4.17) B 56 r i + Ω 56 r i B 56 r i { x 1 > 128 r i δ }

and

(4.18) B 56 r i A ( x ) ( A ) B 56 r i p d x δ p ,

where we make A ( x ) to be zero extension from B 56 r i Ω to B 56 r i . Moreover, we also remark that y i y i y 0 + y 0 44 r i , which yields

(4.19) Ω 12 r i ( y i ) Ω 56 r i Ω 70 r i Ω 120 r i ( y i ) .

From Lemma 4.1, the definition of Φ r ( y ) and the fact η > 1 , it follows that

(4.20) Ω 70 r i D m u p d x Ω 120 r i ( y i ) Ω 70 r i Ω 120 r i ( y i ) D m u p d x 2 d κ

and

(4.21) Ω 70 r i f p d x Ω 120 r i ( y i ) Ω 70 r i Ω 120 r i ( y i ) f p η d x 1 η 2 d κ δ p .

Let us make the scaling functions on Ω 35 again as follows:

(4.22) A ˜ i ( x ) = A ( 2 r i x ) 2 d κ p , u ˜ i ( x ) = u ( 2 r i x ) 2 d κ p , f ˜ i ( x ) = f ( 2 r i x ) 2 d κ p .

Then, u ˜ i is the weak solution of

(4.23) α = β = m D α A ˜ i ( x ) D m u ˜ i , D m u ˜ i p 2 2 A ˜ i ( x ) D β u ˜ i = α = m D α ( f ˜ i p 2 f i ˜ α ) in Ω 35 , γ m 1 D γ u ˜ i = 0 on w Ω 35 ,

and by equations (4.18)–(4.21), we conclude

(4.24) B 35 A ˜ i ( x ) ( A ˜ i ) B 35 p d x δ p

and

(4.25) Ω 35 D m u ˜ i p d x + 1 δ p Ω 35 f ˜ i p d x 1 .

In similar way, we also check that the Dirichlet problem (4.23) still satisfies the hypotheses of Lemmas 3.8 and 3.10, which implies that there exist small δ > 0 and function v ˜ i defined in B 32 + such that

Ω 28 D m v ˜ ¯ i d p ( 1 + σ ) d 2 ( 1 + σ ) d x N 2 and Ω 30 D m ( u ˜ i v ˜ ¯ i ) p d x ε ,

where N 2 = N 2 ( d , m , p , λ , Λ , σ 1 ) is a position constant, and 0 < σ σ 2 as in Lemma 3.8. Here, we made an extension of v ˜ i by zero from B 32 + to B 32 and denote it by v ˜ ¯ i . Scaling these functions back and denoting v i by the translated function of

v ˜ i ( x ) = v ( 2 r i x ) 2 d κ p ,

we conclude that

(4.26) Ω 56 r i D m v ¯ d p ( 1 + σ ) d 2 ( 1 + σ ) d x N 3 κ d ( 1 + σ ) d 2 ( 1 + σ ) and Ω 60 r i D m ( u v ¯ ) p d x κ ε .

Furthermore, using the expression of domain relation (4.19), we obtain

(4.27) Ω 12 r i ( y i ) D m v ¯ d p ( 1 + σ ) d 2 ( 1 + σ ) d x N 4 κ d ( 1 + σ ) d 2 ( 1 + σ ) and Ω 16 r i ( y i ) D m ( u v ¯ ) p d x κ ε .

In short, combining the interior estimates (4.16) with the boundary estimates (4.27), we write N = max { N 1 , N 4 , 1 } , which is independent of the index i . In the following, we write

(4.28) T = 2 p N > 1 and B = 2 1 δ d Ω B 1 ( y ) 128 R d .

Lemma 4.3

Let κ B κ 0 be a fixed number satisfying equation (4.2). Assume that u W 0 m , p ( Ω ) is the weak solution to the Dirichlet problem (1.1) with (1.2). Then, for any ε > 0 there exists a small constant δ = δ ( d , m , p , λ , Λ , Ω ) > 0 such that if ( A ( x ) , Ω ) satisfies ( δ , R 0 ) -vanishing for some R 0 > 0 as shown in Assumption 1.1, then we have

(4.29) E ( ϑ T κ ) C ε ϑ 1 + ϑ d ( 1 + σ ) d 2 ( 1 + σ ) E ( κ 4 ) + 1 ( A κ ) η A κ λ η { x Ω : f p > λ } d λ λ ,

where the parameter ϑ = ϑ ( d , m , p , λ , Λ ) > 1 is determined later, and C = C ( d , m , p , λ , Λ , R 0 , Ω ) is a positive constant.

Proof

By Lemma 4.1 and the fact that E ( ϑ T κ ) E ( κ ) for ϑ > 1 and T > 1 as in equation (4.28), we obtain that { Ω 8 r i ( y i ) } i = 1 cover almost all E ( ϑ T κ ) . This deduces that

(4.30) E ( ϑ T κ ) = { x Ω : D m u p > ϑ T κ } i = 1 { x Ω 8 r i ( y i ) : D m u p > ϑ T κ } i : interior case { x Ω 8 r i ( y i ) : D m u p > ϑ T κ } + i : boundary case { x Ω 8 r i ( y i ) : D m u p > ϑ T κ } .

For the interior setting: we first note that d ( 1 + σ ) d 2 ( 1 + σ ) > 1 for some 0 < σ σ 1 mentioned as in Lemma 3.4. Using the facts that D m u p 2 p 1 ( D m ( u v ) p + D m v p ) , Ω 8 r i ( y i ) = B 8 r i ( y i ) , and the interior estimates (4.16), we find that

(4.31) { x B 8 r i ( y i ) : D m u p > ϑ T κ } { x B 8 r i ( y i ) : D m ( u v ) p > N 1 ϑ κ } + { x B 8 r i ( y i ) : D m v p > N 1 ϑ κ } 1 N 1 ϑ κ B 24 r i ( y i ) D m ( u v ) p d x + 1 ( N 1 ϑ κ ) d ( 1 + σ ) d 2 ( 1 + σ ) B 12 r i ( y i ) D m v ¯ d p ( 1 + σ ) d 2 ( 1 + σ ) d x C ε ϑ 1 + ϑ d ( 1 + σ ) d 2 ( 1 + σ ) B r i ( y i ) ,

where C = C ( d , m , p , λ , Λ ) is a positive constant.

For the boundary setting: we carry out the same procedure as in equation (4.31) with the boundary estimates (4.27) instead of the interior estimates (4.16) to discover that

(4.32) { x Ω 8 r i ( y i ) : D m u p > ϑ T κ } C ε ϑ 1 + ϑ d ( 1 + σ ) d 2 ( 1 + σ ) Ω r i ( y i ) .

Inserting equations (4.31) and (4.32) into equation (4.30), we obtain

E ( ϑ T κ ) C ε ϑ 1 + ϑ d ( 1 + σ ) d 2 ( 1 + σ ) i = 1 Ω r i ( y i ) .

Furthermore, using Lemma 4.2 yields

E ( ϑ T κ ) C ε ϑ 1 + ϑ d ( 1 + σ ) d 2 ( 1 + σ ) i = 1 Ω r i ( y i ) E ( κ 4 ) + 1 ( A κ ) η A κ λ η { x Ω r i ( y i ) : f p > λ } d λ λ

for all κ B κ 0 and A = δ p 4 . Note that { Ω r i ( y i ) } i = 1 are nonoverlapping, which leads to the required result.□

Now we are ready to give the proof of Theorem 1.3.

Proof of Theorem 1.3

Case 1: q . Note that

D m u p L γ , q ( Ω ) q = γ 0 ( κ γ { x Ω : D m u p > κ } ) q γ d κ κ .

Recalling the definition of the upper-level set E ( κ ) as equation (4.3) and using the change of variables with ϑ > 1 and T > 1 as in equation (4.28), we obtain

D m u p L γ , q ( Ω ) q = γ ( ϑ T ) q 0 ( κ γ E ( ϑ T κ ) ) q γ d κ κ .

Let κ 1 > 0 be a positive constant in equation (4.2), then we have

(4.33) D m u p L γ , q ( Ω ) q = γ ( ϑ T ) q 0 κ 1 ( κ γ E ( ϑ T κ ) ) q γ d κ κ I 1 + γ ( ϑ T ) q κ 1 ( κ γ E ( ϑ T κ ) ) q γ d κ κ I 2 .

To estimate I 1 : we first recall the definition of κ 1 in equation (4.2) and obtain that

I 1 γ q ( ϑ T ) q Ω q γ κ 1 q C ϑ q Ω q γ Ω D m u p d x + 1 δ p Ω ( f p + 1 ) η d x 1 η q .

This combined with the standard L p estimate (1.4) and the Hölder inequality yields

I 1 C ϑ q Ω q ( 1 γ 1 ) Ω ( f p + 1 ) η d x q η .

Recalling η = 1 + σ with 0 < σ σ 0 as in the proof of Lemma 4.2 and taking σ sufficiently small such that 1 < η = 1 + σ < γ , by Theorem 2.1, we conclude that there exists a positive constant C = C ( d , m , p , γ , q , λ , Λ , δ , R 0 , Ω ) such that

(4.34) I 1 C ϑ q ( f p L γ , q ( Ω ) q + 1 ) .

To estimate I 2 : using the power decay estimate of upper-level set in Lemma 4.3, we obtain that

(4.35) I 2 C ϑ q ε ϑ 1 + ϑ d ( 1 + σ ) d 2 ( 1 + σ ) q γ κ 1 κ γ E ( κ 4 ) + 1 ( A κ ) η A κ λ η { x Ω : f p > λ } d λ λ q γ d κ κ C ε q γ ϑ q γ ( γ 1 ) + ϑ q γ ( γ d ( 1 + σ ) d 2 ( 1 + σ ) ) 0 ( κ γ E ( κ 4 ) ) q γ d κ κ + 0 κ γ 1 ( A κ ) η A κ λ η { x Ω : f p > λ } d λ λ q γ d κ κ C ε q γ ϑ q γ ( γ 1 ) + ϑ q γ ( γ d ( 1 + σ ) d 2 ( 1 + σ ) ) ( I 21 + I 22 ) .

For the estimate I 21 : a simple change of variable shows

I 21 C D m u p L γ , q ( Ω ) q .

For the estimate I 22 : we consider following two cases:

If q γ > 1 , then we use Hardy’s inequality shown as Lemma 2.6 to obtain

I 22 C 0 ( κ γ { x Ω : f p > A κ } ) q γ d κ κ C f p L γ , q ( Ω ) q .

If 0 < q < γ , then we use the reverse-Hölder inequality shown as Lemma 2.7 to obtain

A κ λ η { x Ω : f p > λ } d λ λ q γ C ( ( A κ ) η { x Ω : f p > A κ } ) q γ + C A κ ( λ η { x Ω : f p > λ } ) q γ d λ λ .

Again using a simple change of variable and Hardy’s inequality as Lemma 2.6, we obtain

I 22 C 0 ( κ γ { x Ω : f p > A κ } ) q γ d κ κ + 0 κ γ ( A κ ) η q γ A κ ( λ η { x Ω : f p > λ } ) q γ d λ λ d κ κ C 0 ( κ γ { x Ω : f p > A κ } ) q γ d κ κ C f p L γ , q ( Ω ) q .

Putting above two estimates I 21 and I 22 into equation (4.35), we obtain that

I 2 C ε q γ ϑ q γ ( γ 1 ) + ϑ q γ ( γ d ( 1 + σ ) d 2 ( 1 + σ ) ) ( D m u p L γ , q ( Ω ) q + f p L γ , q ( Ω ) q ) .

Combining the estimates I 1 and I 2 with equation (4.33) yields

(4.36) D m u p L γ , q ( Ω ) q C ε q γ ϑ q γ ( γ 1 ) + ϑ q γ ( γ d ( 1 + σ ) d 2 ( 1 + σ ) ) ( D m u p L γ , q ( Ω ) q + f p L γ , q ( Ω ) q ) + C ϑ q ( f p L γ , q ( Ω ) q + 1 ) .

Since d t d t is monotonically increasing in t , we can find ε 1 = ε 1 ( d , m , p , λ , Λ ) > 0 such that

d ( 1 + σ ) d 2 ( 1 + σ ) = d d 2 ( 1 + ε 1 ) .

Thus, taking ε = d ε 1 d 2 and using the fact that 1 < γ < d d 2 + ε , we obtain

γ d ( 1 + σ ) d 2 ( 1 + σ ) < 0 .

Furthermore, we take ϑ > 1 sufficiently large and ε > 0 sufficiently small such that

C ε q γ ϑ q γ ( γ 1 ) + ϑ q γ ( γ d ( 1 + σ ) d 2 ( 1 + σ ) ) 1 2 .

Therefore, it follows from equation (4.36) that

(4.37) D m u p L γ , q ( Ω ) C ( f p L γ , q ( Ω ) + 1 ) .

Case 2: q = . Note that

D m u p L γ , ( Ω ) = sup κ > 0 ( κ γ { x Ω : D m u p > κ } ) 1 γ .

Also, using the definition of the upper-level set E ( κ ) in equation (4.3) and the change of variables with ϑ > 1 and T > 1 in equation (4.28), we obtain

D m u p L γ , ( Ω ) = ϑ T sup κ > 0 ( κ γ E ( ϑ T κ ) ) 1 γ .

Let κ 1 > 0 be a positive constant as in equation (4.2), we have

(4.38) D m u p L γ , ( Ω ) ϑ T sup 0 < κ < κ 1 ( κ γ E ( ϑ T κ ) ) 1 γ I 3 + ϑ T sup κ κ 1 ( κ γ E ( ϑ T κ ) ) 1 γ I 4 .

To estimate I 3 , by the definition of κ 1 in equation (4.2), we obtain that

I 3 ϑ T Ω 1 γ κ 1 C ϑ Ω 1 γ Ω D m u p d x + 1 δ p Ω ( f p + 1 ) η d x 1 η .

This in combination with the standard L p estimate (1.4) and the Hölder inequality yields

I 3 C ϑ Ω ( 1 γ 1 ) Ω ( f p + 1 ) η d x 1 η .

Recalling that η = 1 + σ with 0 < σ σ 0 as in the proof of Lemma 4.2 and taking σ sufficiently small such that 1 < η = 1 + σ < γ . Thanks to Theorem 2.1, there exists a positive constant C = C ( d , m , p , γ , q , λ , Λ , δ , R 0 , Ω ) such that

I 3 C ϑ ( f p L γ , ( Ω ) + 1 ) .

To estimate I 4 : using the power decay estimate of upper-level set as in Lemma 4.3, we deduce that

(4.39) I 4 C ϑ ε ϑ 1 + ϑ d ( 1 + σ ) d 2 ( 1 + σ ) 1 γ sup κ κ 1 κ γ E ( κ 4 ) + 1 ( A κ ) η A κ λ η { x Ω : f p > λ } d λ λ 1 γ C ε 1 γ ϑ 1 γ ( γ 1 ) + ϑ 1 γ ( γ d ( 1 + σ ) d 2 ( 1 + σ ) ) sup κ 0 ( κ γ E ( κ 4 ) ) 1 γ + sup κ 0 κ γ ( A κ ) η A κ λ η { x Ω : f p > λ } d λ λ 1 γ C ε 1 γ ϑ 1 γ ( γ 1 ) + ϑ 1 γ ( γ d ( 1 + σ ) d 2 ( 1 + σ ) ) ( I 41 + I 42 ) .

For the estimate I 41 : a simple change of variable shows

I 41 C D m u p L γ , ( Ω ) .

For the estimate I 42 : we use the relation 1 < η < γ to show that

I 42 = sup κ 0 κ γ ( A κ ) η A κ λ η γ ( λ γ { x Ω : f p > λ } ) d λ λ 1 γ f p L γ , ( Ω ) sup κ 0 κ γ ( A κ ) η A κ λ η γ d λ λ 1 γ = 1 ( γ η ) 1 γ A f p L γ , ( Ω ) .

Putting I 41 and I 42 into equation (4.39) deduces

I 4 C ε 1 γ ϑ 1 γ ( γ 1 ) + ϑ 1 γ ( γ d ( 1 + σ ) d 2 ( 1 + σ ) ) ( D m u p L γ , ( Ω ) + f p L γ , ( Ω ) ) .

Combining the estimates I 3 and I 4 with equation (4.38), we obtain

D m u p L γ , ( Ω ) C ε 1 γ ϑ 1 γ ( γ 1 ) + ϑ 1 γ ( γ d ( 1 + σ ) d 2 ( 1 + σ ) ) ( D m u p L γ , ( Ω ) + f p L γ , ( Ω ) ) + C ϑ ( f p L γ , ( Ω ) + 1 ) .

Once we obtain the above estimation, the remainder is similar to that of the case q . This completes the proof of Theorem 1.3.□

Acknowledgements

The authors would like to thank the anonymous referees for their valuable comments and suggestions that help to improve and clarify the article greatly.

  1. Funding information: H. Tian was supported by the National Natural Science Foundation of China (Grant No. 11901429), and S. Zheng was supported by the National Natural Science Foundation of China (Grant Nos 12071021 and 12271021).

  2. Author contributions: Both authors contributed equally and significantly to this article.

  3. Conflict of interest: The authors state that there is no conflict of interest.

  4. Data availability statement: Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.

References

[1] E. Acerbi and G. Mingione, Gradient estimates for a class of parabolic system, Duke Math. J. 136 (2007), no. 2, 285–320. 10.1215/S0012-7094-07-13623-8Search in Google Scholar

[2] R. A. Adams, Sobolev Spaces, Academic, New York, 1975. Search in Google Scholar

[3] K. Adimurthil and N. C. Phuc, Global Lorentz and Lorentz-Morrey estimates below the natural exponent for quasilinear equations, Calc. Var. Partial Differ. Equ. 54 (2015), 3107–3139. 10.1007/s00526-015-0895-1Search in Google Scholar

[4] A. Alberico, I. Chlebicka, A. Cianchi, and A. Zatorska-Goldstein, Fully anisotropic elliptic problems with minimally integrable data, Calc. Var. Partial Differ. Equ. 58 (2019), 186. 10.1007/s00526-019-1627-8Search in Google Scholar

[5] A. Alvino, V. Ferone, and G. Trombetti, Estimates for the gradient of solutions of nonlinear elliptic equations with L1 data, Ann. Mat. Pura Appl. 178 (2000), 129–142. 10.1007/BF02505892Search in Google Scholar

[6] P. Baroni, Lorentz estimates for degenerate and singular evolutionary systems, J. Differential Equations 255 (2013), 2927–2951. 10.1016/j.jde.2013.07.024Search in Google Scholar

[7] P. Baroni, Lorentz estimates for obstacle parabolic problems, Nonlinear Anal. 96 (2014), 167–188. 10.1016/j.na.2013.11.004Search in Google Scholar

[8] P. Baroni and J. Habermann, New gradient estimates for parabolic equations, Houston J. Math. 38 (2012), no. 3, 855–914. Search in Google Scholar

[9] A. Bensoussan and J. Frehse, Regularity Results for Nonlinear Elliptic Systems and Applications, Springer-Verlag, Berlin, 2002. 10.1007/978-3-662-12905-0Search in Google Scholar

[10] V. Bögelein, Higher integrability for weak solutions of higher-order degenerate parabolic systems, Ann. Acad. Sci. Fenn. Math. 33 (2008), 387–412. Search in Google Scholar

[11] S. S. Byun, L. H. Wang, and S. Zhou, Nonlinear elliptic equations with BMO coefficients in Reifenberg domains, J. Funct. Anal. 250 (2007), no. 1, 167–196. 10.1016/j.jfa.2007.04.021Search in Google Scholar

[12] S. S. Byun and S. J. Ryu, Gradient estimates for higher-order elliptic equations on nonsmooth domains, J. Differential Equations 250 (2011), 243–263. 10.1016/j.jde.2010.10.001Search in Google Scholar

[13] S. Campanato and P. Cannarsa, Differentiability and partial Hölder continuity of the solutions of nonlinear elliptic systems of order 2m with quadratic growth, Ann. Scuola Norm. Sup. Pisa 8 (1981), 285–309. Search in Google Scholar

[14] S. Campanato, Differentiability of the solutions of non-linear elliptic systems with natural growth, Ann. Mat. Pura Appl. 131 (1982), 75–106. 10.1007/BF01765147Search in Google Scholar

[15] L. A. Caffarelli and I. Peral, On W1,p estimates for elliptic equations in divergence form, Comm. Pure Appl. Math. 51 (1998), 1–21. 10.1002/(SICI)1097-0312(199801)51:1<1::AID-CPA1>3.3.CO;2-NSearch in Google Scholar

[16] J. Choi and D. Kim, Weighted Lp,q-estimates for higher-order elliptic and parabolic systems with BMOx coefficients on Reifenberg flat domains, Calc. Var. Partial Differential Equations 58 (2019), no. 3, Art. 90. 10.1007/s00526-019-1537-9Search in Google Scholar

[17] H. J. Dong and D. Kim, On the Lp-solvability of higher-order parabolic and elliptic systems with BMO coefficients, Arch. Ration. Mech. Anal. 199 (2011), no. 3, 889–941. 10.1007/s00205-010-0345-3Search in Google Scholar

[18] H. J. Dong and D. Kim, Higher order elliptic and parabolic systems with variably partially BMO coefficients in regular and irregular domains, J. Funct. Anal. 261 (2011), 3279–3327. 10.1016/j.jfa.2011.08.001Search in Google Scholar

[19] H. J. Dong and D. Kim, The conormal derivative problem for higher-order elliptic systems with irregular coefficients, Recent advances in harmonic analysis and partial differential equations, Vol. 581, Contemporary Mathematics, 2012, pp. 69–97. 10.1090/conm/581/11534Search in Google Scholar

[20] H. J. Dong and C. Gallarati, Higher-order elliptic and parabolic equations with VMO assumptions and general boundary conditions, J. Funct. Anal. 274 (2018), no. 7, 1993–2038. 10.1016/j.jfa.2018.01.011Search in Google Scholar

[21] E. DiBenedetto and J. Manfredi, On the higher integrability of the gradient of weak solutions of certain degenerate elliptic systems, Amer. J. Math. 115 (1993), no. 5, 1107–1134. 10.2307/2375066Search in Google Scholar

[22] F. Duzaar, A. Gastel, and J. F. Grotowski, Optimal partial regularity for nonlinear elliptic systems of higher-order, J. Math. Sci. Univ. Tokyo 8 (2001), 463–499. Search in Google Scholar

[23] M. Eleuteri and J. Habermann, Calderón-Zygmund type estimates for a class of obstacle problems with p(x)-growth, J. Math. Anal. Appl. 372 (2010), no. 1, 140–161. 10.1016/j.jmaa.2010.05.072Search in Google Scholar

[24] L. C. Evans, Partial Differential Equations, Vol. 19, American Mathematical Society, Providence, 1997. Search in Google Scholar

[25] M. Giaquinta and G. Modica, Regularity results for some classes of higher-order nonlinear elliptic systems, J. Reine Angew. Math. 311/312 (1979), 145–169. 10.1515/crll.1979.311-312.145Search in Google Scholar

[26] M. Giaquinta, Multiple Integrals in the Calculus of Variations and Nonlinear Elliptic Systems, vol. 105, Princeton University Press, Princeton, 1983. 10.1515/9781400881628Search in Google Scholar

[27] L. Grafakos, Classical Fourier Analysis, Graduate Texts in Mathematics, vol. 249, Springer-Verlag, New York, 2008. 10.1007/978-0-387-09432-8Search in Google Scholar

[28] J. Habermann, Regularity Results for Functionals and Calderón-Zygmund Estimates for Systems of Higher Order with p(x) growth, Doctoral Thesis, 2006. Search in Google Scholar

[29] J. Habermann, Calderón-Zygmund estimates for higher-order systems with p(x)-growth, Math. Z. 258 (2008), no. 2, 427–462. 10.1007/s00209-007-0180-xSearch in Google Scholar

[30] J. Habermann, Partial regularity for minima of higher-order functionals with p(x)-growth, Manuscripta Math. 126 (2008), no. 1, 1–4010.1007/s00229-007-0147-6Search in Google Scholar

[31] K. Ho and I. Sim, On sufficient “local” conditions for existence results to generalized p(⋅)-Laplace equations involving critical growth, Adv. Nonlinear Anal. 12 (2023), 182–209. 10.1515/anona-2022-0269Search in Google Scholar

[32] T. Iwaniec, Projections onto gradient fields and Lp-estimates for degenerated elliptic operators, Studia Math. 75 (1983), no. 3, 293–312. 10.4064/sm-75-3-293-312Search in Google Scholar

[33] D. Jerison and C. Kenig, The inhomogeneous Dirichlet problem in Lipschitz domains, J. Funct. Anal. 130 (1995), 161–219. 10.1006/jfan.1995.1067Search in Google Scholar

[34] J. Kinnunen and S. L. Zhou, A local estimate for nonlinear equations with discontinuous coefficients, Comm. Partial Differential Equations 24 (1999), 2043–2068. 10.1080/03605309908821494Search in Google Scholar

[35] J. Kinnunen and S. L. Zhou, A boundary estimate for nonlinear equations with discontinuous coefficients, Differential Integral Equations 14 (2001), 475–492. 10.57262/die/1356123316Search in Google Scholar

[36] S. Leonardi and N. S. Papageorgiou, Positive solutions for a class of singular (p,q)-equations, Adv. Nonlinear Anal. 12 (2023), 20220300. 10.1515/anona-2022-0300Search in Google Scholar

[37] G. M. Lieberman, A mostly elementary proof of Morrey space estimates for elliptic and parabolic equations with VMO coefficients, J. Funct. Anal. 201 (2003), no. 2, 457–479. 10.1016/S0022-1236(03)00125-3Search in Google Scholar

[38] V. Maz’ya, The Wiener test for higher-order elliptic equations, Duke Math. J. 115 (2002), no. 3, 479–512. 10.1215/S0012-7094-02-11533-6Search in Google Scholar

[39] N. G. Meyers, An Lp-estimate for the gradient of solutions of second-order elliptic divergence equations, Ann. Sc. Norm. Super. Pisa Cl. Sci. 17 (1963), no. 3, 189–206. Search in Google Scholar

[40] T. Mengesha and N. C. Phuc, Global estimates for quasilinear elliptic equations on Reifenberg flat domains, Arch. Ration. Mech. Anal. 203 (2012), 189–216. 10.1007/s00205-011-0446-7Search in Google Scholar

[41] G. Mingione, Gradient estimates below the duality exponent, Math. Ann. 346 (2010), 571–627. 10.1007/s00208-009-0411-zSearch in Google Scholar

[42] Y. Miyazaki, Higher order elliptic operators of divergence form in C1 or Lipschitz domains, J. Differential Equations 230 (2006), 174–195. 10.1016/j.jde.2006.07.024Search in Google Scholar

[43] Y. Pang, V. D. Rǎdulescu, and R. Z. Xu, Global existence and finite time blow-up for the m-Laplacian parabolic problem, Acta Math. Sin. (Engl. Ser.) 39 (2023), no. 8, 1497–1524. 10.1007/s10114-023-1619-7Search in Google Scholar

[44] E. R. Reifenberg, Solution of the Plateau problem for m-dimensional surfaces of varying topological type, Acta Math. 104 (1960), 1–92. 10.1007/BF02547186Search in Google Scholar

[45] M. V. Safonov, Harnack inequality for elliptic equations and the Hölder property of their solutions, J. Soviet Math. 21 (1983), 851–863. 10.1007/BF01094448Search in Google Scholar

[46] I. V. Skrypnik, Methods for Analysis of Nonlinear Elliptic Boundary Value Problems, Translations of Mathematical Monographs, vol. 139, American Mathematical Society, Providence, 1994. 10.1090/mmono/139Search in Google Scholar

[47] Y. Su and Z. Feng, Lions-type theorem of the p-Laplacian and applications, Adv. Nonlinear Anal. 10 (2021), 1178–1200. 10.1515/anona-2020-0167Search in Google Scholar

[48] G. Talenti, Elliptic equations and rearrangements, Ann. Sc. Norm. Super. Pisa Cl. Sci. IV (1976), no. 3, 697–718. Search in Google Scholar

[49] L. Tavares, Solutions for a class of problems driven by an anisotropic (p,q)-Laplacian type operator, Comm. Anal. Mech. 15 (2023), no. 3, 533–550. 10.3934/cam.2023026Search in Google Scholar

[50] H. Tian and S. Z. Zheng, Lorentz estimate with a variable power for parabolic obstacle problems with non-standard growths, J. Differential Equations 266 (2019), 352–405. 10.1016/j.jde.2018.07.049Search in Google Scholar

[51] L. H. Wang and F. P. Yao, Global regularity for higher-order divergence elliptic and parabolic equations, J. Funct. Anal. 266 (2014), 792–813. 10.1016/j.jfa.2013.10.018Search in Google Scholar

[52] S. Z. Zheng, X. L. Zheng, and Z. S. Feng, Regularity for a class of degenerate elliptic equations with discontinuous coefficients under natural growth, J. Math. Anal. Appl. 346 (2008), 359–373. 10.1016/j.jmaa.2008.05.059Search in Google Scholar

Received: 2022-10-01
Revised: 2023-11-01
Accepted: 2024-01-03
Published Online: 2024-02-23

© 2024 the author(s), published by De Gruyter

This work is licensed under the Creative Commons Attribution 4.0 International License.

Articles in the same Issue

  1. Regular Articles
  2. Existence of normalized peak solutions for a coupled nonlinear Schrödinger system
  3. Orbital stability of the trains of peaked solitary waves for the modified Camassa-Holm-Novikov equation
  4. Global boundedness in a two-dimensional chemotaxis system with nonlinear diffusion and singular sensitivity
  5. Existence and uniqueness of solution for a singular elliptic differential equation
  6. The bounded variation capacity and Sobolev-type inequalities on Dirichlet spaces
  7. Vanishing and blow-up solutions to a class of nonlinear complex differential equations near the singular point
  8. Existence, uniqueness, localization and minimization property of positive solutions for non-local problems involving discontinuous Kirchhoff functions
  9. Concentration phenomena for a fractional relativistic Schrödinger equation with critical growth
  10. Beyond the classical strong maximum principle: Sign-changing forcing term and flat solutions
  11. Monotonicity of solutions for parabolic equations involving nonlocal Monge-Ampère operator
  12. On a nonlinear Robin problem with an absorption term on the boundary and L1 data
  13. Multiple solutions for the quasilinear Choquard equation with Berestycki-Lions-type nonlinearities
  14. Gradient estimates for a class of higher-order elliptic equations of p-growth over a nonsmooth domain
  15. Global existence and decay estimates of the classical solution to the compressible Navier-Stokes-Smoluchowski equations in ℝ3
  16. Global existence and finite-time blowup for a mixed pseudo-parabolic r(x)-Laplacian equation
  17. Multiple positive solutions for a class of concave-convex Schrödinger-Poisson-Slater equations with critical exponent
  18. A minimization problem with free boundary for p-Laplacian weakly coupled system
  19. Multiplicity of semiclassical solutions for a class of nonlinear Hamiltonian elliptic system
  20. k-convex solutions for multiparameter Dirichlet systems with k-Hessian operator and Lane-Emden type nonlinearities
  21. Infinitely many solutions for Hamiltonian system with critical growth
  22. On optimal control in a nonlinear interface problem described by hemivariational inequalities
  23. Variational–hemivariational system for contaminant convection–reaction–diffusion model of recovered fracturing fluid
  24. Potential and monotone homeomorphisms in Banach spaces
  25. Critical fractional Schrödinger-Poisson systems with lower perturbations: the existence and concentration behavior of ground state solutions
  26. Supercritical Hénon-type equation with a forcing term
  27. Concentration of blow-up solutions for the Gross-Pitaveskii equation
  28. Mountain-pass-type solutions for Schrödinger equations in R2 with unbounded or vanishing potentials and critical exponential growth nonlinearities
  29. Regularity for critical fractional Choquard equation with singular potential and its applications
  30. Double phase anisotropic variational problems involving critical growth
  31. Normalized solutions of NLS equations with mixed nonlocal nonlinearities
  32. Nontrivial solutions for resonance quasilinear elliptic systems
  33. Standing waves for Choquard equation with noncritical rotation
  34. Low regularity conservation laws for Fokas-Lenells equation and Camassa-Holm equation
  35. Modified quasilinear equations with strongly singular and critical exponential nonlinearity
  36. Limit profiles and the existence of bound-states in exterior domains for fractional Choquard equations with critical exponent
  37. Nests of limit cycles in quadratic systems
  38. Regularity of minimizers for double phase functionals of borderline case with variable exponents
  39. Boundedness, existence and uniqueness results for coupled gradient dependent elliptic systems with nonlinear boundary condition
  40. Ground state solutions for magnetic Schrödinger equations with polynomial growth
  41. Nonoccurrence of Lavrentiev gap for a class of functionals with nonstandard growth
  42. Dynamics for wave equations connected in parallel with nonlinear localized damping
  43. Boussinesq's equation for water waves: Asymptotics in Sector I
  44. Optimal global second-order regularity and improved integrability for parabolic equations with variable growth
  45. Normalized solutions of Schrödinger equations involving Moser-Trudinger critical growth
  46. Normalized solutions for the double-phase problem with nonlocal reaction
  47. Normalized solutions for Sobolev critical fractional Schrödinger equation
  48. Well-posedness of Cauchy problem of fractional drift diffusion system in non-critical spaces with power-law nonlinearity
  49. Improved results on planar Klein-Gordon-Maxwell system with critical exponential growth
  50. Existence and multiplicity of solutions for a new p(x)-Kirchhoff equation
  51. Quasiconvex bulk and surface energies: C1,α regularity
  52. Time decay estimates of solutions to a two-phase flow model in the whole space
  53. Bounds for the sum of the first k-eigenvalues of Dirichlet problem with logarithmic order of Klein-Gordon operators
  54. The properties of a new fractional g-Laplacian Monge-Ampère operator and its applications
  55. A fractional profile decomposition and its application to Kirchhoff-type fractional problems with prescribed mass
  56. Cauchy problem for a non-Newtonian filtration equation with slowly decaying volumetric moisture content
  57. Multiplicity of normalized semi-classical states for a class of nonlinear Choquard equations
  58. The second Yamabe invariant with boundary
  59. Peaked solitary waves and shock waves of the Degasperis-Procesi-Kadomtsev-Petviashvili equation
  60. Normalized solutions for the Choquard equations with critical nonlinearities
  61. Boundedness and long-time behavior in a parabolic-elliptic system arising from biological transport networks
  62. Initial boundary value problem and exponential stability for the planar magnetohydrodynamics equations with temperature-dependent viscosity
  63. Nonexistence of mean curvature flow solitons with polynomial volume growth immersed in certain semi-Riemannian warped products
  64. Analysis of a vector-borne disease model with vector-bias mechanism in advective heterogeneous environment
  65. Normalized solutions for the Kirchhoff equation with combined nonlinearities in ℝ4
  66. Nonexistence of the compressible Euler equations with space-dependent damping in high dimensions
  67. Multiplicity of solutions for a nonhomogeneous quasilinear elliptic equation with concave-convex nonlinearities
  68. On semilinear inequalities involving the Dunkl Laplacian and an inverse-square potential outside a ball
  69. Besov regularity for the elliptic p-harmonic equations in the non-quadratic case
  70. Uniqueness and nondegeneracy of ground states for Δ u + u = ( I α u 2 ) u in R 3 when α is close to 2
  71. Existence of solutions for a class of quasilinear Schrödinger equations with Choquard-type nonlinearity
  72. Weighted Hardy-Adams inequality on unit ball of any even dimension
  73. Existence and regularity for a p-Laplacian problem in ℝN with singular, convective, and critical reaction
  74. Temporal periodic solutions of non-isentropic compressible Euler equations with geometric effects
  75. Nodal solutions for a zero-mass Chern-Simons-Schrödinger equation
  76. The modified quasi-boundary-value method for an ill-posed generalized elliptic problem
  77. Nonlocal heat equations with generalized fractional Laplacian
  78. Choquard equations with recurrent potentials
  79. Local and global well-posedness of the Maxwell-Bloch system of equations with inhomogeneous broadening
  80. The Riemann problem for two-layer shallow water equations with bottom topography
  81. Global stability and asymptotic profiles of a partially degenerate reaction diffusion Cholera model with asymptomatic individuals
  82. On Kirchhoff-Schrödinger-Poisson-type systems with singular and critical nonlinearity
  83. Energy-variational solutions for viscoelastic fluid models
  84. Stability on 3D Boussinesq system with mixed partial dissipation
  85. Existence and multiplicity results for non-autonomous second-order Hamiltonian systems
  86. Erratum
  87. Erratum to “Normalized solutions for the Kirchhoff equation with combined nonlinearities in ℝ4
Downloaded on 7.9.2025 from https://www.degruyterbrill.com/document/doi/10.1515/anona-2023-0132/html
Scroll to top button